Preprint papers of

The 14th International Saga Conference

Uppsala, 9th15th August 2009

Volume 1
 Papers from the Department of Humanities and Social Sciences 14

austrvega
Saga and East Scandinavia

Preprint papers of
The 14th International Saga Conference
Uppsala, 9th15th August 2009

Volume 1

Edited by

Agneta Ney, Henrik Williams and Fredrik Charpentier Ljungqvist

in cooperation with

Marco Bianchi, Maja Bckvall, Lennart Elmevik, Anne-Sofie Grslund, Heimir Plsson,
Lasse Mrtensson, Olof Sundqvist, Daniel Svborg and Per Vikstrand

http://www.saga.nordiska.uu.se

Gvle: Gvle University Press, 2009

These preprint volumes, and the conference itself, has been made possible by very generous
grants from the following sponsors:

Swedish Research Council

The Royal Swedish Academy of Letters, History and Antiquities
Riksbankens Jubileumsfond
Samfundet Sverige-Island
Sven och Dagmar Salns stiftelse
Kungl. Gustav Adolfs Akademien fr svensk folkkultur
Ortnamnssllskapet i Uppsala
Uppsala kommun
Helge Ax:son Johnsons stiftelse
Kungl. Humanistiska Vetenskaps-Samfundet i Uppsala

The conference is hosted by and arranged through:

Department of Scandinavian Languages at Uppsala University

University of Gvle
Institute for Language and Folklore
Islndska sllskapet

The logotype of The 14th International Saga Conference on the cover and titlepage is derived from a
detail in the pictoral scene of Sigurr Ffnisbani on the runic rock at Ramsund in the province of
Sdermanland, Sweden. Drawing: Arkeobild.

ISSN: 1653-7130.
ISBN: 978-91-978329-0-8.

2009, Contributors
http://urn.kb.se/resolve?urn=urn:nbn:se:hig:diva-4837

Layout: Fredrik Charpentier Ljungqvist.

Printed in Sweden by Universitetstryckeriet, Uppsala, 2009.
 Table of contents

Plenary lectures
Elena Gurevich....................................................................................................................1
Lars Lnnroth......................................................................................................................2
John McKinnell ...................................................................................................................3
Svanhildur skarsdttir.......................................................................................................5

Papers
Sirpa Aalto and Ville Laakso ..............................................................................................6
Aalheiur Gumundsdttir................................................................................................13
Joonas Ahola .......................................................................................................................21
Theodore M. Andersson......................................................................................................29
Anders Andrn ....................................................................................................................34
rmann Jakobsson ..............................................................................................................35
David Ashurst......................................................................................................................43
Hugh Atkinson ....................................................................................................................44
Auur Ingvarsdttir .............................................................................................................45
Robert Avis .........................................................................................................................52
Maja Bckvall .....................................................................................................................60
Patricia A. Baer ...................................................................................................................62
Sverre Bagge .......................................................................................................................71
Massimiliano Bampi ...........................................................................................................78
Bjrn Bandlien ....................................................................................................................85
Geraldine Barnes .................................................................................................................92
Eleanor Rosamund Barraclough..........................................................................................99
Karen Bek-Pedersen............................................................................................................106
Chiara Benati.......................................................................................................................110
Lisa Bennett ........................................................................................................................119
Maths Bertell .......................................................................................................................127
Kjersti Bruvoll.....................................................................................................................136
Katrina Burge ......................................................................................................................144
Hannah Burrows..................................................................................................................151
Jrg Bschgens....................................................................................................................160
Jesse L. Byock.....................................................................................................................167
Betsie A.M. Cleworth..........................................................................................................176
Margaret Clunies Ross ........................................................................................................185
Jamie Cochrane ...................................................................................................................193
Tommy Danielsson .............................................................................................................201
Matthew J. Driscoll .............................................................................................................207
Amy Eichhorn-Mulligan .....................................................................................................213
Leif Einarson .......................................................................................................................221
Eln Bra Magnsdttir .......................................................................................................229
Alexey Eremenko................................................................................................................237
Stefka G. Eriksen.................................................................................................................242
Fulvio Ferrari.......................................................................................................................250
Svante Fischer .....................................................................................................................257
Rune Flaten .........................................................................................................................259
Charlotte Frantzdatter..........................................................................................................268
Lisa Fraser ...........................................................................................................................269
Frog .....................................................................................................................................270
Natalija Ganina....................................................................................................................279
Leszek Gardea....................................................................................................................285
Gsli Sigursson ..................................................................................................................295
Jrg Glauser.........................................................................................................................296
Galina Glazyrina .................................................................................................................303
Erin Goeres..........................................................................................................................309
Bo Grslund.........................................................................................................................318
Jonathan Grove....................................................................................................................327
Gurn Nordal.....................................................................................................................336
Guvarur Mr Gunnlaugsson ............................................................................................343
Terry Gunnell ......................................................................................................................346
Jan Ragnar Hagland ............................................................................................................354
Agathe M. Hahn ..................................................................................................................359
Richard L. Harris.................................................................................................................360
Eldar Heide..........................................................................................................................361
Heimir Plsson ....................................................................................................................369
Helgi Skli Kjartansson ......................................................................................................373
Pernille Hermann.................................................................................................................379
Kate Heslop .........................................................................................................................380
Olof Holm ...........................................................................................................................390
Silvia Hufnagel....................................................................................................................398
Anders Hultgrd ..................................................................................................................405
Lise Hvarregaard .................................................................................................................411
Michael Irlenbusch-Reynard ...............................................................................................420
Tsukusu It..........................................................................................................................429
Tatjana N. Jackson ..............................................................................................................438
Judith Jesch .........................................................................................................................443
Jhanna Katrn Fririksdttir ..............................................................................................452
Karl G. Johansson ...............................................................................................................460
Vera Johanterwage .............................................................................................................. 469
Regina Jucknies...................................................................................................................477
Anna Kaiper ........................................................................................................................482
Merrill Kaplan .....................................................................................................................483
Kri Gslason.......................................................................................................................484
John Kennedy......................................................................................................................491
Laila Kitzler hfeldt ...........................................................................................................498
Karoline Kjesrud .................................................................................................................506
Elise Kleivane .....................................................................................................................513
Lydia Klos ...........................................................................................................................521
James E. Knirk ....................................................................................................................528
Annette Kruhffer ...............................................................................................................536
Hans Kuhn...........................................................................................................................545
Ann-Drte Kyas ..................................................................................................................550
Hendrik Lambertus.............................................................................................................. 551
Paul S. Langeslag ................................................................................................................560
Carolyne Larrington ............................................................................................................ 568
Annette Lassen ....................................................................................................................576
Helen F. Leslie ....................................................................................................................584
Emily Lethbridge.................................................................................................................585
Cecilia Ljung .......................................................................................................................595
Nanna Lkka .......................................................................................................................603
Maria Cristina Lombardi.....................................................................................................611
Lorenzo Lozzi Gallo............................................................................................................620
Emily Lyle...........................................................................................................................628
Magns Hauksson ...............................................................................................................635
Mikael Males.......................................................................................................................643
Tommaso Marani ................................................................................................................649
Lasse Mrtensson ................................................................................................................657
Suzanne Marti .....................................................................................................................659
Inna Matyushina..................................................................................................................667
Bernadine McCreesh ...........................................................................................................675
Rory McTurk.......................................................................................................................682
Bernard Mees ......................................................................................................................683
John Megaard ......................................................................................................................691
Elena Melnikova .................................................................................................................692
Savva Mikheev....................................................................................................................694
Stephen Mitchell .................................................................................................................695
Jakub Morawiec ..................................................................................................................696
Else Mundal.........................................................................................................................704
Gudlaug Nedrelid ................................................................................................................712
Andreas Nordberg ...............................................................................................................721
Hans Jacob Orning .............................................................................................................. 729
Rune Palm ...........................................................................................................................738
Alexandra Petrulevich .........................................................................................................742
Jonatan Pettersson ...............................................................................................................751
Carl Phelpstead....................................................................................................................761
Alessio Piccinini..................................................................................................................768
Debbie Potts ........................................................................................................................776
Judy Quinn ..........................................................................................................................783
Ragnheiur Msesdttir ......................................................................................................788
Slavica Rankovi.................................................................................................................795
Margareta Regebro..............................................................................................................802
Lena Rohrbach ....................................................................................................................897
Philip Roughton...................................................................................................................816
Carrie Roy ...........................................................................................................................823
Giovanna Salvucci...............................................................................................................832
Daniel Svborg....................................................................................................................837
Werner Schfke ...................................................................................................................845
Jens Peter Schjdt................................................................................................................851
Brittany Schorn ...................................................................................................................852
Michael Schulte...................................................................................................................860
John Shafer..........................................................................................................................867
Tatiana Shenyavskaya .........................................................................................................875
Leszek P. Supecki ..............................................................................................................876
Terje Spurkland ...................................................................................................................884
Rolf Stavnem.......................................................................................................................891
Nichole Sterling...................................................................................................................897
Marjolein Stern....................................................................................................................898
Mathias Strandberg .............................................................................................................906
Ilya V. Sverdlov ..................................................................................................................913
Sverrir Jakobsson ................................................................................................................918
rur Ingi Gujnsson ......................................................................................................925
orleifur Hauksson..............................................................................................................926
Bernt yvind Thorvaldsen ..................................................................................................932
Maria-Claudia Tomany .......................................................................................................940
Torfi H. Tulinius .................................................................................................................948
Matthew Townend...............................................................................................................956
Zanette Tsigaridas Glrstad ................................................................................................957
lfar Bragason ....................................................................................................................958
Jens Ulff-Mller ..................................................................................................................966
Fjodor Uspenskij .................................................................................................................975
Tereza Vachunov...............................................................................................................983
Helena Victor ......................................................................................................................991
Francesco Vitti ....................................................................................................................998
Vilmos Voigt .......................................................................................................................1006
Sabine H. Walther ...............................................................................................................1008
Elisabeth Ida Ward ..............................................................................................................1009
Jonas Wellendorf.................................................................................................................1016
Diana Whaley......................................................................................................................1024
Tarrin Wills .........................................................................................................................1032
Kendra Willson ...................................................................................................................1039
Kirsten Wolf........................................................................................................................1047
Lars Wollin..........................................................................................................................1054
Bryan Weston Wyly ............................................................................................................1055
Yelena Sesselja Helgadttir-Yershova................................................................................1064
Torun Zachrisson.................................................................................................................1072
Kristel Zilmer ......................................................................................................................1074
 Plenary lecture
From Accusation to Narration: The Transformation of senna
in slendinga ttir
Elena Gurevich, Institute of World Literature, Russian Academy of Sciences
The paper deals with the transformation of the genre of verbal duelling (senna) in the stories
of Icelanders abroad, the so-called tanferar ttir. Senna is a stylised verbal duelling,
which from the first phrase to the last develops according to a traditional pattern elaborated in
every detail. One of the most vivid examples of the senna-pattern in Old Icelandic literature is
provided by lkofra ttr. In Sneglu-Halla ttr (ch.6), where the quarrel of two rivals, the
skalds of Haraldr Harri, is described, the senna acquires some novel features. First of all,
the important role in the conflict is given to the representative of the audience, whereas the
audience in the senna is always silent and only watching the development of the debate. In
this case the spokesperson happens to be the konungr himself. The whole argument of the
skalds is guided by the kings questions and remarks and their assaults take place only with
his sanction. However, the most significant innovation in Sneglu-Halla ttr is that, instead of
the not always completely articulated remarks about the opponents past (or hints about
some shameful past events) which are typically found in senna, a detailed linear retrospective
story is introduced, similar to other first-person narratives which can occur in tanferar t-
tir. There are good grounds for believing that one of the reasons for this transformation con-
sists in the impossibility of appealing to the collective memory. This arises as a consequence
of the inclusion of a verbal duel between heroes who are Icelanders into a narrative whose
action is set in Norway; hence the other characters of the story, who are witnesses of the un-
folding scene, do not possess any preliminary information about the relevant past events. A no
less important influence on the transformation of the senna in Sneglu-Halla ttr is exerted by
another genre, the anecdote.

1
 Plenary lecture
Old Norse Texts as Performance
Lars Lnnroth, Department of Literature, History of Ideas and Religion,
University of Gothenburg, Sweden
In recent years some scholars, inspired by speech act theory or performance theory, have ana-
lysed Old Norse texts as performative acts in which the words are used not only to convey
meaning but to achieve specific results with people in a social context. One may, for example,
analyse a court scene in a saga as an exemplary piece of judicial action, an exchange of insults
as a method of provoking violence, a skaldic drpa as an act of rhetorical celebration, or a
galdr as a magic ritual aimed at destroying ones enemy.
Such studies may become even more interesting, however, if they are combined with some
consideration of how the Old Norse texts were originally performed in front of an audience.
Were they, for example, recited, chanted, or sung to the accompaniment of a musical instru-
ment? Were they dramatically enacted by one or more actors? Who were the performers, and
how did these performers relate to their audience? In my book Den dubbla scenen: Muntlig
diktning frn Eddan till ABBA (1978), I tried to answer such questions, and other scholars
have since that time provided interesting new answers. A new and partly updated edition of
Den dubbla scenen came out in 2008, but there are still many unsolved problems that deserve
to be further discussed with regard to the oral performance of Eddic poems, skaldic poems,
and sagas during the early Middle Ages.
My lecture will discuss some saga passages in which the performance of the saga charac-
ters is likely to have merged with the performance of the actual saga as it was told or read
aloud to its audience, thus creating what I refer to as a double scene, in which the narrative
and its narration become almost indistinguishable.

2
 Plenary lecture
Ynglingatal: A Minimalist Interpretation
John McKinnell, Durham University, England
Ynglingatal is usually read alongside the prose account of the early Swedish and Norwegian
kings in Ynglinga saga, in which it is embedded, and sometimes also in conjunction with
other prose sources that date from the twelfth and thirteenth centuries. Understandably, most
commentators have tended to interpret the poem though the eyes of Snorri (or whoever the
writer of Ynglinga saga was), but in this paper I will try to limit my discussion to the poetic
text itself. Admittedly, we have to depend on the prose writers for the order of stanzas, since
jlfr (as I shall call the poet) does not explicitly say that any of the kings he names is the
immediate successor of any other; there are also some textual variants which seem to have
been intended by later medieval revisers to modify the reputations of particular kings. How-
ever, I shall suggest that these can be detected, that the overall structure of the poem itself can
still be studied, and that what it does not say may be as important as what it includes. For ex-
ample, its early stanzas do not claim that any king is directly the son of a god, nor do its final
ones make any connection with the lineage of Haraldr hrfagri, or, therefore, with any of the
subsequent kings of Norway.
I will suggest that jlfr probably used a number of different oral sources:
1. A list of kings of Vestfold, including details of where each one is buried (and in
some cases, also where he died); this may derive from a claim of inheritance.
2. A single legendary figure (lfr of Vermaland), who is used to link the kings of
Vestfold to the ancient kings of the Swedes at Uppsala.
3. A list of legendary kings of the Swedes at Uppsala, each of whose names begins
with a vowel, and most of whom are also known in other legendary sources that
seem not to be dependent on Ynglingatal.
4. A rather disparate sequence of quasi-mythological stories, beginning with Fjlnir
and probably ending with Agni, into which is inserted
5. A sequence of four kings whose names begin with d (Dmaldi to Dagr).
Together, these provide a total of twenty-seven generations of kings, a number that I shall
argue is probably significant and deliberate. It is noticeable that jlfr says almost nothing
about three of his Swedish kings (Dmarr, Dyggvi and nundr). No deeds or manner of death
are attributed to them, and while two of them seem at first sight to have the sites of their fu-
nerals mentioned, even these are probably either commonplace or poeticism. It is possible that
jlfr knew only their names, but perhaps more likely that he invented them in order to
make up a predetermined number of generations.
There has been much debate about the intended function of Ynglingatal and jlfrs atti-
tude towards the kings who are his subject. Despite its (rather faint) praise of Rgnvaldr
heiumhri in the final stanza, the poem as a whole certainly does not look like praise of his
glorious ancestors; but while some kings are viewed in a hostile or ironic way, many are not,
so it seems equally difficult to take the whole poem as an attack or satire on the kings. In the
last part of my paper I will consider jlfrs probable personal contribution to the tradition;
I will try to show:
1. That he makes no attempt to valorise either death in battle (as is clearly done in
Hleygjatal, for example) or death by human sacrifice.
2. That most of his expressions of disapproval and his most biting examples of irony
are reserved for those who commit violence against their own kin; this attitude is

3
 consistent whether the king is the agent or the victim of such violence.
3. That in at least two cases, jlfr probably alters his received stories in order to
express this viewpoint more forcefully.
4. That the epithets applied to the more recent Norwegian kings seem generally more
complimentary than those given to the more remote Swedish ones; some of them
may be intended ironically, but we cannot assume this without some hint to that ef-
fect.
I shall conclude by suggesting a possible structure and function for the poem as a whole.

4
 Plenary lecture
To the letter: Philology as a core component
of Old Norse studies
Svanhildur skarsdttir, Stofnun rna Magnssonar slenskum frum, Iceland
At the beginning of a new century it is time to take stock and review the situation of textual
editions in our field. A lot remains to be done. Many of us still rely on the fruits of C.R.
Ungers heroic labours (to name but one of that productive generation), and although the ar-
namagnan institutes have ventured to carry on the ambitious projects begun in the heyday of
the cultural-political strife over the Icelandic manuscripts, they have not managed to keep up
the impressive output of the 1960s. Meanwhile, almost everything around us has changed.
There are drastically different views on what constitutes a text, manuscripts are scrutinized in
new ways, the printed book is no longer the only not even the preferred medium for edi-
tions, and last, but not least, the concept of collaboration has a whole new significance in the
age of the internet. In the paper I will explore these issues and the consequences they ought to
have for Old Norse studies, for the scholar and for the student; or rather: for the community of
scholars and students.

5
 Karelia, Finland and Austrvegr
Sirpa Aalto and Ville Laakso, University of Joensuu and University of Turku
The lack of written sources from within Finland and Karelia (see for example Uino 1997: 13
16)1 during the Viking Age and the early Middle Ages also called the Crusade period by
Finnish archaeologists and historians has led Finnish scholars to rely on archaeological re-
cords to perceive the regions settlement and culture during this time. However, in spite of the
continual accumulation of material from excavations, it is still impossible to get a full picture
of settlement and culture in the Viking Age. The sagas may offer some help depicting these
places since they have a few mentions of a Finnland and of a Kirjlaland, which is generally
acknowledged to denote Karelia. This in itself is remarkable because they are one of the first
written documents mentioning Karelia (Karelians in Erikskrnikan, Lind 1981: 174177)2.
The sagas, however, do not give detailed descriptions of these lands and their peoples, but as
such they are interesting. The purpose of this article is to present the possible role Finnland
and Kirjlaland had on Austrvegr and to impart what new information the latest excavations
can provide on this matter. The question then is: How were Finnland and Kirjlaland con-
nected to Austrvegr?
In the Finnish archaeological record, Viking Age contacts with Scandinavia are repre-
sented by artifacts of Scandinavian origin. These artifacts have been found in several parts of
the country, but the areas with the greatest contact are recognised to be Southwestern Finland
and Karelia, in the East. In the historical province of Finland Proper, which refers to the most
southwestern area of Finland, the people had their own distinctive culture, which is demon-
strated by the indigenous jewelry. In Western Finland, Scandinavian artefacts are undoubtedly
a result of direct contacts with Sweden: weapons found in graves confer contacts to Gotland
and to Middle Sweden. Artifacts deemed to be of foreign origin have been found to be con-
centrated in the coastal area, but they also have been found spread to some parts of the inland.
Settlement in Western Finland by the Viking Age was already long established, with no traces
of a Scandinavian population visible (Lehtosalo-Hilander 1984: 296298). Trade contacts
must thus have existed, but other contacts we cannot be sure of confirmed other contacts. It
seems that the Finns in Finland Proper did not adopt the hierarchical society system of the
Svear, and the lack of typical graves of the type found in Svealand suggest that the Svear did
not have a foothold in Finland (Lehtosalo-Hilander 2001: 101. Still, it seems that some of the
rune stones imply that the Svear made plundering expeditions to Finnland and Tafeistaland
(Fi. Hme) (Palm 2004: 55).
In Karelia, the Viking Age was a dynamic period of growth: the number of known settle-
ments was much higher than in the previous periods of the Iron Age (Uino 1997: maps on pp.
104 and 110). Archaeological finds have been concentrated on the western shore of Lake
Ladoga. Further to the west, there was also settlement in the Savo region, but large uninhab-
ited or sparsely populated territories separated these centres from each other. Important new
archaeological evidence, mainly in the form of cemeteries, has also come to light for exam-
ple in the Kymenlaakso area of southeastern Finland, which is at the westernmost edge of
Karelia (Miettinen 1998: 93129). In addition, recent palaeoecological studies from lake and
1
The area of Karelia can be defined in several ways, but here the term refers to the historical province that now
belongs partly to Russia and partly to Finland. After the Second World War the most central areas of the prov-
ince have been a part of Russia.
2
The Karelians are mentioned in the Swedish Erikskrnikan: they were supposed to have attacked Sigtuna in the
end of 1100s. This view is disputed, and for example John Lind has stated that we cannot be sure who the attack-
ers were. Erikskrnikan, p. 43. Swerige haffde mykin vadha aff karelom ok mykin onadha[]En tima fil them
en luna, at the brndo Sightuna[].

6
bog sediments have revealed pollen evidence of Viking Age agriculture in several areas where
archaeological evidence of permanent settlement is lacking ( ee e.g. Vuorela et al. 2001;
Alenius et al. 2004; Alenius & Laakso 2006; and Tomminen 2006).

Finnar and Kirjlar

It is important to point out that in the Viking Age people who lived in the areas of what are
now present day Finland and Karelia were not a homogenous people. The people in Karelia
were, according to present understanding, a mixture of an autochthonous population and im-
migrants from the Western part of Finland. In the eyes of the contemporary Scandinavians,
Karelians did not differ from Western Finns; for example the Karelian women wore orna-
ments and brooches that were fashionable in Western Finland in the 8th11th centuries (Uino
1997: 176). They also spoke a Baltic Finnic language as did the Finns in Finland. The biggest
difference would have been in their language and belief system. These are not, however, men-
tioned in the sagas. In the sagas, the name of the people, Kirjlar, seems not to be mentioned
as often as that of the place, Kirjlaland.3
Finnar in the sagas do not usually refer to the Finns, but to the Smi people (Aalto 2005;
Aalto 2003; Mundal 1996). However, in some a few cases it is possible that the word Finnar
actually refers to the Finns: in lfs saga ins helga the young lfr Haraldsson is plundering
with his men in Finnland. The people are called Finnar in the prose text, but in the poem that
is connected to this episode, they are interestingly refered to as Finnlendingar. The poem also
mentions two place names: Blagarssa and Herdalar.4 The meaning of these place names
has brought about much discussion (Galln 1984: 256; Holmberg 1976: 175176). for ex-
ample Inger Zachrisson has suggested that Herdalar actually refers to Hrjedalen in Sweden
(Zachrisson 1991: 192). It is probably impossible to pinpoint the exact geographical locations
on the basis of these place names, but maybe they should be viewed in another light; they
show that there was a need to give names to these places and not just refer to the area as
Finnland. The problem with the sagas and the scaldic poetry is that they were produced by
Norwegians and Icelanders who had little or no knowledge of the areas east of Svj. If we
had written sources from the Svear the situation might be different. In chapter 80 of the previ-
ously mentioned lfs saga ins helga the Swedish lagmar orgnr mentions how the earlier
kings of the Svear made plundering expeditions to Finnland ok Kirjlaland, Eistland ok Kr-
land ok va um Austrlnd.5 It seems that Finnland and Kirjlaland were interesting and
prosperous enough to be targets for plunder. Kirjlaland is also mentioned in Fagrskinna,
when jarl Svein made a plundering expedition in Austrrki and also in Kirjlaland.6 Egils
saga Skalla-Grmssonar mentions Kirjlaland, and gives a geographical location for it: En
austr fr Naumudal er Jamtaland, ok Helsingjaland ok Kvenland, Finnland, Kir-
jlaland; en Finnmrk liggr fyrir ofan essi ll lnd[]7 As indeterminate as the geographi-
cal descriptions tend to be in the sagas and other medieval texts, this description seems to
place Kirjlaland in Karelia.

Kirjlaland Kurkijoki?
3
For example Heimskringla and Fagskinna do not mention the Kirjlar but Egils saga and Hkonar saga Hk-
onarsonar do. The ethnic background of the Kylfingar has also raised questions, but this will be not dealt with in
this article. See e.g. Egils saga ch. 10, p. 27 and references.
4
Hkr II lhelg ch. 9, pp. 1011. The poem also calls the inhabitants of Finnland as Finnlendingar.
5
Hkr II lhelg ch. 80, p. 115.
6
Fsk ch. 29, p. 178.
7
Egils saga ch. 14, p. 36.

7
Kirjlaland is generally accepted to refer to Karelia (Fi. Karjala), meaning approximately the
area of Vyborg (Fi. Viipuri) region and the Karelian Isthmus during the Viking Age. Informa-
tion on a more exact location of Kirjlaland is somewhat contradictory, and several reason-
able possibilities have been presented (Uino 1997: 185 and cited literature). The etymology of
the word Kirjlaland has been explained to derive from the place name Kurkijoki, which is
situated on the north-western shore of Lake Ladoga.8 The Russian chronicles mention it for
the first time in AD 1396 as an adjective, Kjurjeskij or Kirjeskij pogost. This name is also
found on a birch-bark document (no. 248) from Novgorod, discovered in an excavation layer
dated to the period 13961422. The Tax Book of 1500 for the Vodskaja pjatina gives the
name Kirjakoi (Uino 1997: 185). However, this etymology cannot be confirmed with cer-
tainty. Considering the possible etymology of the word, it would be interesting if the word
really derives from a place name given to it by its inhabitants and not by some outsiders. The
information that the sagas give on Kirjlaland and its inhabitants should not be seen as facts.
Educated guesses can be made for its location and background, but the truth is that the infor-
mation that the sagas give should actually be seen to reflect the vast network of trade in Aus-
trvegr (Korpela 2004: 56).
In the archaeological materialrecord, there is actually nothing that contradicts the connec-
tion between Kirjlaland and the area of Kurkijoki. On the contrary, there are several interest-
ing Viking Age sites in and around Kurkijoki, and the density of sites and finds is greater
there than elsewhere in Karelia (Uino 1997: 114, Fig. 4:6). Especially noteworthy items come
from the cremation cemetery of Lopotti, which had been excavated already by the 1880s.
These items include several artifacts of Scandinavian origin: a pair of oval tortoise brooches, a
bracelet, and a tongue-shaped fire striker. In fact this is the largest concentration of Scandina-
vian artifacts in Karelia. According to Pirjo Uino, this type of oval brooche is particularly
common in Norway; however, as a whole it is difficult to distinguish the exact nature of the
connections with Scandinavia reflected by these artifacts (Uino 1997: 182; see also Uino
2003: 327331, 354357). Interestingly enough, the burial form at the Lopotti cemetery is not
Scandinavian, but in fact has been identified as West Finnish. Thus, even the community that
used the cemetery was probably Finnic, not Scandinavian (Uino 1997: 115). It is also note-
worthy, that the village of Lopotti was later, in the 15th century at the latest, the center of the
Kurkijoki pogost, or parish. All in all, based on the archaeological information, the area of
Kurkijoki seems to be the best candidate for the location of Kirjlaland. For example in the
Vyborg area, which is another candidate for its location, there have been practically no ar-
chaeological finds from the Viking Age (Uino 1997: 114, Fig. 4:6).
One explanation for the importance of Kurkijoki in the Viking Age clearly is its geo-
graphical position at the mouth of rivers leading to the inland areas in the west and north.
These water routes have undoubtedly played an important role for trading, especially for the
acquisition of furs. Kurkijoki may have been a trading place, at least temporarily, and would
thus have been known by the Scandinavian traders and raiders. Kurkijoki never became an
important trading station, maybe because of its relative proximity to Staraja Ladoga, which
also seems to be the most probable place of manufacture for at least some of the Scandinavian
artifacts found in Karelia.
Even if Bjarmaland and the Bjarmians do not belong to the scope of this article, they can-
not be totally neglected because they are mentioned in the sagas as one of the peoples inhabit-
ing the vast northern area of Fennoscandia. The location of Bjarmaland is disputed, as is the
ethnic background of the Bjarmians (Jackson 1992 and 2002). Mervi Koskela Vasaru has
made an extensive study of the subject lately. Her conclusion was that Bjarmaland was situ-
ated on the Kola Peninsula, and that the Bjarmians were a people speaking a Baltic Fennic

8
See already Mikkola 1942, p. 26. Since the 17th century the parish also has had a Swedish name, Kronoborg.

8
language (very close to Finnish and Karelian). Also according to her theory the Bjarmians
were assimilated into the Karelians, because they disappear from written sources in the 13th
century (Koskela Vasaru 2008). The activity of the Karelians in Northern Fennoscandia and
Finnmark supports this theory. In this area the Norwegians and the Karelians/Novgorodians
competed with each other. Both parties for example levied taxes from the Smi people. In
Hkonar saga Hkonarsonar this problem of spheres of interest intersecting is raised when
the envoy of the King of Novgorod9 arrives at the court of King Hkon Hkonsson. The pas-
sage mentions that Kirjlar, who were tributary to the King of Novgorod, and Norwegian
sysslumenn had murdered and robbed each other. The King of Novgorod wanted to put an end
to this and suggested peace. He also suggested a marriage alliance between Hkons daughter
Kristin and his own son.10

Part of a trading network

The main Austrvegr route was the River Neva, and Karelia lay outside of this central passage
(Uino 1997: 184). However, the effects of trade in Austrvegr reached even the inland of pre-
sent day Finland. For example, in Mikkeli Orijrvi a Viking Age silver hoard was found at the
end of 1990s. The hoard consisted of 136 coins or fragments of coins. One of the coins was of
Arabic origin, 22 were English, and 83 German. Terminus post quem for the hoard is AD
1014. The excavations also revealed a rather big field that had been in use for agriculture be-
tween AD 800 and 1300 (Mikkola & Tenhunen 2003: 57, 7071). For a long time, the num-
ber of Scandinavian artifacts in Karelia had been estimated to be so small, that the easiest ex-
planation for their distribution seemed to be the contacts that took place within the Lake
Ladoga area (Nordman 1924: 186; Uino 1997: 181). At present, no graves of a clearly Scan-
dinavian type have been found in Karelia, unlike, for example, the burial mounds that do exist
on the southeastern coast of Lake Ladoga (Uino 1997: 182).
In his book The Northern Crusades Eric Christiansen gives a rather depressing picture of
the natural conditions and preconditions for livelihood in Fennoscandia in the Viking Age
(Christiansen 1997: 810). It is true that the climate can be harsh and crop failures were and
still are not uncommon. However, the inhabitants learned to cope with the environment. Agri-
cultural and pastoral livelihood was supplemented with game, fish, mushrooms, and berries.
Hunting as a livelihood was important in two ways: it was an addition to daily nutrition, and it
also gave extra income when furs were sold. The areas of Karelia, as well as Finland, were
part of Austrvegr in the Viking Age as the sources of furs. However, despite Karelias advan-
tageous position (with other river routes to Gararki the Old Rus that went along the
southern shore of Lake Ladoga) archaeologists have not found evidence of any permanent
trading stations there (Uino 1997: 179). It is possible that the trading stations were temporary
and/or the trade was concentrated at Staraja Ladoga. Since no Scandinavian graves have been
found in Karelia, this suggests that Scandinavian contacts and settlement were not permanent
and that this area was not of central interest to the Viking Age traders (Uino 1997, p. 182).
But actually, in the beginning of the 11th century, when the Caucasia portion of the eastern
trade route suffered from attack by the Seldjuks and trade here began to decline, the areas of
Southeastern Finland, Ladoga, Karelia, and Viena Karelia became centres for the fur trade.
This situation lasted a few hundred years (Korpela 2004: 41 and references). Thus, even
though Karelia was not situated on the trade route of Austrvegr, it was in its sphere of influ-
ence.

9
Aleksandr Nevskyi of Novgorod.
10
Konunga sgur, p. 419420.

9
 The Finns and Karelians were not just passive peddlers of furs; they also made trading and
plundering voyages into Northern Fennoscandia as previously described. They also competed
with the Norwegians in Finnmark where they collected tributes from the Smi people. This
competition is apparent for example in Egils saga, which mentions the Kvenir, Krjlar, and
Kylfingar, who competed with each other and with the Norwegians.11 During the Crusade
Period, the presence of Karelian traders and raiders can be verified by archaelogical evidence,
which shows that the Karelians distributed at least their own type of axe to Northern Fenno-
scandia. It is possible also that all kinds of small objects were transported, such as Orthodox
cross pendants and other small metal ornaments (Uino 1997: 199).
No definite evidence has been presented of any Karelians participating in the eastern voy-
ages of the Scandinavians during the Viking Age, but this has been considered quite possible
(Uino 1997: 183; Uino 2003: 354357). The presence of Western Finns seems very plausible;
items pointing to this possibility have been found for example in the Luistari cemetery in Eura
(Lehtosalo-Hilander 2001: 97).

Conclusion
Finland and Karelia are mentioned in written sources as part of the Austrvegr, even if the ref-
erences are short and few in number. The archaeological record also confirms that there were
contacts with the Scandinavians in these areas. Unfortunately, neither the written sources nor
the excavations reveal the nature of the contacts; we may only assume that the inhabitants of
Finland and Karelia were part of a trading organization where they acted as the suppliers of
furs. Maybe this trade provided them with enough wealth that they were also considered lu-
crative targets for raiding.

11
Egils saga ch. 10, p. 27; ch. 14, pp. 35 37.

10
 Abbreviations
Fsk = Fagrskinna
Hkr = Heimskringla
F = slenzk fornrit
lhelg = lfs saga ins helga

Sources
Egils saga Skalla-Grmssonar. F 2. Sigurur Nordal gaf t. Reykjavk, 1933.
Fagrskinna. Bjarni Einarsson gaf t. F 29. Reykjavk 1985. [Fsk]
Konunga sgur. Sagaer om Sverre og hans efterflgere. [Eirspennill] udgivne af C.R. Unger.
Christiania, 1873.
Heimskringla II, F 28. Bjarni Aalbjarnarson gaf t. Reykjavk 1979. [Hkr]

Cited literature:
Aalto, Sirpa 2003. Alienness in Heimskringla: Special Emphasis on the Finnar, in: Scandinavia and
Christian Europe in the Middle Ages. Papers of the 12th International Saga Conference
Bonn/Germany, 28th July 2nd August 2003, Simek, R. & Meurer, J. (eds.), Bonn, 17.
Aalto, Sirpa 2005. Muukalaisuus Snorri Sturlusonin Heimskringlassa tapaus Finnar (Historiallinen
Aikakauskirja 1/2005), 2733.
Alenius, Teija & Grnlund, Elisabeth & Simola, Heikki & Saksa, Aleksandr 2004. Land-use
history of Riekkalansaari Island in the northern archipelago of Lake Ladoga, Karelian
Republic, Russia. (Vegetation History and Archaeobotany 13), 2331.
Alenius, Teija & Laakso, Ville 2006. Palaeoecology and Archaeology of the Village of
Uukuniemi, Eastern Finland. (Acta Borealia 2/2006), 145165.
Christiansen, Eric. 1997. The Northern Crusades. Printed in England.
Galln, Jarl 1984. Lnsieurooppalaiset ja skandinaaviset Suomen esihistoriaa koskevat lhteet. In:
Suomen vestn esihistorialliset juuret. Bidrag till knnedom av Finlands natur och
folk. (Utgivna av Finska Vetenskaps-Societeten, H. 131) Helsinki, 249262.
Holmberg, Maj-Lis 1976. Om Finland och vriga finnlnder i den islndska fornlitteraturen. (Arkiv
fr Nordisk Filologi 91), 166191.
Jackson, Tatjana 1992. Location of Bjarmaland. In: Suomen varhaishistoria, Julku, K. (ed.) (Historica
Septentrionalia 21) Jyvskyl.
Jackson, Tatjana 2002. Bjarmaland revisited. (Acta Borealia 2), 165179.
Korpela, Jukka 2004. Viipurin Linnalnin synty. (Viipurin lnin historia 2) Jyvskyl.
Koskela Vasaru, Mervi 2008. Bjarmaland. (Unpublished PhD-thesis, University of Oulu. Forthming in
2009).
Lehtosalo-Hilander, Pirkko-Liisa 1984. Suomen nuoremman rautakauden esineist kansallisuusolojen
heijastajana. In: Suomen vestn esihistorialliset juuret. Bidrag till knnedom av Finlands natur och
folk. (Utgivna av Finska Vetenskaps-Societeten, H. 131) Helsinki, 283301.
Lehtosalo-Hilander, Pirkko-Liisa 2001. Luistarin yhteisst ymprivn maailmaan. In:
Viikinkej Eurassa? Pohjoismaisia nkkulmia Suomen esihistoriaan. Lehtosalo-Hilander, P-L. &
Wahlqvist, S. (eds.), Vammala, 89 101.
Lind, John 1981. Varger, nemcer og novgoroder r 1188. Hvor var Chorek og Novotorcek?
(Historisk Tidskrift fr Finland 2/1981, rg. 66), 145 177.
Miettinen, Timo 1998. Kymenlaakson esihistoriaa. (Kymenlaakson maakuntamuseon julkaisuja no
26).
Mikkola, J. J. 1942. Lnnen ja idn rajoilta. Historiallisia kirjoitelmia. Porvoo.
Mikkola, Esa & Tenhunen, Tanja 2003. Uusimpia tutkimustuloksia Mikkelin Orijrven hopea-aarteen
lytpaikalta. (Sihti 5 (2003). Savonlinnan maakuntamuseon julkaisuja), 5573.
Mundal, Else 1996. The Perception of the Saamis and their religion in Old Norse sources. In: Shaman-
ism and Northern Ecology. Pentikinen, J. (ed.). (Religion and Society 36) Berlin, New York,
97116.
Nordman, C. A. 1924. Karelska jrnldersstudier. (Finska Fornminnesfreningens Tidskrift 34:3) Hel-
singfors.

11
Palm, Rune 2004. Vikingarnas sprk. Falun.
Tomminen, Tarja 2006. Paleoekologisia tutkimuksia: Luumen Niemenkyl ja Lappeenrannan
Kauskila. (Suomen Museo 2005), 139155.
Uino, Pirjo 1997. Ancient Karelia. Archaeological Studies. Muinais-Karjala. Arkeologisia tutkimuk-
sia. (Suomen Muinaismuistoyhdistyksen aikakauskirja 104) Helsinki.
Uino, Pirjo 2003. Viikinkiaika n. 8001100 jKr. Karjalan synty. (Viipurin lnin historia 1), 313382.
Vuorela, Irmeli & Lempiinen, Terttu & Saarnisto, Matti 2001. Land use pollen record from
the Island of Valamo, Russian Karelia. (Annales Botanici Fennici 38), 139165.
Zachrisson, Inger 1991. The South Saami Culture: In Archaeological Finds and West Nordic Written
Sources from AD 8001300. In: Social Approaches to Viking Studies, Samson, Ross (ed.).

12
 Dancing Images from Medieval Iceland
Aalheiur Gumundsdttir, Hskli slands, Reykjavk, Iceland
The Icelandic manuscript Codex Upsaliensis, De la Gardie 11 (written in the period 1300
1325),1 is interesting not only because it preserves the Uppsala-Edda but also on account of
the marginalia and drawings it contains, which date from various times. The drawings have
not yet received critical attention from art historians, but according to Olof Thorell, in his in-
troduction to the diplomatic edition of the manuscript in 1977, the oldest drawing depicts
Gangleri with Hr, Jafnhr and rii as described in Gylfaginning in Snorris Edda. Thorell
considers this picture, which is on fol. 26v, as dating from the 14th century, making it the
most remarkable one in the manuscript (1977:xviii).2 The other drawings in the margins or
blank spaces following the text appear to be of a more recent date than the manuscript itself,
though this is not necessarily the case.
Apart from the Gangleri illustration, six of the drawings in the manuscript which show
people in a variety of poses are of particular interest. In this talk I intend to concentrate on
these drawings and take a close look at them, not only to add to the codicological information
about this manuscript, which is already considerable, but also to consider the question of
whether all these drawings depict people dancing, and if so, then what evidence there is for
this view. If it proves possible to answer this question, and if the conclusion is that the draw-
ings are of people dancing, then they would rank among the oldest dance images that have
been preserved in the Nordic countries. My aim is therefore to describe what the drawings
depict and to propose likely dates for them.
To throw light on these six drawings, I shall give particular attention to four points: a) the
poses in which the figures are depicted, b) their clothing, c) the ink used in the drawings, and
d) a comparison with other illustrations in the manuscript. As regards clothing, I shall make
comparisons with illustrations in other Icelandic manuscripts,3 and even in manuscripts from
elsewhere in Europe, since it is generally accepted that there is considerable foreign influence
in Icelandic manuscript illumination.
Here follows an account of the main features of the six drawings.

1. Fol. 24v
The first drawing depicts a man and a woman, showing the upper half of their bodies. They
are facing each other. The man, who is wearing a hat with a narrow brim, is bending back-
wards at a sharp angle. The woman is wearing an upper garment with a square neckline, and
her hairstyle is similar to that of the three kings in the Gangleri picture (from the 14th cen-
tury). The same hairstyle is common in European medieval manuscripts, and is by far the
most common type found in Icelandic 14th-century manuscript illuminations, e.g. in Stjrn
and AM 344 fol., and also in Belgsdalsbk and Svalbarsbk, where people are shown wear-
ing long tunics with necklines that are reminiscent of that shown on the drawing of the

1
The manuscript is now in the Uppsala University Library.
2
The folio numbers in the references given here are different from those in the manuscript itself, the facsimile of
1962 and the diplomatic edition of 1977.
3
The following manuscripts, which I mention more than once each below, will be referred to by their appropri-
ate catalogue numbers or names as follows. From the 14th century: AM 350 fol. (Skarsbk), AM 227 fol.
(Stjrn), AM 343 fol. (Svalbarsbk), AM 344 fol., AM 347 fol. (Belgsdalsbk), GKS 1005 fol. (Flateyjarbk),
AM 135 4to (Arnarblisbk) and Holm. perg. 16 4to (Helgastaabk). 14th to 15th centuries: AM 673a III 4to
(slenska teiknibkin, the Icelandic Drawing-Book), which was compiled over a period of about 150 years,
from c. 1350 to1500 (Gubjrg Kristjnsdttir 1997:9598). 16th century: AM 345 fol. (Reykjabk).

13
woman (Halldr Hermannsson 1935:Pl. 14, 8, 1625, 5354, 5557 and 71).4 The same
hairstyle is very much in evidence in slenska teiknibkin (the Icelandic Drawing-Book).
Necklines similar to that of the womans garment can be found in illustrations from continen-
tal Europe, e.g. from the 14th and early 15th centuries (see Liepe 2003:156 and Wagner
2000:Part I, Pl. 4 and 6). There are similarities between the mans hat and hats in Flateyjar-
bk, AM 132 4to (from the 15th century; Halldr Hermannsson 1935:Pl. 46 and 73), and in
slenska teiknibkin (Bjrn Th. Bjrnsson 1954:122). Hats appear in many manuscript illus-
trations, e.g. in Reykjabk (from the 16th century), where they are generally shown with
broader brims, and the hats themselves are rather taller (see, e.g., Jnas Kristjnsson 1993:86
and 124). A Scandinavian example of a hat with a narrow brim can e.g. be found in a Norwe-
gian carving from the early 14th century (Vedeler 2006:212213).
The mans backward-bending pose strongly suggests that the pair are dancing. But what
sort of dance could they be dancing?
There exists a considerable body of historical sources on dancing in Iceland, some of them
of great age, and it is generally accepted that dancing was a common amusement in Iceland in
the 12th and 13th centuries. Actual descriptions of the dances, or of the movements involved,
are of rather later date: from about 1600 or later. Furthermore, each of the surviving descrip-
tions has certain unique features. Most of them, however, distinguish between two types of
dance: dans (ballad dances) and vikivaki. The difference appears to be as follows, taking ac-
count of the style of singing that accompanied the dance.5
1) Dans. In the dans, a precentor appears to have led the singing, either alone or with the
help of singers; in which case these did not take part in the actual dance. This appears to have
applied to round dances and/or dances where pairs of dancers chose particular places. It seems
that participation by the dancers took one of two forms. In one, the dancers did not participate
in the singing, in which case the dancing tended to be more lively, even though the dancers
did not move to new places but simply moved to and fro on the same spot. In the other type of
dance, the dancers joined in the singing. The precentor sang the main text, and the body of the
dancers, moving in a circle, joined in the refrains. It is believed that men and women also
danced separately.
2) Vikivaki. In the vikivaki, the dancers probably sang or chanted in turn, moving in a
round dance until all the dancers had finished their verses. These dances would have been
accompanied by vikivaki verses, in which a man and a woman recited verses in turn. Accord-
ing to an anonymous account dating from the 17th or 18th century there seem to have been
special mens and womens vikivaki dances, as there were mens and womens dans;6 the
sources also mention another type of vikivaki in which a man and woman hold hands and sing
to one another without moving from their basic position, though various movements were
involved, such as steps forwards and backwards and even movements of the type described in
this vikivaki verse (Jn Samsonarson 1964:cxxviii): egar vikivaka / vil eg sprundin taka, /
xlum gjra a aka / og vkja sr til baka. (When in the vikivaki / I wish to take the girls, /

4
Comparable hairstyles can be found in AM 249c fol. from c. 1300, in the 14th-century manuscripts AM 679
4to, AM 68 fol., AM 241a fol., GKS 3268 4to, AM 233a fol., Holm. perg. 5 fol., AM 127 4to, GKS 3269a 4to,
Belgsdalsbk, Skarsbk, Flateyjarbk, Helgastaabk, AM 249 e fol. and AM 545ab 4to (Halldr Hermanns-
son 1935:Pl. 10, 12 and 1415, 3139, 4347, 4952 and 5568 and Liepe 2006:73 og 76), in AM 126 4to and
Thott 1280 fol., from c. 1400, and AM 132 4to from the mid-15th century; the same hairstyle appears in two
manuscripts from about 1500 (Halldr Hermannsson 1935:Pl. 40, 7274 and Jnas Kristjnsson 1993:106 and
118119).
5
On the types of dance, see in particular Jn Samsonarson 1964:xxviiicxliii.
6
It is also possible that contra dances formed part of the entertainment in the vikivaki (Jn Samsonarson
1964:xxix).

14
move my shoulders to and fro / and bend backwards.) Presumably, therefore, the vikivaki was
a type of dance that fundamentally involved interaction between a man and a woman.
In the light of this brief definition, therefore, the pair shown in this drawing could either be
participants in a dans or a vikivaki. However, the mans posture may indicate that he is danc-
ing a vikivaki, since it conforms to the description in the fragment quoted above.

2. Fol. 25r
The second drawing is of a woman in a long-sleeved full-length dress or a close-fitting upper
garment with a pleated skirt, with a bordered hem. She is wearing a wimple, a belt and
pointed shoes. She is standing, but seems to be leaning slightly forwards, perhaps stepping
forwards on her left foot, and is pointing forwards with both hands.
Long dresses that are close-fitting at the top but wide at the bottom are frequently encoun-
tered in medieval manuscript illuminations.7 The dress in this drawing is plain and very simi-
lar to dresses in 14th-century manuscript illuminations, in some of which the women wear
wimples, as is the case here; on the other hand, the dress is very unlike womens best or
ceremonial clothing, or at least that of upper-class women, as illustrated in the drawing of
women on the way to a wedding in the 16th-century Reykjabk (Jnas Kristjnsson 1993:86).
The design is reminiscent of the dresses found in Herjlfsdalur in Greenland, and it is be-
lieved to have been common in the Nordic countries from the 12th century down to the early
15th century, and particularly in the period 12501400 (Vedeler 2007:113). Dresses of similar
types can be seen in continental European manuscript illuminations, e.g. in a 15th-century
manuscript in which the woman is also wearing a belt and a wimple (Wagner 2000:Part I. Pl.
6). Wimples, including ones of the simple type shown here, were common throughout the
Middle Ages (see, e.g., Gutarp 2000:2830).
The natural interpretation of the womans pose is that it represents her dancing, possibly a
type of dance in which the dancer does not move to a new place, but steps alternately back-
wards and forwards. Of course, the pose by itself does not indicate what type of dance could
be involved, but the fact that the woman is shown on her own might indicate either a dans or a
vikivaki in which only women took part. Her hand gestures, on the other hand, cannot be ex-
plained in terms of known descriptions of dancing in previous centuries, except insofar as
they mention hand movements and gestures. But it is more likely that they indicate vikivaki
rather than a dans, since when people danced a vikivaki, they also played vikivaki games
(vikivakaleikir); the entire entertainment generally involved a threefold entertainment consist-
ing of dans, vikivaki and vikivakaleikir. Some of these games, such as rhildarleikur and
Hindarleikur, involved the dancers forming man-and-woman pairs. In some cases the women
chose their partners; in others, it was the men who did the choosing (see, e.g. Jn Samsonar-
son 1964:cxxvii and cxcviiccix); it seems natural to assume that games of this sort would
have involved a lot of hand gestures, though no descriptions of these are to be found in writ-
ten sources.

3. Fol. 25r
The third drawing shows a man in a closely-fitting buttoned doublet or jacket with long
sleeves and a collar reaching to the neck. He is wearing a short pleated skirt with a belt, and
presumably close-fitting hose (tights) underneath it. He is wearing shoes with a narrow toe;
7
See, for example, the 14th-century manuscripts Skarsbk, Belgsdalsbk, Helgastaabk and Flateyjarbk, in
which women are almost always depicted wearing wimples with dresses, as is the case here (Halldr Hermanns-
son 1935:Pl. 36, 47, 5557, 59 and 68). The dresses in slenska teiknibkin are generally more substantial,
though it also contains examples of dresses that are close-fitting on the upper body, in addition to which wimples
are worn with them (Bjrn Th. Bjrnsson 1954:153).

15
on his head he has a pointed cap hanging down on the right-hand side. In his left hand he is
holding a short sword upright; his right hand is resting on his hip, and the position of his legs
suggests that he is in motion, dancing on his toes; in addition, his trunk is curved backwards.
As regards his clothing, the buttons, short skirt and cap are particularly interesting. Tightly-
buttoned doublets are to be found in illuminations in other Icelandic manuscripts, e.g. in Thott
1280 fol., dating from about 1400 (Halldr Hermannsson 1935:Pl. 74), and also in Reykjabk,
dating from the 16th century (Jnas Kristjnsson 1993:9 and 86). Both tightly-buttoned kirtles
and doublets were in fashion in many parts of Europe in the Middle Ages, at least from the
13th down to the 15th century (see, e.g., Gutarp 2000:18 and Wagner 2000:Part I Pl. 4, 63, 73
and 91).
Pictures of men in short skirts and hose can be found, for example, in Belgsdalsbk (14th
century) and in AM 132 4to (mid-15th century; Halldr Hermannsson 1935:Pl. 57 and 73).
Stjrn, Skarsbk, Helgastaabk, GKS 3269a 4to and Flateyjarbk contain examples of men
dressed in short knee-length tunics (kirtles) with a belt, and in hose. Sometimes it is difficult
to decide whether what is depicted is a skirt with an upper garment or a kirtle, since in some
cases the lower part of the kirtle (or the skirt) is pleated. One of the figures in the sources
cited is holding a short sword similar to the one in no. 3 (Halldr Hermannsson 1965:Pl. 1 and
2, 10, 1630, 39, 4546, 5051 and 6568).
Although the point on the dancers cap is not very conspicuous, it presumably reflects the
fashion of pointed hoods (see, e.g., Halldr Hermannsson 1935:Pl. 51), which became com-
mon from about the middle of the 14th century. To some extent, this fashion developed in
tandem with the fashion in footwear, reflecting tapering shoes known as poulaines. While the
dancers shoes in this drawing do not have a long toe, it could perhaps be said that the woman
in the previous figure is wearing such shoes. Shoes of this type are to be found in Helga-
staabk, (14th century), slenska teiknibkin (Halldr Hermannsson 1935:Pl. 36 and 39 and
Bjrn Th. Bjrnsson 1954:60, 65, 7780 and 128), and in European manuscript illuminations
from the 13th, 14th and 15th centuries (Wagner 2000:Part I. Pl. 76, 87 and 91). Most of the
dignitaries depicted in the illustrations in Reykjabk (16th century) wear shoes that are very
different from these, with square toes (Jnas Kristjnsson 1993:5657).
The most interesting thing about No. 3 is, without doubt, the sword that the man is holding;
it suggests that he is dancing some sort of sword dance. Though there is no mention in written
sources of such a dance having been danced in Iceland, it can be deduced from other manu-
script illustrations that Icelanders at least knew of some sort of games involving weapons in
the 14th century (Aalheiur Gumundsdttir 2008:73). Furthermore, the lawcode manuscript
Belgsdalsbk (AM 347 fol., c. 135070) contains a picture that is reminiscent in many ways
of the drawing in DG 11. The picture (on p. 94v), that is probably of a sligtly younger date
than manuscript text, shows a man with a stick or a sword projecting out to the right; his right
hand is in front of his chest. The man appears to be in motion, as if dancing, and his clothing
is similar to that of the sword-dancer in DG 11, i.e. a short skirt with a belt, and he seems to
be wearing a cap or hood that hangs down on the right.

4. Fol. 25r
The fourth drawing in DG 11 is of a man, shown in half-length. He has a goatee beard and is
wearing a kirtle or long upper garment, a belt and a pleated skirt. There are borders on the
neckline and cuffs of the upper garment. The man, who seems to be wearing a tight-fitting
cap, is holding in his left hand what could be a rope or a stick with a loop at the top, while his
right hand is pointing forwards. His posture seems to suggest that he too is dancing, his ges-
tures having certain similarities with the pose of the woman in No. 2.

16
 Although it is not possible to state with certainty whether this man is wearing a kirtle or a
skirt, the latter seems more likely in view of how distinct the pleating is beneath the belt; this
would seem to distinguish it from the material of the upper garment. A plain cap of similar
design can be seen in Flateyjarbk (Halldr Hermannsson 1935:Pl. 46) and slenska teikni-
bkin (Bjrn Th. Bjrnsson 1954:121122).
If it is accepted that this picture is of a man who is dancing, like the other figures on this
leave of the manuscript, then it would seem that what is depicted here is some sort of rope-
dance. According to Tobias Norlind, writing about Nordic dances in 1911, there were consid-
erable similarities between the Faroese rope-dances, which were danced to the accompani-
ment of ballads and verses, and dances with weapons. He regarded the Faroese rope-dance as
a developed variant of the bow- or sword-dance, and as being most closely related to the Finn-
ish bow-dance. If one goes further and ignores the role of the rope in the Faroese rope-dance,
then similarities with a Norwegian dance are revealed; in this, the dancers change their orien-
tation in the ring as they pass under the raised hands of two men (Norlind 1911:750751).
This description by Norlind calls to mind an Icelandic dance known as hringbrot (breaking
the ring), which in turn has been compared to the Finnish bow-dance. Hringbrot is men-
tioned in written sources as early as the 15th century; in the earliest description (Niurraan
og undirvsan, see above) it is assumed that six pairs take part in the dance, using a rope link-
ing them all (Jn Samsonarson 1964:lii and clxvii; Aalheiur Gumundsdttir 2008:6263).
In view of the relationship between rope-dances and weapon-dances, or certain variants
thereof, one could perhaps conclude that the two dances depicted in Nos. 3 and 4 go back to
the same origin.

5. Fol. 26r
The fifth drawing shows a woman dressed in a long, long-sleeved dress which is close-fitting
in the upper body, with a belt and a broad skirt. The dress is bordered along the bottom, at the
neckline and on the cuff, and is buttoned down the front. As with Nos. 13, the womans
clothing and hairstyle reflect models in manuscript illuminations from the 14th and 15th cen-
turies. Her hand gestures are similar to those of the woman in No. 2, and she appears to be
stepping forwards and bending her upper body backwards, as if dancing.

6. Fol. 56r
In the sixth and last drawing, two figures are drawn beneath the text: a man on the left and a
woman on the right. The pair face each other and both are pointing forwards, the man with his
left hand, the woman with both hands.
The man is dressed in a costume similar to the sword-dancers costume, except that he is
wearing a brimmed hat with two seam lines along the length of the crown. His shoes are simi-
lar to the poulaine shoes of which there are examples in Helgastaabk and slenska teikni-
bkin, and also in manuscript illuminations from outside Iceland in the period from the 13th
to the 15th centuries, as has been mentioned above in the discussion of No. 3. The womans
costume is similar to the one in No. 2, and her hairstyle is similar to those of the women in
Nos. 1 and 5. The man is standing with his feet wide apart, his body is curved well back and
he has his right hand on his hip; the womans pose is the same as those of the women in Nos.
2 and 5.
As before, the physical poses here suggest that the figures depicted are dancing a dance
characterised by stepping forwards or sideways and arching ones upper body backwards, and
by hand gestures. The simplest explanation is to see these drawings as depicting vikivaki
dances or games, as was the case with No. 1 and 2.

17
 Features in common
If these six drawings are examined together, it appears certain that they all illustrate dancing
or games with a dance element, and in particular vikivaki and the sword-dance, and possibly
also a type of rope-dance. The body postures are of various types, and it seems that the artist
took pains to depict a range of different types of dances and games; this makes his or her
drawings an independent source about dancing in the past.
It is difficult to say how old vikivaki dances are, but there is nothing to rule out the possi-
bility that they, or dances of a comparable type, were danced in Iceland as early as in the 12th
century. The oldest written references to dance-related games on the other hand, are probably
those found in literary works from the 14th and 15th centuries (Aalheiur Gumundsdttir
2008:6263).
The Nordic sword-dance is basically an ancient dance genre, regarded by some as forming
an unbroken tradition reaching all the way back to Viking times or even earlier. On the other
hand, sword-dance seems to have been revived, in a new form, in Germany in the 15th cen-
tury, and to have become reasonably widely disseminated in this form during the 16th cen-
tury. Rope-dances or stick-dances are probably to be seen as reflexes of the sword-dance, or
of other European dances of the 15th or 16th centuries. Therefore, no accurate dating of the
drawings in DG 11 is possible on the basis of dance history, but the period from the 15th and
16th centuries is a framework that could cover all the drawings. This conclusion is not far out
of keeping with the framework suggested by the clothing; comparison with other manuscripts
indicates that the fashions involved were in vogue from the 14th century to the early 15th cen-
tury, with some leeway at each end.
A comparison of the dance images in DG 11 with other manuscripts is somewhat restricted
by the fact that the majority of Icelandic illuminated manuscripts date from the 14th century,
with a relative dearth of material for comparison from the 15th and 16th centuries. Anna Zan-
chi, author of a Ph.D. thesis on clothing in the Icelandic sagas and ttir, considers that while
medieval manuscripts are highly useful for gaining an insight into clothing and fashion at the
time they were written, it must be remembered that in some cases illuminations were drawn
from foreign originals (2006:13). This is very much the case, for example, with the illumina-
tions in lawcode manuscripts, as has been demonstrated in the case of Stjrn, in which the
illuminations were probably based on English models. For this reason I have chosen to make
comparisons with illuminations in manuscripts from other European countries as a source
regarding the general outlines in the development of clothing fashions. As an example of
comparable fashions, we can take the Flemish brass engraving of about 14151420 in the
church in Nousis, Finland, showing men in short pleated skirts, tightly-buttoned doublets and
poulaines, women in long-sleeved full-length dresses that are broad at the bottom but close-
fitting around the upper body, even with buttons and bordered necklines, and wearing wim-
ples (Liepe 2003:118 et sqq.).
I mentioned at the beginning that in addition to the features examined and compared
above, it was desirable to examine the ink used in the drawings and also to compare them
with the other illuminations in the manuscript. As it proved impossible to examine the origi-
nal manuscript, my conclusions regarding the ink are based on an examination of digital im-
ages, which imposes a limit to their validity. However, it seems that the ink of the drawings is
not very different from that used in the text, though if anything, it seems to be slightly lighter.
This need not indicate a difference in the composition of the ink, however, since the shorter
strokes in script would probably use more ink than the longer strokes of the drawings.8 But a
8
There is also a very slight variation in hue between the drawings themselves, but here it may be necessary to
allow for the possibility that this may lie in the digital images themselves.

18
comparison of the dance images with the other illuminations and marginalia in this manu-
script enables us to rule out at least some of the marginalia, written in the 15th or 16th cen-
tury. The ink in some of the drawings, on the other hand, is more like that in the dance im-
ages, but it is evident from the artistic style of the Gangleri picture that a different artist was
at work, and Thorells claim that the Gangleri picture is the oldest one in the manuscript is
probably well-founded.
In two of the drawings, the colour of the ink and the artistic style are comparable with
those of the dance images; these are the bishop on fol. 1v and the knight on fol. 37v. The
knight is wearing a costume that corresponds to what we find in illuminations from continen-
tal Europe in the late 14th and 15th centuries (Wagner 2000:Part II, III, VI and IX). The draw-
ing of the bishop is characterised by light pen-strokes similar to those in No. 6; this, and his
features, suggest that the same artist was at work here as in the dance images (cf., for exam-
ple, No. 5).
The bishop is dressed in a cope, and is also wearing a mitre and holding a crozier in his left
hand; his apparel suggests that the drawing pre-dates the Reformation, since it is believed that
in some countries, including Iceland, bishops abandoned the use of mitres and croziers at the
time of the Reformation, i.e. about 1550. The composition of the picture and the bishops
clothing are reminiscent of older pictures of bishops, such as the one believed to be of Bishop
orlkur helgi in Arnarblisbk, dating from the 14th century (Halldr Hermannsson
1935:Pl. 58 and Gunnar F. Gumundsson 2000:26). The mitre is short, as is the case in older
pictures, and consequently unlike the tall mitres that came into fashion later, at least from
about 1500 (see Gubjrg Kristjnsdttir 2000:167) and possibly earlier, at least outside Ice-
land (see the illumination in a French manuscript from the late 14th century in Gunnar F.
Gumundsson 2000:180). According to Thorell, this picture can scarcely be older than from
the 15th century (Thorell 1977:xvii).
The question arises whether the bishop on fol. 1v was drawn there for a particular purpose,
or whether it is simply an exercise in draughtsmanship that happens to be there. Could it be
that the bishop was intended as blessing the heathen content of the manuscript, or is he per-
haps here to admonish the dancers who seem to be enjoying themselves in the margins of DG
11? At first sight this may seem improbable, but if we bear in mind the history of dancing in
Iceland, there is no avoiding the fact that the bishops played a certain role in it. It seems that
clerical opposition to dance gatherings arose as early as the 12th century, when Jn gmunds-
son, who became Bishop of Hlar in 1106 (d. 1121), fought against, and had a prohibition
imposed on, the game (leikur probably dance) that was commonly played in which a man
and a woman sang improper verses to each other (Jn Samsonarson 1964:ixx). For a long
time thereafter, the church was opposed to dancing, and more bishops followed Jn g-
mundssons example. At least five of the pre-Reformation bishops showed their disapproval
of dances, speaking out against them or banning them, and the Lutheran bishops and clergy
continued this after the Reformation. Thus, it is by no means far-fetched to see the drawing of
a bishop in DG 11 as being related to the dance images in the manuscript; this could be fur-
ther evidence of how closely they reflect Icelandic reality.
At the outset I stated my intention to examine the drawings in DG 11 not least because
they might prove to be among the oldest images of dancing in the Nordic countries. The old-
est Nordic dance image that I know of is a Danish church painting in rslev, near Sklskr in
Sjlland, which is thought to date from about 1325.9 Most of the evidence suggests that the
9
This image, showing dignitaries dancing in a group and holding hands, is very different from the drawings in
DG 11 and is consequently of little help in dating them. Others propose dates of c. 1350 or c. 1380 for the paint-
ing. A church painting in the church at Vigersted in Roskilde (15th century) shows some sort of knife-
game/dance; other church paintings depicting dancing are from the 16th century or later (Saxtorph 2004:49, 146
and 159).

19
dance images of DG 11 are not quite so old, though the time-frame, in the broadest sense,
runs from the 14th to the 16th century. As far as I have been able to establish in this examina-
tion, the most likely date for them is the 15th century, which is still enough to place them
among the earliest dance images from the Nordic countries.

Bibliography
Aalheiur Gumundsdttir, 2008: Fra balladedans til hringbrot og svrddans. In Balladdans i Nor-
den. Symposium i Stockholm 89 November 2007. Ed. by N. S. Hermansson. Meddelanden frn
Svenskt visarkiv 48. Pp. 6178.
Bjrn Th. Bjrnsson, 1954: slenzka teiknibkin rnasafni. Reykjavk.
Gubjrg Kristjnsdttir, 2000: Bnaur dmkirknanna. In: G. F. Gumundsson. slenskt samflag og
Rmakirkja. Kristni slandi II. Reykjavk. Pp. 164171.
Gutarp, Else Marie, 2000: Hurusom man sig kldde en bok om medeltida drkt. Visby.
Gunnar F. Gumundsson, 2000. slenskt samflag og Rmakirkja. Kristni slandi II. Reykjavk.
Halldr Hermannsson, 1935: Icelandic Illuminated Manuscripts of the Middle Ages. Corpus Codicum
Islandicorum Medi Aevi VII. Copenhagen.
Jn Samsonarson, 1964: Kvi og dansleikir I. slensk jfri. Reykjavk.
Jnas Kristjnsson, 1993: Handritaspegill. Reykjavk.
Liepe, Lena, 2003: Den medeltida kroppen. Kroppens och knets ikonografi i nordisk medeltid. Lund.
Liepe, Lena, 2006: Marginalteckningarna i AM 595 ab 4to. Rmverjasaga. Bilder i marginalen. Ed.
K. Markus. Tallin. Pp. 7183.
Saxtorph, Niels M, 2004: Danmarks kalkmalerier. Kbenhavn.
Norlind, Tobias, 1911: Svrdsdans och bgdans. Festskrift til H.F. Feilberg. Stockholm. Pp. 738756.
Thorell, Olof, 1977: Inledning. In: Snorre Sturlassons Edda. Uppsala-handskriften DG 11, II. Uppsala.
Zanchi, Anna, 2006. Manuscript illumination: a reliable source for medieval Scandinavian dress? In:
Vari reistur Guvari M Gunnlaugssyni fimmtugum 16. september 2006. Reykjavk.
Wagner, Eduard, Zoroslava Drobn and Jan Durdk. 2000. Medieval Costume, Armour and Weapons.
Mineola and New York. [2nd ed.]

20
 Outlaws, women and violence. In the social margins of saga
literature
Joonas Ahola, Institute for Cultural Studies, University of Helsinki, Finland
In the society that the Icelandic family sagas depict, whose public sphere was ruled by men,
violence was an extraordinary extent of action for women but it takes place. The image of
women in sagas responded to the ideas that prevailed in the context. Representations of the
image were necessarily if not acceptable, at least conceivable but within the restrictions of the
saga genre. In this paper, I will focus on social factors that would guide the interpretation of
occurrences of female violence in the saga literature.
Since womens possibilities to social influence were quite limited in the past as depicted in
sagas, the final target of their actions often required an intermediate of the opposite sex. Their
contribution to conflicts often was limited to arbitration or whetting.
Solidarity as a social act had wide consequences since every Icelander belonged to compli-
cated social networks. In a state of emergency, like in a raving blood feud, solidarity towards
a party was easily interpreted as hostility towards the other. Outlaw figures in saga literature
often are described seeking and finding protection from women. If a person was condemned
to outlawry he was not only an enemy of his prosecutor, the plaintiff, but simultaneously an
enemy of the law: by outlawry, he was denied any protection from the law. This made any
assistance of an outlaw a highly risky and morally questionable deed: one which required
strong reasons.
It is remarkable to which extent the accounts of women taking violent acts in saga litera-
ture are connected to their expressions of solidarity towards an outlaw. Repetition of a narra-
tive element such as this connection indicates its significance to the saga writers and their
audiences. Repeated narrative elements cannot be considered as mere empty literary motifs,
or clichs, without expressive power. They are meaningful expressions with a narrative func-
tion.
My initial research question is if there can be found explaining factors to the literary con-
nection of outlaw figures and women who express solidarity to an extreme degree in social
structures that are expressed in the saga literature. Any correspondences could provide hints
of how these occasions ought to be interpreted and, in a general sense, they can indicate
something about the narrative techniques of the saga writers.
I have adopted in this paper the view that textual referentiality within the corpus of medie-
val Icelandic literature was a comprehensive process rather than a mere literary convention.
Individual similarities between narrative elements may have origins in different phases of
saga production process from oral narratives to a written manuscript and its copies. Direct
borrowing from another written source, rittengsl, can be considered intertextuality in a philol-
ogical sense. When a saga writer applies literary commonplaces, or topoi, without a precise
source, the result can be called intertextuality in a literary sense. If similarities between narra-
tive elements in distinct sagas have their origins in the corpus of respective oral tradition we
can talk about traditional referentiality.1 The original source of similarity may have differed
but passages that had recognisable similarities were associated with each other and contrib-
uted to each others meanings. Narrative elements with recognizable similarities had a seman-
tic connection.
Whose conceptions the sagas reflect is another matter and will not be addressed here any
further than briefly by the binary gender aspect. Sagas in principle discuss issues that con-
1
Term coined by John Miles Foley (1991: Immanent Art: from structure to meaning in traditional oral epic.
Bloomington: Indiana.); discussed in connection with saga literature by Slavica Rankovic (2005), esp. 108116.

21
cerned the highest social strata in the time of writing: feuding was an occupation of the social
elite in the 13th century and the most sophisticated legal and political turns in saga literature
required vast corresponding knowledge and this indicates that they were written by / for rep-
resentatives of the same strata; and, it should be emphasized, its male representatives. Like
Else Mundal (1992, 108) has put it: man was the norm in the society that sagas depict.
In this paper, answers to the research question are looked for in relation with gender roles
that underlie the sagas: violence in relation with female roles and outlawry in relation with
male roles. The occasions of violent acts of a woman in assistance or even on behalf of an
outlaw have different functions in the plots of different sagas but as a repetitive narrative ele-
ment, it supposedly has a semantic logic which makes it applicable in different narrative con-
texts. Additionally, I will briefly discuss how this narrative element makes sense against the
context of saga production and, likewise, reception.

Women and Outlaws

The strong image of women in the family sagas, especially compared to contemporary conti-
nental literature, is a commonplace2. However, womens ways of acting were limited by a
strict normative social order.3 Gender roles and the corresponding social expectations were
quite clear-cut: women were responsible of managing the farm-stead whereas men took
charge of activities outside it (Jochens 1995, 116117; Grgs 1b: 44; 2:173). Womens range
of responsibility was limited to the private field of the society: homestead and immediate rela-
tives. Women were excluded of responsibilities and, hence, power in the public field of the
society: in legislation and practice of law, administration and politics, central institution of
which were the assemblies.4 Neither did women travel abroad, except for pilgrimage or set-
tlement to a new land.
Fights and battles, feuding, were the ultimate way of negotiating relationships in the public
field. Women did not have an immediate role in feuding (Jochens 1989, 109). Women natu-
rally had their share of interests in issues such as family honor or rights of inheritance, but
they lacked direct means to promote them. The principal means women had in feuding was to
goad / whet men into action (Mundal 1992, 103; Mundal 1994, 78; Heller 1958, 117118;
Bagge 1992, 1415). Goading as a means of action under such circumstances is not excluded
to women but sagas represent also goading men who lack sufficient means to act otherwise
(Byock 1983, 9495; Miller 1990, 212213).
Sagas narrate some occasions when women force a truce between fighting parties by, for
instance, casting clothes over their weapons. Peaceful enterprises are far less frequent in the
family sagas than in contemporary sagas. According to Else Mundal it is due to differences in
the narrative genre and not differences in actual practices in different times that these groups
of sagas describe (Mundal 1992, 104). Peaceful deeds are not a favoured topic for family sa-
gas unless they underline the following disaster.5 Else Mundal has pointed out, based on the
testimony of saga literature, that the honorary qualities for women were much the same as
for men, but they were used in a different way (Mundal 1994, 10). These qualities were such

2
For instance, Else Mundal (1993), 723724.
3
The conceptions of Saga Age circumstances that the writing period Icelanders had, and which reflect in saga
literature, was probably based not only on existing historical texts (sagas and Landnmabk) and respective
historical traditions but also on law texts that were produced still after subordinance to Norwegian crown and
introduction of the new law code.
4
The division private public is applied here as descriptive of the division of labour and sphere of influence
between the sexes. I am aware that as a societal system, the division private public makes sense only in the
context of a coercive state (Miller 1990, 305). However, the material at hand indicates that direct power and
hence, access to immediate action was limited for the female members of the society to only one of these fields.
5
E.g. Andersson 1967, 3840.

22
as intelligence, health, beauty and toughness (Miller 1990, 305). Strong, initiative deeds of a
woman were admired.
Male virtues in sagas are largely connected to preservation and gaining of social capital,
honour, for the individual and the social network he represents. Such virtues are expressed in
sagas mainly in terms of physical strength, bravery, intelligence and verbal skills. Honour and
manliness are closely related in saga vocabulary: the virtues of drengskapr, mannleikr, ger-
vileiki all express general honorary properties which are already by the kernel of the first
two terms gender-specific. The relation is at least as visible in the negation of the concepts:
the usage of words mannleysi or drengr refer not only to lack of additional manly virtues but
rather to total absence of them, unmanliness in an insulting sense (Meulengracht Srensen
1983, 7677).
Outlaws are a distinct group of narrative actors in the sagas. Outlawry was the severest
punishment which could be condemned for most flagrant legal offences, mainly for offending
anothers physical inviolability. According to the law, an outlaw lost his whole property and
was not to be supported, or transported, or helped in any way (Grgs I, 139), and anyone
could kill him without a threat of retribution. In principle, an outlaw was totally expelled from
the society. He was basically a dead man walking.
However, the sagas dont show that literal following of the letter of law. The position of
outlaws was desperate but the final execution of their sentence was up to the plaintiff: with a
plaintiff weak enough, or with an ally strong enough, an outlaw could well survive for a while
(Amory 1992, 194196). In sagas, outlaws visit farms like any other representatives of social
margins who wander from a farm to another.6
As main characters of a saga, outlaws are provided with the narrative perspective and con-
sequently, sympathies of the audience. Three sagas have been nominated as outlaw sagas for
the sake of the outlawry of their main characters and the concentration on their adventures and
close escapes during the time in outlawry. The so-called outlaw sagas Gsla saga, Grettis
saga and Harar saga provide an account of positive male qualities of an outlaw figure.
The outlaws in sagas are depicted according to the requirements of their unforgiving state.
Big size and great strength are typically stressed. Grettirs huge size and feats of strength are
accounted along the whole saga and a descriptive is for instance his superiority over the two
rurs sons at wrestling games at Hegranessing: eir vri eigi sterkari tveir en Grettir
einn, en hvrr eira hafi tveggja manna megin eira sem gildir vru (Grettis saga ch 72:
236). Gsli is not depicted as a man of special strength as is actually no-one else in this saga
but the saga tells that he proofed stronger than others at ball-games and that on an occasion,
he threw a stone to a distant skerry og kom ar enn at fram a Gsli var betr at rttum
binn en flestir menn arir (Gsla saga ch. 20: 66). In the discourse of the saga, this proba-
bly counts as a proof of superior strength. He was able to convincingly act Ingjaldrs retarded
son who was mikill vexti, nr sem troll (ch 25: 79). Hrur var hrur manna best og
rammur a afli[] og a llu vel vaxinn (ch 11) The outlaw heroes show extraordinary mar-
tial skill against overbearing enemy (Gsli, Hrr) and even supernatural creatures (Grettir).
Resourcefulness of the outlaw heroes is stressed to a considerable degree. Gsli is depicted as
an especially handy man in his saga (hann var hagari en flestir menn arir: ch 25: 79) and
even though Grettir is less enthusiastic about labour of any sort he is able to use his physical
strength to overcome difficult situations. Cleverness helps an outlaw to avoid his pursuers,
and they are all named vitr (Harar saga ch 36; Grettis saga (ch 93: 289); Gsla saga (ch 22:
70). Poetic skills can be understood as a proof of mental capacity, too. Actually, all the virtues
of an outlaw as well as the stock description correspond to male virtues in the sagas in gen-
eral.

6
Cochrain 2002; Cris Callow (2004) discusses Norwegians as similar extraneous narrative agents.

23
 Assemblies, which have a relatively central place in saga narration, were a forum excluded
to men: the status of a free male guaranteed one the right to attend. Attending was a central
manly function. Outlawry stripped one of all protection of the law, not to speak of attending
the gathering of the central institution of law. According to the law (but not the sagas, it has to
be noted) if a man was expected to be outlawed at an assembly for a manslaughter, he was not
even allowed to attend the court in which his own case was treated. Consequently, an outlaw
was stripped of a significant share of his masculinity along with his right to fulfill male obli-
gations. Outlawry was a male state. The law applied to women as well but no female outlaws
appear in the family sagas.
The outlaw lost his home and the world outside was hostile and forbidden for him. In order
to find refuge, he had two directions to turn to: either to the deserted, uninhabitable inland or
to the private field.7

The Connection
Solidarity in saga literature seems to require taking its social consequences in consideration:
assistance is an act in the interests of a party and all its reference groups, but often it is simul-
taneously an act against those of another. Family is an obvious solidarity group. Grettir is pro-
tected by orbjrg digra for the sake of their kinship (Grettis saga ch 52: 167169). As Else
Mundal (1992, 104) has noted, Women had a double loyalty: to their own family and that of
their husbands. Even an ideally loyal wife, as Aur is depicted in Gsla saga, may act
against the interest of her own husband if it is in conflict with her obligations by blood rela-
tionship. In ch 29 (9293), Aur gives shelter to her two nephews after they have killed
Gslis brother orkell. Gsli gets furious when he hears of it but Aur is able to calm him
down. The absurdity of the settings, outlaws covered from a pursuing outlaw, indicates stabil-
ity of the motif.8 According to the law, a killer was an outlaw right after the deed and had no
protection from the law: the sentence passed at the court was its formal announcement. This
makes Aurs nephews outlaws in the passage.9 In Gunnars ttr irandabani, Gunnarr kills
irandi and escapes the revenge. Gunnar goes to Helgi sbjarnarson who accommodates
him in an outhouse. Helgi asks his wife rds to take good care of Gunnarr while he is away.
rdss brother Bjarni uses the opportunity together with other pursuers of Gunnarr and de-
mands rds to deliver Gunnar to them. rds assures her will to cooperate and asks the
men to return next morning when she would hand over Gunnar. During the night, she gathers
a large troop of her relatives to the farm, and in the morning the pursuers have to withdraw.
After returning later, Helgi says to his wife: Vissa ek, at ek var vel kvntr, ok er at vel, at
hon sagisk tt sna (ch 6: 209) which resembles to a great degree the words uttered by
Gsli to his wife Aur for her support. Aur joins his husband in his final battle with a club
and hits the leader of the attackers, Eyjlfr in the arm and disables him from continuing the
fight. Gsli expresses his admiration: at vissa ek fyrir lngu, at ek var vel kvntr, en
vissa ek eigi, at ek vra sv vel kvntr sem ek em. (Gsla saga ch 34: 112)
It was not the first time Aur hit Eyjlfr. Earlier in the saga, Eyjlfr offered Aur a purse
full of silver for handing over Gsli. Aur responds by hitting him on the nose with the purse,

7
Kirsten Hastrup (1985) divided the conceptual (horizontal) world view of medieval Icelanders into the inside
(innan) and outside (tan) of civilisation (147154) and placed the outlaws to the latter sphere. The division is
telling in many ways but as an approach to the following materials, I suggest a not binary but tripartite division
to Private Public Outside.
8
Also Laxdla saga, ch 14: 31.
9
Likewise, in Vatnsdla saga (ch 44: 118119), a man called orkell kills a man called Gli and, conse-
quently, is pursued by his followers. He has to seek shelter inside a farm and a woman called Helga allows him
to escape through her bed-closet which she seals by swinging an axe in a threatening way.

24
emphasising the humiliating nature of the deed: Haf n etta ok bi skmm ok klki.
(Gsla saga ch 32: 101)
The bloody motif of a woman defending an outlaw with a purse is met elsewhere as well:
in Laxdla saga (ch 15) a woman called Vigds had given a shelter to her relative rlfr
who was dodging consequences of a killing. Her husband is ready to give in the outlaw to his
pursuers against silver but Vigds spoils his plot by warning rlfr who escapes. When the
head of the pursuers, Ingjaldr, demands his silver to be returned, she hits him on the nose with
the purse so that blood spilled on the ground and refuses to return it. Ingjaldr retreats in-
deed, and considers his journey a failure (unir illa vi sna fer).10
Quite contrary to this, also in Laxdla saga, is the willingness of an anonym maid to hand
over her outlaw lover Stgandi and to be bought to freedom in exchange (ch 38: 107108).
Her behaviour is not praised: she is not even given a name.11
To avenge a persons death was an act of posthumous solidarity, recognition of responsi-
bility. Avenging was strictly limited to male members of the society. However, women may
have got short of other means when they had a personal interest in a matter but no-one to
goad. Gsla saga tells that after the death of Gsli Srsson, his sister rds is wining and din-
ing the killers who were at the service of her husband, Brkr. rds drops a tray of spoons in
front of the table at which the head of the killers, Eyjlfr gri sits. When she bends down to
pick them up she grasps Gslis sword that Eyjlfr had placed by him, and attempts to push it
into his stomach under the table. The hilt of the sword hits the table and the sword hits Ey-
jlfrs thigh. He survives even though the wound is serious. Brkr pays compensation for the
wound, and rds declares herself divorced. (ch 38: 116117) None of the killers of Gsli is
very happy about this outcome, and especially Eyjlfr, once more, unir illa vi sna fer.12
In Harar saga, after the death of Hrur, his sister orbjrg is frustrated of not having
anyone to avenge her brothers death. In the evening, in the bed she places a sword on her
husband but he grasps the weapon by the edge and has his hand cut (ch 38).13 It seems like
orbjrg strikes him: Indrii blocks the movement of the sword (tk mti) and cuts himself
in the process, which can be considered not likely to happen if the movement is not fast. In-
drii was among the killers of Hrr which made him liable and a target of revenge. He at-
tempted to buy the peace from his wife by agreeing to get the head of orsteinn gullknappur
who gave Hrr his death-blow. orbjrgs avenging blow on Indrii with the sword reached
the target, indirectly, although the ultimate victim, orsteinn, was crippled instead of killed.
However, she lost her husband in the process.
Stabbing in bed is a motif which has been studied carefully. Hermann Plsson (1974)
stressed the nature of this stock motif as a heroic motif and Preben Meulengracht Srensen
(1986) as a form of a crime of passion. The stock scene has been applied in outlaw biogra-
phies in several ways / functions: in the most famous instance in Gsla saga, it is revenge in a
reciprocal manner, and crucial one for the saga plot as the initial crime which led to Gslis
outlawry. In a similar way, and to the degree of a literary loan in similar details, in Droplau-
garsona saga it is a manner of revenge of a brother that leads to outlawry of the responsible
(ch 13: 169171). In Bandamanna saga, it is a passionate deed of the outlawed spkr who
kills the lover of his former wife (ch 12: 360362). The stock scene has a mythic dimension

10
Compare the expression in Gsla saga: Eyjlfr[] unir illa sinni fer ch 32: 101.
11
However, Stgandi was a witch, which appears to be the factor that made his deceiving if not praised at least
appropriate and unpunished.
12
William Ian Miller (1990) considers rds failure as an indirect expression of disapproval by the saga author
(355) but the humiliating location of the wound right below the stomach is to my mind just as easily read as a
successful one.
13
En er au komu sng um kveldi br orbjrg saxi og vildi leggja Indria bnda snum en hann tk
mti og var sr mjg hendi.

25
as the idea of death of a settled wandering hero in a hostile alien house: Sigurr Ffnisbanis
death as a revenge for the honour of sister but in reality for the sake of a womans (Bryn-
hilds) intrigues (which are motivated by passionate love) as depicted in Sigurarkvia in
skamma (Plsson 1974). Bed as the setting of violence stresses the helplessness and unaware-
ness of the victim. Its connotations extend to sexuality, and being the receiving party of a
thrust was a shameful position (Meulengracht Srensen 1983, 5960).
It is worthwhile noticing that these women only wound their victims. To my knowledge,
there are no reported killings by women in saga literature. This is perhaps where the limit of
acceptability was. Killing would be a direct contribution to a feud, and require revenge in the
structural balance of saga narration, whereas mere assistance is obviously seen as interference
of an exterior agent.
When a narrative element establishes it starts to function as if given: with its face value.
Connotations seize a larger share and the denotative meaning, that the propositions it consists
of expressed as their sum, becomes less significant. Connotative meanings may even take a
leading role. Here follows an example.
Laxdla saga tells that rr and Gurn are both unhappy in their marriages and fond of
each other. rr advices Gurn to get a lawful reason for divorce: by sewing her husband a
shirt with so wide neck opening that his nipples show. She does this and divorces success-
fully, and reciprocally advices rr to get divorced on grounds that his wife brka-Aur uses
mens pants skarsk setgeirabrkr sem karlkonur (ch 35: 96). Aur avenges this humilia-
tion after the divorce by stabbing rr in his sleep.
According to Aurs introduction in the saga, ekki var hon vn kona n grvilig and this
made her less feminine by saga standards (Jochens 1991, 21). Initially, the accusation of Aur
wearing pants is indicated as false. However, when she takes off for the avenging journey on
horse-back, the saga writer sees necessary to remark that at that time she must have worn
pants. Her masculinity is also stressed by telling that she rides the horse so fast that her male
companion can hardly follow (ch 35: 97).
The detailed, tense description of her approach to rrs bedcloset resembles the stab-
bings in bed in Gsla saga and Droplaugarsona saga. Aur stabs rr in the arm and both
his nipples are wounded; the blow is so powerful that the sword is stuck in the bed. This also
is an element of the stock scene of stabbing in bed (Plsson 1974, 1921; Meulengracht
Srensen 1986, 250251). Her brothers admired her deed (eir ltu vel yfir) (98); and it was
also accepted by rr as a compensation to his previous deed (thus avoiding any following
feuding, potentially unhonourable when a woman is involved). Jochens (1991, 910) consid-
ers this scene an ironic variation of the slaying in bed-theme, the passionate ethos revealed by
the term describing rrs turning in his wakening towards the attacker, snask at which
indicates sexual initiative and is used in Gsla saga as well (Jochens 1995, 203 n.43). Accord-
ing to William Ian Miller (1990, 354355), the irony is emphasised by the fact that Aur cut
the humiliating neckline on his flesh.
Female violence does not take place in connection with an outlaw in this passage. How-
ever, it has clear referential connections to central outlaw narratives. The narrative elements
are no longer attached to outlawry only but to social margins in a more general sense: a figure
that is depicted in a sexually marginal, less feminine light resembles an outlaw her unability
to function openly according to the norms of the public field (her brothers seem no help)
forces her to commit the deed in secrecy, within the private field, her previous home.
Reciprocal solidarity of the outlaws towards women as a stereotyped narrative element in
the saga literature indicates some degree of fixedness of the narrative characters and the con-
nection between them. Outlaws in sagas often help women in connection to the stock motif of
women, and especially their honour threatened by male ogres such as berserks. Within an

26
outlaw biography, the scene functions as a trial of the hero. 14 The tension of the motif is cre-
ated by the solitude and helplessness of the women and the unexpected appearance of the
hero, though not always an outlaw. Occasionally the meeting of an outlaw and a woman cre-
ates romantic tension uncommon to saga narrative in general; already the appearance of a
strange man in the domestic sphere, in absence of male family members, creates the tension.
A socially independent agent is potentially free to act within the society without restrictions it
has set on its members; norms as preventing people from following their instincts do not tie
those outside.

Conclusion
Narrative in general is interested in the extraordinary: in this context we can talk about saga-
worthy material. Both assistance of an outlaw and violence of a woman exceeded the limits
of conventionality within the scope of saga literature, and it is no wonder that narrative ele-
ments combining them were attractive to saga writers. From the perspective of the central
institutions of social interaction in medieval Iceland, the law and its manifestation in the
aling, both women and outlaws belonged to the social margins. Their mutual solidarity in the
sagas, narratives that had this perspective, was enabled by belonging to the same social stra-
tum. The private field, ruled by women, was an alternative refuge for the outlaw, and the one
with social consequences. These consequences were the materials for saga literature and
hence recorded. The emphasis on social marginality in the encounters enabled also extreme
acts. Violent acts of women were way beyond the norms of feuding. A mans reliance on the
support of a woman was inappropriate but accepted in the state of outlawry just like other
normally unacceptable deeds such as robbery and thievery (Andersson 1984, 501502). The
common elements in the passages dealt with in this paper indicate a semantic connection be-
tween the violence of women and an outlaw figure.15 In more general terms, they indicate that
orientation of a scene (to social margins) was closely tied to the selection of narrative materi-
als and that their usage in an ironical purpose stresses the fixedness of the semantics.

Bibliography
Bandamanna saga. Jnsson, Guni (ed.) 1936: slenzk fornrit VII: Grettis saga smundarsonar. Hi
slenzka fornrtaflag: Reykjavk. P. 291363.
Droplaugarsona saga. Jhannesson, Jn (ed.) 1950: slenzk fornrit XI: Austfiringa sgur. Hi slenzka
dornrtaflag: Reykjavk. Pp. 135180.
Gsla saga Srssonar. rlfsson, Bjrn & Jnsson, Guni (eds.) 1943: slenzk fornrit VI: Vestfiringa
sgur. Hi slenzka fornrtaflag: Reykjavk. Pp. 1118.
Grgs. Karlsson, Gunnar; Sveinsson, Kristjn; rnason, Mrur (ed.) 1992: Grgs. Lagasafn slen-
ska jveldisins. Ml og menning: Reykjavk.
Grettis saga. Jnsson, Guni (ed.) 1936: slenzk fornrit VII: Grettis saga smundarsonar. Hi slenzka
fornrtaflag: Reykjavk. P. 1290.
Gunnars ttr irandabana. Jhannesson, Jn (ed.) 1950: slenzk fornrit XI: Austfiringa sgur. Hi
slenzka fornrtaflag: Reykjavk. Pp. 193211.
Harar saga. Hast, Sture 1960: Harar saga. Ejnar Munksgaard: Kbenhavn.
Laxdla saga. Sveinsson, Einar l. (ed.) 1934. slenzk fornrit V: Laxdla saga. Hi slenzka
fornrtaflag: Reykjavk.

14
The fight with a monster who claims a woman is an international medieval motif, especially popular in the
fornaldarsgur. See Holtsmark 1951, 7; Liestl 1930, 154.
15
Actually, Anne Heinrichs (1986) analysis on similarities between heroic female characters resemble those of
an outlaw hero to a considerable degree. Her hypothesis is that the categorical characteristics of isolation, active
life, sexual initiative, wisdom and power over life and death in saga heroines derive from prepatriarchal concep-
tions and are adapted from the Niebelungen tradition to the saga narratives. It is worth considering whether the
categories were general prerequisites and characteristics for any heroic action, regardless of sex.

27
Vatnsdla saga. Sveinsson, Einar l. (ed.) 1939: slenzk fornrit VIII: Vatnsdla saga. Hi slenzka
fornrtaflag: Reykjavk. P. 1131.

Amory, Frederic 1992: The medieval Icelandic outlaw: life-syle, saga, and legend. From Sagas to
Society: comparative approaches to early Iceland. Hisarlik Press: Chippenham, Wiltshire. Pp.
189204
Andersson, Theodore M 1967: The Icelandic Family Saga. Harvard University Press: Cambridge,
Massachusetts.
1984: The Thief in Beowulf. Speculum vol 59, no 3. pp. 493508.
Bagge, Sverre 1992: Menn og kvinner i Heimskringla. In Sellevold, BJ; Mundal, E; Steinsland, Gro
(eds.): Fokus p kvinner i middelalderkilder. Viktoria Bokfrlag: Skara.
Byock, Jesse 1982: Feud in the Icelandic Saga. University of California Press: Berkeley, Los Angeles.
Callow, Chris 2004: Narrative, Contact, Conflict, and Coexistence: Norwegians in Thirteenth-Century
Iceland. In Adams, Jonathan and Holman, Katherine (eds.): Scandinavia and Europe 8001350:
Contact, Conflict, and Coexistence. Medieval texts and cultures of Northern Europe 4. Brepols:
Turnhout.
Clover, Carol 1986: Hildigunnrs lament. In: Lindow, John, Lnnroth, Lars and Weber, Gerd Wolf-
gang (eds.): Structure and meaning in Old Norse Literature. Odense University Press: Odense. Pp.
141183.
Cochrain, Jamie 2002: Hawkers, beggars, assassins and tramps : Fringe characters in the slendinga
sgur. In Stefanie Wrth, Tnno Jonuks, Axel Kristinsson (eds.): Proceedings of the conference
Sagas and Societies Section 1: Societies and Social Circumstances ; 17 http://tobias-lib.ub.uni-
tuebingen.de/volltexte/2004/1071/ 3.4.2009
Frank, Roberta: Why Skalds Address Women. In: Proli, Teresa (ed.): Poetry in the Scandinavian
Middle Ages: The Seventh International Saga Conference. Centro Italiano di Studi sullAlto Me-
dioevo: Spoleto. Pp. 6783.
Hastrup, Kirsten 1985: Culture and history in Medieval Iceland. Clarendon Press: Oxford.
Heinrichs, Anne 1986: Annat var vrt eli. The type of the prepatriarchal woman in Old Norse litera-
ture. In: Lindow, John, Lnnroth, Lars and Weber, Gerd Wolfgang (eds.): Structure and meaning
in Old Norse Literature. Odense University Press: Odense. Pp. 110140.
Heller, Rolf 1958: Die literarische Darstellung der Frau in den Islndersagas. Veb Max Niemeyer
Verlag: Halle.
Holtsmark, Anne 1951. Studies in the Gsla Saga, Studia Norvegica Ethnologica et Folkloristica 6,
355.
Jochens, Jenny 1989: The Medieval Icelandic Heroine: Fact or Fiction? In: Tucker, John (ed.): Sagas
of the Icelanders. A Book of Essays. Pp. 99125.
1995: Women in Old Norse Society. Cornell University Press: Ithaca.
1996: Old Norse Images of Women. University of Pennsylvania Press: Philadelphia.
Meulengracht Srensen, Preben 1983: The Unmanly man. Concepts of sexual defamation in early
Northern society. Odense University Press: Odense.
1986: Murder in marital bed. An attempt at understanding a crucial scene in Gsla saga. In: Lindow,
John, Lnnroth, Lars and Weber, Gerd Wolfgang (eds.): Structure and meaning in Old Norse Lit-
erature. Odense University Press: Odense. Pp. 235263.
Miller, William Ian 1990: Bloodtaking and Peacemaking. Feud, Law, and Society in Saga Iceland.
University of Chicago Press: Chicago.
Mundal, Else 1992: Norrn litteratur som kjelde til nordisk kvinnehistorie. In: Lvkrona, Inger (ed.):
Kvinnospor i medeltiden. Lund University Press: Lund. Pp. 93114.
1993: Women in sagas. In Phillip Pulsiano, Kirsten Wolf (ed.): Mediaeval Scandinavia. An Ency-
clopedia. Taylor & Francis
1994: The position of women in Old Norse society and the basis for their power. In: NORA vol. 2, nr
1. Pp. 311.
Plsson, Hermann 1974: Death in Autumn. Tragic Elements in Early Icelandic Fiction. Bibliography
of Old Norse-Icelandic Studies 1973. The Royal Library: Copenhagen. Pp. 739.
Rankovic, Slavica 2005: The Distributed Author and the Poetics of Complexity: a Comparative Study
of the Sagas of Icelanders and Serbian Oral Poetry. An unpublished doctoral thesis. University of
Nottingham.
Turville-Petre, Gabriel 1977: Outlawry. Sjtu ritgerir helgaar Jakobi Benediktssyni 20. Jl 1977.
Stofnun rna Magnssonar, Reykjavk. Pp. 76978.

28
 The Formation of the Kings Sagas
Theodore M. Andersson, Germanic, Indiana University, USA
What I propose in this paper is merely a footnote to Tommy Danielssons second volume,
Sagorna om Norges kungar (2002b). After a detailed consideration of many specialized prob-
lems in the kings sagas Tommy concludes with a brief chapter (pp. 38595) on how the
kings sagas evolved into the form we now have. He reminds us of the prominent place occu-
pied by the Norwegian kings in Laxdla saga, and a number of other sagas, and goes on to
review the meetings of prominent Icelanders with Norwegian monarchs particularly in the
ttir. These contacts could have served as the point of departure for the Icelandic interest in
the kings and the growth of oral narrative. Tommy also reviews the named Icelandic tradition
bearers: orgeirr afrskollr, Oddr Kolsson, and Hallr rarinsson, all of whom were among
Aris sources, and the young Icelander who learned the story of Haraldr harris early ad-
ventures from Halldrr Snorrason and performed it at Haraldrs court. We do not know the
exact form of such transmissions, but Tommy notes the comments made by Theodoricus and
Saxo indicating that the Icelanders cultivated rich traditions, a reputation confirmed by the
prologues in Heimskringla and by the ttir in Morkinskinna. This narrative material is gen-
erally assumed to have provided a rough basis for the written accounts later shaped by writers,
but Tommy asks whether this assumption is necessarily correct and whether the underlying
narrative could not have been in the form of polished storytelling (p. 392: ett ytterst
avancerat berttande). These stories do not surface in the early period because there would
have been no reason for Ari or Smundr or the later synoptic historians to reproduce stories
that everybody knew.
At about the same time as these stories were circulating, domestic Icelandic sagas would
have been evolving on the basis of legal disputes and feud stories, as Tommy argues in his
first volume. A likely venue for the exchange of such stories would have been the Icelandic
thingmeetings, just as the young Icelandic storyteller in Morkinskinna learned the story of
Haraldr harri at thingmeetings over a series of summers. The evolution of royal stories is
perhaps less easy to grasp than the evolution of native stories, but Tommy suggests several
possibilities. There could have been a tradition of comparing kings, or the kings could have
been of ongoing and central importance to the Icelanders, or there could have been a concreti-
zation of royal stories analogous to the ttir. The kings sagas could also have been modeled
on the agonistic patterns of the evolving Icelandic sagas. In turn, the growth of the
kingssagas into large books could have paved the way for the large Icelandic sagas such as
Laxdla saga and Njls saga.
The central issue in this argument is the existence of fully developed kings sagas in oral
tradition. This is indeed a new perspective on the kings sagas, and we may ask ourselves why
it has not been aired before. One reason is surely that the very idea of an oral saga fell out of
favor in Icelandic circles throughout the twentieth century. A leading project of the Icelandic
School was to diminish our faith in the existence of full oral sagas about early Iceland, and it
was only to be expected that the generations engaged in this project would not contradict
themselves by advocating oral kings sagas. On the contrary, they focused on the development
of the sagas as a purely literary enterprise, perhaps ultimately based on scattered oral tradi-
tions but carried out exclusively with quill and ink. Naturally the same assumption would
have carried over to the kings sagas.
But an analogy with the native Icelandic sagas is not the only justification for believing in
the piecemeal literary composition of the kings sagas. Our information about the latter begins
in fact almost a century earlier than the information on the native sagas and gives every ap-

29
pearance of suggesting a gradual literary evolution from smaller written denominations to
larger denominations. The process began with Smundr and Ari at the beginning of the
twelfth century and culminated in the Norwegian synoptics at the end of the century. That this
was a literary sequence is supported by what seems to be a growing consensus that there is a
continuity between the early epitomes and the later ones. Despite Theodoricuss protestations
that he based himself not on visa but on audita, it seems likely that he also used written
sources and that these sources are most likely to have been Smundr and Ari.
The picture that emerges from the twelfth century is therefore a puzzling together of in-
formation, including oral sources but collected by writers who converted what they could
learn into little digests and summaries, not stories. This picture is reinforced by the shape of
the first full-length kings sagas, Oddr Snorrasons lfs saga Tryggvasonar and The Oldest
Saga of Saint Olaf largely represented by The Legendary Saga. To be sure, these are full-
blown biographical stories, but they are quite awkwardly composed. They do not suggest au-
thors recording flowing narratives but rather writers who are trying to fit and join scraps of
tradition. That might lead us to believe that writers in the twelfth century began by condens-
ing the main points provided by the oral transmissions and ended by trying to expand these
early indications somewhat artificially into real books, an entirely literary project. The tacit
assumption might then go on to stipulate that when the master narratives appear, largely in
Morkinskinna and Heimskringla, they again perfect the form of the older written narratives
using strictly literary methods.
How does Tommy Danielssons suggestion of ready-made, full-fledged oral narratives
about the kings comport with this picture of writers struggling to achieve a literary form for
the royal biographies from scattered traditions? If the first biographers were faced with the
simple task of setting down well articulated oral stories in writing, why did they perform the
task so poorly? Perhaps an analogy will help us out of this dilemma. Since the publication of
Gsli Sigurssons and Tommy Danielssons books no one seems any longer to have diffi-
culty with the idea that there were fully developed sagas about early Iceland, but we must
remind ourselves that these sagas also had an awkward beginning.
The question of which slendingasgur came first is of course a subject of dispute, and I
can only say which sagas I think came first. I think that all the skald sagas, including
Gunnlaugs saga, were early, and to that group of four I would add Fstbrra saga, Vga-
Glms saga, and Reykdla saga. What these sagas have in common is that they are not grace-
fully composed, unlike the great sagas of the next generation, Egils saga, Gsla saga, and
Laxdla saga. The early sagas are in some cases quite short and in other cases rather me-
chanically constructed around skaldic stanzas.
Accordingly we find both among the sagas about early Iceland and the kings sagas a
prefatory period of experimental and rather problematical composition before the perfected
form emerges. There can now be little doubt that the domestic Icelandic sagas were drawn
from oral tradition. It therefore seems clear that the transposition from oral stories to written
stories was by no means straightforward. It required practice. By analogy we can suppose that
oral kings sagas would have been no easier than the slendingasgur to convert smoothly
into written sagas at the first attempt. That means that the awkward first biographies of Olaf
Tryggvason and Olaf Haraldsson do not exclude the possibility that there existed full oral
sagas about these and other kings. It was only a question of learning to recast these oral proto-
types into written sagas.
Nor should we forget that there were stories intermediate between the domestic sagas and
the kings sagas, to wit the ttir, in which equal space is given to the Norwegian kings and
the Icelandic adventurers. The ttir are very much at the center of Tommy Danielssons dis-
cussion and are at least one secure key to the operations of oral transmission, inasmuch as
they can hardly be explained by any other conveyance. They provide information on the kings

30
and their attitudes, character, and politics, as well as on their contacts with the Icelanders. The
warrant that they were circulated in the earliest period of saga writing (1200 to 1220) is the
preservation of thirteen examples in Morkinskinna. Their focus is the Icelandic experience of
the outside world, and they must therefore have been handed down in Iceland, perhaps in the
families of those who experienced them. The dual focus on kings and Icelanders assures us
that at least some memory of the kings would have stayed alive in Iceland.
As Tommy Danielsson points out, they also illustrate the general Icelandic preoccupation
with Norwegian kings. In the early twelfth century, both Smundr and Ari directed their at-
tention to the nearest kings in Norway. If their books had been preserved, the task of under-
standing Icelandic thinking about the Norwegian kings would perhaps have been facilitated,
but even the bare existence of these books tells us something. The kings seem to have been
Smundrs sole preoccupation, and though we may be apt to think of Aris konunga vi as
a supplement to his slendingabk, simply because we have one and not the other, the situa-
tion may have been reversed. Perhaps the konunga vi were the primary undertaking, and
perhaps we should consider slendingabk as the supplement. In either case the Norwegian
kings were a dominant factor when the Icelanders first began to write.
We are not told much about the interaction between the Norwegian kings and Iceland un-
der the early kings down to 995, but after the advent of the conversion kings the interaction
becomes charged. Olaf Tryggvason appears to have been an energetic prosylitizer well be-
yond the shores of Norway, and that may perhaps understate the case. He was credited with
the conversion of five lands (chapter 17 of Historia Norwegiae and chapter 19 of grip), and
texts such as Aris slendingabk, Oddr Snorrasons lfs saga Tryggvasonar, Kristni saga,
and Laxdla saga lead us to believe that he exerted strong pressure on the Icelanders to con-
vert. Indeed, it seems not unlikely that the chief reason for Icelands conversion was Olafs
mission. This is the point at which Norway becomes a real, not to say a menacing, factor in
the political life of Iceland.
The threat materializes palpably under Olaf Haraldsson, who, according to Heimskringla
(F 27.21418, 240), not only tries to cajole the Icelanders into making him a gift of the island
Grmsey but later holds distinguished Icelanders hostage to exert pressure. Subsequently Har-
ald Hardrule is said to have been a great friend of the Icelanders (Morkinskinna, p. 170), but
given his record of deceitfulness and his aggressive foreign policy, we would like to know
what motivated his friendship. Adam of Bremen (Book 3, chap. 17 [p. 159]; Book 4, scholion
146 [p. 267]) states that Harald extended his rule as far as Iceland. This corresponds to noth-
ing in the indigenous sources, but we may well wonder where Haralds contemporary Adam
may have gotten the idea. In the twelfth century the Norwegian kings were sufficiently preoc-
cupied with other matters that they did not pose much of a threat, but the very fact that the
Icelanders had such a clear memory of Norwegian aspirations under the Olafs indicates that
they must have had a watchful eye on Norway. Add to this that, whatever the actual history of
immigration to Iceland may have been, the Icelanders clearly thought of themselves as kin to
the Norwegians by lineage and culture. The national umbilical cord seems not to have been
severed, and Norway remained much more than just a horizon.
We can be in no doubt that information on Norway was plentiful in Iceland, but the ques-
tion to be dealt with is not one of information but of literary form. The Icelanders could of
course have known a great deal about Norway without ever casting anything in narrative
form. That they did think in terms of literary form is sufficiently demonstrated by the ttir
with their identifiable morphology, but the oral existence of short ttir may not justify the
assumption of longer sagas. Even so the evidence for oral kingssagas is rather better than the
evidence for oral slendingasgur. This evidence resides largely in the tferarsaga of Harald
Hardrule that Halldrr Snorrason teaches to a young Icelander, who in turn recites it at Har-
alds court. Tommy Danielsson refers to this recital in both of his volumes, but it may lend

31
itself to further exploitation. At the very least the episode suggests that such stories were for-
mally composed with enough detail so that they had to be learned, that they were formally
recited to a large group, and that they were long enough to be presented for two weeks. They
were formal stories, not just random accounts.
Not only that, but the story of Haralds adventures in the Mediterranean, as it is told in
Morkinskinna and by extension in Heimskringla, was clearly a highly dramatic story of in-
trigue in the Byzantine court, military prowess and ingenuity, and the accumulation of fabu-
lous wealth, a thirteenth-century counterpart to The Count of Monte Cristo. The oral version
that held the attention of King Haralds court for two weeks must have shared some of these
qualities; it too must have been a rousing tale of derring-do and high romance.
I have indicated that there seems to be a tacit assumption that the evolution of the
kingssagas from notes and summaries in the twelfth century to epic canvases in the thirteenth
century was a strictly literary process. That is to say, people simply learned to write better and
better and more fully as time went on. At the same time we have evidence that there were
fullblown, dramatic tales in oral form. The awkward formulations in the twelfth-century
epitomes and the first attempts at biography teach us that the ostensibly simple option of tran-
scribing oral stories was not adopted. The first efforts at duplicating what may have been
rather good oral stories fell short and converted good stories into not very successful books.
The art of capturing good stories on parchment was a gradual process, learned slowly and a
little painfully. It seems to have combined a knowledge of stories with a faltering acquisition
of writing skills.
Vsteinn lason has recently used the word imitation to describe this process and refers
to Preben Meulengracht Srensens earlier use of the same term (2007:34): The narrative
style and technique of the sagas shows every sign of being an imitation, conscious or uncon-
scious, of oral narrative. Imitation may well be as close as we can get to a resolution of
this problem. Vsteinn uses it with reference to the slendingasgur, but, following Tommy
Danielsson, I have no difficulty in extending the usage to the kings sagas as well. Indeed, it
seems to me that the kings sagas reveal the nature of the imitation more clearly and more
fully by making the stages in the development more palpable. The first stage was to skim the
highpoints by way of a summation. The second stage was to add detail in order to approxi-
mate at least the length of the oral sagas. The third stage was then to imitate the narrative style
as well as the narrative dimensions of the oral stories.
The progress from brief summary of the main points in the oral transmission to a fuller re-
capitulation in the first biographies and finally to a recreation of the dramatic story line is
clearer in the kings sagas than in the slendingasgur, but the same line can also be detected,
though more tentatively, in the latter. The famous summary of Hnsa-ris saga found in
Aris slendingabk is analogous to the epitomes on the Norwegian kings and represents the
first stage in the narrative development. The second stage is more difficult to match because
the differences in the quality of composition among the early slendingasgur are less palpa-
ble than in the kings sagas. We can nonetheless make it plausible that the earliest slendin-
gasgur were less well assembled than the later masterpieces. If I were to choose one slend-
ingasaga to illustrate the original defects of composition, it might be Kormks saga, a saga
that does not so much tell the story as it extracts the main moments of the biography from a
large collection of stanzas. Dialogue and drama are largely missing.
There is also a good match in the chronology of these developments. If Egils saga was
written as early as the 1220s, we might infer that the third stage in the slendingasgur was
reached in the same time frame as the culmination of the kings sagas in Morkinskinna and
Heimskringla. These were the highpoints in both genres and they represent a level seldom
attained again.

32
 Bibliography
Texts
Adam of Bremen. Gesta Hammaburgensis Ecclesiae Pontificum; Hamburgische Kirchengeschichte.
Scriptores Rerum Germanicarum in Usum Scholarum. Ed. Bernhard Schmeidler. 3rd ed. Hannover
and Leipzig: Hahn, 1917; rpt. 1977.
grip af Nregskonungasgum. A Twelfth-Century Synoptic History of the Kings of Norway. Ed. and
trans. M. J. Driscoll. [London]: Viking Society for Northern Research, University College London,
1995.
Ari orgilsson. slendingabk. Ed. Jakob Benediktsson. slenzk fornrit I.1. Reykjavk: Hi slenzka
Fornritaflag, 1968.
Egils saga Skalla-Grmssonar. Ed. Sigurur Nordal. slenzk fornrit II. Reykjavk: Hi slenzka Fornri-
taflag, 1933.
Fstbrra saga. In Vestfiringa sgur. Ed. Bjrn K. rlfsson and Guni Jnsson. slenzk fornrit
VI. Reykjavk: Hi slenzka Fornritaflag, 1943.
Gsla saga Srssonar. In Vestfiringa sgur. Ed. Bjrn K. rlfsson and Guni Jnsson. slenzk
fornrit VI. Reykjavk: Hi slenzka Fornritaflag, 1943.
Gunnlaugs saga ormstungu. In Borgfiringa sgur. Ed. Sigurur Nordal and Guni Jnsson. slenzk
fornrit III. Reykjavk: Hi slenzka Fornritaflag, 1938.
Heimskringla. Ed. Bjarni Aalbjarnarson. slenzk fornrit XXVIVIII. Reykjavk: Hi slenzka Fornri-
taflag, 194151.
Historia Norwegie. Ed. Inger Ekrem and Lars Boje Mortensen; trans. Peter Fisher. Copenhagen: Mu-
seum Tusculanum Press, 2003.
Hnsa-ris saga: In Vestfiringa sgur. Ed. Sigurur Nordal and Guni Jnsson. slenzk fornrit III.
Reykjavk: Hi slenzka Fornritaflag, 1938.
Kormks saga: In Vatnsdla saga. Ed. Einar l. Sveinsson. slenzk fornrit VIII. Reykjavk: Hi
slenzka Fornritaflag, 1939.
Kristni saga: In Kristni saga, ttr orvalds ens vfrla, ttr sleifs biskups Gizurarsonar,
Hungrvaka. Ed. Bernhard Kahle. Halle an der Saale: Niemeyer, 1905.
Laxdla saga. Ed. Einar lafur Sveinsson. slenzk fornrit V. Reykjavk: Hi slenzka Fornritaflag,
1934.
Morkinskinna. Ed. Finnur Jnsson. SUGNL 53. Copenhagen: J. Jrgensen, 192832.
Oddr Snorrason. Saga lfs Tryggvasonar af Oddr Snorrason munk. Copenhagen: Gad, 1932.
Olafs saga hins helga; Die Legendarische Saga ber Olaf den Heiligen (Hs. Delagard. saml. nr.
8II). Ed. and trans. Anne Heinrichs, Doris Janshen, Elke Radicke, Hartmut Rhn. Heidelberg: Win-
ter, 1982.
Reykdla saga: In Eyfiringasgur. Ed. Jnas Kristjnsson. slenzk fornrit IX. Reykjavk, 1956.
Theodoricus monachus. Historia de Antiquitate Regum Norwagiensium. In Monumenta Historica
Norvegiae. Ed. Gustav Storm. Kristiania: A. W. Brgger, 1880; rpt. Oslo Aas & Wahl Boktrykkeri,
1973.
Vga-Glms saga: In Eyfiringasgur. Ed. Jnas Kristjnsson. slenzk fornrit IX. Reykjavk: Hi
slenzka Fornritaflag, 1956.

Danielsson, Tommy. 2002a. Hrafnkels saga eller fallet med den undflyende traditionen. Hedemora:
Gidlunds Frlag.
. 2002b. Sagorna om Norges kungar. Frn Magns gi till Magns Erlingsson. Hedemora: Gid-
lunds Frlag.
Gsli Sigursson. 2002. Tlkun slendingasagna ljsi munnlegrar hefar. Tilgta um afer. Reyk-
javk: Stofnun rna Magnssonar slandi. Trans. Nicholas Jones, The Medieval Icelandic Saga
and Oral Tradition: A Discourse on Method, Publications of The Milman Parry Collection of Oral
Literature 2. Cambridge, MA: The Milman Parry Collection of Oral Literature, Harvard University,
2004.
Meulengracht Srensen, Preben. 1993. Fortlling og re. Studier i islndingesagaerne. Aarhus: Aar-
hus University Press.
Vsteinn lason. 2007. The Icelandic Saga as a Kind of Literature with Special Reference to Its Rep-
resentation of Reality. In Learning and Understanding in the Old Norse World: Essays in Honour
of Margaret Clunies Ross. Ed. Judy Quinn, Kate Heslop, and Tarrin Wills. Turnhout: Brepols,
2007. Pp. 2747.

33
 Whirls, horses and ships: Towards an interpretation of the
early picture stones on Gotland
Anders Andrn, Dept. of Archaeology and Classical studies,
Stockholm University, Sweden
The early picture stones on Gotland have not been discussed to the same extent as the later
ones, mainly due to the formalized and repetetive character of the images. However, in recent
years archaeological investigations have proved that some of these monuments are much ol-
der than previously assumed. This new chronological context of the early picture stones opens
up for new comparisons in time and space. From such comparisons new attempts of interpre-
ting the iconogarphy of the early picture stones will be presented in the paper.

34
 Why be afraid?
On the practical uses of legends
rmann Jakobsson, Dept. of Icelandic, University of Iceland

1. The practical side of dragon-slaying

Sigurr Ffnisbanis great achievement was slaying a dragon and as the abundance of texts
about this Germanic hero indicate, that was not a trivial feat. In the Germanic North, dragon-
slayers seem to have been in a heroic class of their own, albeit a class with only two members:
Sigurr Ffnisbani and Ragnarr lobrk. There are several texts about each, pictures as well
as narratives, but there is also the legend a different kind of text which materialises in
these medieval texts, of which Vlsunga saga and Ragnars saga lobrkar will be studied
below. The scholar who wishes to say something about the heroic dragon-slayer myth is try-
ing to interpret this text but it is not tangible. A myth does not exist on paper; thus it becomes
necessary to work from its versions in narratives such as the fornaldarsgur and use them as a
pathway to the essence of the myth.
There are two reasons for this. Myths and legends always express themselves through lan-
guage; there is no clear separation between beliefs or ideas and their linguistic expression.1
My second reason is an interest in the practical uses of myths to an imagined audience, in this
case 13th, 14th and 15th century Icelanders, the audience of Snorra-Edda, Reginsml, Ffnis-
ml, Vlsunga saga and Ragnars saga. It seems logical to approach the myth as they did,
through texts such as the fornaldarsgur.
What kind of text is a legend or a myth? There is no shortage of definitions and I will keep
myself to the functionality of myths and legends. This is the aspect of the myth lost to a mod-
ern audience that does not believe in the myth and starts out impervious to its possible ex-
planatory value; it tends not to regard stories such as Vlsunga saga as practical literature
and will miss some of its value to its audience. I speak here of functionality of the myth rather
than its meaning; there is no real need to distinguish between the two when looking at the
meaning of the myth from the perspective of an audience that wants to put it to some use. The
functionality of myths entails that a myth always exists in two ages: on one hand in the an-
cient past where it has been placed and on the other in the present, in the lives of its audience.
The myth is very distant, as deities and venerated figures have to be, and yet it exists within
ourselves and thus everywhere. Myths can be complex but their essence tends at the same
time to be very simple, even mundane. Myths are supposed to explain the world and invent a
harmony between the inner and the outer, the vast and the small, thus helping a simple hu-
man, in his smallness, to grasp a complex world. Life is not static, neither are myths. They are
narratives about movement, a quest with a clear purpose that is often absent from our every-
day lives, and where the hardships of the hero provide the myth with an intensity that may be
lacking in our daily existence.

1
I see no reason to distinguish between myths and legends in this study. Demarcation between the two is far
from clear and definitions vary. Bascom (1965) defines myths as having non-human principal characters and
belonging to the ancient past, whereas legends are closer in time and have human principal characters. This defi-
nition has been criticised by Csapo (2005, 39) who questions the need for such a clear demarcation. To clarify
my stance, I understand legend as a traditional narrative, not necessarily historically accurate (though purport-
ing to be). Myth is used mainly about cosmological narratives with an explanatory function. It could be argued
that legends serve a similar function, although less overtly. And as I understand it, both the Sigurr and the Rag-
narr legends are a part of a larger unity which is really a myth, that of the dragon-slaying youth.

35
 Myths are a paradox; on the one hand they have to be lofty and cosmological, explaining
the biggest things imaginable to men (god, the sky, time, life), and on the other hand they give
meaning to the small and insignificant private lives of ordinary people. If myths and legends
did not address the ordinariness of existence, they would lose much of their force. And this is
what one is faced with when studying Vlsunga saga and other narrative versions of the
Sigurr legend: the meaning of a extraordinary hero such as Sigurr and a huge, mythical
beast such as a dragon to the existence of, for example, poor farmers and their families in a
peaceful Icelandic countryside. As the myth is ubiquitous, it assumes that there is a Sigurr
inside every man and that the legend has a function for everyone. But Sigurr is a king as
well, and that is another important function of the legend, to sustain the charisma of ruler-
ship.2 The legendary past is always two-dimensional: it concerns both society and the life of
the individual.
The mythical hero is gone but still present, the legend is not just storytelling about the past
but also an afterlife for the hero who keeps serving his didactic function: this hero matters to
the everyday life of his audience. From the 17th century onwards, Sigurr Ffnisbani became
more and more distant, first as a figure from a very distant heroic past, someone who repre-
sented what we were instead of what we are. Then, after this heroic past had been dismantled
and was no longer considered true history,3 he became at best a part of a cultural heritage that
had stagnated and is no longer vibrant, at worst a fabrication, a myth in the negative sense of
the word which all rationalists must uphold, something untrue and consequently not very in-
teresting to the historian.
But in the Middle Ages the legendary hero is both dead and alive. The medieval view of
the past was not grounded in a firm belief in evolution or a sense of change where the past is
seen as alien to the present; thus legends could serve as examples and guidelines (see e.g.
Burke 1969, 16). Medieval men projected themselves back on to the men of the past, these
men were described as contemporaries and their ideals were those of the High Middle Ages.
In the culture of feudal society in Europe the heroes of old become medieval knights: Achilles
and Hector, Alexander the Great, Caesar, King David, King Arthur and Charlemagne. And
this is how the Sigurr legend works in 13th and 14th century Iceland; it is historical and yet
topical and timeless. Therefore it has a didactic function and is far more intimate than it later
became.
In what follows I will focus on the personal rather than the public function of the legend al-
though it probably had practical value for its West Nordic audience both as an analysis of
society and of the psychology of the individual. My main subject will be how the legend ex-
presses, but also to a degree problematises, the concepts of youth and courage, through the
figures of the hero and the dragon.

2. Killing a dragon in the North

The dragons killed by Sigurr and Ragnarr are not the only two dragons in the medieval
Norse-Icelandic textual corpus; indeed there are several serpents of various types to be found
there. And yet Ragnarr and Sigurr stood out among Northern European dragon-slayers, es-
pecially Sigurr who is the principal Germanic example of the dragon-slaying myth that Wat-
kins (1995, 297303) has located all over the Indo-European world.
The dragon provides the Sigurr legend with its core. Thus understanding the legend
means understanding the meaning of dragons to its late medieval audience. Here I will focus
on the narrative purpose and the practical function of the monster, as explained above. There
2
I have written about the myth of rulership in the medieval North at some length: rmann Jakobsson 1997, 89
154.
3
Whereas Vlsunga saga can be regarded as a part of a tradition of historical writing (see e.g. Andersson 1999).

36
is also the possibility of a religious purpose that I will not discuss any further here but which
has been discussed by other scholars (see e.g. sds Egilsdttir 1999). The draconic function I
am mostly interested in is fear. While dragons are not a part of the daily existence of most
people, fear certainly is, and I will argue that fear provides the dragon-slaying legend with a
clear purpose.
J.R.R. Tolkien (1936, p. 11) exaggerated perhaps when he said that in the North, dragons
were as rare as they are dire,4 but he is right in that the flugdrekar that Gull-rir and his
companions slay in orskfiringa saga when stealing their hoard (pp. 18588) do not seem as
terrible as the mighty Ffnir whom Sigurr kills.5 The dragon that Bjrn Htdlakappi slays
in his story hardly seems worth a mention, neither in this study nor indeed in Bjarnar saga
itself where it is referred to most perfunctorily (p. 124),6 and even though the bully orkell
hkr in Njls saga boasts of having killed another flugdreki, he is put in his place by Skarp-
hinn: killing a dragon does not seem to compare to the heroism of skating over a frozen
river to kill a chieftain in his sixties accompanied by seven men (Brennu-Njls saga, 303).7
Even though Haraldr harris mettle is put to the test when making short work of the em-
peror of Constantinoples dungeon dragon in Saxos Gesta Danorum and Morkinskinna (a
type of monster familiar not only to medievalists but to everyone who has seen The Return of
the Jedi), he emerges unchanged from the pit. His dragon is terrible but not the making of the
man (Morkinskinna, 8082; Gesta Danorum: Danmarkshistorien 2, 10). The dragon-slayings
of Sigurr Ffnisbani and Ragnarr are thus the only clear representations of the powerful
dragon-slaying myth,8 and the word dreki may not be the best guide to the draconitas of
Sigurr and Ragnarrs antagonists.9
The dragon which Bjrn Htdlakappi kills earns him no special status in the Mrassla.
On the other hand, Sigurr Ffnisbani and Ragnarr lobrk became the most celebrated heroic
figures of the medieval North. Ragnarr pales by comparison to Sigurr yet his dragon-slaying
is not only the subject matter of Ragnars saga lobrkar but is also referred to in several
other Old Norse texts: Hervarar saga ok Heireks, Bsa saga ok Herraus, Hlfdanar saga
Eysteinssonar and Norna-Gests ttr. He also figures in Saxos Gesta Danorum and in
Hauksbk.10
Sigurr is even more distinguished and makes an appearance all over the Germanic world:
in Nibelungenlied, in Beowulf, in images carved on Swedish runestones, and in Old Norse
texts, including Snorra-Edda, ireks saga and Vlsunga saga. His story is worthy of being
retold at length alongside the mythical narratives of the Codex Regius of the Elder Edda, such
as Vlusp, Hvaml and Vafrnisml, and he even makes it into the late 14th century or-
steins ttr skelks in Flateyjarbk as a prime example of a heroic heathen, before becoming
the hero of several post-mediaeval ballads in various parts of Scandinavia.11

4
On this exaggeration, see Evans 2005, see esp. pp. 21821 and pp. 24148.
5
These dragons are so large they can carry a man in their jaw, they fly, spew fire and poison, and yet the narra-
tive is devoid of any sense of wonder or danger, which may be regarded as typical of this saga type; see Svborg
2009. At the end of the saga, it is suggested in an equally offhand fashion that rir himself may have changed
into a dragon instead of dying (p. 226).
6
After Bjrn has killed the dragon, it is never referred to again.
7
According to the saga, orkell has also fought a finnglkn (a chimera).
8
Tolkien also included the Beowulf dragon but, for some reason, not Ragnarrs dragon (p. 11). Since space is
limited, the dragon in Beowulf will not be dealt with here. At first glance this narrative might seem to provide a
useful counter-example to my main argument here, since Beowulf does not encounter the dragon in his youth.
On the other hand, he does not survive the battle either.
9
The word is not Germanic but Greek (see e.g. Evans 2005, 217), and neither is there a clear separation between
the Germanic dragon and its Indo-European counterparts (Evans, 22130).
10
On the origins and the popularity of the Ragnarr legend, see McTurk 1991, esp. pp. 5362.
11
See e.g. Rowe 2006, who provides a good review of the diversity of how Sigurr functions in the texts.

37
 orsteins ttr skelks is preserved in Flateyjarbk which presents its audience with a rig-
orously Augustinian world view in which the heathen past is outlawed (Rowe 2005, 6597).
But who is Sigurr? Why is his legend so popular and why does the late 14th century editorial
team of Flateyjarbk care about this prehistoric heroic figure? As outlined in Ragnars saga,
Sigurr acquires some significance as the mythical ancestor of the perhaps equally mythical
King Harald Fairhair of Norway and other Northern kings.12 However, that is hardly enough
to explain his elevated status in the culture of the medieval North. It seems more likely that
his importance lies in the dragon-slaying itself, myth rather than history, a feat which also
manages to elevate Ragnarr lobrk over most other prehistoric viking kings.

3. Youth and the dragon-slayer

Even though Ragnarr lobrk is actually Sigurr Ffnisbanis son-in-law in the preserved
Ragnars saga lobrkar, the two heroes are quite dissimilar. There are also significant differ-
ences in the most detailed narratives of the two killings. With Ragnarr, the emphasis is on his
ingenuity and on the hairy breeches which he uses to escape the poison of the worm and
which provide him with a lasting identity. In the Sigurr narrative, the emphasis is on his de-
sire for revenge and the influence from his fosterfather Reginn. Still, there are shared ele-
ments. We find evidence for this in the fornaldarsgur variations of the myth, in Vlsunga
saga and Ragnars saga, presumably composed in the 13th or early 14th century but preserved
together in the early 15th century manuscript NKS 1824 b 4to.13 I use these texts as represen-
tative for the myth in this paper, not because they are the oldest or the most original variant
but they do demonstrate a possible function of the myth for a late medieval audience who
encountered it through these texts.
The first important common denominator, emphasised in both sagas, is that the dragon-
slayers are youths. When Reginn first presents Sigurr with the task of killing a dragon,
Sigurr remarks that he is still little more than a child (vr erum enn ltt af barns aldri) (p.
33), and it is only a short while later that he avenges his father before going to face the
dragon. His youth is also made clear in the ensuing conversation between the dragon and his
slayer. Ffnir calls him sveinn and keeps asking about his father (p. 42).14 In Ragnars saga,
Ragnarr claims to be 15 years of age when he kills the dragon and the earls daughter he has
liberated finds him more like an ogre than a man of such a young age: ykkisk hn eigi vita
hvrt hann er mennskr mar ea eigi, fyrir v at henni ykkir vxtr hans vera sv mikill sem
sagt er fr vttum eim aldri sem hann hafi (p. 119). It is an important factor in both
stories, in their fornaldarsgur form, that the hero is young, still a teenager, a man between
childhood and adulthood.15
The youth of the hero means that the climactic event of the heros life is placed early in the
narrative. What happens after the dragon-slaying is often a long decline. Ragnarr survives but
relinquishes the place of honour in his story to his wife and sons after having killed the
dragon. Sigurr makes a mess out of his life, gets entangled with two different women, one
too many, and ends up getting killed by his in-laws. But why must the dragon-slayer be a
youth? To address this, we have to go to the second common denominator of the Sigurr and
Ragnarr narratives: the bravery necessary to confront the worm.
12
As evidenced by the manuscript AM 415 4to from the early 14th century where Ragnarr is the purported ances-
tor of the kings of Denmark, Norway and Sweden (on this manuscript, see Sverrir Jakobsson 2005, 50).
13
This manuscript is used in the edition of Olsen (Vlsunga saga ok Ragnar saga lobrkar); all references to
both Vlsunga saga and Ragnars saga are to this edition but I have normalised the spelling.
14
This is even more evident in the Ffnisml version of their conversation, see Norrn fornkvi, 21926, esp.
stanzas 18 and 1213.
15
This makes perfect sense if the myth is seen as an initiation ritual (see e.g. Eliade 1974, 1718) but that is not
the aspect which interests me here.

38
 4. Fear is the key
In Ragnars saga, when the worm that dwells on ras casket starts to grow, people start to
become terrified of it: orir engi mar at koma til skemmunnar fyrir essum ormi (p. 117),
making her a virtual recluse. Fear also follows Ffnir the dragon; after he is first mentioned
Sigurr remarks: Kann ek kyn essa omrs tt vr sim ungir ok hefi ek spurt at engi orir at
koma mt honum fyrir vaxtar sakar ok illsku (p. 33). When Reginn has extracted from
Sigurr a promise to kill the dragon, he keeps expressing doubts that Sigurr will make good
on his promise and when they have come to the heath where Ffnir dwells, he starts goading
his young apprentice again: Eigi m r r ra er ert vi hvatvetna hrddr (p. 41). The
text is full of talk of fear and not by chance; what the two dragons have in common is the fear
they inspire in others.
Getting back to a possible symbolic role for the dragon in the myth, both dragons may be
said to represent, even embody, terror, and in Vlsunga saga this terror is objectified in the
Helmet of Fear (the gishjlmr) that Ffnir possesses. As the dragon remarks in Vlsunga
saga: Hafir eigi frtt at hversu allt flk er hrtt vi mik ok vi minn gishjlm? (p.
42). The dragon seems almost vexed that the young hero is not suitably scared by him, but the
Helmet of Fear has to be taken seriously. If this part of Vlsunga saga is compared to Ffnis-
ml, one notes a change from a Helmet of Fear which might be a metaphor or an expression16
to an actual concrete helmet that Sigurr can carry away with him, along with a golden byrnie
and the sword Hrotti (p. 47). What does not change is the symbolic meaning of the helmet.
The dragon has a Helmet of Fear because it is terror itself.
Tolkien believed that the Beowulf poet did not like dragons as a sober zoologist (1936,
11), indirectly warning against regarding a dragon as a mere beast. It is, of course, a hybrid of
several actual animals, with wings and its scales, its claws and its serpent-like length, but
there is also the terrible fire that it breathes (in the preserved Vlsunga saga the emphasis is
more on its venom) which is not taken from the animal kingdom but from the human mind,
from our fear of the destructive power of fire. As Vlsunga saga indicates a dragon is both
poisonous and has magical powers, two attributes greatly feared in the Middle Ages (pp. 41
44). It is no accident that fear is referred to in both narratives, Vlsunga saga and Ragnars
saga, right before the young hero accomplishes his feat, and that a dragon should be in the
possession of a Helmet of Fear that causes all to cower.
In his pivotal study of North European dragons, Jonathan Evans sees the main mythic
function of dragons as being metaphors of avarice (2005, 26169). It is true that in both these
legends (Sigurr and Ragnarr) there is a clear connection between dragons and gold and thus
with greed, both the dragons own and that of others. Although it can hardly be ascertained
what the most important mythic function of a mythic narrative might be or whether its func-
tionality changed through the ages, one can at least say that in the late medieval variant in
Vlsunga saga and Ragnars saga fear seems to be accentuated above everything else in the
dragon-slaying narratives and greed is hardly mentioned in connection with the two heroes.
Although desire for gold may be a motivation for Reginn, desire for vengeance is more
prominently voiced and gold seems to provide no motivation for Sigurr Ffnisbani, even
hough he takes Ffnirs treasure when he sees it. It is quite unclear what possesses Ragnarr to
fight his dragon; although he ends up in deep mourning for ra, he has never seen her before
the fight, and it seems more logical that his motivation is heroism for its own sake, since the
key fact in the narrative preceding his killing seems to be how terrifying the worm is and how
nobody dares to approach it.

16
Bera gishjlm is indeed an expession in Icelandic (see Jn Frijnsson 1993, 736) but it is hard to say
which comes first: the metaphor or an actual helmet. The present authors money would be on the expression.

39
 Although a dragon can be both an embodiment of its own savage greed and the fear of oth-
ers, both Ragnars saga and Vlsunga saga indicate that for the youthful hero, the first is not
very important but the second all-important. When Sigurr has killed the dragon, Vlsunga
saga describes him with loving attention to detail, his armour and his weapons, his gracious
manners, his chestnut hair and curls, his sharp eyes and his powerful shoulders. And it ends
with this statement: Eigi skorti hann hug ok aldri var hann hrddr (p. 57). If we see the
dragon as an embodiment of terror, it is clear that this is why young Sigurr defeats it. For the
fearless youth, fear does not exist and thus it can be vanquished. In this myth, overcoming the
fear of the dragon means its automatic destruction.
It is fitting that Sigurr should later make an appearance in the orsteins ttr skelks, a late
14th century adaption of the folktale The Boy Who Knew No Fear (AT 326). This tale is a
reminder that there are actually two kinds of fearlessness: that one which is a handicap, a de-
fect in a young man too simple to know fear, too limited to understand what it is (Aarne/
Thompson 1961, 11415). 17 This is not how Sigurrs lack of fear is defined; his bravery
makes him more rather than less of a man.

5. Youth, bravery and dragons

One might not expect to find a place in the lives of ordinary people for dragon-slaying, but
fear provides that place. Although Sigurr is exceptional, his courage is something that every-
one in the audience can relate to, since the audience is composed of people who have known
fear and had to rely on bravery, though it also seems likely that their relationship with it var-
ied quite a bit.18
On a personal level, this legend also concerns the ages of man, an important medieval
theme (see esp. Burrows 1986). There are all kinds of fear, and indeed many kinds of bravery:
existential, moral and physical. The fear of the dragon can be characterised as a strong physi-
cal fear. In fact, the dragon is intensely physical, savage and bestial and its threat is of death
itself: instant, brutal and sudden. It is thus logical that the man who may defeat a dragon
should be far removed from death and full of vitality and zest, the life-force that some call
Eros.19 In fact, the perfect person to conquer this image of death is a youth, a teenager like our
heroes.
Sigurr somewhat insolently says to Ffnir as the latter lies dying: Frr er gamall harr, ef
hann er bernsku blautr (p. 43). The youth nonchalantly regards courage as his own property
and the disregard for physical fear is indeed a well-known characteristic of youth or at least
the myth of youth. Youths may ignore consequences, scorn danger and brave death in various
ways. This is all part of an erotic existence: being far removed from death, it poses no danger.
Youths often possess great physical courage but are on the other hand given to social fears:
being unpopular among peers, talking to strangers at parties, being uncool, being the object of
scorn. Killing a dragon seems easy by comparison.
In feudal society such tempestuous youths formed a social group; Georges Duby has spo-
ken of bands of aristocratic youths in 12th century France who formed the cutting edge of
feudal aggressiveness (1968, 200), and from 13th century Iceland we have the example of the
youthful band of the orvaldssynir of Vatnsfjrr who go to conquer their own dragon, the

17
On the history of this folktale in Iceland see Lindow 1978.
18
On the cultural importance and the representations of fear in the Middle Ages, see e.g. Dinzelbacher 1996.
19
In Freudian psychology (from Freuds Beyond the Pleasure Principle (1920) onwards), it is customary to
acknowledge two opposing forces; the life force (Eros) and the death force (Thanatos), although Freud himself
did not use these concepts. See esp. Marcuse 1972, 3554. As the death force involves repetition and conserva-
tive behaviour, it makes sense to see the aggressiveness of youth, including the fearlessness and courtship of
death, rather as a part of its erotic energy.

40
mighty chieftain Sturla Sighvatsson, but whose spirited attack fails in becoming heroic, since
all they encounter are women and unarmed men whom they kill and wound instead with all
the frenzy that might have come in handy against a dragon (rmann Jakobsson 2003).
Youth and fear go hand in hand in the dragon-slaying legends of Sigurr and Ragnarr. In
both instances, the dragon-slaying takes place in the heros youth and is the climax of the
heros life. The fortitude the hero needs is the fortitude of youth, that zest for life and belief in
ones invincibility that leads to disregard for death and fearlessness in face of physical danger,
and in both fornaldarsgur youth and fearlessness are the heros main attributes. As previ-
ously related, neither Sigurr Ffnisbani nor Ragnarr lobrk do so well after their dragon-
slaying. The sagas version of the myth seem to reflect a youthful point of view: killing drag-
ons is something one can accomplish but relationships with in-laws are complicated and
messy and beyond ones skills.
Sigurr is partly superhuman, descended from inn and glamourous beyond everything
the audience of Vlsunga saga is likely to have experienced. And yet he is also everyman at
a certain age, ruled by his lust for life and contempt for death. His fearlessness may be lofty
but it also something all may experience. The legend is not about something else; it is about
us.

Bibliography

Aarne, Antti and Stith Thompson 1961: The Types of the Folktale: A Classification and Bibli-
ography. Helsinki.
Andersson, Theodore M. 1999: Goafri ea sagnfri?: Dmi Vlsunga sgu. Heiin
minni: Greinar um fornar bkmenntir. Ed. Haraldur Bessason and Baldur Hafsta. Reyk-
javk, 91101.
rmann Jakobsson 1997: leit a konungi: Konungsmynd slenskra konungasagna. Reyk-
javk.
rmann Jakobsson 2003: Snorri and His Death: Youth, Violence, and Autobiography in Me-
dieval Iceland. Scandinavian Studies 75, 31740.
sds Egilsdttir 1999: Drekar, slngur og heilg Margrt. Heiin minni: Greinar um fornar
bkmenntir. Ed. Haraldur Bessason and Baldur Hafsta. Reykjavk, 24156.
Bascom, William 1965: The Forms of Folklore: Prose Narratives. Journal of American Folk-
lore 78, 320.
Bjarnar saga Htdlakappa 1938: slenzk fornrit 3. Ed. Sigurur Nordal and Guni Jnsson.
Reykjavk.
Brennu-Njls saga 1954: slenzk fornrit 12. Ed. Einar lafur Sveinsson. Reykjavk.
Burke, Peter 1969: The Renaissance Sense of the Past. London.
Burrows, J.A. 1986: The Ages of Man: A Study in Medieval Writing and Thought. Oxford.
Csapo, Eric 2005: Theories of Mythology. Oxford.
Dinzelbacher, Peter 1996: Angst im Mittelalter: Teufels-, Todes- und Gotteserfahrung, Menta-
littsgeschichte und Ikonographie. Paderborn.
Duby, Georges 1968 [1964]: In Northwestern France: The Youth in Twelfth-Century Aris-
tocratic Society. Lordship and Community in Medieval Europe: Selected Readings. Ed.
Fredric L. Cheyette. New York, 198209.
Eliade, Mircea 1974 [1954]: The Myth of the Eternal Return or, Cosmos and History. Transl.
by Willard R. Trask. Princeton.

41
Evans, Jonathan 2005: As Rare As They are Dire: Old Norse Dragons, Beowulf, and the
Deutsche Mythologie. The Shadow-Walkers: Jacob Grimms Mythology of the Monstrous.
Ed. Tom Shippey. Tempe, Ariz., 20769.
Gesta Danorum: Danmarkshistorien 2005: Ed. Karsten Friis Jensen and Peter Zeeberg. Co-
penhagen.
Jn G. Frijnsson 1993: Mergur mlsins: slensk oratiltki, uppruni, saga og notkun.
Reykjavk.
Lindow, John 1978: Hreiars ttr heimska and AT 326: An Old Icelandic Novella and an
International Folktale. Arv 34, 15279.
Marcuse, Herbert 1972: Eros and Civilization: A Philosophical Inquiry into Freud with a new
preface by the author. London.
McTurk, Rory 1991: Studies in Ragnars saga lobrkar and its Major Scandinavian Ana-
logues. Oxford.
Morkinskinna 1932: Samfund til udgivelse af gammel nordisk litteratur 53. Ed. Finnur
Jnsson. Copenhagen.
Norrn fornkvi 1965 [1867]: Ed. Sophus Bugge. Oslo.
Rowe, Elizabeth Ashman 2005: The Development of Flateyjarbk: Iceland and the Norwe-
gian Dynastic Crisis of 1389. Odense.
Rowe, Elizabeth Ashman 2006: Qvid Sigvardus cum Christo: Moral Interpretations of Sigurr
Ffnisbani in Old Norse Literature. Viking and Medieval Scandinavia 2, 167200.
Svborg, Daniel 2009: Avstnd, grns och frundran: Mten med de vernaturliga i isln-
ningasagan. Greppaminni: Rit til heiurs Vsteini lasyni sjtugum. Reykjavk, 32349.
Sverrir Jakobsson 2005: Vi og verldin: Heimsmynd slendinga 11001400. Reykjavk.
Tolkien, J.R.R. 1936 [repr. 1977]: Beowulf: The Monsters and the Critics. London.
Vlsunga saga ok Ragnar saga lobrkar 1908: Samfund til udgivelse af gammel nordisk
litteratur 36. Ed. Magnus Olsen. Copenhagen.
Watkins, Calvert 1995: How to Kill a Dragon: Aspects of Indo-European Poetics. Oxford.
orskfiringa saga 1991: slenzk fornrit 13. Ed. rhallur Vilmundarson and Bjarni
Vilhjlmsson. Reykjavk.

42
 Sigurr Ffnisbani as Nineteenth-Century Man.
David Ashurst, Dept. of English Studies, Durham University, England
The figure of Sigurr Ffnisbani takes a central position in the works of Richard Wagner and
William Morris, two revolutionary writers who themselves stand at the centre of the nine-
teenth century. Both these men looked back on the Middle Ages in order to comment on their
own times and to look forward into the future that they hoped would soon come about.
This paper examines the medieval sources used by Wagner and Morris and asks what they
found there that made the figure of Sigurr resonate for them as a contemporary man. It
shows how eddic and saga accounts of Sigurr feed into nineteenth-century debates on power
and anarchism, the nature of masculinity, concepts of heredity, and finally, above all, the loss
of certitude in ethics and epistemology.

43
 upp ek r verp ok austrvega:
death overseas and the dead in the east
Hugh Atkinson, Department of Scandinavian Studies, University College London, England
In this paper I will explore the idea, once raised by Gabriel Turville-Petre, that the complex of
beings made up by jtnar, ursar and trll, as they are described in Old Icelandic mythologi-
cal texts, are the devouring demons of death [and] may even be the dead. This hypothesis is
tenable inasmuch as the jtnar, the forefathers of the sir, seem to have occupied a position
within the Norse cosmos in relation to the sir commeasurable with that of the ancestral dead
in relation to the living. A corollary of this is that the sphere of the cosmos occupied by the
jtnar may be reckoned a realm of death and of the dead, one of several accommodated
within Norse cosmography.
Taking the poem Lokasenna as my point of departure, I will first test this hypothesis in the
light of ideas expressed in certain mythological texts (Snorra Edda, rymskvia, Eirksml).
Drawing on direct and indirect cosmographical statements in these sources, a structural analy-
sis suggests that the identification is valid. Applying an anthropological perspective, I will ask
whether the attitude of the sir towards the often malevolent, incursive jtnar is congruous
with attitudes of the living towards the unquiet ancestral dead, attitudes suggested both by
texts (saga accounts of revenants) and by archaeological finds (burial practices).
Snorri located the jtnar and their ilk in the east (rr var farinn austrvega at berja trll).
I will ask whether there was an element of recursivity in medieval ideas of the lands to the
east of Scandinavia, whether they may have derived in part from cosmographic conceptions;
did Norse cosmography and Norse geography accord on this point? Here I draw evidence
from runic inscriptions and picture stones with a provenance in coastal communities of east-
ern Scandinavia, for whom the death of loved ones in the lands on the far shore of the Baltic
was a commonplace. In their religious response to this everyday experience of death, contem-
plated against the background of pan-Scandinavian(?) cosmographical conceptions, these
communities may have mythicised the East as a place where the dead took up permanent resi-
dence, and the austrvegar as something akin to Helvegr.
In exploring these themes I will problematize the validity of comparing the statements and
images on eastern Scandinavian commemorative stones raised in the late heathen era with the
perhaps over-rationalised accounts of cosmography found in western Scandinavian manu-
scripts of the Christian era.

44
 Vibtur Hauks Erlendssonar ea hefbundi
Landnmuefni
Auur Ingvarsdttir, Reykjavk , Iceland

Inngangur
Sustu r hefur mnnum veri tamast a lta Landnmu sem stutta og sgulega. Hr hafi
veri um a ra nytsama skr sem san hafi smtt og smtt fengi sig annars konar svip
vegna dugnaar eirra Sturlu, Styrmis, Hauks og jafnvel Melamannsins svokallaa v allir
eiga eir a hafa skoti inn Landnmugerir snar sgulegu efni. Hr er a sjlfsgu um a
ra tilgtu manna um kvena run, en ekki heilagan sannleik, stundum bregist menn
vi eins og um umbreytanlegar stareyndir s a ra. r mialdagerir af Landnmu sem
varveittar eru .e. Sturlubk, Hauksbk og Melabk bera ess allar merki a vera sgulega
uppbyggar. S tlkun a textinn hafi rast r stuttum skrarkenndum texta sgulegar
frsagnir styst v ekki vi raunverulegar gerir. Hr er um a ra tilgtu um nytsama skr
12. aldar manna. Einni ger, Melabk sem eingngu er varveitt brotum, hefur veri
hampa af frimnnum sem besta fulltra upprunalegustu gerar Landnmu. Hn er
vissulega styttri en hinar gerirnar og me ltilshttar annars konar efnisrun. Melabk hefur
v a einhverju leyti falli betur a hugmyndum manna um elstu ger. Jn Jhannesson var
hrifarkur frimaur svii Landnmurannskna. Niurstaa hans um hina tndu
Styrmisbk sem forrit allra varveittra gera hefur last tluvera hylli. Forsendan sem hann
gekk t fr var jafnframt s a Melabk vri s ger sem nst sti upprunagerinni. S
hugmynd hefur veri rkjandi fr sari hluta 19. aldar en hefur veri andmlt m.a af hfundi
essarar greinar (Auur Ingvarsdttir 2004:91118). Menn vitna enn til ess sem stareyndar
a Haukur hafi btt vi texta sinn fr ekktum heimildum, munnmlum og tndum sgum.
Fyrir nokkrum rum tk frimaur dmi af essu vinnulagi Hauks til ess a rkstyja a
munnmli fr landnmsld hefu veri lifandi a minnsta kosti fram yfir 1300. Hann vsar
essa hugmynd um verklag Hauks Erlendssonar eftirfarandi htt: tekur inn ntt efni sna
ger Landnmu sem tla m a hann hafi ekki geta stt sr til ritara bka. (Gsli
Sigursson 2002:58).
Haukur Erlendsson gefur greinargar upplsingar um forrit sitt og segist notast vi tvr
Landnmur .e. eftir Sturlu rarson og hins vegar eftir Styrmi hinn fra. N er
Sturlubk varveitt afriti og Hauksbk a hluta til eiginhandriti Hauks sjlfs, s texti sem
Haukur hefur fram yfir Sturlubk tti samkvmt hans eigin orum a vera fr Styrmsbk
kominn. Vegna eirrar upprunakenningar sem veri hefur vi li hafa menn efast um or
Hauks. sta ess a telja srtexta Hauks kominn fr Styrmisbkarforriti hans er gert r fyrir
msum heimildum ekktum og ekktum. t fr essari hugmynd hefur veri fullyrt: a
miklu skiptir um heimildargildi Landnmutexta hvort hann er fr Styrmisbk runninn ea er
vibtur Sturlu og Hauks. (Jakob Bendiktsson 1968: 53) Hr vera svokallair viaukar
Hauks um orstein tjaldsting skoair. Getum vi lykta hvort eir eru r Styrmisbk ea
gltuum tti?

Styrmisbkartexti Hauks.
Er hgt a gera sr einhverja hugmynd um hina gltuu Styrmisbk? Er ekki sennilegt
megni af umframtexta Hauks fram yfir Sturlubk s fr Styrmi komi? S skilningur er
fyllilega samrmi vi or Hauks Erlendssonar um verklag sitt, hafa ek at r hvorri er

45
framar greindi (H.3541). Eru einhver rk fyrir v a draga or Hauks sjlfs efa? snum
tma taldi Landnmutgefandinn Finnur Jnsson vafalaust a de vrige plus-stykker hos
Haukr m med strst sandsynlighed henfres til Styrme. (1900:11). Jn Jhannesson taldi
Hauk hafa notast vi fjlda annarra heimilda og fullyrti a Haukur hefi: va fari eftir
rum heimildum en bkum Sturlu og Styrmis, munnlegum ea rituum. (Jn Jhannesson
1941:175). N er Styrmisbk gjrsamlega gltu og v verur a telja etta heldur hpna
fullyringu hj Jni. essi skoun hefur fengi byr frunum (Jakob Benediktsson
1968:52). lit margra frimanna verklagi Hauks er tilkomi vegna ess a Melabk er
metin of mikils. Jn Jhannesson jafnar hinni gltuu Styrmisbk t.d. vi Melabk, segir a
hn veri a teljast bezti fulltri Styrmisbkar, sem n er vl . (1941:174). etta vihorf
hans kemur hva eftir anna fram t.d. Melabk var a mestu eftirrit Styrmisbkar[]
(1941:225). ar sem a Melabk var hans huga besti fulltri Styrmisbkar er srtexti Hauks
metinn eftir v hvort sj m merki um hann Melabk ea ekki. Haukur Erlendsson er svo
talinn hafa breytt forriti snu eftir fjlmrgum heimildum og a fullyrt a frsagnir r, sem
eru eftir hann sjlfan, en ekki teknar upp r rum ritum, eru ekki vel til ess fallnar, a vekja
traust honum sem heimildarmanni. (Jn Jhannesson 1941:207). Hvernig er rskura um
upprunalegan texta forrita Hauks? Hr eru a vslega brotin af Melabk sem gefa
upprunalegan texta forrits Hauksbkar til kynna. Ekki er hgt skv. Jni Jhannessyni a
greina srstakan Styrmisbkartexta hj Hauki nema ar sem brotin af Mb. eru til
samanburar. (1941:175). Melabkartextinn hefur annig rskurarvald hvaan texti
Hauks er ttaur. essi falska vitneskja um Styrmisbk hefur leitt af sr mis konar
vandri. Ekki er hgt a fallast tengingu a Melabkartextinn s nr skeikull
vitnisburur um Styrmisbkartextann. essar getgtur um verklag Hauks eru heldur ekki
samrmi vi a sem vi sjum texta eim sem hann hefur eftir ru forriti snu Sturlubk.
a er slandi hve mikinn hluta af Landnmutextanum hann hefur teki breyttan ea lti
breyttan r Sturlubkarforriti snu eins og hgt er a sannfrast um me samanburi vi
Sturlubkarhandriti sem varveitt er afskrift fr 17. ld (Jakob Benediktsson 1968:51).

Hvaan er orsteinn tjaldstingur upprunninn?

a er upplsandi a lta sgulega frsgn af landnmsmanninum orsteini tjaldstingi til
ess a greina hvernig Jn Jhannesson reiknar t viauka Hauks Erlendssonar. orsteinn
tjaldstingur var landnmsmaur Rangringi og segir af honum llum varveittum
Landnmugerum (H.314; S.356358,og Sk.356 bls.169171). enn einum sta er frsgn
af orsteini tjaldstingi, en a er Flateyjarbk. ar er frsagnarttur um hann eim
hluta Flateyjarbkar sem telst til bkarauka fr lokum 15. aldar (Jonna Louis-Jensen
1969:235250). essi bkarauki fjallar um konungana Magns og Harald harra og svo er
ar a finna nokkra tti, sem eiga a sameiginlegt a fjalla um samskipti manna vi
konunga, ar er m.a. tturinn um orstein tjaldsting (Flateyjarbk IV:183230). N er a
ljst a um rittengsl er a ra milli Landnmu og ttarins eins og auvelt er a sannfrast
um me samanburi. Fyrir utan nstum samhlja upphafskafla og landnmslsingar, er a
finna ll aalatriin bum gerum2 Landnmu og ttinum. Eftirfararandi atria er geti a)
skattheimtu Haralds hrfagra, b) sendiferar rorms, c) drps sgrms, d) hernaar
orsteins, e) furhefnda, f) farar eirra brra og mursystur eirra til slands, g) landnms
orsteins ar; h) hjlpsemi orsteins vi sttveika skipverja sem komu Rangrs, i) fgraftar
ess skipverja sem lengst lifi (H.:358). Hr m sj ll skilyri ess a um bein rittengsl s a
ra. snum tma rskurai Finnur Jnsson a a frsgn Hauksbkar vri greinilega
tdrttur r tti sem vri eldforn (1927:188). ar tti hann a sjlfsgu vi glataan X-tt
1
Hr er vsa til kafla Landnmabk,og stafsetningin samrmd.
2
Melabk er hr aeins varveitt afbrigum rarbkar.

46
sem einnig hefi veri stust vi Flateyjarbk og ar vri, fremstillingen sikkert nrmere
originalen (Finnur Jnsson 1927:188). essi skilningur tndri og fornri frumsgu bak vi
sguna var algengt vihorf manna en hefur stt gagnrni sari rum. a er essi horfni X-
ttur bak vi orsteins tt tjaldsting sem g vil hr gera athugasemd vi. Er sta til
ess a gera r fyrir essum horfna tti fr v um 1200 til ess a skra r frsagnir sem
varveittar eru dag? rauninni jnar essi tndi ttur aeins eim tilgangi a a skra t
tlaa notkun Sturlu, (og Styrmis) og Hauks heimildum. Ef vinnubrg Hauks og Sturlu
hafa ekki veri lund sem tast er a halda fram, blasir vi allt nnur mynd af ritinu
Landnmu.
eir frimenn sem tldust til hins svokallaa slenska skla veltu iulega fyrir sr
notkun Landnmu egar eir af nkvmni tiltku heimildir sagnanna. ar virist s
umalfingursregla vera gildi a telja beinar frsagnir af landnmum eiga uppruna r
Landnmu (Einar l. Sveinsson 1934:3940). Aftur mti voru sgulegar frsagnir sem
krkkt er af llum varveittum gerum Landnmu iulega afgreiddar sem innskot og
viaukar fr ekktum og ekktum heimildum. essi skipting hefbundi Landnmuefni
og sgulegar vibtur veldur neitanlega msum erfileikum. Jni Jhannessyni fannst
nausynlegt a gera r fyrir v a s sem setti saman ttinn um orstein tjaldsting hefi
notast vi forna Landnmu og innskot r essari fornu Landnmu s a finna ttinum
(1941:199). Aftur mti var a andsttt hugmyndum manna um tilur Landnmu og run
a telja mgulegt a hin sgulega frsgn hafi veri upprunaleg Landnmu.
Vissulega vri hugsanlegt a frsgnin ll vri komin r essum sgulega tti upphafi,
og tind inn Landnmu vegna ess a ar vri fjalla um landnmsmenn. Helst virist
tgefandi ttarins rhallur Vilmundarson hallast skoun, a.m.k telur hann lklegast a
tdrttur r ttinum hafi veri llum varveittum gerum Landnmu (1991:192). verur
a gera ttinn bsna gamlan v rhallur fylgir Jni Jhannessyni a mlum um samhengi
geranna og sameiginlegt forrit Melabkar og Sturlubkar. Ea eru bara sgulegar frsagnir
sem koma ekki heim vi tla form Landnmu innskot? Er landnmsfrsgnin fr fornri
Landnmu? En hin sgulega atburarrs fr skrifari sgu? verur hfundur ttarins a
hafa notast vi essa fornu Landnmu a.m.k. til ess a f upplsingar um landnmsmenn.
Sgulegar frsagnir jafnvel a r komi heim vi varveittar Landnmugerir urfa
tskringa vi enda ekki hefbundi landnmsefni a dmi frimanna. etta vihorf til
efnis Landnmu kemur va fram, til dmis er essu haldi fram nlegu riti: Efni
lfljtslaga fellur a.m.k. ekki a hefbundnu Landnmuefni. (Sveinbjrn Rafnsson
2001:168).

Frsagnir af landnmsmnnum ttinum og Landnmu

a hefur veri rkjandi skoun meal frimanna a gera r fyrir v a efni Landnmu
hafi veri mjg markvisst og einskora. Efni sem ekki kemur heim vi essa hugmynd er
gjarnan rskura sem innskot r lkum ttum. Landnmsfrsagnirnar eru efni sem er
vikennt sem dmigert landnmsefni, .e. eru frsagnir af landnmsmanni, landnmi hans og
afkomendum. tturinn um orstein tjaldsting Flateyjarbk hefur einmitt ess httar efni.
Notaist hfundur ttarins vi forna Lannmu?
eim sgum sem fjalla um landnm velta tgefendur fyrir sr notkun Landmu. N er
a svo a ttinum margumtalaa er minnst landnm tveggja manna, eirra Flosa
orbjarnarsonar sem er titlaur landnmsmaur og svo hins vegar orsteins tjaldstings.
ess konar efni hefur venjulega mefrum frimanna veri viurkennt sem hefbundi
landnmsefni.

47
 Hann nam land at ri Flosa fyrir ofan Vkingalk ok t til mts vi Svnhaga; bj Skari inu
eystra (orsteins ttur: 430)

orsteinn nam land at ri Flosa, er numit hafi r Rangrvllu, fyrir ofan Vkingslk til mts
vi Svnhaga-Bjrn [ok] bj Skarinu eystra (S.:358)

nam land at ri Flosa fyrir ofan Vkingslk ok t til mts vi Svnhaga-Bjrn ok bj Skari
enu eystra (H.:314)

Hr eru enn meiri lkindi me Hauksbk, og lsingunni ttinum. Notai Haukur hr enn
ttinn til ess a f aeins annars konar oralag? Ekki er a lklegt. yri samt sem ur
a gera r fyrir v a s sem setti saman ttinn hefi fengi efni einhvers staar fr v
greinileg rittengsl eru lka vi Sturlubk. Jn Jhannesson velur einmitt lei a telja a
hafi hfundur ttarins nota forna Landnmu. a verur hann a gera til ess a skra t
lkindin vi Sturlubk (Jn Jhannesson 1941:199).

er framar greindi
Haukur notaist vi tv forrit af Landnmu, .e. Styrmisbk og Sturlubk sem a orra til var
eins a hans sgn en fr eftir eirri er framar greindi eins og alkunnugt er (H.:354) Haukur
hefur kaflanum um orstein tjaldsting umtalsveran textaauka fram yfir Sturlubk sem a
mestu kemur heim vi ttinn Flateyjarbk. Eftirfarandi atrii er ar a finna: a) tilraun til
burar, b) hvtvoungur fer me vsu, c) mir orsteins er nefnd, d) sgrmur sendi
konungi gjf hest gauskan ok silfr mikit. (H.314). Smu atrii er einnig a finna ttinum,
m.a. er tala um gjfina til konungs ennan htt: hestr gauzkr ok ar me mikit silfr.
(orsteins ttr:427). essi lkindi og greinilegu rittengsl sanna ekki a Haukur hafi notast
sjlfsttt vi tt um orstein tjaldsting. Frsgnin er meginatrium lka Sturlubk eins
og fyrr segir. a hltur a teljast bsna mikil tilviljun ef Haukur, og Sturla og hfundur
ttarins og jafnvel Melamaurinn hafi allir notast vi ttinn sjlfsttt. Er ekki elilegast a
lta svo a arna hafi Styrmisbk haft tarlegri texta en Sturlubk? Kemur a ekki heim vi
or Hauks sjlfs um vinnulag sitt a hann hafi einmitt notast vi fyllri textann essari
frsgn af orsteini? v er ekki a heilsa, a mati Jns Jhannessonar, en hann er ess
fullviss a Sturla hafi arna upprunalegri texta; er engum blum um a a fletta, a texti
Stb3. er frumlegri, tt hann hafi sennilega ori fyrir smbreytingum. (Jn Jhannesson
1941:199). Af hverju fllst Jn ekki skiljanlegu lyktun a textaauki Hauks vri fr
Styrmi kominn? Kenningu Jns Jhannessonar verur ekki ger skil essari grein. Hr eins
og var miar hann vi texta Melabkar og gerir hann hika a besta fulltra X-
Landnmunnar svoklluu sem hann taldi vera sameiginlegt forrit Sturlubkar og Melabkar
(Styrmisbk). v er nefnilega annig vari a sj m a Melabk hefur haft texta um
orstein tjaldsting sem kemur betur heim vi Sturlubk. lyktun Jns verur v ann
veg a Haukur geti ekki hafa haft textaaukann fr Styrmi. etta kemur fram ar sem hann
bendir a sendifera rorms hafi veri geti Melabk samkvmt texta rarbkar en
engin heimild er til a tla, a a hafi veri gert rkilegar en Stb (Jn Jhannesson
1941:199). a skiljanlegast s a gera r fyrir v a frsgn af orsteini landnmsmanni
og hremmingum hans hafi veri a finna forriti Hauks (.e. Styrmisbk) og i forriti Sturlu
(hinni ekktu X-Landnmu hans) er v vsa bug. Gert er r fyrir tdrtti Hauks r
tndum tti, en jafnframt v tla a Sturla hafi notast eitthva vi ttinn. Enn eykur
flkjuna essum rittengslafrum a hinni forni ttargerarmaur hefur lka notast vi
forna Landnmu a hyggju Jns Jhannessonar ea a.m.k er ar a finna. innskot einhvers

3
Stb. er stytting Jns Jhannessonar Sturlubk.

48
ritara ttarins r fornri Landnmu (Jn Jhannesson 1941:199). rhallur Vilmundarson
gerir reyndar r fyrir tdrtti r ttinum llum ekktum gerum Landnmu og ar me
Styrmisbk en afneitar engan veginn hugmynd Jns um innskotin ttinn (rhallur
Vilmundarson 1991;192).

stuttu mli sagt, Styrmisbk

Ef upprunakenning Jns Jhannessonar er lg fyrir ra er hgt a skoa samband ttarins
og frsagnarinnar Landnmu me rum htti. Ekkert bendir til annars en forrit ttarins
geti veri a sama og Hauks, og er auvelt a benda Styrmisbk. Aldursins vegna
kemur a harla vel heim vi Styrmisbk, v eftir hefbundum leium hefur tturinn veri
tmasettur um 1200 (rhallur Vilmundarson 1991:199200).
a vill reyndar svo heppilega til a ttartlu lok ttarins er hgt a bendla vi Styrmi
hinn fra, ar er nefnilega ttartala fr orsteini tjaldstingi til Lopts fur Gunnlaugs
smis (orsteins ttr:431) sem mun hafa veri upp um aldamtin 1200.4 essi ttartala gefur
v vsbendingu um aldur forritsins og kemur alveg heim vi tlaan aldur Styrmisbkar.5
Ennfremur m sj Hauksbkartexta Landnmu, ttartlu til essa sama Gunnlaugs. ar er
ekki um langfegatal fr orsteini a ra, heldur rakin tt fr landnmsmanninum
Eyvindi til smisins fyrrnefnda (H.:351). N er a ekkt a ritarar Landnmugeranna bttu
vi ttartlum til sinna manna, og jafnan oftar en einu sinni. Haukur tundar
samviskusamlega ttir Sturlunga og btir einstaka sinnum vi til sinna manna. a er v
engin sta til annars en ttrakninguna til Gulaugs smis hafi Haukur fengi r ru forriti
snu enda ekki a sj a hann hafi nokkur persnuleg tengsl vi ennan ekkta smi. Ekki
hefur veri rannsaka hvort sj megi skipulega ttrakningu sem rekja megi til Styrmis
srtexta Hauks en vissulega gefur etta dmi grun um tengsl Gunnlaugs/Gulaugs smis vi
Styrmi.

Mlska og leikrn uppsetning og rkrn atburars

Er form og efni ttarins ess elis a ar hljti frsgn Hauks a vera iggjandi? Ekki er
unnt a koma auga a. Strsti munurinn ttinum og frsgn Hauks er skipulg
atburarrs ttarins. Skipulagari uppsetning ttarins hefur veri talin til aldursmerkja.
Haukur hefur svo a mati Jns Jhannessonar o.fl. gert frsgnina ruglingslegri egar hann
var a notast vi heimildir snar, annig hafi rin raskazt hj Hauki skum ess, a hann
hefur rita frsgina um fyrri ferina eftir Stb. ur en hann fr fyrir alvru a gera tdrtt r
ttinum. (Jn Jhannesson 1941:200). Ef etta vri rtt tti textinn byrjun frsagnarinnar
hj Hauki a vera hreinn Sturlubkartexti. Svo er ekki, byrjun frsagnarinnar eru einmitt
fjlmrg atrii sem ekki er sagt fr Sturlubk t.d. er sagt fr systur orsteins og mur,
nnari lsing landafri og ennfremur kemur skringin gjf sgrms beinu framhaldi
sem skring v hvers vegna sgrmur galt ekki skattinn (H.314; S.356). essi tkni til a
greina aldur og uppruna frsagna hefur mtt vaxandi efasemdum friheiminum. Rklegri
og betri frsgn er ekki endilega s eldri eins og fyrri tar mnnum var gjarnt a halda
fram.
Anna atrii hefur tturinn srstu me en a er mikil hersla beina ru.
Persnurnar ra mlin gjarnan hnyttnum setningum. Konungurinn segir t.d. um andf
sgrms vi a borga skatt: Skjtt munu vr gjra skiptin . Vr munum eignask land hans
og lausaf, en tla honum leng af jru. orsteinn segir egar hann frttir drp fur sns:

4
etta er hgt a reikna t me upplsingum um ekktan frnda Gunnlaugs sem kemur fram ttartlu
Hauksbkar, .e. a Finnur Hallson lgsgumaur (11391145) hafi veri mmubrir hans. (H.351)
5
Hr er mia vi lklegan aldur Styrmis v hann hefur a.m.k. veri uppkominn 1206 og deyr 1245.

49
Bitu hann enn rin Haralds konungs, en brtt mun eptir vera tt vra, ef Haraldr konungr
skal einn fyrir sj. (orsteins ttr:427; 428429). Mrg fleiri oraskipti formi beinnar ru
eru rakin ttinum, mr taldist til 25 slk oraskipti. eir sem oftast tala eru konungur,
sgrmur, orsteinn tjaldstingur, og rormur sendimaur konungs, en einnig eru hf
bein oraskipti eftir rli, sendimnnum, bndum og skipverjum og einu sinni er ltil setning
hf eftir landnmsmanninum Flosa. Frsgnin Landnmu Hauks er laus vi slkt en
bregur fyrir a oraskipti manna eru hf eftir beint: spuri bndur, ef eir vildi greia
konungi slkan skatt sem beizk var. (H.314). Gefur leikrn uppsetning og mlska ttarins
til kynna a s frsgn s upprunalegri? Mialdarmenn tkuu a oft a betrumbta
sgur, og tldu sr a jafnvel rtt og skylt. Vissulega hefur frsgnin Hauksbk boi upp
slkar lagfringar og lkt er frsgnin lsilegri eirri mynd sem vi hfum Flateyjarbk.
Ef Haukur hefi haft ttinn svipari mynd og Flateyjarbk vi hndina er trlegt anna
en hann hefi notfrt sr lipulega ritaan textann ar og sem smekkmaur texta hefi hann
lti einhver hnyttiyrin fylgja me. a er ekkt a Haukur stytti og skerpti texta og eyddi
t arfa skrafi. Hann var orsins maur og greinilega gefinn fyrir gar sgur. Lagfringar
hans texta jnuu fremur eim tilgangi a gera frsgnina skrari og lsilegri (Jansson
1945:114). a er v skjn vi mynd sem vi hfum af honum, a hann hefi fari a
rugla atburarsinni fr gtlega skilmerkilegri r eins og ttinum.

Annars konar texti Hauks, betra handrit, frleikur og almlt

sannindi?
a eru nokkur atrii hj Hauki sem hvorki er a finna ttinum Flateyjarbk n
Sturlubk. ar er ekki sagt fr orlaugu systur orsteins, ekki hvaan r Noregi au sigla, en
skv. Hauki bjuggust au til slands r Grenmar fyrir austan Landisnes (H.314). Einnig er
Hauksbk ttartala fr orgeiri brur orsteins til Smundar fra (H.314). Sj m texta hj
Hauki sem er fornlegri og eins og oralagi rllinn hvatti gref og kemur a heim vi
vsuna en etta atrii vantar ttinn. En hver er skringin essu? Jn svarar eftirfarandi
htt: Um sumt hefur Haukur ekki urft neinar ritaar heimildir. nnur voru alkunn, svo sem
tt Smundar fra, forfur Steinunnar, konu Hauks. Loks hefur Haukur auslega haft
betra hdr. af ttinum en n er til. (Jn Jhannesson 1941:201). essar skringar eru ess
elis a hvorki er hgt a sanna r n hrekja. Skiljanlegra er a gera r fyrir notkun Hauks
Styrmisbk og hfundur ttarins hafi fengi sguna um landnmsmanninn orstein
tjaldsting r Landnmu, gengur skyldleiki essara rita upp n allra essara getgtna um
vinnulag Hauks.

Niurstur
Hugmyndir manna um upprunalegt efni Landnmu hefur gert a a verkum a efni hennar
hefur veri flokka niur hefbundi Landnmuefni og hins vegar sgulegar vibtur.
tturinn um orstein tjaldsting er einn eirra innskotstta sem Haukur a hafa notast
vi Landnmu sinni. a er raun ftt sem styur a verklag Hauks v tturinn er til
muna yngri og unglegri a flestu leyti enda er forrit Hauks af ttinum tali betra og
upprunalegra. a er hr eins og var upprunakenning Jns Jhannessonar sem strir
mnnum essa tt. Melabk er grunnvimi upprunalega ger og lkist Sturlubk
essum sta annig verur niurstaa Jns Jhannessonar s a Styrmisbk hafi veri eins.
getur a sem framar greinir hj Hauki ekki veri fr Styrmi komi. tt dmi urfi a
vera bsna flki til a ganga upp v a Sturla arf lka a hafa stust vi ttinn sjlfsttt
og san er innskot r fornri Landnmu ttinum. Mun elilegra er a gera r fyrir v a
s sem setti saman ttinn Flateyjarbk hafi einmitt notast vi sams konar rit og Haukur og
essu tilviki vntanlega Styrmisbk. a er ljst a a eru rittengsl vi allar gerir

50
Landnmu. Sturla hefur mun forari texta en samt eru orrttir eins upphafskaflar og frsgn
af landnmi. Einnig eru tv merkingaratrii sem fylgjast a hj Sturlu og ttinum sem
Haukur hefur ekki. Sennilegast er a hfundur ttarins hafi notast vi Styrmisbk og
jafnframt hafi forrit Sturlu haft essa sgulegu frsgn sem og forrit Melabkar. S mynd
sem hefur veri vi li, af hinni fornu Landnmu sem stuttri skr snist mr v tpast
standast. Allar ekktar gerir Landnmu hafa veri sgugerar og svo hefur vntanlega
veri me hina upphaflegu smi. a er einnig hgt a benda athyglisverar
ttarupplsingar orsteins tti sem gefa til kynna tengsl vi Styrmisbk. ar er ttrakning
til orsteins smis, ttartlu til essa sama orsteins smis er smuleiis a finna srtexta
Hauksbkar allt rum sta og bendir v til Styrmisbkartexta.

Heimildir
Auur Ingvarsdttir, 2004: Sagnarit ea skr? Staa Melabkar sem upprunalegustu gerar
Landnmu. In: Saga.
Einar l. Sveinsson, 1934: Formli. In: Laxdla saga. Halldrs ttir Snorrasonar. Stfs ttr. slenzk
fornrit. Reykjavk.
Finnur Jnsson, 1900: Indledning. In: Landnmabk IIII. Hauksbk. Sturlubk. Melabk.
Kaupmannahfn,), bls. ilx.
Finnur Jnsson, 1927: Flateyjarbk. In: Aarbger for nordisk Oldkyndighed og Historie.
Flateyjarbk IIV, 1944. Akranesi.
H = Hauksbk = Landnmabk IIII. Hauksbk. Sturlubk. Melabk. Udgiven af Det kongelige
nordiske oldskrift selskab [Finnur Jnsson]. Kaupmannahfn, 1900.
Gsli Sigursson, 2002: Tlkun slendingasagna ljsi munnlegrar hefar. Tilgta um afer.
Reykjavk.
Jakob Benediktsson, 1968:Formli. In: slendingabk. Landnmabk. slenzk fornrit I. Jakob
Benediktsson gaf t. Reykjavk.
Jn Jhannesson, 1941: Gerir Landnmabkar. Reykjavk.
Jansson, Sven B.F. 1945: Sagorna om Vinland. Handskrifterna till Erik den rdes saga I (Kungl. Vit-
terhets historie och antikvitets akademiens handlingar. del. 60:1). Stockholm.
Lois-Jensen, Jonna,1969: Den yngre del af Flateyjarbk. In: Afmlisrit Jns Helgasonar 30. jn 1969.
Reykjavk, bls. 23550.
Sveinbjrn Rafnsson, 2001: Sguger Landnmabkar. Um slenska sagnaritun 12. og 13. ld. Rit-
safn Sagnfristofnunar 35. Reykjavk.
rhallur Vilmundarson, 1991: Formli. In: Harar saga Brar saga, orskfiringasaga, Flamanna
saga, rarins ttr Nefjlfssonar, orsteteins ttr uxafts, Egils ttr Su-Hallssonar, Orms ttr
Strlfssonar, orsteins ttr tjaldstings, orsteins ttr forvitna, Berba ttr, Kumlba ttr,
Stjrnu-Odda draumr. slenzk fornrit XIII. rhallur Vilmundarson og Bjarni Vilhjlmsson gfu t.
Reykjavk, bls.5228.
Landnmabk. slenzk fornrit I. Jakob Benediktsson gaf t. Reykjavk. 1968.
S = Sturlubk = Landnmabk IIII. Hauksbk. Sturlubk. Melabk. Udgiven af Det kongelige
nordiske oldskrift selskab [Finnur Jnsson]. Kaupmannahfn. 1900.
Sk = Skarsrbk. Landnmabk Bjrns Jnssonar Skars. Jakob Benediktsson gaf t. Reykjavk.
1958.
orsteins ttr tjaldstings. slenzk fornrit XIII. rhallur Vilmundarson og Bjarni Vilhjlmsson
gfu t. Reykjavk, 1991.

51
 Writing origins: the development of communal identity in
some Old Norse foundation-myths and their analogues in Guta
saga
Robert Avis, St Johns College, Oxford
slendingabk and Landnmabk have long been regarded as the lynchpins of an Icelandic
foundation-myth, a myth which pervades not only these purportedly historical works but also
the slendingasgur and, arguably, much other Old Norse literature produced in Iceland. An
understanding of the use of this myth is essential to our wider understanding of the productive
use of the idea of historicity present in the essentially fictional accounts of Icelandic society
found in the family sagas. Certain texts, slendingabk and Landnmabk among them, are
dominated by a single narrative interest in the literary establishment of the conceptual
boundaries of a community, the foundation-myths of the title. Conversely, a typical slend-
ingasaga contains multiple narratives, not only those internal to the narrative logic of the saga
but also those pertaining to what Jrg Glauser has described as the groe Erzhlung (great
narrative, Glauser 2006:41) of Iceland itself. Rather than looking exclusively at the myth of
Icelandic settlement across such generic boundaries, this paper will begin by looking to-
wards Guta saga. Although this medieval Scandinavian settlement myth was written in Old
Gutnish rather than Old Icelandic, it nevertheless exhibits productive similarities with our
Icelandic examples. We will also examine Snorri Sturlusons redaction of the foundation of
Scandinavian identities in Ynglinga saga, the prologue, essentially, to the konungasgur of
Heimskringla. These texts are drawn from across spatial and temporal boundaries: whilst
slendingabk dates to the first third of the twelfth century (Grnlie 2006:xiii), Heimskringla
and Guta saga have both been dated to the first half of the thirteenth century (Heimskringla
1941:xxix, Guta saga 1999:xiii). Whilst Snorri Sturluson states his admiration for Ari inn
fris work in the preface to Heimskringla (and, in all probability, was familiar with some
form of Landnmabk),1 there is no evidence of awareness of Guta saga in Iceland. Neverthe-
less, a comparison of these texts may illuminate a common act of writing the origin of a cul-
tural community, an act essential to an understanding of the literature of a specific group of
people in time and space.

The frequently fantastic content of Guta saga or Ynglinga saga, such as the account that be-
fore the arrival of fire Gotland by dagum sank ok natum var uppi (sank by day and rose up
by night Guta saga 1999:2), or the story of the creation of Seeland by Gefjuns four sons
(Heimskringla 1941:15), initially betrays little in common with slendingabk or Land-
nmabk beyond a common interest in origins. These narratives of Gotland and Seeland, lo-
cated in a mythic past, contain territories whose very physicality is unstable. This is not the
case for Iceland, since the temporal location of the settlement is scrupulously established
within human history with reference to the wider European (and Christian) context in slend-
ingabk (dating according to the martyrdom of St Edmund, for example) and Landnmabk
(which dates the settlement according to the reigns, amongst many others, of the Pope, Byzan-
tine emperor, and the kings of Norway, Sweden and Denmark). If, however, one is to leave to
one side the literal plausibility of the events described in Guta saga, on a mythic level there is
more unity in design and purpose than first meets the eye: the text sits between the implausi-

1
[O]k ykkir mr hans sgn ll merkiligust (and his whole story seems to me most noteworthy, Heimskringla
1941:6). Of course, we cannot be entirely sure whether Snorri is referring specifically to slendingabk or other
works by Ari now lost.

52
ble (and thus potentially allegorical) narratives of Ynglinga saga and the ostensible factuality
of slendingabk and Landnmabk, moving between two modes of historical writing.
Like Iceland, the community of Gotland is first established by settlement from abroad. The
discoverer of Gotland, ieluar, is not established within a genealogical sequence; in this re-
spect, he is not dissimilar from the Naddodd vking who first named Iceland Snland ac-
cording to Landnmabk (1968:34). Neither, perhaps significantly, is ieluar granted a prior
ethnicity: Stephen A. Mitchell interprets this as follows:

The implicit value of having ieluar appear from no specific point of origin is, of course, that
the Gotlanders then owe no a priori allegiance to any of the national kingdoms. By thus begin-
ning the tale in media res, the redactor avoids any possible constraints on the future loyalties
and political freedom of the Gotlanders (Mitchell 1984:171).

This is a considerable point of deviation from the settlement-myth of Iceland, in which Nor-
ways significance looms large. Nevertheless, ieluar establishes a dynasty whose genealogy
is inscribed on the new land itself. His three grandsons split their fathers inheritance and thus
the origin of Gotlands three districts is established. This marks the creation of the community
of Gotland, as distinct from the geographical feature. ieluars role as the father of the con-
cept of Gotland is cemented by his role in stabilising the very island itself, and preventing it
from sinking back into the sea: En ann mar quam fyrsti eldi a land, ok sian sank et aldri
(And that man first brought fire to the land, and it never sank afterwards Guta saga 1999:2).
Settlement the creation of a community here has a direct physical effect on the land itself;
Gotland becomes inhabitable, indeed simply thinkable as the physical dimension of a com-
munity, through the Promethean action of the first settler. This action transforms the land
from the malleable, non-physical, mythic landscape we find in the story of the creation of
Seeland to a more concrete model, closer to the tangible landscape of Iceland.
Whether a moment exists at which a group of settlers becomes a coherent community, at
which a label such as Icelanders or Gotlanders becomes meaningful, is a fundamental
question in assessing the validity of these texts as foundation myths. The Gotlanders (the
fulk i Gotlandi Guta saga 1999:2) fast transcend the physical dimensions of the island of
Gotland when the text begins a substantial digression to describe the fate of a third of the
population who had to leave, because the land elpti aim ai alla fya (land was not able to
support them all Guta saga 1999:2). The significance of this part of the text, which includes
the exiled Gotlanders nomadic existence until they settle in the Byzantine Empire, lies in the
fact that it forms part of this text at all. Despite the fact that all these events take place in an
ill-defined mythic past, the texts concern for the fate of these people demonstrates the con-
ceptual birth of a community: after relating an anecdote in which the Gotlandic exiles trick the
Byzantine emperor into granting them permanent permission to reside in his realm, the author
observes that hafa air sumt af varu mali (they retain some of our language Guta saga
1999:4).2 The use of the second-person plural possessive adjective varu betrays the authors
implicit expectation of address to a particular community. slendingabk demonstrates the
same phenomenon in its opening line, slendinagbk gra ek fyrst byskupum rum, orlki
ok Katli (I first wrote the Book of the Icelanders for our bishops orlkr and Ketill slend-
ingabk 1968:3).
In her analysis of the development of an Icelandic ethnicity from an anthropological per-
spective, Kirsten Hastrup has observed that:

2
The argument for some form of identification here between Gotlanders and Goths has been much rehearsed; E.
V. Gordon gives a brief summary (1956:175). If so this would only reinforce the view taken here that founda-
tion-narratives such as Guta saga readily utilise legend and folktale in order to reinforce communal identity.

53
 At first there was only a community of settlers, but after a few generations this community was
turned into a society by way of collective social action (Hastrup 1990:79).

We have observed how Guta saga demonstrated at a very early stage in the narrative that a
self-defined community had come into existence, one that was no longer simply a short-hand
for those individuals or family groups who happened to inhabit the same physical space, but
an identity that had become intrinsic to the individual to the extent that the concept of a Got-
lander abroad was meaningful. Hastrups observation is important in taking this further, since
the concept here of society necessitates not only collective identity but also collective ac-
tion. In Guta saga, Hastrups society is arguably born at the same time as the community
first becomes evident, for the mechanism by which a third of the Gotlanders become Gotland-
ers abroad is by the casting of lots, which suggests the presence of some form of administra-
tive system; we have already, after all, witnessed the creation of the administrative districts of
Gotland. Later in the text the author observes that the people of Gotland e iemlika sigri ok ret
sinum (always held the victory and their rights Guta saga 1999:6) in their dealings with
foreign powers, which seems to establish incontrovertibly a discrete Gotlandic ethnicity, soci-
ety and community, by differentiation from foreign attackers. We will see later that Gotland
comes to exist in a special relationship with Sweden, which is analogous to that between Ice-
land and Norway; but first we will examine how Icelandic foundation myths operate with a
more complex understanding of the foreign and the tland, which is not predicated, as so
many foundation myths are, on the violent struggle against exterior forces, but instead inte-
grates multiple named points of origins for its settlers who go on to acquire a degree of indi-
geneity.
Migrations occupy a special status at the heart of the myths of origin examined here. The
first hint of collective identity in Icelandic literature is related to the common situation of the
settlers as emigrants from a different, established community: here Norway assumes a role as
the mother country. But even if it might have been true that the early occupants of Iceland
were simply Norwegians abroad, the literature depicts migrations as an inherently transforma-
tive act. Snorri Sturlusons Ynglinga saga, within Heimskringla, postdates slendingabk by a
considerable number of years, but its source material, Ynglingatal, is dated to the late ninth
century (Pulsiano 1993:665), although it only survives embedded within much later texts.3
Snorris saga includes an account of a forced westward migration which recalls the conven-
tional myth of Icelandic emigration following the aggressive centralisation of Haraldr hrfa-
gris reign, but with the greater complication of direct familial identification between oppres-
sor and emigrants. The Swedish king Ingjaldr burns down his hall containing six rival kings
whom he had invited to a feast, and

Eptir etta lagi Ingjaldr konungr undir sik ll essi rki, er konungar hfu tt, ok tk skatta af
(Heimskringla 1941:67).

After that King Ingjaldr brought under his control all these kingdoms, which the kings had
ruled, and made them pay tribute.

This aggressive accumulation of authority over former petty kingdoms and the levying of
taxes on his new acquisitions is strikingly redolent of the portrait of Haraldr hrfagri found in
several instances of the migration-myth, especially in those prefacing slendingasgur; for
example, in Egils saga:

3
The dating of Ynglingatal is somewhat contentious: kerlund (1939) has argued for an earlier date, whilst more
recently Krag (1991) has put forward a date close to the composition of Ynglinga saga. The weight of critical
opinion seems to lie with the former view.

54
 Haraldr konungr eignaisk hverju fylki ul ll ok allt land, byggt ok byggt, ok jafnvel sjinn
ok vtnin, ok skyldu allir bendr vera hans leiglendingar, sv eir, er mrkina ortu, ok saltkar-
larnir ok allir veiimenn, bi sj ok landi, vru allir eir honum lskyldir (Egils saga
1933:12)

King Haraldr took into his own hands in every district all the estates and all the land, settled and
uninhabited, and even the seas and lakes, and all the farmers were obliged to be his tenants, and
so those who worked in the forests, and the salt-driers and all the huntsmen, both of the sea and
the land, were all then subject to him.

The narrative of emigration in Ynglingatal is stretched over the reign of more than one king,
however: although Ingjaldr pursued territory and tax in a similar manner to Haraldr, he was
not ultimately successful, and burned himself in his hall when he realised that resisting the
Danish King varr would be futile. varr, a greater consolidator of power than Ingjaldr or, per-
haps, Haraldr, went on, according to Snorri, to rule Denmark and Sweden, precipitating the
flight of the Ynglings from their traditional power-base around Uppsala westwards, in the
direction of Norway. The similarities in these two narratives only go so far: Snorris interest is
primarily in the genealogical history of a certain supposed family, the Ynglings, rather than
the history of a people; and the degree of identification between lfr trtelgjas new prov-
ince and a discrete identity from the Swedes is complicated by the very fact that lfr, the
architect of emigration, was himself the son of Ingjaldr, the first oppressive and centralising
tyrant. Myths of exodus, of course, are not exclusive to a Scandinavian cultural context: the
Bible surely provides an archetype. These are not, therefore, directly related narratives, but
instead Ynglinga saga and its primary source Ynglingatal provide a number of analogues for
one of the most important aspects of the Icelandic foundation myth: its definition of itself
against an other, namely, centralised kingship. Whether medieval Iceland before 1262 really
was a state or just a rebellious province of Norway is an irrelevance in the face of a literary
tradition which repeatedly alludes to the transformative power of emigration.4

We can see, therefore, that the settlement-phase of myths of origin frequently contains an im-
plicit development of a social identity. The central piece of collective action which cements
this identity in slendingabk is the conversion to Christianity. Like the migration-myth, the
conversion-myth transcends any single text; Sin Grnlie describes the multiplicity of forms
in which it is found:

[I]t appears in different contexts and genres and therefore in different guises: as a key moment
in the history [of] the Icelandic people (in slendingabk), as a successful missionary effort on
the part of the Norwegian king lfr Tryggvason [] and as a focus for the historical fiction
of many of the family sagas, most famously Njls saga (Grnlie 2006:vii).

There are several reasons why the conversion ought to be considered a key moment. It is the
clearest example of Hastrups collective action, insofar as Christianity is incorporated into
the self-definition of Icelandicness. It also demonstrates, conversely, the peripheral nature of
religion to identity. The Icelanders remain Icelanders before and after, and despite a Norwe-
gian kings role in initiating the conversion, the ability of the Alingi peacefully to resolve to
follow the new faith affirms the capability of Icelandic society to absorb a wholesale change
4
We should observe that these migrations follow a westward trajectory, a concept which appears integrated into
an Icelandic perception of space: the east is Norway, the west new territory: Greenland, Markland and Vnland.
There are further instructive analogues to the establishment of a migration-myth in Iceland in the sagas relating
to the settlement of Greenland, Grnlendinga saga and Eirks saga raua, whilst an analogue to the ambiguous
relationship between Iceland and Norway is also found in Freyinga saga.

55
of values without compromising their perceived independence. Essentially the same event
occurs in Guta saga, with an even more overt claim to the exclusive decision of the Gotland-
ers to change their faith:

Sian gutar sagu kristna manna sii, a lydu air Guz bui ok lerra manna kennu. Toku a al-
mennilika vir kristindomi mi sielfs vilia sinum utan uang, so at engin uang aim til kristnur
(Guta saga 1999:10).

After the Gotlanders saw the customs of the Christians, they then obeyed Gods command and
the teaching of learned men. They then received Christianity generally of their own will, with-
out duress, such that no-one forced them into Christianity.

The emphatic rejection of the idea of the Gotlanders being forced into conversion naturally
highlights the role of their own vilia, again, despite the fact that a Norwegian king in this
instance lfr inn helgi had begun the process of conversion external to Gotland itself. This
form of narrative, composed by the converted, must necessarily strike a balance between the
essential rectitude of the discovery of the true faith and a determination to ensure that the
converted community is shown to have made this decision freely, through its own enlighten-
ment, rather than solely by external coercion. But even this exterior pressure betrays the exis-
tence of the Gotlandic community that the text seeks to affirm. In the same way that the an-
cient Gotlanders retained a communal identity even after being forced to leave the island due
to overpopulation, a Gotlander abroad is converted by lfr Ormika af Hainaim who be-
gins the process of general conversion. This necessity for conversion to begin within the
community, even if supported from outside, is highlighted in slendingabk. lfr
Tryggvasons role is certainly important, but it is notable that angbrandr, lfrs missionary,
remains in Iceland for only a few, rather ignoble, years:

En es hann hafi hr verit einn vetr ea tv, fr hann braut ok hafi vegit hr tv menn
ea rj, es hann hfu ntt (slendingabk 1968:14).

And once he had been here for a year or two, he then went away, having killed here two or three
men who had libelled him.

It is left, therefore, to orgeirr to reconcile the Christian and pagan parties at the Alingi an
external force was successful in introducing disorder into the system, but it takes indigenous
individuals and institutions to re-establish harmony. Harmony is not here a plurality of be-
liefs: it is of course orgeirr who argues for the imperative of lg ein landi hr (one law
here in this land, slendingabk 1968:17); this echoes Guta sagas insistence that Christianity
was accepted allmenilika (generally Guta saga 1999:10). The narrative structure of slend-
ingabk itself binds the new faith to the existing socio-political apparatus: from the eighth
chapter onwards, immediately post-conversion, the naming of Icelandic bishops and law-
speakers are juxtaposed, placed in discrete clauses but interwoven: for example, Ari con-
cludes his eighth chapter (Fr byskupum tlendum, on foreign bishops slendingabk
1968:4) with a list of law-speakers culminating in Gellir Blverksson; he then returns to mat-
ters ecclesiastic with a discussion of the first Icelandic churchman Bishop sleifr, before pick-
ing up the listing of law-speakers again with the observation Gunnarr enn spaki hafi tekit
lgsgu, es Gellir lt af, ok hafi rj sumur (Gunnarr the wise had become law-speaker
when Gellir left off, and held the post for three summers slendingabk 1968:20). Church
and state are thus presented as equal and compatible elements of a single community. This
myth acquires a sense of tragic wish-fulfilment when compared to the probable reality:
Kirsten Hastrup observes that

56
 [T]he introduction of tithes paved the way for an increasing accumulation of wealth by a rela-
tively small number of people, even if still in the name of the Church. This again laid the foun-
dation for serious conflicts over church lands, which contributed to the general breakdown of
Icelandic society (Hastrup 1985:193).

The myth of successful conversion is vital to the creation of a coherent and useful Icelandic
identity for two central reasons which operate in different directions: it legitimises the poten-
tial for the Alingi and the legal institutions of medieval Iceland to settle disputes, since the
stand-off between Christian and non-Christian blocs acts as something of a worst-case sce-
nario that an institution designed to ensure harmony might encounter, and it places Iceland
and Icelanders squarely within Christendom and thus within a wider European tradition of
culture and learning, which leaves its mark on every piece of extant Old Norse-Icelandic lit-
erature, if through nothing more than the use of the Latin alphabet. Orri Vsteinsson
(2000:18) rightly observes that the conversion-narrative of slendingabk was not so much a
matter of salvation as political unity, and the omission of any great concern for the effect of
the Alingis decision on the Icelanders eternal souls affirms the text as speaking to a myth
of the creation of a community or even a nation rather than a myth solely of Christianisa-
tion. Guta saga shares this exploitation of a religious event as an opportunity to establish or
reinforce a specific communal identity within the entry into wider Christendom.
The role of Norway in the conversion of Iceland and Gotland was nevertheless significant,
despite the extent to which these cultures respective literatures used conversion to affirm
discrete identities; the two continental powers, Norway and Sweden, figure largely in these
insular literatures far beyond this one event. The relationship between Norway and Iceland, in
history and as represented in the literature, is both highly complex and decidedly difficult to
discern: it has no direct modern or ancient analogues. Guta saga does, however, provide some
illuminating points of comparison in its description of the relationship between Gotland and
Sweden, in which Gotland nevertheless appears considerably more subservient than the liter-
ary depiction of pre-1262 Iceland, insofar as Gotland was obliged to provide men to fight with
the Swedish king if so demanded, albeit under certain conditions. Guta saga presents the es-
tablishment of a subordinate relationship with Sweden through the decision of the Gotlanders
to petition the Bishop of Linkping to reskep giera (give support Guta saga 1999:10),
before describing the bishops obligations towards Gotland. Nevertheless, although presented
as a willed decision by the Gotlanders, their decision to place their church under the aegis of
an external authority has immediate implications for the independence of the island itself:

Sian gutar toku sir biskup ok presti ok vir fulkumnum kristindomi, a toku air ok vir at fyl-
gia suia kunungi i herfer mi siau snekkium ufan a haiin land, ok ai ufan kristin (Guta saga
1999:12).

After the Gotlanders took for themselves bishop and priest and completely accepted Christian-
ity, they then also took it upon themselves to follow the Swedish king in military expeditions
with seven long-ships against heathen lands, but not against Christian ones.

The final qualification to the above sentence, ai ufan kristin, seems to reassert Gotlandic
sovereignty over their political arrangements, despite the bare facts of their subordinate rela-
tionship to the Swedish monarch. Stephen A. Mitchell has argued that

the compiler of GS was also a propagandist: he wanted to create an historical overview of the is-
land which would demonstrate its traditional independence (Mitchell 1984:173).

57
In slendingabk, a clear differentiation is made between the list of the byskupa eira, es
verit hafa slandi tlendir (the foreign bishops who have been in Iceland slendingabk
1968:18) at the beginning of the eighth chapter and the opening of the ninth chapter with the
first bishop of Iceland, sleifr, the son of Gizurr enn hvti Teitsson who had agreed with King
lfr Tryggvason to help bring Christianity to Iceland. To a certain extent we see here, there-
fore, the inverse of the conversion-myth of Gotland: rather than seeking external legitimacy in
a Norwegian bishop, Norway itself sends converted Icelanders such as Gizurr to effect a
simulacrum of the foundation of an indigenous church.
We have observed some ways in which these foundation-texts work to utilise both history
and myth to construct a literary point of origin for various cultures. But we must not observe
these texts in a vacuum they existed, and, with the exception of Guta saga, continue to exist
within a corpus of texts relating to the same communities. In her analysis of Ynglinga saga,
Marlene Ciklamini describes the function of the text within Heimskringla thus:

[t]o provide mythical models of events and human behaviour for intellectual guidance in the
more familiar but confusing historic era. Interestingly, themes and motifs from Ynglinga saga
recur throughout Snorris account of the historic era, which suggests that Ynglinga saga served
not only as an introduction to the historical part of the saga, but also as a thematic presentation
of mythic and social verities (Ciklamini 1975:90).

This conception of Ynglinga saga as a functional prologue which informs the text it precedes
forms a microcosm of the literary function of slendingabk (and, in a different manner,
Landnmabk) in relation to the later literary products of Iceland pertaining to Iceland,
throughout the free state period and right into the thirteenth century. The myths of origin of a
community developed and explored in these two texts are reified as the origin of a literary
corpus which consistently interacts with the social constructs enumerated in slendingabk:
settlement, conversion, and the politico-legal system the Alingi and its subordinate assem-
blies embody. These myths are interrogated in much Old Icelandic literature, from the family
sagas to the law codes (if they can be labelled as such) and to the ttir of Icelandic skalds.
Guta saga provides us with an analogous text in which mythic origin directly precedes an
explication of how Gotland became a vassal of Sweden: as such, it presents a dramatically
compressed story of a community that Icelandic literature plays out over many texts and sev-
eral centuries. It constitutes a useful example of the way in which literary foundation myths
transform historical events, such as settlement and conversion, into transformative events in
themselves, which create new identities. Whilst they are certainly not accurate as accounts of
the foundations of real societies, they are of fundamental importance as accounts of the foun-
dation of the literary manifestations of these societies. As in The Tempest, whats past is pro-
logue; it is these texts of origins which form the prologues to the literary corpus we study.

Bibliography
Ciklamini, Marlene. 1975. Ynglinga saga: Its Function and Its Appeal. Medieval Scandinavia 8. pp.
8699.
Egils saga Skalla-Grmssonar. 1933. Ed. Sigurur Nordal. slenzk fornrit 2. Reykjavk: Hi slenzka
fornritaflag.
Glauser, Jrg. 2006. Begrndungsgeschichten: Der Mythencharakter der islndischen Literatur.
Skandinavische Literaturgeschichte. Ed. Jrg Glauser. Stuttgart: Metzler. pp. 4050.
Gordon, E. V. 1956. An Introduction to Old Norse. 2nd ed. Oxford: Clarendon Press.
Grnlie, Sin. 2006. slendingabk, Kristni saga: The Book of the Icelanders, the Story of the Conver-
sion. London: Viking Society for Northern Research.
Guta saga: The History of the Gotlanders. 1999. Ed. Christine Peel. London: Viking Society for
Northern Research.

58
Hastrup, Kirsten. 1985. Culture and history in medieval Iceland: An anthropological analysis of struc-
ture and change. Oxford: Clarendon Press.
Hastrup, Kirsten. 1990. Island of Anthropology: Studies in past and present Iceland. Odense: Odense
University Press.
Heimskringla. 1941. Ed. Bjarni Aalbjarnarson. slenzk fornrit 26. Reykjavk: Hi slenzka fornritaf-
lag.
slendingabk, Landnmabk. 1968. Ed. Jakob Benediktsson. slenzk fornrit 1.i. Reykjavk: Hi slen-
zka fornritaflag.
Krag, Claus. 1991. Ynglingatal og Ynglingasaga. En studie i historiske kilder. Oslo: Norges allmenvi-
tenskapelige forskningsrd.
Mitchell, Stephen A. 1984. On the Composition and Function of Guta Saga. Arkiv fr nordisk
filologi 99. pp. 151174.
Orri Vsteinsson. 2000. The Christianization of Iceland: Priests, Power, and Social Change 1000
1300. Oxford: Oxford University Press.
Pulsiano, Phillip, Kirsten Wolf, Paul Acker and Donald K. Fry eds. 1993. Medieval Scandinavia: An
Encyclopedia. New York: Garland Publishing.
kerlund, Walter. 1939. Studier ver Ynglingatal. Lund: C. W. K. Gleerup.

59
 Eddadikterna i Codex Upsaliensis DG 11
En projektpresentation
Maja Bckvall, Dept. of Scandinavian Languages, Uppsala University, Sweden
Jag r knuten till projektet Studier i Codex Upsaliensis som doktorand, och min del av projek-
tet handlar om eddadiktcitaten i handskriftens Gylfaginning. Utgngspunkten fr min under-
skning r att i frsta hand se p dikterna utifrn handskriftens egna premisser, snarare n att
som i tidigare forskning avgra vad som r rtt eller fel i frhllande till andra handskrifter. I
andra hand kommer jag dock ocks att jmfra DG 11 med de resterande tre huvudhandskrif-
terna av Edda, men frmst fr att belysa de skrivningar som mter i DG 11. Med denna ut-
gngspunkt hoppas jag kunna nrma mig handskriftens lsare. Om man frestller sig en me-
deltida islnning som i motsttning till oss inte har tillgng till andra handskrifter n DG 11,
hur kan hon (eller han) ha uppfattat de ord och formuleringar som nutida forskning avfrdar
som felskrivningar? Fr att ta ett exempel ur Vlusp 9, s skriver Codex Regius av Edda
(GKS 2367 4to) at skyldi dverga drtt of skepja (Faulkes 2005:16), medan DG 11s skrivning
lyder hverr skyldi dverga drtt um spekja.1 DG 11 r ensam bland handskrifterna om att ha
spekja hr, och det r inte svrt att frklara: skrivaren har frmodligen rkat byta plats p <k>
och <p>, en inte ovanlig typ av skrivarfel. Sett ur ett skrivarperspektiv r problemet s att
sga lst men vad kan lsaren ha lst? Ordet spekja r i sig inte srskilt problematiskt; det
betyder bl.a. gjre vis, begave med Visdom (betydelse 2 i Fritzner s.v.). Att gra dvrgarna
visa kan ocks ses som ett stt att skapa dem p, srskilt om man jmfr med Vlusp 17
18, dr de frsta mnniskorna skapas genom att f bl.a. nd och r, ande och tanke. Trots
att det allts finns starka indicier fr att skrivaren av DG 11 (eller dess frlaga) har rkat skri-
va fel, s har en hypotetisk samtida lsare goda mjligheter fr att frst strofen som den str.
Drmed anser jag att man inte kan tala om en skrivning som spekja som ett fel, eftersom den
fortfarande r begriplig i sitt sammanhang.
Jag kommer i min underskning att operera med ett antal olika tolkningsniver, dr de tre
viktigaste r skrivarniv, lsarniv och egentliga fel. Hur dessa niver kan anvndas har ex-
emplifierats i det ovanstende. P skrivarniv r det rimligt att misstnka att ordet spekja i
Vlusp 9-citatet inte har valts medvetet av skrivaren, vilket gr att man br se om skrivning-
en kan tolkas p lsarniv vilket jag som synes anser att man kan. Det r frst nr ven en
tolkning p lsarniv r omjlig eller mycket osannolik som man kan beskriva en frn majori-
teten avvikande skrivning som ett egentligt fel. Men ven de egentliga felen har ngot att sga
oss. Ett tydligt exempel p en skrivning som br tolkas som ett egentligt fel finns i Vlusp
57-citatet, dr DG 11 har Sl mun sortna <sigrfolldinnar> (Grape et al. 1977:34), vilket r
nrmast jmfrbart med strofen som den str i Codex Regius av eddadikterna (GKS 2365
4to): Sl tr sortna, sgr fold mar (Jn Helgason 1955:13). Att <sigrfolldinnar> har sitt ur-
sprung i ett missfrstnd av <nn> fr <m> torde st klart, och det r heller inte mjligt att f
ut ngon sprklig mening av skrivningen (se Bckvall 2007:42 f.). nd behver man inte
nja sig med att som Jn Helgason (angende andra diktcitat i DG 11) tala om en forvirret
skrivers vilkrlige pfund (1961:X). Elementet <sigr> visar till exempel att DG 11 hr ligger
nra formuleringen i de tv handskrifter som har dikten i sin helhet (GKS 2365 4to och
Hauksbk), vilka bda har sgr fold mar. I de vriga huvudhandskrifterna av Edda (GKS
2367 4to, Codex Wormianus och Codex Trajectinus) inleds strofen Sl mun sortna, skkr fold
mar. Detta r inte det enda exemplet p en strof dr DG 11s citat har mer gemensamt med
dikthandskrifterna n med vriga eddahandskrifter, men precis vad detta innebr, om ngot,

1
Skrivet h. skylldi dverga drott vm spekia (Grape et al. 1977:8).

60
terstr att se. Det faktum att fyra ord har skrivits ihop till en enhet tyder dessutom p att skri-
varen av DG 11 eller dess frlaga har analyserat ordfljden som ett ord; medeltida skrivare
skriver vanligtvis inte ihop mer n tv ord t gngen. En hypotes r att ett hopskrivet <imar>,
s som det till exempel ser ut i bde GKS 2365 och Hauksbk, kan ha setts av avskrivaren
som ett -innar och uppfattats som slutet p ett particip i f.sg.gen., vilket r vad <sigrfolldin-
nar> liknar rent morfologiskt. Kanske kan missfrstndet ha att gra med att marr inte har
varit ett levande ord i skrivarens ordfrrd; det finns i stort sett bara belagt i eddadikter (Fritz-
ner s.v.).
En annan frga som vcks i samband med eddadikterna i DG 11 rr vad som frvntas
och/eller krvs av lsaren. Handskriften har till exempel generellt samma strofcitat som de
vriga huvudhandskrifterna av Edda, men ett mrkbart undantag utgrs av det lngsta sam-
manhllna eddadiktcitatet i Edda. I de vriga handskrifterna citeras nio Vluspstrofer i rad
(strof 46/58, 47/14, 48, 5053, 5557); i DG 11 endast tre. Men det r inte tre slumpms-
sigt utvalda strofer av dessa som citeras i DG 11, utan de tv frsta (om man rknar kombina-
tionen av strof 46 och 47 som en strof) och den allra sista. Det verkar inte helt orimligt att
tnka sig att stroferna inte r utvalda p grund av innehllet, utan snarare fr att de utgr in-
ledning och avslutning p ett lngt citat. En lsare med knnedom om Vlusp skulle drfr
kunna supplera de strofer som kommer dremellan, srskilt som denna del av dikten (som
handlar om vilka som ska slss vid Ragnark) r ett av de mer sammanhngande avsnitten i
Vlusp.
Detta r bara ngra f exempel p vad studiet av eddadikterna i DG 11 kan ge. Jag kommer
ocks att gna mig t frhllandet mellan prosasammandragen av dikterna och de direkt drp
fljande citaten, liksom frgor omkring namnformer och metrik. Mnga av dessa frgor
kommer frmodligen mynna ut i en diskussion om eddadikternas liv i den medeltida isln-
ningens medvetande. I mina frsta frsk att gra ngot s skenbart enkelt som att lsa texten
i DG 11 som den str har jag sett hur uppgiften knoppar av sig och ger upphov till frgestll-
ningar av vitt skilda slag. Jag ser drfr mycket fram emot att arbeta vidare och kunna ter-
komma med resultaten inom de kommande ren.

Litteratur
Bckvall, Maja Marsling, 2007: Vlusp i Uppsalaeddan. En nyfilologisk underskning. (D-uppsats i
svenska sprket/nordiska sprk vid Uppsala universitet).
Faulkes, Anthony, 2005 (ed.): Edda. Prologue and Gylfaginning. London.
Fritzner: Fritzner, Johan, 1973: Ordbog over det gamle norske sprog. 4 uppl. Oslo. Bergen. Troms.
Grape, Anders, et al., 1977 (utg.): Snorre Sturlassons Edda. Uppsala-handskriften DG 11. II. Transkri-
berad text och Paleografisk kommentar. Uppsala.
Jn Helgason, 1955 (udg.): Eddadigte. I. Vlusp, Hvaml. (Nordisk filologi. A: 4). 2. ndrede
udgave. Kbenhavn. Oslo. Stockholm.
1956 (udg.): Eddadigte. II. Gudedigte. (Nordisk filologi. A: 7). 3. gennemsete udgave. Kbenhavn.
Oslo. Stockholm.

61
 Individuality and Iconography: Jakob Sigurssons Render-
ings of Codex Upsaliensis f.26v
Patricia A. Baer, University of Victoria, Canada

1. Introduction
Anthony Faulkes and other scholars have thoroughly documented the textual transmission of
Snorri Sturlusons thirteenth-century Prose Edda. However, the transmission and reception of
Edda illustrations in manuscripts and early print sources has received scant critical attention.1
This paper will examine the earliest rendering of an Edda scene2 which is the Gylfaginning
(The Deluding of Gylfi) illustration on f.26v in the early fourteenth-century Icelandic Codex
Upsaliensis [hereafter U] and the eight renderings which stem from it. My paper will clarify
how a cycle of illustrations occurred that resulted in the transmission of Us illustration over a
four-hundred-year period from Iceland to Sweden and back to Iceland. The papers major
focus is on four full-page renderings of the Gylfaginning scene that were produced in the pe-
riod 1760 to 1765 in three hand-copied paper manuscripts by Jakob Sigursson [hereafter JS].
JSs four renderings include two renderingsNKS 1867 4to [hereafter N] f.111v and B 299
4to [hereafter ] f.59vthat are very similar to Us illustration. However, this paper will estab-
lish that JSs renderings were inspired by Olaus Vereliuss copperplate rendering of Us
Gylfaginning illustration in a Swedish print edition of Gautreks Saga in 1664. In addition to
his two rather close renderings of Verelius copperplate, JS also created two idiosyncratic ren-
deringsN. f.98r and SM [hereafter S] f.78rthat are part of his two sets of sixteen Edda
scenes in N and S.3 All four of JSs Gylfaginning renderings differ from each other and from
that of Verelius rendering of the scene, and not surprisingly the two idiosyncratic renderings
feature major differences from the copperplate. Apart from the light it casts on medieval Ice-
landic illustrative practices, my study offers insights into illustrator- and patron-relationships
in book production and culture in eighteenth-century Iceland, as well as in seventeenth-
century Sweden. As my paper will demonstrate, illustrators through the ages have essentially
adhered to the description of Gylfaginning in Snorris text and to the basic composition of Us
illustration. However, illustrators of this scene, from U to the present day, have also individu-
alized their renderings in ways that reveal fascinating aspects of the transmission and recep-
tion of Us illustration, thus clarifying an important chapter in the textual reception of Snorris
Edda.

2. The Illustration of Gylfaginning in Codex Upsaliensis

The well-known illustration of Gylfaginning in U depicts the Swedish King Gylfidisguised
as Gangleristanding before three regal figures seated on high seats hierarchically arranged so
that they tower above him. The seriousness of the situation is only fully discernable to those
familiar with the narrative. Gylfi has come to discover if the formidable abilities of the
strangers from Asia are due to the gods that they worship. He is immediately ensnared in a
1
See Hans Kuhns Greek Gods in Northern Costumes concerning paintings and other art works based on Norse
mythology in nineteenth-century Scandinavia (2000: 209 219), and Margaret Clunies Ross examination of the
illustrations of the Poetic Edda for Thomas Grays Norse odes: The Fatal Sisters and The Descent of Odinn
(1988: 105 118).
2
AM 738 4to from 1680 features twenty-three illustrations of individual figures, as well as Valhll and Yggdra-
sil, but does not contain illustrations of narrative scenes.
3
I will present a thorough discussion of the two sets of Edda illustrations and their differences in Chapter Four
of my forthcoming dissertation.

62
wisdom contest and is threatened with bodily harm if he loses. This contest serves as a narra-
tive frame for the Gylfaginning section of Snorris Edda. It explicitly reinforces the process of
euhemerization that was introduced in the Prologue, and subtly raises the question as to ex-
actly who is being deluded. Does Gylfi merely act dumb and play along, or does he actually
come to believe that these men are gods? The text does not describe the seated figures but
simply states that they are kings and identifies them with names from the large list of ins
heiti (poetic synonyms for inn) as Hr, Jafnhr, and rii (High, Just-as-high, and Third),
Despite Snorris statement that the seated figures are all kings, it is intriguing that Us illustra-
tor depicted the lower figure as a female, as evidenced by her feminine face and the contours
of the robe outlining her breasts. It is possible that Us illustrator may have chosen to use a
feminine figure in order to represent a negative hypostatic representation of inns true
character. inn was a practitioner of the type of magic known as seir, which was so
strongly associated with women that it was considered to be unmanly even in pagan times and
was demonized in the Christian era.

63
 U eventually came into the possession of the manuscript collector Bishop Brynjlfur
Sveinsson who sent it to Denmark in 1639 as a gift to Stephanus Johannis Stephanius. The
Bishop had a copy made of U before it left the Iceland, namely Marsh. 114 [hereafter M], but
M also left the country when it was taken to England as part of Thomas Marshalls collection
in 1690. The Gylfaginning scene on f.23v in M is of interest to my argument because it is not
an exact copy of U. Ms illustrator portrayed the three seated figures as bearded kings and
explicitly identified them within the illustration as renning ins (a trinity of inns).
Ms rendering of Gylfaginning does not appear to have inspired any renderings in England
and consequently did not participate in the further transmission of Us illustration.
Us illustration did not engender any renderings in Denmark and consequently its transmis-
sion might well have ended there as well. Danish scholars were not interested in copying U
because it was not considered to be the best text to base a translation on. The Danish edition,
Peder H. Resens Edda Islandorum, was published in Copenhagen in 1665 and was based on
the Laufs Edda. Resens Edda made a print version of Snorris Edda accessible for the first
time in Icelandic, Danish and Latin but it was not illustrated. However, Resens introduction,
which took a metaphysical approach to the Edda, was included in hand copied manuscripts in
Iceland such as N, , and S, and JS illustrated a cover page of his own devising for it in .

3. Verelius Copperplate and Other Swedish Renderings

U was acquired by the Swedish collector, Magnus Gabrielle de la Guardie after Stephanius
death in 1650. U arrived in Sweden during a period of intensely patriotic antiquarian scholar-
ship, and it was a welcome resource, given that Gylfi was a Swedish king and that the events
of Gylfaginning took place in the vicinity of Uppsala. Verelius created his full-page copper-
plate rendering of Us Gylfaginning illustration in 1664, which was sometimes4 inserted into
his notes accompanying his translation of Gautreks Saga (1664:42a). The reason for includ-
ing or excluding Vereliuss copperplate from editions may represent a subscription option
offered by its publisher. U does not appear to have circulated but Verelius copperplate en-
gendered further renderings. Johannes Schefferus placed his rendering of Vereliuss copper-
plate onto a page crowded with other representations of triple crowns (1668:fig. 32) in re-
sponse to competing Danish claims to the crest. Olaus Rudbeck included a similar rendering,
also based on Verelius, on a page with other illustrations whose connections to Gylfaginning
are not readily apparent (1679:309 fig. 29). However, Rudbecks rendering of the three gods
in Gylfaginning was part of his efforts to prove that Sweden was in fact the lost Atlantis and
the cradle of civilization. These early print renderings of Gylfaginning reflect the fact that
Sweden was the first Scandinavian country to develop the printing press and also the first to
use an Edda illustration, in the patriotic spirit of the times, to promote their nationalistic agen-
das in print.
The lower seated figure in all of the Swedish renderings is very close to that of U but does
not necessarily indicate a visualization of a hypostatic representation of inn. Verelius and
Schefferus were minimalists when depicting folds in the figures clothing, but Rudbeck em-
phasized the contour of her left breast with a triple line. However, for Verelius, and his fellow
scholars, the temple trio at Uppsala would have been composed of inn, Thor, and the god-
dess Frigg. In Sweden, Frigg had supplanted Freyr in Adam of Bremens description of the

4
Anders Grape (1962:29) notes that the copperplate was rarely inserted into Vereliuss notes. However, I dis-
covered that Roll 366 of the Scandinavian Culture Series contains two editions of Verelius notes and both of
them contain the copperplate.

64
Temple of Uppsala, due to an error in the transmission of Adams text.5 Schefferus appears to
have been the first to claim that the trio of enthroned figures in U could be traced back to the
temple gods in Uppsala (1678: 157). Consequently, the lower seated figure simply represents
Frigg when it is depicted as a woman in seventeenth-century Swedish renderings of U.

5
In Adams description, Freyrs name was rendered as Fricco, which became Frigh in early Swedish versions
and was later misinterpreted in the writings of Johannes and Olaus Magnus as representing Frigg. See Magnus
1555:185 endnote 33.

65
 It is significant that the Swedish renderings included a detail that was originally a pen trial
in U, a face with a crown that the Swedish renderings transformed into an icon of the sun with
a human face. In his text, Verelius makes a connection between inn and the Sun based on
inn being monoculus. It is odd that Verelius remained faithful to Us illustration and did
not depict inn as one-eyed in his rendering of Gylfaginning. However, Verelius and Rud-
beck were determined to establish a link between Norse and Classical mythology and asserted
that the Temple of Uppsala had originally been the Temple of Apollo, and therefore both
temples could have been associated with a sun icon. Schefferus opposed the association of the
temple with that of Apollo on the basis of archaeological evidence (Ellenius 1957:6264).
However, he may have retained the sun icon because the sun was considered to be the king of
celestial bodies and inn, being one of the sir, was an astral deity. Consequently, when the
sun icon is present in renderings it indicates that the illustrator was not copying directly from
U but from a rendering of Verelius copperplate.

4. Vereliuss Copperplate and its Icelandic Renderings

Verelius notes to Gautreks saga were often included in eighteenth-century hand-copied pa-
per manuscripts of that saga in Iceland, and it was no doubt through a print edition of his text
that his rendering of Us illustration came to the attention of JS (1729 1779). JS was a ten-
ant farmer and a prolific copier and illustrator of texts6, as well as a poet. He was fostered at
Kirkjubr in north-eastern Iceland and spent his life as a tenant farmer in the surrounding
district. Lutheran pietism insured that all children at the time were taught to read in order to
be confirmed, but neither writing (Olafsson 2009:6) nor drawing would have been considered
a necessary part of their education. JSs informal education would have been enabled by the
clergyman, lafur Brynjlfsson, who was also a scribe and illustrator and was in charge of
Kirkjubrs farmstead and church. JS supplemented his livelihood by producing hand-copied
paper manuscripts which were part of an informal system of book production in Iceland from
the sixteenth to the early twentieth century.
It is significant that JSs four renderings of Gylfaginning all contain the sun icon from the
Swedish renderings, as well as the same manner of depicting Gylfis clothes so that they gen-
erally conform to the outlines of his robe in U. The basic layout of all of JSs renderings are
mirror images of the Swedish renderings and the reversal of the layout indicates that Vere-
liuss copperplate was his exemplar. Unlike the other Swedish renderings, Verelius copper-
plate was printed on only one side of a page, and it sometimes bled through the paper thereby
producing a mirror-image.7 The renderings by Schefferus and Rudbeck were printed on heav-
ier paper and have images on both sides of the page. However, Vereliuss copperplate was
printed on only one side of a page because, as previously mentioned, it was not inserted into
every edition. Access to an edition with the copperplate and its bleed-through would have
given JS the choice of copying the reverse image, which obviously appealed to him artisti-
cally because he used the reversed image for all of his renderings.
As previously mentioned in my introduction, the N manuscript, which is the oldest of the
three manuscripts under discussion, is unusual because it contains both a close copy (f.111v)
and an idiosyncratic rendering of Vereliuss copperplate (f.98r). N is also unusual because JS
signed the close copy in N J. Sigurdsson with my own hand as well as adding a verse:
Hrs er lygin hrna snd me hvopta pri linu, en ins kunungs talin og tnd tign
hsetinu. (Highs lie is shown here with strong eloquence. But the dignity of King Odin in

6
See Hrafnkelsson (2004:13) for a list of JSs extant mss.
7
See Roll 366 of the Scandinavian Culture Series: the copperplate does not bleed through in #2355 but it does in
#2563.

66
the high seat is described and lost.)8 The verse is not unusual because JS included it in all four
of his renderings, however in N it contains a minor correction changing og (and) to en (but).
The placement of the verse, squeezed onto the bottom of the page in N, along with the correc-
tion, suggests that this was the first Edda scene that JS illustrated, and that he simultaneously
recorded it on the page as he composed it. The spontaneous nature of the composition and
recording of JSs poem in N suggests that it was Verelius copperplate that initially inspired
him to create his close renderings, and subsequently his idiosyncratic renderings. JS did not
sign his other three renderings of Gylfaginning and the verse is more carefully placed and
lettered in the latter renderings.
JSs attention to detail in his close renderings of Verelius suggests that he regarded the
copperplate to be an accurate rendering of U, and accorded it the respect that he would have
given to the original illustration. However, JS does vary somewhat from Verelius in the close
renderings as to the major detail in his depiction of all of the seated figures as bearded and the
minor detail of his inclusion of a tiled floor in . The status of the close rendering in N
(f.111v) is confirmed by its placement in a group of renderings of historical artefacts com-
prised of rune stones and spears. Moreover the fact that the compiler of N also included one
of JSs idiosyncratic renderings (f.98r) indicates that close rendering was perceived differ-
ently than the idiosyncratic rendering.
Verelius did not label his figures and JSs labelling of the three kings varies in his render-
ings. In the two rendering in N, the labels follow the order given in the text but he reversed
them in and S. The confusion regarding the labels indicates that eighteenth-century readers
in Iceland struggled, then as we do now, to make sense of Snorris description in which Third
is the topmost figure, Just-as-High the middle figure, and High the lowest.
JSs idiosyncratic renderings in N (f.98r) and S (f.78r) feature many differences, both mi-
nor and major, from Verelius rendering and represent a major break in the tradition of copy-
ing U. For instance, a major change occurs when JS depicts all of the seated figures as being
one-eyed, thereby visually indicating that his figures are hypostatic representations of inn.
Moreover, JS also changed all of inns declamatory hand gesturesa standard oratory ges-
ture since antiquityto a two-fingered gesture, which in the Christian tradition is associated
with the conveyance of blessings or absolution. Possibly, by depicting the figures gesturing in
a way that is inappropriate to them, JS is reminding his audience (which would have been
familiar with the gestural conventions of their Lutheran pastors) that the gods are engaged
in a sort of fraud. Thus, as we can see, JSs alterations to inns gestures in the idiosyncratic
renderings gives greater emphasis to the verse in all four of his renderings concerning inns
lie and his consequent loss of dignity.
JS also changed Gylfis declamatory gesture in the idiosyncratic renderings to an open
handed gesture, and his arm is thrown up over his head. This exaggerated gesture suggests
enthusiasm and gullibility, and JS labelled Gylfi with text that describes him as gapir (gap-
ing) and as gleypir sig lygi (swallowing the lie). Taken all together the change in gestures
along with the verse and the text indicates the manner in which JS and his patrons perceived
the dynamics of the scene. Thus Gylfi was viewed as having been thoroughly deluded by
inns eloquence, but eighteenth-century Lutheran Icelanders no longer viewed pagan myths
as material that they might fall into believing. They read the Edda despite the disapproval of
the Church and used its contents in the composition of ballads known as rmur.
JS also altered the three figures of inn and that of Gylfi in his idiosyncratic renderings.
The inn figures are less dignified in their body language but Gylfi undergoes the greatest
change. In the close renderings (N f.111v and f.59v), Gylfis disguise is that of a beggar, or
possibly a paganised pilgrim, but in the idiosyncratic renderings (N f.98r and S f.78r), he ap-

8
I consulted various friends and colleagues while doing the translations for this paper; any mistakes are my own.

67
pears to be a simpleton with a deformed body and face that seems less than human. Conse-
quently, it appears that inn and Gylfi were both viewed as foolish figures in eighteenth-
century Christian Iceland, which is also indicated by his verse denigrating inn as a liar and
his text identifying Gylfi as a gullible fool.
There are indications in JSs idiosyncratic sets of sixteen Edda scenes in N and S that he
tailored his work to suit his clients interests or level of education, but only his renderings of
Gylfaginning falls within the scope of this paper. The kings are empty-handed in the idiosyn-
cratic rendering (f.78r) in S but in the idiosyncratic rendering in N (f.98r), which was owned
by the clergyman at Kirkjubr, the highest king is holding an orbis terrarum. Moreover, the
middle figure in the idiosyncratic rendering in N is holding an object that represents a pagan-
ised orbis cruciger, with the head of Thors hammer, Mjllnir, replacing the Christian cross.
The orbs in N make it possible to identify the topmost figure as inn, the middle figure as
Thor, and the lower figure as a pagan version of the Holy Spirit. Rory McTurk has observed
that the three figures can be seen as offering support to Anne Holtsmarks suggestion that
Snorri presents the heathen religion partly as an inverted Christianity, and he further sug-
gests that Snorris three kings represent three figures of inn as a pagan version of the Holy
Trinity (1994:11). In S, whose provenance and textual contents indicate that its owner had
less esoteric interests than the clergyman who owned N, the three hypostatic depictions of
inn are empty-handed. It appears that Ss owner was not interested in subtleties of a pagan-
ised Trinity or in creative anachronisms.

5. Conclusion
JS stands out among illustrators of Gylfaginning because he is the only illustrator to have cre-
ated more than one rendering of the scene and also because his illustrations represent the most
recent renderings of Gylfaginning for almost two hundred years. Moreover, as my work indi-
cates, JS idiosyncratic renderings (N f.98r and S f.78r) represent a fascinating chapter in the
reception and transmission of the Edda because they move beyond the ambiguous description
in Snorris text by depicting the three figures of inn as one-eyed bearded males and in de-
picting Gylfis enthusiastic gullibility. JSs compilations preserve evidence of the reading
interests of eighteen-century Icelanders and his illustrations of Gylfaginning offer insights as
to their engagement with the text of the Edda. JSs labours as a scribe and illustrator insured
that his clients were not restricted to reading the material deemed appropriate by the Church
which owned the only printing press in Iceland during this period. The enthusiasm with which
JS (presumably at the behest of his patrons) took up the challenge of revisualizing Us medie-
val image that had returned to Iceland by means of a seventeenth-century engraving indicates
a culture which at that particular moment was keen to engage with its mythological heritage.

Works Cited and Consulted

1) Illustrations of Gylfaginning
Codex Upsaliensis f.26v. c. 1325. In: Snorre Sturlasons Edda; Uppsala-Handskriften. DG 11. Ed. A.
Grape. Uppsala.
B 299 4to f.59v. 1764. Landsbkasafn slands. The National Library of Iceland.
Marshall 114 f.23v. c. 1638. In: Eddurit Jns Gumundssonar Lra[]Samantektir Um Skilning
Eddu. Reykjavk: Stofnun rna Magnssonar slandi. Front Cover.
NKS 1867 4to. ff.98r. and 111v. 1760. Det Kongelige Bibliotek. The Royal Library, Copenhagen.
<http://www.kb.dk/permalink/2006/manus/738/dan/> (Last accessed on 04/26/2009).
Rudbeck, Olaus. 1938. Olf Rudbeks Atland, Eller Manheim[]Olaus Rudbecks Atlantica. Uppsala
Almqvist & Wiksells boktryckeri-aktiebolag. Fig.29:309.

68
SM 66 f.78r. 1765. Stofnun rna Magnssonar slandi. The rni Magnsson Institute in Iceland
Schefferus, Johannis. 1678. Johannis Schefferi Argentoratensis De Antiquis Verisque Regni Sueciae
Insignibus: Liber Singularis, Holmiae: Excudit Nicolaus Wankiif. Fig. 32.
Verelius Olaus. 1664. Gothrici & Rolfi[] Gothica Conscripta. P. 42a.

2) Primary Sources and Editions

Codex Upsaliensis. c. 1325. In: Snorre Sturlasons Edda; Uppsala-Handskriften. DG 11. Ed. A. Grape.
Uppsala.
Einar G. Ptursson, Jn Gumundsson lri, 1998. Eddurit Jns Gumundssonar Lra[]1574
1658.; Samantektir Um Skilning Eddu. Reykjavk: Stofnun rna Magnssonar slandi.
B 299 4to. 1764. Landsbkasafn slands. The National Library of Iceland.
Jakob Sigursson & rn Hrafnkelsson. 2004. Inngangur tgefenda. In: Handarlnulist og hfube-
inafri. Hafnarfiri: Sguspekingastifti. Pp. 7 14.
Magnus, Olaus, 1996. Historia De Gentibus Septentrionalibus: Rom 1555, London: Hakluyt Society.
Marshall 114. F.23v. C. 1638. In: Eddurit Jns Gumundssonar Lra[]Samantektir Um Skilning
Eddu. Reykjavk: Stofnun rna Magnssonar slandi.
NKS 1867 4to. 1760. Det Kongelige Bibliotek. The Royal Library, Copenhagen.
<http://www.kb.dk/permalink/2006/manus/738/dan/> (Last accessed on 04/26/2009).
Rudbeck, Olaus, 1938. Olf Rudbeks Atland, Eller Manheim[]Taflor Till Olaus Rudbecks Atlantica.
Uppsala Almqvist & Wiksells boktryckeri-aktiebolag.
Saganet. SM 66. <http://sagnanet.is>. (Last accessed on 04/26/2009).
SM 66. 1765. Stofnun rna Magnssonar slandi. The rni Magnsson Institute in Iceland
Schefferus, Johannis, 1678. Johannis Schefferi Argentoratensis De Antiquis Verisque Regni Sueciae
Insignibus: Liber Singularis, Holmiae: Excudit Nicolaus Wankiif.
Snorri Sturluson, Magns lafsson, & Faulkes, A., 1979. Two versions of Snorra Edda from the 17th
century. Reykjavk, Stofnun rna Magnssonar.
Snorri Sturluson, 1179?1241., 1987. Edda. Ed. A. Faulkes. London: Dent. P. 6.
Snorri Sturluson,1179?1241., 1962. Snorre Sturlasons Edda; Uppsala-Handskriften DG 11. Ed. A.
Grape. Uppsala.
Verelius, Olaus, 1664. Gothrici & Rolfi Westrogothiae Regum Historia Lingua Antiqua
Gothica Conscripta: Quam E.m.s. Vetustissimo Edidit Scheffer, Johannes.; 16211679; Ed.
by Henricus Curio. Uppsala

3) Secondary Sources
Benedikt Gslason, 1967. Jakob sgu-skrifari. In jviljinn, Reykjavik. Pp. 40 45.
Bksted, A., 1986. Go Og Hetjur Heinum Si: Allegt Fririt Um Goafri Og Hetusgur,
Reykjavik
Ellenius, A., 1957. Johannes Schefferus and Swedish Antiquity. In: Journal of the Warburg and Cour-
tauld Institutes, 20(1/2). Pp. 59 74.
Gsli Sigursson, 2004. Melsteds Edda: The last manuscript sent home? In: The Manuscripts of Ice-
land. Ed. by V. Olason. Reykjavk: rn Magnsson Institute in Iceland. Pp. 179 185.
Kuhn, Hans, 2000. Greek gods in Northern costumes: Visual representations of Norse mythology in
19th century Scandinavia. In: Old Norse myths, literature and society: the proceedings of the 11th
International Saga Conference 27 July 2000, University of Sydney. Sydney, Centre for Medieval
Studies. Pp. 209 219.
Malm, M., 1991. Improving History with Old Norse Poetics: A 17th Century Theory of Interpretation.
In: The Audience of the Sagas. II. Pp. 46 53. Olafsson, David, Handwritten books in the [sic]
19th Century Iceland in Ennen & nyt, Vol. 1: The Papers of the Nordic Conference on the His-
tory of Ideas, Helsinki (2001). <http://www.ennenjanyt.net/2-01/olafsson.pdf > (Accessed: April
22, 2009 ) Pp.1 15.
McTurk, R., 1994. Fooling Gylfi: Who Tricks Who? In: Alvssml 3. Pp. 3 18.
Olafsson, David, Handwritten books in the [sic] 19th Century Iceland in Ennen & nyt, Vol. 1: The
Papers of the Nordic Conference on the History of Ideas, Helsinki (2001).
<http://www.ennenjanyt.net/2-01/olafsson.pdf > (Accessed: April 22, 2009 ) Pp.1 15.
Ross, Margaret Clunis, and Lnnroth, L., 1999. The Norse Muse: Report from an International Re-
search Project. In: Alvssml 9. Pp. 3 27.

69
Ross, Margret Clunis, 1998. The Verbal and the Visual Sublime: the Reception of Thomas Grays
Norse Odes. In: The Norse Muse in Britain: 17501820 / Collins, Amanda J., Trieste: Parnaso. Pp.
105118.

70
 St. lfr and his Enemies in the Saga Tradition
Sverre Bagge, Centre for Medieval Studies, University of Bergen, Norway
The following is a part of a longer article dealing with the medieval literature about St lfr
Haraldsson (king 101530). Its aim is neither to discover the truth about lfr nor to solve the
difficult problem of the textual relationship between the various works about him, but to ex-
amine the tradition as such, from the vague references to lfrs life and reign in Passio Olavi
(c. 1175) to the detailed narrative in Snorri Sturlusons Separate Saga (c. 1225) and Heim-
skringla (c. 1230). My conclusions from this examination can be summarised as follows.
If we consider the difference between the first and the last of these works, it is easy to leap
to the conclusion than an enormous change has taken place during these fifty years, leading to
the invention of the lfr known by most modern readers of the sagas. This impression is
wrong. Quite a substantial part of this story must have been known to our earliest author,
Theodoricus, who, in his terse and succinct way, renders a fair number of the episodes told in
greater detail by his successors and shows that much of what was included in the later sagas
was known to him and his contemporaries, at least concerning the first and last phase of
lfrs reign. As Theodoricus was very selective in what he included, he may also have
known some episodes of which there is no trace in his work, although this is of course impos-
sible to prove. From Legendary Saga we can conclude that a great variety of traditions, partly
oral, partly written, must have existed, particularly concerning the early and late phases of
lfrs reign. The age of these traditions is difficult to determine, but the existence of skaldic
poetry, partly combined with background narrative (Begleitprosa), may suggest that at least
some of them go back to lfrs own lifetime.
In this paper, I shall discuss the great conflict between lfr and his internal and external
enemies, which ended in his death at Stiklestad. All the Norwegian-Icelandic sources, includ-
ing Passio Olavi, agree that lfr was killed in battle, and, with the exception of Passio
Olavi, all sources list Cnut as well as a number of Norwegian magnates as his enemies. How-
ever, only two sources give more details, Legendary Saga and Snorri. Despite altogether three
references to Cnuts attempts to gain Norway, the author of Legendary Saga shifts the focus
from him to the internal Norwegian opposition. For the first time, we get information about
individual motives for resisting lfr.

Challenge and Response: the Individual Motives

The main example of this is the story of sbjrn selsbani, which occurs for the first time in
the Oldest Saga and then in Legendary Saga and Snorri (Otte Brudstykker, pp. 3 f.; Leg. Saga
ch. 4749; HkrOH ch. 11720, 123). sbjrn sails from Northern Norway to Sola to buy
grain from his uncle Erlingr Skjlgsson. As lfr has banned the export of grain from South-
ern Norway, his rmar Selrir confiscates sbjrns cargo and sends him home empty-
handed. Next year, sbjrn kills Selrir in lfrs presence, is taken captive and sentenced
to death, but is saved by Erlingr, who forces lfr to accept compensation, after which lfr
demands that sbjrn take Selrirs place. When sbjrn fails to fulfil the condition for his
release, he is killed by one of lfrs men. This story follows immediately upon a comment
about lfrs strict justice which caused the revolt against him (Leg. Saga ch. 46, p. 108) and
is clearly intended as an example of this. It is also followed by a comment that this was one of
many conflicts between lfr and Erlingr. However, the author does not mention any direct
effects of sbjrns death.
The story of sbjrn contains no skaldic stanzas, which, combined with the fact that it oc-
curs only in three sources, may give rise to suspicions that it is a late invention. However, it is

71
hardly invented by the author of Oldest Saga. Nor is it difficult to explain that it does not oc-
cur in Theodoricus who usually omits or abbreviates such stories and who may have found it
sufficient to point to various magnates hatred of lfr without going into detail about its ori-
gin. Its absence from Fagrskinna may have a similar explanation. This work deals briefly
with lfrs reign and in general contains little information about the inner struggles in Nor-
way. As the story deals with dramatic events that are likely to have been remembered locally,
I am inclined to believe that it contains a kernel of truth, although some dramatic details have
probably been invented.
In the introduction to the story of sbjrn, both Oldest Saga and Legendary Saga briefly
mention a series of other episodes leading to conflicts between lfr and individual magnates,
mainly rir hundr, who only with difficulty manages to get reconciliation after having killed
Karli, a good man (gan mann) in Bjarmaland (Otte Brudstykker, p. 2; Leg. Saga ch. 46).
In the latter context, the author states that lfr, learning about the disloyalty towards him, let
four men be killed, including rir hundrs nephew (the son of his sister) and Grjtgarr,
whose wife was later married to Klfr rnason. The two others are anonymous. Afterwards,
rir killed three men for lfr who were his closest friends. The author of Legendary Saga
may well have intended these killings as examples of lfrs strict justice, to which the mag-
nates reacted, but the connection is not obvious; the wish for revenge leading to a series of
killings as in the feuds described in the Icelandic sagas would seem an equally likely motive.
This statement in Legendary Saga is most probably derived from a written source. In
Theodoricuss case, we can easily imagine a basis in either oral or written storytelling that is
condensed in the brief references to factual events, as Theodoricus is not particularly inter-
ested in narrative. This is not the case in Legendary Saga, which contains a number of stories,
some of which are even well narrated. Consequently, it is unlikely that the author knew stories
without rendering them in his text. By contrast, he may well have known Aris or Smunds
lost works, both of which were probably very brief. Thus, he may have taken over the infor-
mation about the four men killed by lfr from one or both of these predecessors, despite
being able to identify only two of them.
If Snorris source was Oldest Saga or another source similar to Legendary Saga, this
somewhat cryptic presentation of lfrs conflicts with the magnates must have represented a
challenge for him. Snorri devotes more space to these conflicts than any other writer does. His
starting-point is the story of sbjrn where he largely follows his predecessor, although add-
ing a few more details. However, the main difference between the two works is that Snorri is
more precise regarding the consequences of this episode for the relationship between lfr on
the one hand and Erlingr and rir hundr on the other.
Erlingr has been the leading man in Western Norway since the reign of lfr Tryggvason,
a position lfr attempts to weaken (Bagge, Society and Politics, pp. 78 f., 1258). A com-
promise is reached between the two adversaries just before sbjrns fatal expedition to the
south, an agreement Erlingr does not want to break, although he also feels obliged to aid his
kinsman sbjrn. Snorri does not explain in detail what happened after sbjrn had been
reconciled and had broken the agreement, but he makes it clear that the relationship between
lfr and Erlingr had deteriorated and that Erlingr was ripe for Cnuts overtures. Thus, in
Erlingrs case we are dealing with a conflict of interests which according to Snorri could be
partly solved by compromises, but was exacerbated by sbjrns foolish actions. The fact that
Snorri, following his predecessors, also tries to acquit lfr of the responsibility for Erlingrs
death instead blaming Erlingrs kinsman slkr points in the same direction: lfr and
Erlingr respected each other and would have been able to cooperate, had not their friendship
been destroyed, first by sbjrn, then by slkr.
Concerning rir hundr, only Snorri specifies the kinship between him and sbjrn, stat-
ing that he was sbjrns paternal uncle. Consequently, sbjrns mother turns to rir to get

72
revenge for her dead son, which, according to Snorri, leads to rir killing Karli, who had
taken part in the killing of sbjrn (HkrOH ch. 123, 133). Snorri here reverses the sequence
in Legendary Saga, according to which rir had already killed Karli at the time of sbjrns
fatal expedition to the south. A modern observer may well forgive Snorri for doing this, given
Legendary Sagas record of inconsistency and confused sequences. However, there are also
other reasons for being suspicious of Snorris version. Karli is not killed until he and rir
have spent the whole summer together on a combined trading and Viking expedition to Bjar-
maland, after which they run into quarrel over the booty. Thus, rir apparently has another
motive to kill Karli. Moreover, why would Karli join rir in an expedition after having par-
ticipated in the killing of his relative? Despite the fact that Snorri tries to answer this question,
the story seems to have a tenuous link with that of sbjrn, which suggests that it may origi-
nally have had nothing to do with it but simply been a story of quarrel over booty leading to a
killing. Whereas a factual or at least a traditional basis may have existed for the expedition to
Bjarmaland, possibly also for Karli being one of lfrs men, the story of rir avenging
sbjrn by killing Karli is likely to be Snorris own invention. By contrast, Karlis death may
well be one of the reasons for the enmity between rir and lfr.
Apparently, there were also others. Legendary Saga identifies two of the men lfr killed
for rir as his sisters son and Grjtgarr. Snorri repeats the statement about lfr killing
four men for rir in the speech immediately before the Battle of Stiklestad in which rir
explains his reasons for fighting lfr (Leg. Saga ch. 62; HkrOH ch. 219). rir here names
sbjrn, his brothers son; rir and Grjtgarr, his sisters sons, and lvir, their father.
Grjtgarr is thus in Heimskringla the son of lvir whom lfr killed early in his reign be-
cause of his participation in pagan cult and whose wife he married to Klfr rnason who was
then his friend. On this occasion, however, Snorri does not mention that this wife was rirs
sister (HkrOH ch. 110, cf. 1079). According to Heimskringla, lfr later killed both rir
and Grjtgarr, the former because he had accepted gold from Cnut to betray him, the latter
because he wanted to avenge his brother.
Curiously enough, however, Snorri makes little use of this motive in rirs case; he only
mentions it on this occasion. By contrast, the death of the two young men has a decisive in-
fluence on Klfr rnason, as they are his stepsons (HkrOH ch. 165, 166, 183). This identifica-
tion is not to be found in any other source. Has Snorri simply invented this story in order to
find a reasonable explanation for Klfrs defection? Or has he even invented the defection
itself? In Legendary Saga Klfr is all the time lfrs adversary and fights against him al-
ready at Nesjar. Nor is he said to have any reason for being grateful to lfr. In the dialogues
with his enemies, including Klfr, before the Battle of Stiklestad, lfr blames two of them
for forgetting the benefits he has conferred on them, but does not direct this accusation at
Klfr. Even stranger, there is no such accusation in the corresponding passage in Heimskrin-
glas version; only rir blames lfr for the death of lvir and his sons (Leg. Saga ch. 62;
HkrOH ch. 219). It therefore seems that the story of Klfrs conflict of loyalty after the death
of his stepsons is Snorris invention on the basis of the information in Legendary Saga or a
similar source about lfr killing Grjtgarr and Klfr marrying his widow.1
Finally, there are some reasons to suspect the identity of the two women named Sigrr in
Snorris narrative, as they do not occur anywhere else (Jochens 1996: 176 f.). Sigrr is also
the name of the prototype of an aggressive woman, Sigrr the Haughty, who killed lfrs
father. Thus, when there is no other evidence than Snorris for Karli being involved in

1
See Schreiner 1926: 77 f. who also regards this story as Snorris invention and suggests that the name lvir is
derived from one of Sigvatrs poems about a pagan named lvir who denied him hospitality for the night. By
contrast, the name Grjtgarr seems appropriate for an adversary of lfr, as it indicates descent from the earls
of Lade.

73
sbjrns death and a perfectly reasonable alternative explanation for rir killing him exists,
it is possible that the whole story of Sigrr with the bloody spear and rir taking revenge for
sbjrn is Snorris invention. The corresponding lack of evidence for the existence, not only
of the other Sigrir, later married to Klfr, but also of her two sons, suggests a similar conclu-
sion in this case as well. It must be admitted, however, that some kind of kinship between
rir and sbjrn did exist according to Legendary Saga which might have given rir a mo-
tive for turning against lfr. Moreover, Sigrr does not ask rir to kill Karli but to kill
lfr, which he does by piercing him, apparently with the spear he received from her, at Stik-
lestad. In a similar way, Finnr rnasons violent hatred against his brother is better explained
by Klfr having defected from lfr than by the brothers just having chosen different sides
(Leg. Saga ch. 73, 85; HkrOH ch. 231).
Snorris main reason for inventing or changing these stories is rirs and Klfrs central
role in the opposition against lfr and above all the fact that they were or were suspected of
being lfrs killers. They therefore needed a strong motive, and the strongest motive Snorri
could imagine was revenge. This is the motive of all three killers, although the first one, or-
steinn knarrasmir, who wants revenge for the ship lfr has confiscated, seems almost like a
parody compared to the two others. By contrast, both Klfr and rir are important magnates
who are mentioned several times earlier in Legendary Saga and are very prominent in Heim-
skringla.

Explaining lfrs Fall: from Legendary Saga to Snorri

The concentration of lfrs failures to his five last years enables Snorri to create a consistent
plot of his conflicts with the chieftains. By contrast, the vague chronology of Legendary Saga
suggests to the reader that the enmity was there all the time. Nor does the author give much
information about individual motives for turning against the king. This picture neither sup-
ports nor contradicts the authors generalization about lfrs strict justice combined with
Cnuts gold as the reason for the opposition against him. Although both explanations also
occur in Heimskringla, the detailed account of lfr engaging in one conflict after the other
during the last five years of his reign points to additional and more complex motives. By his
inventions and changes in the tradition, Snorri manages to create a strong network of the men
opposing lfr, all of whom have good reasons for fighting him, which also serves to explain
the turning-point in lfrs reign in his eleventh year.
I have earlier claimed that Snorri essentially depicts the conflicts as a series of power
struggles between lfr and individual magnates (Bagge 1991: 6670). Power is important in
the case of all the men mentioned above but it is not the only factor. Neither rir nor Klfr
wants a conflict with lfr; they are both forced by women demanding revenge who appeal to
their sense of shame and honour. rir is almost out of his mind, having received the bloody
spear, and Klfr has good reasons to be grateful to lfr, besides risking the friendship with
his brothers. In the case of Erlingr, his conflict with lfr might in Snorris opinion have been
solved, were it not for his loyalty to sbjrn. An additional argument for the importance of
revenge as a motive in Snorris thought is the fact that these episodes are likely to have been
his own invention. It would therefore seem that he has reduced the importance of power as a
motive in favour of revenge. On the other hand, none of the three magnates takes up arms
against lfr until Cnut has made his claim and a strong alliance can be formed against him.
The concern for honour and revenge is combined with political realism.
This emphasis on the individual motives weakens the two general explanations Snorri has
taken over from his predecessors. The leaders of the opposition willingly accept Cnuts
bribes, but this is not their decisive motive; their experience with lfr is more important.

74
Generally, Snorri attaches less importance to Cnut than most of his predecessors do; the brave
Norwegians cannot be conquered by the Danes; they are themselves able to depose their king
(Bagge 2002: 191). Nor can the conflicts be explained as the result of lfrs strict justice.
The detailed accounts of lfrs behaviour towards the men who later became his enemies
hardly confirm the picture of a king acting out of concern for strict justice, neither from a
modern nor from a thirteenth century point of view. Although it is more than a conventional
piece of religious rhetoric, it is not Snorris real explanation of lfrs fall.
Where does Snorris sympathies lie? In contrast to his predecessors, he not only gives a de-
tailed account of why lfrs adversaries turned against him, but also deals with their prepara-
tions for the Battle of Stiklestad without any word of condemnation. He even attributes a
speech to the Danish bishop Sigurr, condemning lfr as a robber and evildoer. The fact that
Snorri lets people present their arguments in speeches does not necessarily mean that he
agrees with them. However, his sympathy clearly lies in what in later terminology would be
called a balanced constitution, the king ruling in cooperation with the people, represented by
the aristocracy, and listening to the advice of the leading men in the country. His two famous
examples of confrontations between the king and the people, sbjrn of Mealhs against
Hkon the good and orgnr lgmar against King Olof of Sweden, both in all likelihood his
own constructions (Hkr. Hkonar ga, ch. 15; HkrOH ch. 80), illustrate this ideal quite well.
Particularly the latter example has been regarded as an expression of the Icelandic magnate
Snorris attitude to the Norwegian king (Moberg 1941: 20715). However, Snorri makes it
clear that orgnrs accusation against King Olof of Sweden cannot be directed against the
Norwegian lfr, who listens to his people when they want peace with their neighbour. Nor
does Snorri depict a constant conflict between lfr and the aristocracy. He lists a number of
magnates on lfrs side (Bagge 2002: 1847), and he gives specific reasons for the individ-
ual magnates who turn against him. Snorris opinion of kingship in general may be better il-
lustrated by the dialogue between the two kings Hrrekr and Hringr: On the one hand, a
strong king can easily reduce the power and independence of the magnates. On the other
hand, such a man is also able to reward his friends and punish his enemies. As an Icelander,
Snorri may in addition have taken consolation from the argument he attributes to Hrrekr
about the advantages of a distant king (HkrOH ch. 36).
Nevertheless, it is difficult to avoid the impression that Snorri blames lfr. A characteris-
tic expression of his attitude is the words he attributes to Erlingr Skjalgsson during one of
their meetings: I serve you best when I serve you voluntarily.2 Snorri seems to agree with
Erlingr that it would have been in both mens interest if lfr had allowed Erlingr to keep his
position in Western Norway rather than trying to reduce his power. Snorri may here have had
in mind another great magnate whose power the king wanted to reduce, namely his friend and
patron Earl Skli. Moreover, lfrs behaviour in the series of conflicts during the last five
years of his reign probably seemed incredibly stupid to Snorri. He alienates Erlingr
Skjlgsson by insisting on the death penalty for his nephew for killing a lowborn man, de-
scending from slaves, despite the fact that Erlingr is willing to pay whatever lfr wants in
return for sbjrns life. He then insists on sbjrn taking over the position as royal represen-
tative, which, according to the view expressed in Snorris narrative, was an extreme humilia-
tion and hardly likely to lead to lasting peace. After sbjrns death and rirs revenge, he
lets Finnr rnason humiliate rir who, like Erlingr, would probably have been willing to pay
compensation to retain lfrs friendship. At Erlingrs surrender in the Battle of Tunga, lfr
insists on humiliating him before pardoning him, although this leads to slkrs fatal blow
and would probably in any case have made it difficult for lfr to gain Erlingrs friendship.
Finally, at the time when Cnut prepares his attack on Norway and several of the leading men

2
s mun r mn jnosta hallkvmst, er ek veiti r me sjlfri (HkrOH ch. 60, p. 89).

75
have joined him, he kills both Klfrs stepsons, thus making another important magnate and
old friend join the enemy camp. A last episode completes the picture, the story of the skald
Steinn Skaptason who kills lfrs rmar but is protected by orbergr rnason (HkrOH ch.
138, cf. Leg. Saga ch. 58, p. 138). This story is referred briefly in Legendary Saga, where it is
just an episode, illustrating Steinns difficult character. In Heimskringla it completes the pic-
ture created by the concentration of all lfrs conflicts with the chieftains to his last five
years: lfr is a stubborn king who challenges too many enemies at the same time and refuses
to accept reasonable compromises. By insisting on the death penalty for Steinn, lfr almost
makes his closest friends, the rnasons, turn against him.
We do not know what Snorri really thought about these episodes after all, lfr was not
any tyrant from the old days, but the eternal king of Norway, resting in a shrine in the Cathe-
dral of Nidaros but it is understandable that Snorri needed an excuse for such behaviour,
which he found in the statement about lfrs strict justice. However, it must be added, in
defence of the real lfr, that some of these stories are Snorris constructions. Did Snorri need
an excuse for the magnates to turn against lfr, particularly for those who killed him? Did he
find it psychologically unlikely for such men to betray their king just for gold and silver? Or
did he simply examine the available sources for any trace of motives, developing those he
found into complete stories explaining the actions of lfrs main adversaries? In any case,
Snorri gives both a more complex account of the rebellion against lfr and shows greater
understanding for his adversaries. Ultimately, however, he shows them to have been wrong.
lfrs alleged tyranny was replaced by an even worse exercised by the Danes, and lfrs
holiness which Snorri did not doubt was used to throw off the Danish yoke and place
lfrs son on the throne.

Conclusion
Whereas the examination of the story of lfr as a whole shows a considerable amount of
continuity from Theodoricus to Snorri, the present account of his conflicts with the magnates
points to Snorris almost revolutionary intervention in the tradition. He is the first to attempt a
consistent or almost consistent interpretation of the conflicts that led to lfrs fall. He is also
the first to create a consistent chronology out of the mass of separate stories, most of which
were unrelated to one another in the earlier tradition. From a present-day point of view of his-
torical truth, this revolution has not been without costs: originally totally unrelated stories are
linked together and not only speeches, but individual persons and their actions have been in-
vented. The result, however, is an entirely new kind of narrative.
How do these observations fit it with the general development of the saga literature? This
development has recently been dealt with by Theodore M. Andersson, whose focus is mainly
on the Icelandic family sagas and for whom the final stage in the evolution is represented by
Njls saga from the late thirteenth century (Andersson 2006: 2159, 86101 and 183210).
Although the family sagas and the kings sagas have much in common, there are also differ-
ences between them. The kings sagas show closer similarity to the classical and contempo-
rary Latin historiography, through features like prologues, invented speeches and above all
chronology. Theodoricuss work is an example of advanced, theological historiography al-
ready around 1180, whereas Historia Norwegie represents a more classicizing Latin tradition.
However, the influence from these traditions on vernacular historiography is more difficult to
ascertain. The dry, terse style of grip and to some extent Fagrskinna may have been influ-
enced from Theodoricuss Latin prose, but Ari, who writes in the same style in his extant
slendingabk, seems a more likely source of inspiration. The step from this listing of facts to
epic narrative is taken in Oldest Saga, today mainly known from Legendary Saga, and the
approximately contemporary saga of lfr Tryggvason by Oddr munkr, in both with serious

76
costs regarding coherence and consistency. Most of the epic material is clearly derived, di-
rectly or indirectly, from oral tradition, but its organisation in a longer narrative has presented
a problem. We may nevertheless wonder whether the chaotic narrative of these works can be
explained by the relationship between written text and oral performance at this stage. We
know that texts at this time, and largely also later, were meant to be read aloud, perhaps also
to serve as a source for oral storytelling. Could we imagine the author of Legendary Saga in-
cluding a number of different versions in his text, not because of carelessness, but in order to
have a variety of material available for various oral performances? The development from this
saga to Snorri would then mean a change in the status of the text, from a raw material for oral
performance to literary prose, intended to be performed in one particular way.
Between these two stages, we meet an early masterpiece, Sverris saga, not dealt with by
Andersson, probably because of its limited importance for the development leading to Njls
saga. Here the combination of vivid narrative and precise chronology is already perfect and
the individual episodes serve to explain major changes in the relationship between Sverrir and
his adversaries. Sverris saga also, with the exception of the very early part, Gryla, probably
only covering the period until 1178, represents the same objective narrative as Snorris works
and is likely to have served as his model. There is, however, the great difference that Sverris
saga deals with contemporary history where at least a relative chronology was easy to estab-
lish, whereas Snorri had no evidence for his chronology of lfrs reign. The significance of
Sverris saga for the development of the sagas is difficult to establish, because of the uncer-
tainty about its date. Gryla can be dated to 118588, and at least a major part of the saga may
have been written already during Sverrirs lifetime, but most of the saga may also be as late as
from around 1220 (Bagge 1996 1518; Krag 2005: 4648 and orleifur 2007: LX f.). This
uncertainty, together with the general uncertainty about the dates of the kings sagas and the
fact that most of them are after all written within a relatively short period of time, should warn
us against drawing too firm conclusions about their development from one stage to another;
we may also imagine the coexistence of various approaches. Nor is the latest necessarily the
best.

Bibliography
Andersson, Theodore M, 2006: The Growth of the Medieval Icelandic Sagas (11801280). Ithaca.
Bagge, Sverre, 1991: Society and Politics Society and Politics in Snorri Sturlusons Heimskringla.
Berkeley etc.
Bagge, Sverre, 1996: From Gang Leader to the Lords Anointed. Kingship in Sverris saga and Hko-
nar saga Hkonarsonar. Odense.
Bagge, Sverre, 2002: Mellom kildekritikk og historisk antropologi. Olav den hellige, aristokratiet og
rikssamlingen, Historisk tidsskrift 81, pp. 173212.
Hkr. Hkonar ga = Heimskringla, Hkonar saga ga, ed. Finnur Jnsson. Copenhagen 18931901,
I, pp. 16522.
HkrOH = Heimskringla, lfs saga helga, ed. Finnur Jnsson. Copenhagen 18931901, II.
Jochens, Jenny. 1996. Old Norse Images of Women. Philadelphia.
Krag, Claus, 2005: Kong Sverre. Norges strste middelalderkonge. Oslo.
Leg. Saga = lfs saga hins Helga. Die Legendarische Saga ber Olaf den Heiligen (=Leg. Saga),
edd. and transl. A. Heinrichs et al. Heidelberg 1982.
Moberg, Ove, 1941: Olav Haraldsson, Knut den store och Sverige: studier i Olav den helliges frhl-
lande till de nordiska grannlnderna. Lund.
Schreiner, Johan, 1926: Tradisjon og saga om Olav den hellige. Oslo.
Otte Brudstykker af Den ldste Saga om Olav den hellige, ed. and intr. G. Storm. Christiania 1893.
orleifur Hauksson. 2007: Sverris saga, slenzk fornit 30, ed. and intr. Reykjavk.

77
Gfuct dr ec heiti: Deer Symbolism in Sigurr Ffnisbani?
Massimiliano Bampi, Dipartimento di Scienze del Linguaggio, Universit Ca Foscari
Venezia, Italy
One of the most intriguing aspects of the complex and fascinating figure of Sigurr Ffnisbani
is that in a number of Old Norse texts which tell in various ways of the heros deeds, he is
compared, more or less explicitly, to a stag, or his life is connected in some way to this ani-
mal. As is widely known, these texts are Gurnarkvia II, Vlsunga saga, Ffnisml and
ireks saga. In the first two texts the hero is explicitely likened to a stag endowed with ex-
traordinary features, while in ireks saga it is recounted how the parentless Sigurr is raised
up and nourished by a hind in the woods. More controversial is the interpretation of stanza 2
of Ffnisml, where the hero replies to a question posed by the dying dragon by defining him-
self as a gfuct dr: this phrase has given rise to a prolonged and lively discussion in Old
Norse scholarship and will be commented on in the sections to follow. In the same eddic
poem, however, the reference to Hindarfjall (the mountain of the hind), where Sigurr will
meet the sleeping valkyrie, is part of the same deer imagery. In addition to these occurrences,
one should also mention that one of the descendants of Sigurr the dragon-slayer is called
Sigurr hjrtr (the stag).
An obvious major question arising from the occurrences of the image of the stag is why the
hero is presented as such. Undoubtedly, any possible answer to this question depends first and
foremost on the identification of the symbolic values which are likely to have been attached to
the image of this animal in the Middle Ages. Indeed, the occurrence of this comparison in all
the texts cited above certainly invites a symbolic interpretation. With regard to this point, one
should ask against which cultural background one should interpret this image to work out its
symbolic meanings. Should it be assessed as part of a heathen imagery or is it rather a Chris-
tian motif, or an elder motif that acquires new meanings according to Christian symbolism? In
the present paper some reflections on one possible interpretation of the stag imagery in the
depiction of Sigurr will be proposed. Since the scholar who is willing to embark on such a
problem-ridden interpretive enterprise is faced with a considerable number of issues, extreme
caution will be necessary in dealing with this topic. Indeed, a major problem is certainly rep-
resented by the multi-layered nature of symbols.
Let us now present each single occurrence of the deer imagery in the texts mentioned
above.
In Gurnarkvia II (henceforth Gr. II) Gurn complains to irekr about her sorrow
and looks back to the time when she was married to Sigurr. In st. 2 she describes her hus-
band as follows:

Sv var Sigurr uf sonom Gica

sem vri grnn laucr r grasi vaxinn,
ea hiortr hbeinn um hvssom drom,
ea gull glrautt af gr silfri (Neckel 1983: 224)

From this passage it becomes clear that the three terms of comparison used by Gurn corre-
spond to the intention to describe Sigurr as an outstanding hero, especially in comparison
with the sons of Gjki. Interestingly, this description finds a thorough comparison in the
words uttered by Sigrn to praise her husband, Helgi Hundingsbani (Helgakvia Hundings-
bana II, 38):

78
 Sv bar Helgi af hildingom
sem trscapar ascr af yrni,
ea s drklfr, dggo slunginn,
er fri ferr llom drom,
ok horn gla vi himin silfan. (Neckel 1983: 158)

Here Helgi is compared, among other things, to a young stag whose antlers shine towards the
sky. What is worth noticing is that the rhetoric and stylistic pattern is the same as the one used
in Gr. II: the extraordinariness of the hero is expressed and underlined by making use of im-
ages taken from the world of plants and animals. Furthermore, in both of them the image of
the stag is used, whereas the plant names used are different.
Closely connected to Gr. II are the relatively numerous attestations to be found in Vl-
sunga saga. In ch. 27 [25] Gurn tells Brynhildr about an ominous dream she had the night
before:

at dreymdi mik, sagi Gurn, at vr gengum fr skemmu margar saman ok sm einn

mikinn hjrt. Han bar langt af rum drum. Hr hans var af gulli. Vr vildum allar taka drit,
en ek ein na. Drit tti mr llum hlutum betra. San skauztu drit fyrir knjm mr (Guni
Jnsson 1950: 173174).

Brynhildr herself contributes to the interpretation of this dream by making Gurn understand
that the big stag (mikinn hjrt) in the dream is Sigurr himself. As in the case of Gr. II
briefly presented above, also here one notices that the occurrence of the image of the stag as a
representation of the hero responds to the intention to underline his greatness and uniqueness.
This is made clearer through the use of expressions aiming at enhancing the status of Sigurr
(hann bar langt af rum drum; Drit tti mr llum hlutum betra).
In ch. 34 [32] Gurn gives voice to her sorrow following the same rhetoric pattern and
drawing from the same metaphoric repertoire as in Gr II. Sitting in her own room at Atlis
court, she recalls her husband and the happy times when he was still alive:
Betra var vrt lf, er ek tta Sigur. Sv bar hann af llum mnnum sem gull af jrni
ea laukr af rum grsum ea hjrtr af rum drum. (Guni Jnsson 1950: 194195)
Two further elements in Vlsunga saga clearly point to the stag imagery: one is the refer-
ence to Hindarfjall (especially chs. 20 and 21) and the other is the quotation of the very same
controversial strophe in Ffnisml mentioned above.
Further instances of the same deer imagery are given in ireks saga. In ch. 162 [267] it is
told that Sigurr has no parents and that he has been raised and nourished by a hind in the
forest:

Nv kom ar at t hind oc tecr barnit imunn ser oc berr heim til sins blis ar atti hon .ij. born.
ar legr hon sveinen nir. oc ltr sveinen drecka sic. oc fir hon han sem sin born. (Bertelsen
19051911: 302303).

In addition, in the scene depicting the quarrel between the two queens (ch. 388 [343]) Bryn-
hildr rudely invites her opponent to go in search of Sigurr following the path of the hind.
Last but not least, let us turn to the controversial phrase in Ffnisml. The dying dragon
asks the hero about his own identity. Sigurr replies by defining himself as a gfuct dr
(noble beast?).1 As was briefly hinted above, scholars are not agreed on the interpretation of
this expression (von See et al. 2006: 402404). Most of them explain it as a reference to the
1
It has generally been considered to be implausible that this phrase could refer to the account of Sigurrs being
nourished by the hind to be found in ireks saga. See von See et al. 2006: 402.

79
stag. Others have put forward an ofljst kenning hypothesis according to which the phrase is a
pun on the name Sigurr (Gade 1990: 65). In particular, Gade claims that the name Sigrr,
an alternative form which she assumes as the basis of her interpretation, is a circumlocution
for Hildisvni, Freyjas golden bristled boar from Hyndloli (Gade 1990: 65)2. For reasons
that will be explained later, the stag hypothesis appears to be altogether more plausible.
Starting from all these occurrences, in the following sections the main concern will be to
weigh up one possible hypothesis regarding the approach to a symbolic interpretation of the
image of the stag in the description of the famous dragon-slayer.
Otto Hfler (1961) carried out a thorough investigation into the stag imagery connected
with Sigurr. Yet the results of his study are heavily biased. Indeed, the whole interpretive
architecture of his argument is built on the assumption that Sigurr/Siegfried is to be identi-
fied with Arminius, a chieftain of the Cherusci who defeated a Roman army in 9 A.D near the
Teutoburg Forest. According to Hfler (1961: 27), the ethnonym Cherusci would contain the
word *herut- (deer). Hence, from this viewpoint the occurrence of the stag imagery in the
description of Sigurr/Siegfried alias Arminius would come as no surprise. Furthermore,
Hfler also claims that this imagery has to be seen as a clue to the existence of a stag-cult of
which the stag imagery represents a reminiscence. The whole line of reasoning is based on
weak grounds and has been widely questioned by most scholars (see, among others, von See
1981: 3941). A thorough re-evaluation of Hflers argument is certainly beyond the scope of
this essay. Yet there is at least one major point that needs to be underlined: the textual occur-
rences briefly examined above clearly point to a metaphoric interpretation and do not provide,
as they are, any evidence supporting a cultic hypothesis, as in the case of Hfler. Hence, the
imagery under study should be analyzed as a literary metaphor (von See 1981: 40) employed
to exemplify certain features of the hero and of his life, a means to give symbolic emphasis to
the praise of Sigurrs stature as an outstanding champion. In particular, what has to be de-
termined is whether the symbolic meaning of this animal in this specific cultural context
should be assessed according to a system of pagan cultural coordinates or according to a
Christian vision of the world.
At least one major point is clear: as regards the origins of the image of the stag, there is no
doubt that it ultimately traces back to pre-Christian times. As Steuer (1999: 588) points out,
the stag hat als stattliches Jagdtier und mit seinem prchtigen Geweih als eindrucksvolle
Gestalt immer eine hervorragende Rolle gespielt, was ber die Zeiten und die Kulturen hin-
weg zu mannigfaltiger bildlicher und plastischer Wiedergabe gefhrt hat.
As to its major symbolic values, in pre-Christian times it was associated, among other
things, with prosperity, rebirth and rejuvenation, with regeneration and fertility.3 Furthermore,
it was often connected with sun symbolism (Steuer 1999: 588), mainly because his horns
were compared to sunrays. Within the Germanic world, the whetstone sceptre decorated with
a little stag found at Sutton Hoo has led most scholars to see the stag as a royal symbol, at
least as far as the Anglo-Saxon world is concerned (Simek 2006: 181; Ellis Davidson 1988:
57). This animal is also present in Old Norse mythology, where it appears among the cosmic
animals (Heizmann 1999: 604). Stags appear both in the mythological section of the Poetic
Edda (Grmnisml, sts. 26, 33, 35) and in Snorra Edda. Four stags (Grmnisml, st. 33;
Gylfaginning 16) are described as feeding on the ash tree Yggdrasill. Furthermore, another
stag, called Eikyrnir, stands on top of Valhll (Grmnisml, sts. 2526; Gylfaginning 39) and
bites from the branches of the tree. Copious drops falling from his antlers reach Hvergelmir.
Whatever the origin and the hypothetical original meaning of the image of the stag in pre-
Christian times in Scandinavia, here I will focus on one possible explanation that may help us

2
On this interpretation see von See et al. 2006: 403.
3
For a discussion of the the stag as a symbol of regeneration and fertility see Heizmann (1999: 598-600).

80
account for the representation of the stag as a paradigm of nobility and excellence in the de-
scription of Sigurr. Indeed, the question that this paper aims at raising is whether it is possi-
ble to propose an interpretation of the symbolic meaning of the stag associated with Sigurr
against the Christian background of the culture within which the writing down of the story of
this hero has taken place. As a matter of fact, the texts containing references to Sigurr as a
stag are dated in their present form to the 13th century. Vlsunga saga is generally dated to
around 1260, i.e. after the coming into being of the Codex Regius of the Poetic Edda (Wrth
2003: 101). As is widely known, the saga draws from the material contained in the heroic lays
of the Poetic Edda, among them Ffnisml and Gr. II. Although their composition dates
back prior to the redaction of Vlsunga saga, these two lays are extant in their oldest written
form in the Codex Regius. ireks saga is generally dated to around the middle of the 13th
century, and is held to have been compiled at the court of Hkon IV Hkonarson in Bergen
(Kramarz-Bein 2002).
A Christian symbolic re-interpretation of the stag is deeply rooted in the Holy Scripture
(especially in the Song of Songs and in Psalm 42) and in the writings of the Church Fathers. In
particular, they see in the young stag appearing in the Song of Songs, among other things, a
representation of Christ himself (Domagalski 1991: 569).
As Pastoureau (2004: 75) points out, [l]aissant volontairement de ct les aspects ngatifs
et sexuels de la symbolique du cerf, les Pres et les thologiens en font un animal pur et
vertueux, une image du bon chrtien, un attribut ou un substitut du Christ.
Also in Old Norse literature the same kind of Christian interpretation of the stag is well at-
tested.
A stag as a clear symbol of Christ appears, for example, in Plctus saga, the Old Norse
version of a hagiographic text telling of the conversion of Eustace to Christianity that was
widely circulated in the Middle Ages. During a hunt, Placidus and his retinue run into a herd
of stags. He decides to run after the biggest of them, which soon reveals itself to be Christ and
converts the Roman warrior.
Another noteworthy text bearing witness to the same kind of symbolic interpretation of the
stag is the Christian didactic poem known as Slarlj from the beginning of the 13th century
(Njrur P. Njarvk 1991: 7; Simek-Plsson 1987: 329). St. 55 reads as follows:

Slar hjrt
leit eg sunnan fara,
hann teymdu tveir saman.
Ftur hans
stu foldu ,
en tku horn til himins.
(Njrur P. Njarvk 1991: 30)

Here the stag is unanimously considered to be an incarnation of Christ (Amory 1990: 259;
see also Njrur P. Njarvk 1991: 84). Interestingly, in the text the tremendous size of the
animal is brought to the fore: the horns reach up to the sky. This invites comparison with Hel-
gakvia Hundingsbana II (st. 38), as Amory (1990: 259) points out. Indeed, as was mentioned
earlier, also in the eddic poem the young stag bears big horns shining towards the sky.
A third text unmistakably mentioning the stag as a symbol of Christ is the so-called Physi-
ologus, of which two fragmentary translations exist in Old Icelandic. Basing on Psalm 42, in
the chapter dedicated to the illustration of the symbolic meaning of the stag the animal is de-
scribed as fighting against the snake. A contraposition between the stag and the snake ulti-
mately traces back to pre-Christian times and is also attested in the Scandinavian world, e.g.

81
on the Skrydstrup B bracteate (Heizmann 1999: 600). Yet, in Christian symbolism the contra-
position is turned into a fight between the good and the evil, between Christ and the devil.
Given these occurrences of a Christian interpretation of the stag as a symbol of Christ, the
question arising from the comparison between the stag imagery connected with Sigurr and
the instances briefly discussed above is whether one may find a connection between them. In
other words, one is led to wonder whether the Christian Scandinavians to whom the narratives
about the dragon slayer were addressed used to interpret the stag imagery connected with
Sigurr against the background of Christian symbolism as attested in the three texts men-
tioned above.
I think that clues to a possible Christian influence can be detected on a formal-stylistic
level. As was mentioned above, in Gr. II and in Helgakvia Hundingsbana II the same pat-
tern to describe the excellence of the hero is deployed. As Klaus von See (1981: 4041) points
out, this type of descriptive scheme has its roots in the Holy Scripture, especially in the Song
of Songs, and is widely attested in the religious literature of the Middle Ages. As a conse-
quence, he comes to the conclusion that this type of metaphor in Gr.II more generally in
the Gurn poems is very likely to have been stylistically influenced by religious literature
(von See 1981: 41).
A further aspect that pertains to the stylistic level regards the way Sigurr as a stag is de-
scribed in the texts. In Gr. II (hirtr hbeinn) and in Vlsunga saga (einn mikinn hjrt.
Han bar langt af rum drum), the emphasis is placed on the size of the animal to symbol-
ize its superiority. In fact, the same can be observed in the case of the description of Helgi as
well. A similar emphasis on the size of the beast is found in Plctus saga, where the stag
incarnating Christ is described as follows:
En er ollum riddurum var skipat til veidarinnar, a syndiz Placido einn hiortr ollum odrum
meiri (Unger 1877: 193; my emphasis)
Here the crucifix-bearing stag is depicted as the biggest of the whole herd into which
Placidus and his retinue run. As was seen above, in Slarlj the sun-stag is described as even
bigger.
Given the relative chronology of the texts examined, I think that a direct influence of
Plctus saga on the heroic description of Sigurr as a stag cannot be excluded.4
Let us now turn to Ffnisml. As was mentioned previously, scholars do not agree on the
interpretation of the self-definition (gfuct dr) given by Sigurr at the opening of the dia-
logue with the dying dragon. Nevertheless, an identification with a stag appears to be plausi-
ble for two major reasons. First, from what has been observed so far it is clear that a connec-
tion with the stag is part of the imagery associated with Sigurr. Indeed, the occurrences in
texts, other than Ffnisml regardless of their genealogic relationships attest to this fact.
Second, as was seen above the stag was widely considered to be the enemy of the snake. In
the Middle Ages, dragons were generally conceived of as big serpents (Homann 1986: 132),
and this is certainly true also of the dragon Ffnir. Indeed, in Reginsml it is told that Ffnir
l Gnitaheii oc var orms lki (Neckel 1983: 176; my emphasis). Hence, it is tempting to
see behind the self-definition of Sigurr as a stag an intention to emphasize the contraposition
with the dragon. Such an interpretation invites a further step towards a Christian reading of
this phrase. Could such a contraposition be interpreted in Christian terms?
As Ashman Rowe points out, Sigurr continued to be a suitable subject in certain Chris-
tian contexts, for series of scenes from his story decorate the portals of five Norwegian stave
churches from the twelfth and thirteenth centuries, and individual scenes are found on Nor-
wegian church sites such as door-jambs, capitals, fonts, chairs, and benches from the same

4
Plctus saga has come down to us in four versions, the oldest of which was written in Trondheim about 1150.
See Tucker (1993: 504).

82
period (2006: 168). Furthermore, as she convincingly demonstrates (Ashman Rowe 2006),
Sigurr and his deeds are subject to ethical and moral interpretations also in a number of Old
Norse texts. In addition, as regards the iconographic material, the hero is mostly depicted in
the act of slaying the dragon. The presence of scenes describing the killing of the dragon on
Norwegian church portals and church sites has brought about a lively discussion about the
possible interpretation in Christian terms. As Byock points out, the fact that Sigurr success-
fully fought against the dragon made him suitable, at least superficially, for reinterpretation
within a Christian context (1990: 624). Furthermore, he claims that the killing of Ffnir
paralleled the Christian understanding of the devil-monster menace, and as such was a cul-
turally mutable symbol, one that was not repugnant to the early Norwegian church (1990:
625).
An identification of Sigurr with Christ has been generally considered to be far-fetched,
especially because he was certainly known as a pagan hero to any Scandinavian in the Middle
Ages. Nevertheless, as a dragon slayer he may have functioned purely as a substitute, a
Scandinavian St George rather than a pagan antetype of St George (Ashman Rowe 2006:
193).
The use of what was employed mainly as a Christological symbol in the Middle Ages (i.e.
the image of the stag) may have been intended as a means of enhancing the exemplarity and
the nobility of the hero in such a way that he could become admirable to a Christian Scandi-
navian. The Church itself was responsible for turning the stag into a noble animal (a gfuct
dr?) by progressively promoting it to the royal game par excellence throughout all Europe.
Such a nobilitation was carried out also by attaching Christian symbolic values to this mild
animal, as the account of the conversion to Christianity of St Eustace and St Hubert clearly
indicates. Interestingly, stags begin to appear as gibier royale et princier (Pastoureau 2004:
76) in the Arthurian literature in the second half of the 12th century. The topos of stag chase
inaugurated in rec et nide by Chrtien de Troyes is taken up again and again in courtly
literature throughout the 13th century (Pastoureau 2004: 76) and finds its way to Scandinavia
through the translated riddarasgur. In its role as royal game, the stag comes to be closely
connected to nobility and kingship.
Furthermore, at this point it must be mentioned that Sigurr was considered as an ancestor
of the Norwegian royal house (Byock 1990: 621; Ashman Rowe 2006: 193). Hence, it is
plausible that the decision to liken the hero to the stag can be seen as a further instance of the
strategy to bestow upon him royal and courtly attributes that has been acknowledged by most
scholars. Such a strategy is evident especially in the account of Vlsunga saga (Wrth 2003:
108) but can also be detected in what has been called the Jungsigurddichtung (comprising,
among others, Ffnisml), where Sigurr is presented throughout as a noble son of king, as a
hero whose behaviour is modelled after the behaviour of the righteous king (Sprenger 2000:
128).
As far as ireks saga is concerned, it is worth noticing that the account of Sigurr being
nourished by a hind in the forest invites comparison with some hagiographic legends, in par-
ticular with the life of St Genoveva of Brabant (Kramarz-Bein 2002: 44) and of St Giles (St
gidius), where the same motif is well attested. However, hasty conclusions should not be
drawn from this analogy. Indeed, the complexity of the question certainly demands further
investigation and calls for caution.
All the thoughts proposed so far lead us to some concluding (albeit tentative) remarks. In
proposing a reading of the symbolic meaning of the image of the stag associated with Sigurr
from within the Christian context in which the texts have been written down, it has been ob-
served that the use of the stag metaphor may be seen as part of a royal imagery intended to
ennoble the hero, especially because of his role as mythic ancestor of the Norwegian monar-

83
chy. The use of rhetorical devices and motifs borrowed from religious texts point to the influ-
ence exerted by Christian symbolism on the representation of Sigurr as the noblest of heroes.

Bibliography
Amory, Frederic, 1990: Norse-Christian Syncretism and interpretatio christiana in Slarlj. In:
Gripla, 7 (1990). Pp. 251266.
Ashman Rowe, Elizabeth, 2006: Quid Sigvardus cum Christo? Moral Interpretations of Sigurr Ff-
nisbani in Old Norse Literature. In: Viking and Medieval Scandinavia, 2 (2006). Pp. 167200.
Byock, Jesse, 1990: Sigurr Ffnisbani: An Eddic Hero Carved on Norwegian Stave Churches. In:
Poetry in the Scandinavian Middle Ages: 12. Congresso internazionale di studi sullalto medioevo.
Ed. by T. Paroli. Spoleto. Pp. 619628.
Domagalski, Bernhard, 1991: Hirsch. In: Reallexikon fr Antike und Christentum. Sachwrterbuch
zur Auseinandersetzung des Christentums mit der antiken Welt. Stuttgart. Cols. 551577.
Edda. Die Lieder des Codex Regius nebst verwandten Denkmlern. Ed. by G. Neckel. 1983. Heidel-
berg.
Ellis Davidson, Hilda Roderick, 1988: Myths and Symbols in Pagan Europe. Early Scandinavian and
Celtic Religions. Syracuse.
Fornaldar sgur Norurlanda.1. Ed. by Guni Jnsson. 1950. Reykjavk.
Gade, Karin Ellen, 1990: Sigurr Gfuct dr: A Note on Ffnisml St. 2. In: Arkiv fr nordisk
filologi 105. Pp. 5768.
Heilagra manna sgur. Fortllinger og legender om Hellige Mnd og Kvinder. Efter gamle
Haandskrifter. Ed. by C.R. Unger. Christiania. 1877.
Heizmann, Wilhelm, 1999: Hirsch. Philologisches. In: Reallexikon der germanischen Altertumskunde.
14. Berlin/New York. Pp. 595612.
Homann, Holger, 1986: Drache. In: Reallexikon der germanischen Altertumskunde. 6. Berlin-New
York. Pp. 131137.
Hfler, Otto, 1961: Siegfried, Arminius und die Symbolik. Mit einem historischen Anhang ber die
Varusschlacht. Heidelberg.
Kramarz-Bein, Susanne, 2002: Die ireks saga im Kontext der altnorwegischen Literatur. Tbingen
und Basel.
Pastoureau, Michel, 2004: Une histoire symbolique du Moyen ge occidental. Paris.
See, Klaus von, 1981: Germanische Heldensage. Stoffe. Probleme. Methoden. Wiesbaden.
See, Klaus von, La Farge, Beatrice, Gerhold, Wolfgang, Picard, Eva,Schulz, Katja, 2006: Kommentar
zu den Liedern der Edda. 5. Heidelberg.
Simek, Rudolf Hermann Plsson, 1987: Lexikon der altnordischen Literatur. Stuttgart.
Simek, Rudolf, 2006: Lexikon der germanischen Mythologie. Stuttgart.
Slarlj. Ed. by Njrur P. Njarvk. 1991. Reykjavk.
Sprenger, Ulrike, 2000: Jungsigurddichtung. In: Reallexikon der germanischen Altertumskunde. 16.
Berlin/New York. Pp. 125129.
Steuer, Heiko, 1999: Hirsch. Archologisches. In: Reallexikon der germanischen Altertumskunde. 14.
Berlin/New York. Pp. 588595.
Tucker, John,1993: Plctus saga. In: Medieval Scandinavia. An Encyclopedia. Ed. by Ph. Pulsiano, K.
Wolf. New York/London. Pp. 504505.
iriks saga af Bern. Ed. by H. Bertelsen. 19051911. 2 vols. Kbenhavn.
Wrth, Stefanie, 2003: The Rhetoric of Vlsunga saga. In: Fornaldarsagornas struktur och ideologi.
Ed. by rmann Jakobsson, A. Lassen, A. Ney. Uppsala. Pp. 101111.

84
 Muslims in Karlamagnss saga and
Elss saga ok Rsamundar
Bjrn Bandlien, University Library, University of Oslo, Norway
By the thirteenth century, Scandinavians had been encountered Muslims for centuries; for
example during raiding in Iberia in the ninth century, trading in Rus in the tenth, as partici-
pants in the Varangian guard in Byzantium in the eleventh, and as pilgrims and crusaders in
the twelfth. Still, although archaeologists and historians have established links between the
Mulism world and Scandinavia from the Viking Age onwards, there have been very few stud-
ies on Scandinavian images of Muslims. In this paper, I will analyse the images of Muslims as
presented in Elss saga ok Rsamundar and Karlamagnss saga. These texts are especially
interesting not only because they were among the first full treatments of this question in the
Norse tongue, but also because the stories about Charlemagne and Elis were popular both in
thirteenth-century Norway, at a time when the Norwegian king promised to go on a crusade,
and in late medieval Iceland that played less of a part in physical battles against heathens.
They then offer an opportunity to understand the renewed use of these images in different
social context.
In recent years, however, the interest in medieval Christians popular and theological im-
ages of the Muslims has been substantial. The recent interest owes much to Edward Saids
perspective in his work on orientialism. Western views on Muslims are less interesting in
what they can reveal about our knowledge about Muslims, but rather with regard to what
these views can tell us about the viewer and the formation of western identity. The unraveling
of the medieval Western images of Muslims can tell us much about the formation of a sense
of Us in stories about encountering the Other. According to this approach, medieval Chris-
tian identity was sustained by elaborate, seemingly intractable racial fantasies centered upon
the supposed absolute otherness of Jews and Saracens (Cohen 2003: 187). Moreover, the
encounters and constructions of Saracens and Jews as Others, in crusades as well as in the
cultural imagination of romances, have been seen as crucial in the creation of a medieval dis-
course on nation (Heng 2003).
Of course, we can deduce several groups of non-Christian Others in Scandinavian
sources. Much has been written about how the heathen ancestors were depicted in the medie-
val sagas, and also some important studies have also been conducted on the Smi in the North
and on heretics. Less has been written on Muslims, even though many scholars in recent years
have emphasised the possible European influence on the Norse worldview. An exception is
John Stanley Martin who has discussed the transmission of attitudes towards Islam from the
chansons de geste to the Norse riddarasgur. These attitudes were a far cry from accurate im-
ages of Islam or Muslims, but rather misrepresenting these infidels as irrational worshippers
of wooden effigies and evil creatures (Martin 1990; 1991).
Martins conclusion is in line with much of the later scholarship on western images of
Muslims and Islam in the Middle Ages, as constructing them as the Others. However, stud-
ies on western attitudes towards Muslims as represented in medieval literature have recently
focused on the more complex and diverse images of the Muslims; punishers of sinful Chris-
tians, heretics, monstrous, irrational, or proto-Christians who might be converted (Tolan
2002). Scholars have also pointed out that many texts show an ambivalent attitude towards
Muslims; on the one hand they could be very human and chivalric, while at the same time
being fierce opponents of Christianity (Bancourt 1982; Jones 2002). In some texts, such as the
Chanson de Roland, it is emphasised that there is a conflict between traditional ways of mak-
ing peace between Muslims and Christians, through the paying of tribute (parias) in a feudal

85
context, and the emergence of a dualism between Christians (or rather Franks) and Pagans
(Kinoshita 2006: 1545)
Also in medieval Norway and Iceland do we find several different images of Muslims
(Bandlien, forthcoming). In this paper, I will focus on the relationship between representa-
tions of Muslims within two texts that have Christian-Muslim encounters as a main theme:
Elss saga ok Rsamundar and Karlamagnss saga. These texts were probably orginally
translated into Old Norse in the thirteenth century. Elss saga is an adaptation of the chanson
de geste Elie de Saint Gille made in Norway in the middle of the thirteenth century,1 but also
preserved in several manuscripts from fifteenth-century Iceland. Karlamagnss saga consists
of adaptations of ten different branches of the Charlemagne cycle. The branches are com-
monly presumed to have been translated independently in the thirteenth century, probably by
Icelanders, or some at the beginning at the fourteenth century, and then compiled into a long
version as they are now preserved.2
These peculiarities in the manuscript transmission of Karlamagnss saga make it difficult
to use as a straightforward remnant of the thirteenth century when the ten different branches
of the saga were probably translated. Still, I think that the evidence for a thirteenth century
translation of most parts of Karlamagnss saga is strong and that it should be interpreted as a
remnant of the great interest in the Charlemagne cycle among both the Icelandic and Norwe-
gian elite in the thirteenth century. Although they are of French or Anglo-Norman origin, it
seems promising to read these texts in their Norwegian and Icelandic setting with regard to a
wider problem: the Europeanization of Scandinavia.
Elss saga ok Rsamundar tells the story of the expulsion of Elis by his father, Duke
Juliens of helge Egidie (Saint Gille). After being knighted, Elis is disinherited by his father
and leaves the court in anger to seek adventures elsewhere. What neither Elis nor his father
are aware of is that the land is being invaded by heathens. Elis alone frees several noble
Christian captives from the heathens, but after killing many of them is captured and brought
to the land of the heathens. He manages to escape and gets help from a repentant robber, Ga-
lopin, only to be wounded outside the walls of the main heathen town, called Sobrieborg. He
is, however, rescued by Rsamunda, the daughter of the heathen king, Maskalbret. She has
fallen in love with the Frankish knight already by her fathers description of his valour and
good looks. When the heathen kingdom is threatened by another heathen king who wants to
1
The oldest version of Elss saga ok Rsamundar is preserved in the Norwegian manuscript known as De La
Gardie 47 fol. As it is preserved now, it contains four texts. First there is Pamfluss saga, a translation of the
story of Pamphils love for the beautiful Galathea. The second is a shorter text, a translation of a part of Guil-
laume de Conchess Moralium dogma philosophorum. Third comes Elss saga ok Rsamundar, and then finally
Strengleikar, a translation of Anglo-Norman lais, most of them by Marie de France. The manuscript is believed
on palaeographical reasons to have been written c. 1270 by a scribe working close to Bergen, the main city in
Norway at that time (Holm-Olsen 1940; Tveitane 1972). It is most likely a copy of an older manuscript, and thus
it is probable that a translation of Elss saga was committed during the reign of Hkon Hkonsson (121763)
2
Although commonly believed to have been mostly translations made in connection to the Norwegian court,
there are convincing arguments for an Icelandic provenance for at least some of these translations. This is espe-
cially the case of the translation of the Pseudo-Turpin Chronicle which Peter Foote dated to the early thirteenth
century (Foote 1959). Also manuscript fragments of parts of Karlamagnss saga preserved from the thirteenth
century points Icelandic scribes, although perhaps made for exportation to Norway (Stefn Karlsson 1992).
There are four Icelandic manuscripts from the fifteenth century that contain Karlamagnss saga, which repre-
sents two versions, usually designated A and B (Halvorsen 1989). The A version is believed to be closest to the
thirteenth century, since it does not contain parts that are believed to have been translated in late thirteenth or
early fourteenth centuries, such as the Anglo-Norman story of Olif and Landres and excerpts of Speculum His-
toriale by Vincent of Beauvais. Elss saga is preserved in a late medieval Icelandic manuscript with some differ-
ences and interesting additions compared to the Norwegian manuscript. The problem is how these stories of
encounters with heathens may have been understood in the changing contexts of the thirteenth Norwegian king-
dom and in late medieval Iceland, both in the light of the social milieu that produced the texts as well as the
context of the crusading and warfare.

86
make Sobrieborg into tax land and marry Rsamunda, the princess tricks her father into let-
ting Elis meet the rival in single combat. Elis wins, and the thirteenth century manuscript then
ends the story by telling how Rsamunda is willing to be baptised in order to marry Elis, since
he refuses to marry a heathen.
The structure of the saga is that of a hero who loses his inheritance, is tested through trials,
and then returns to his lands with a wife. The Muslims are then at first hand depicted as out-
siders to the Frankish kingdom, ruled by King Louis, son of Charlemagne. They are there
primarily to loot the kingdom and cause much distress. They have a huge army, but fear a
united Christian resistance to them. In this way, they are a useful opponent for a young knight
who wants to test his strength and prowess, and at the same time do not cause internal con-
flicts within the Christian kingdom but rather defend it.
In Elss saga, the followers of Maumet are often mocked as heathen dogs, or as devils
limbs.The heathens are identified by their belief in their false gods most notably Maumet,
but also Terrogant, Jupiter and Apollon are mentioned from time to time. These gods are of
course a misrepresentation of the Muslim belief, but during the twelfth and thirteenth centu-
ries this heathen pantheon had become a stock theme in the chansons de geste (Daniel 1984;
Tolan 2002). In Elss saga, Maumet is depicted as a carved image laden with gold and gems.
The heathens pray to Maumet and the other false gods and often swear by them, in order to
get help in the case of battle or to gain favours. Especially Rsamunda is depicted as a very
pious heathen. While her father, king Malkabert seeks Maumets help in defeating Christians,
the heathen princess kneels and prays to Maumet, especially for the health and life of her be-
loved Elis, but without the presence of the carved and decorated image.
The same pattern is visible in much of Karlamagnss saga. In the section Af Agulando
konungi, an adaptation of Pseudo-Turpin and Chanson dAspremont, the heathens bring the
wooden gods with them to the battles against the Christians.3 The wooden statues are also
decorated most lavishly with gold and gems. They become like relics brought to the battle-
field by the Christians, for example in the form of the sword Dyrumdalil that has relics in its
shaft. In this sense, the heathens belief in these gods is depicted as a kind of inversion of the
Christian faith.
In both sagas, a main theme is to show how useless the heathen gods are. Elis, as represen-
tative of the Christians, is offered peace, a high position and the hand of Rsamunda by King
Maskalbret if he shows his allegiance to Maumet. However, Elis, as a good Christian knight,
mocks the gods. They are helpless, he says, and all those who trust in them are fools. Al-
though enraged by this, the heathen king who has captured Elis cannot stop him from jumping
on his horse and escaping Sobrieborg. In a remarkable scene, King Maskalbret throws the
image of Maumet to the ground, denouncing him as helpless and refusing to help him get re-
venge on the Christian knight who has made such damage to him. Maskalbret nearly crushes
the image, but some of his advisors manage to restore him to his senses. Instead, the king
promises to give lavish offerings to the god if he will help him capture Elis.
In Karlamagnss saga, the heathen belief in gods is depicted in a fairly similar way. Hea-
then gods are carried into the battlefield in carved images, supposed to help them the heathens
against the Christians. However, the carved gods are captured and humiliated by the Chris-
tians. Again, the heathens question the power of their gods since they put up with this shame
and cannot help themselves. Still, the heathens mistake is that they do not take the conse-
quences of this and convert because they want revenge. Ultimately, because the heathens have
no help from their gods, the Christians get the upper hand, and in the few cases of conversion

3
An interesting exception is found in the dialogue between the heathen giant Ferakut and and Rollant in Af Agu-
lando konungi about their different faiths. It is based on Pseudo-Turpin but omitted in the B-version of Kar-
lamagnss saga.

87
it is the fact that the heathen gods show themselves unable to help their worshippers that initi-
ates it.4
But while there is, in Karlamagnss saga, much emphasis on the help the Christians get
from God, and especially St James, the success of the Christians in Elss saga is most of all
caused by Elis courage and strength, the love of Rsamunda, as well as the final assembling
forces of Christians that come to the heros assistance. Moreover, King Maskalbrets men are
not really great warriors; they often show cowardliness and only have confident in themselves
when in great multitude.5
Karlamagnss saga, on the other hand, has a much greater emphasis on the fight against
the Muslims as religious warfare. The Pope is time and again depicted as a leader who is
blessing the fights of Charlemagne and his peers, absolving all Christian warriors who fight
against Muslims. Archbishop Turpin the French declares that those who fall will be martyrs
of Christendom and go to heaven before their blood runs cold. In this respect, the crusading
ideology is clearly represented (cf. Stuckey 2008), and the identity of the Muslims is a reli-
gious one.
Still, there are other elements which complicate this picture. Although Muslim belief is
crudely misrepresented in the Norse versions, the Muslim world seems still remarkably alike
the Christian society. Heathen society is depicted as being ruled by regional kings who con-
trolled a fixed hereditary territory, but with over-kings they paid taxes to and that led the
troops in battle. In order to explain the defeat of the heathens, both Elss saga and several of
the branches of Karlamagnss saga seek to understand the internal conflicts and strategies
within the heathen world.
These complex depictions of the internal relations between heathens are noteworthy. In
Elss saga and Af Oddgeiri danska, the Christians support their previous enemies when there
is a third party from another place in the heathen world involved. The fighting between Chris-
tians and heathens suddenly emerges as more complex when other heathen intruders appear,
intent on overthrowing the very antagonist that the Christians are fighting against. Especially
in Elss saga and Af Oddgeiri danska, the Christians suddenly find themselves defending the
very heathens they were about to defeat. In theses cases, the Christians are open for a poten-
tial alliance to a group of Muslims, even though these were not converted.
In Af Agulando konungi, there is a version of the internal struggles within the ranks of the
heathens. Some heathens, who have abandoned a battle against the Christians, are punished
by King Agulandus in a shameful way. This makes one of their kinsmen very angry, and as a
revenge he and his troops leave the final battle against Charlemagne. Furthermore, the son of
the heathen King Agulandus, Jamund, has his own agenda in his fight against the Christians,
as he wants to secure Spain for himself on the advice on his foolish advisors. In these cases,
the rules of the heathen society are quite similar to those of the Christian world. Symptomati-
cally, fear for internal strife between Christian nobles is expressed several times in Kar-
lamagnss saga. The conflict between Girard and Charlemagne in Af Agulando konungi, Rol-
lants troubled loyalty to the emperor in Af Guitalin saxa when he is struck by Charlemagne,
and Elis anger at his father are tensions that are all downplayed when facing the heathen en-
emy. A common Christian cause is thus very useful in order to bring loyalty to the Christian

4
When heathens refuse to convert, it is partly because failures of the Christians to honour priests and poor (the
case of Agulundus), that it would seem to be because of cowardice (Agulandus), or because the heathens will not
abandon the faith of their forefathers and their loyalty to a heathen king (the cases of Karvel in Af Oddgeiri dan-
ska and Balam in Af Agulando konungi who refuses to convert, although he wants to, until his lord Jamund is
defeated).
5
Especially in Elss saga, King Maskalbrets knights are depicted as rather cowardly, refusing to meet the feared
heathen king Julien of Baldursborg.

88
realm (or, in some cases, to outsmoke deceivers, such as Guinelon (Af Runzivals bardaga), or
Milon (Af Olif ok Landres)).
Although most heathens in these two sagas come from Africa or Palestine, they are surpris-
ingly little distinguishable from the Christians. Both skin colour and their outward appear-
ance, such as beards and weapons are depicted similarly. The beautiful Rsamunda is marked
by her white colour; she is even whiter than snow. This also applies to men; sometimes the
warriors on each side are confused. A Christian army in Karlamagnss saga, for instance,
almost makes a grave mistake when they think that an approaching division is heathen. They
even enter battle, and are only stopped when they are near enough to ask about each others
names and family relations. Moreover, one of the distinguishing traits of Charlemagne, his
dignifying white beard, is copied by several heathen kings. Most notably, Af Agulando
konungi repeatedly states how two of the most fierce opponents against Charlemagne would
be the best knights, if only they were Christians. Of course they are ultimately killed and go to
hell, and some are more cowardly than Christians, but in valour and knightly virtues they are
of a similar kind as the Christians.
The outward appearance thus seems to be of no value for identification in these two sagas.
Still, there are blmenn present. In Karlamagnss saga, these are usually Ethiopians, distin-
guishable from other people of Affrica. But they are not the stock type of the magical cun-
ning or near demonic features that sometimes are sometimes applied in legendary sagas (often
in the phrase of berserkir and blmenn; cf. Lindow 1995). The blmenn, as with other Mus-
lims, are more often marked by their multitude and their bold and fierce kings. Blmenn
might be even more fierce in battle than other heathens, but not in a markedly supernatural
way.
In Af Olif ok Landres, there is a remarkable example of the elusive category of blmenn.
A wicked counsellor accuses Queen Olif (falsely) to sleep with a blmar. The counsellor
convinces the king that the boy Landres is the son of this blmar. Landres is even called
enna blmanns son (Kms 1860, p. 64), even though he is not marked by his colour in any
way elsewhere in the ttr.6
The religious difference still remains important in both Elss saga and Karlamagnss saga,
with Christianity as being the hegemonic religion. This religious difference could be, and was,
used to legitimise the killings of, and refusal of paying taxes to the Muslims. Still, Muslims
were not always depicted as being the monstrous Other. Although the category of Other is
crucial to create the identity of Us, people do not categorise others by using a simple Us vs.
Them-dichotomy. The anthropologist Thomas Hylland Eriksen distinguishes between ana-
logue and digital otherness. When others are almost like us or not so different from us,
their otherness is analogue. They are different in degree, not in kind. Digital otherness, on the
other hand, means that outsiders are fundamentally different from us. One example from the
Middle Ages would be the monstrous races; they were indeed various in appearances and
characteristics, but were lumped together as non-humans. Thus, these different groups of
others are then more or less of the same kind of otherness, despite their variation (Eriksen
2002).
If we apply these categories on the two sagas discussed here, it is clear that the Muslims in
these two sagas were analogue, rather than digital, others. They were knights like the Chris-

6
Unlike for example Feirefiz, the son of a Christian king and Saracen queen in Wolfram von Eschenbachs Par-
zifal, and who turns out to have spots of white and black all over his skin. In the later Icelandic versions of EsR
and Kms, the manuscript context may point to another reading. In fourteenth and fifteenth century sagas, the
Muslims and Africans are much more often depicted as irrational and monstrous blmenn and berserks. This
alliterative phrase also creeps in into a passage of Elss saga, when the the heathens are called berserkia ok
blamanna, as in many of late medieval fornaldarsgur and riddarasgur (p. 19). However, later in the same
manuscript Rsamunda still is white.

89
tian warriors, dressed like Christians, and looked and thought very much the same. Religious
difference needs to be emphasised time and again by means of expressions like heathen
dogs or they sure went to hell. Still, sometimes the chivalric identity shines through the text
more visibly than a clear-cut religious one. There is thus a marked negotiation of who the
Muslims really are, drawn between noble opponents and foolish and irrational dog-like hea-
thens. In the late Icelandic redaction of Elss saga, the religious identity of the king of Sobrie-
borg disappears altogether. The alliance between Christians and heathens is being made, al-
though king Maskalbret is never said to be baptised.
Interestingly enough, Norwegians and Icelanders were quite often in peaceful contact with
Muslims during the period of the writing of DG 47 fol., and presumambly also at the time of
the adaptations of the branches of Karlamagnss saga, Norwegians and Icelanders were quite
often in peaceful contact with Muslims. In 1262, Hkon Hkonsson sent an envoy to Soldn
of Tunis. It has been suggested that the two Norwegian messengers tried to ensure Tunis
neutrality for Alfonsos planned crusade. At least partially, the reason for the visit seems to be
to give hunting falcons as gifts to the ruler, possibly to sell a few on the market, but also to
make an alliance in the wake of the crusade of Louis IX. The emir of Tunis was in any case
not seen as a monstrous other, but rather as an exotic and powerful ruler who respected King
Hkon.7
In 1347, a letter from King Magnus Eriksson indicates that falcon trade continued to be
important in Scandinavia. He managed to get papal permission to trade with Soldan of
Babilonia, in order to improve the kingdoms economy. What King Magnus wanted to export
were falcons, something that for long had been very profitable for the Scandinavian kings
(Hofmann 195758). The falcon trade seems quite important both for Norwegian traders and
in diplomatic relations in the thirteenth and the first half of the fourteenth century also in the
Mediterranean. This was the period during and after the translation of chivalric romances and
apostolic vitae into Old Norse. Besides the interest for blmenn and worshippers of Muham-
med as demonic idolaters, then, there seems to have been one for narratives that depicted
Babilonia in a more favourable manner. The cross-cultural contacts in the Mediterranean also
influenced the learned and aristocratic world-view in the North, and the Norse aristocrats
aspiration to courtliness may have made it more appealing to admire the rich culture they met
in, for instance, Tunis and Egypt, than to simply depict them as digital others. This context
may at least offer a partial explanation for the tensions between the two views of Muslims in
thirteenth-century Norway; partly as heathen dogs that should be slain, and partly as exotic
allies that one could trade and make allies with against the bad heathens.
This is a less acute situation in fifteenth century Iceland, the time in which most of the ex-
tant manuscripts were written. In late Icelandic riddarasgur, the will for reconcilation be-
tween antagonists through the exchange of women is very strong (cf. Bagerius 2008). This
may align with the new ending of Elss saga in the late medieval manuscripts (Elss saga, pp.
116139), which strongly differs from the Norwegian (and certainly the Old French) version.
In the Icelandic ending, Elis marries the daughter of the heathen king. King Maskalbert thus
becomes a family member and a close ally to the French kingdom, despite the fact that noth-
ing is mentioned of any conversion. Even though he remains a heathen, he is included in the
network of friends of the French king. The focus in this part of the saga seems to lie less on
religious conflict than on the question of how a marriage can transform former enemies into
peaceful allies in the best interest of community. Possibly the aristocracy in Iceland wanted to
7
Elss saga, p. 4; cf. a similar phrase in Holm perg. 6 fol. where travels of kynstora kaupmenn is emphasised,
Elss saga, p. 9 (D), while Elie de Saint Gille might be more skeptical to traders, cf. Flori 1984. It is also interest-
ing that in the Norwegian Speculum regale, or Konungs skuggsj, written probably in the late 1250s, there is
stated that Norwegian merchants would often find themselves in dangerous situations, both at sea and in heathen
lands. Still, they were advised to respect the local customs wherever they were in order to be well received.

90
emphasise the chivalric values that distinguished the contestants, as well as the exotic setting,
rather than dwelling on the dehumanising images of the worshippers of Maumet.

Bibliography
Aalto, Sirpa, 2006, Categorizing Otherness in Heimskringla. In: The Fantastic in Old Norse / Ice-
landic Literature. Sagas and the British Isles, 13th International Saga Conference. Ed. by J. McKin-
nell, D. Ashurst, D. Kick. Durham. Pp. 1521.
Bagerius, Henric, 2008. En annan Tristram. Homosociala tankemnster bland senmedeltidens aristo-
krater p Island. In: Vi ska alla vara vlkomna!. Nordiska studier tillgnade Kristinn Jhannes-
son. red. Auur G. Magnsdttir, et al. Gteborg. Pp. 105115.
Bancourt, Paul, 1982: Les musulmans dans les chansons de geste du cycle du roi. Aix en Provence.
Bandlien, Bjrn, forthcoming: Images of Muslims in Medieval Norway and Iceland. In: Images of the
Other in Medieval Europe. Ed. by J. Mller Jensen, C. Hgel. Turnhout.
Brunschvig, Robert, 194047: La Berbrie Orientale sous les Hafsides: Des Origines a la fin du XVe
sicle, 2 vols. Paris.
Cohen, Jeffrey Jerome, 2003: Medieval Identity Machines. Minneapolis.
Daniel, Norman, 1984: Heroes and Saracens. An Interpretation of the Chansons de Geste. Edinburgh.
Elis saga ok Rosamundu. Ed. by E. Klbing. 1881. Heilbronn.
Eriksen, Thomas Hylland, 2002: Ethnicity and Nationalism, 2nd edn. London.
Flori, Jean, 1984: Lidologique aristocratique dans Aiol. In: Cahiers de civilisation mdivale. Vol.
27. Pp. 205221.
Foote, Peter, 1959: The Pseudo-Turpin Chronicle in Iceland. A contribution to the study of the Kar-
lamagns saga. London. (London Medival Studies, Monograph 4.)
Halvorsen, Eyvind Fjeld, 1989: Introduction. In: Karlamagnus saga and some religious texts, AM
180a and b fol. Copenhagen. (Early Icelandic manuscripts in Facsimile 18.) Pp. 919.
Heng, Geraldine, 2003: Empire of Magic. Medieval Romance and the Politics of Cultural Fantasy.
New York.
Hofmann, Gisela, 195758: Falkenjagt und Falkenhandel in den nordischen Lndern whrend des
Mittelalters. In: Zeitschrift fr deutsches Altertum und deutsche Literatur. Vol. 88. Pp. 115149.
Jones, Martin H., 2002: Giburc at Orange: The Siege as Military Event and Literary Theme. In: Wolf-
rams Willehalm: Fifteen Essays. Ed. by M.H. Jones, T. McFarland. Columbia, SC. Pp. 97120.
Karlamagns saga ok kappa hans. Ed. by C.R. Unger. 1860. Christiania.
Karlamagns saga, Branches I, III, VI et IX. Ed. by A. Loth. 1980. Copenhague.
Kinoshita, Sharon, 2006: Medieval Boundaries. Rethinking Difference in Old French literature. Phila-
delphia.
Lindow, John, 1995: Supernatural Others and Ethnic Others: A Millenium of Worldview. In: Scandi-
navian Studies. Vol. 67. Pp. 831.
Martin, John Stanley, 1990: Attitudes to Islam from the chansons de geste to the riddarasgur. In:
Parergon. Vol. 8. Pp. 8195.
Martin, John Stanley, 1991: The Transference of Attitudes to Islam from France to Scandinavia in the
Elie de Saint Gille and Elis saga ok Rsamundu. In: The Audience of the Sagas, The Eight Interna-
tional Saga Conference. Gothenburg. Vol. 2. Pp. 6779.
Stefn Karlsson, 1992: Elsta brot Karlamagns sgu og Rekattur ingeyrabkar. Eyvindarbk. Fest-
skrift til Eyvind Fjeld Halvorsen. Ed. by F. Hdneb, J.G. Jrgensen, E.Mundal, M. Rindal,
Vsteinn lason. Oslo. Pp. 302318.
Stuckey, Jace, 2008: Charlemagne as Crusader? Memory, Propagada, and the Many Uses of Charle-
magnes Legendary Expedition to Spain. In The Legend of Charlemagne in the Middle Ages.
Power, Faith, and Crusade. Ed. by M. Gabriele, J. Stuckey. New York. Pp. 137152.
Tolan, John V., 2002: Saracens: Islam in the Medieval European Imagination. New York.
Tveitane, Matthias, 1972, Introduction. In: Elis saga, Strengleikar and other texts. Uppsala University
Library Delagardieska samlingen nos. 47 folio and AM 666b quarto. Oslo. (Corpus codicum
norvegicorum medii aevi 4.), Pp. 934.

91
 Byzantium in the riddarasgur
Geraldine Barnes, University of Sydney, Australia
According to the eyewitness accounts of Robert de Clari (1991: 189) and Geoffroi de Ville-
hardouin (1938: 172), Englishmen and Danes were fighting ferociously alongside the
Greeks when seemingly impregnable Constantinople was sacked by French and Venetian
forces of the Fourth Crusade in April 1204. The events which led up to that striking image
and their influence on the contradictory conceptions of Byzantium in European chivalric ro-
mance and Icelandic riddarasgur are the subject of this paper.
The crusaders sack of Constantinople, a catastrophe from which the city never fully re-
covered (Nicol 1993: 1518) before it finally fell to the Ottoman Turks some two and a half
centuries later, was the horrific culmination of more than a century of escalating tensions be-
tween East and West Christendom, formally marked by the separation between the Church of
Rome and the Church of Constantinople the so-called Great Schism, usually dated to 1054.
From that time on, French and Anglo-Norman romance endorsed the Wests view of Eastern
Christians as religious deviants, sybarites, and arrogant, treacherous, and generally unreliable
allies. Those perceptions were reinforced by the perceived perfidy of the Byzantine emperor
Alexius I (10811118), who extracted oaths of fealty from the leaders of the First Crusade
(109599) and then failed to aid crusaders trapped in the city of Antioch after it was aban-
doned by its Byzantine commander (Angold 1997: 16065). Constantines resplendent New
Rome was reconfigured in French and Anglo-Norman romance as a treacherous place of lux-
ury and double-dealing.
The significance to European romance and pseudo-history of East-West Christian tensions
and of the dynastic alliances, mainly between Byzantine princesses and members of the Euro-
pean nobility, intended to strengthen the Eastern empire (Macrides 1992: 27080) has bur-
geoned as a topic of scholarship in recent years. In Empire of Magic: Medieval Romance and
the Politics of Cultural Fantasy (2003) Geraldine Heng, for example, reads the treacherous,
cowardly and effeminate Romans of Geoffrey of Monmouths Historia Regum Brittaniae
(ca. 1136) as contemporary Byzantines (Heng 2003: 4649) and interpets the declaration in
the Alliterative Morte Arthure that the Roman emperor is a false heretic as covert justifica-
tion for the Fourth Crusade (Heng 2003: 153). Sharon Kinoshitas reading of Chrtien de
Troyess Cligs (ca. 1176) identifies the political agenda of the work as a demonstration of
the hegemony of the Arthurian West over the Byzantine East (Kinoshita 1996: 336). Re-
becca Wilcox discerns a revelation of the Wests lingering anxieties about the questionable
outcomes of the Crusades in the mid-fourteenth-century Guy of Warwick, the English version
of the Anglo-Norman Gui de Warewic, where the Byzantine court is a place of conspiracy and
presumed imperial treachery, and the emperors daughter is dangerously seductive (Wilcox
2004: 220).
Norse-Byzantine relations from the mid-eleventh century onward, however, took a very
different course. Key factors were the apparent irrelevance of the Schism, the cultivation by
Norwegian kings of personal associations with Byzantine emperors, and the prestige associ-
ated with service in the Varangian Guard. After the Schism, whether or not they recognized it
as such or even heard very much about it, as Sverrir Jakobsson (2008: 175, 178) has demon-
strated, Icelanders continued to recognize the Byzantine emperor as the undisputed ruler of
Christendom (Sverrir Jakobsson 2005: 12328). Much maligned by the Anglo-Norman histo-
rians Orderic Vitalis and William of Malmesbury (Angold 2003: 3031), Alexius I is consis-
tently represented favourably in Icelandic sources, and the Norse version of his name, Kiria-
lax, became more or less the generic name for Byzantine emperors in the riddarasgur.
Where Eastern religious practice is mentioned in the riddarasgur, it is with tacit approval,

92
as, for example, in the celebration of the marriage of Kirialax in the Hagia Sophia, church of
the Patriarch of Constantinople s kirkja er mest gr allri Europa (Kirialax saga: 86)
conducted by the patriarch, according to Byzantine custom.
Scandinavians, and Anglo-Saxons, had served emperors of Byzantium long before the First
Crusade (Blndal 1978: 14147), most famously in the case of Haraldr harris service in
the Varangian Guard during the reign of Michael IV (103441). According to Morkinskinna
(2000: 325) and Heimskringla (vol.3: ch.12), after Sigurr Jrsalafari Magnssons visit to
Constantinople on his return trip from Jerusalem in 1110, many Norwegians remained in the
service of Alexius I. Seventy years after that, as Orkneyinga saga (ch.89) tells it, Rgnvald of
Orkney was showered with money by Manuel I (114380) on his arrival in Constantinople
and his men invited to sign on as mercenaries.
It all adds up to a view of Byzantium in the fourteenth- and fifteenth-century riddarasgur
which, as Sverrir Jakobsson (2005) has suggested of the broader Icelandic world view, pro-
jects an image of Constantinople untainted by the tensions of the Schism and their terrible
consequences. From that perspective, an episode in the early fourteenth-century Eirks saga
vifrla might be read as a romance of the establishment of the Varangian guard, which had
its historical origins during the reign of Basil II in the latter part of the tenth century (Blndal
1978: 4153): in the course of his quest for the heathen paradise, the Norwegian prince
Eirkr visits Constantinople, which is beset by vikings. He and his companion, Eirkr of Den-
mark, and their men defeat them and are said to have become the first Northmen to fight for
the Greek king in Constantinople (Eirks saga vifrla: 1415). The emperor, in turn, au-
thoritatively instructs Eirkr vifrla in the Christian faith.
The continuing popularity in Iceland of Haraldr harri may also have contributed to the
romanticization of the soldier of fortune in the riddarasgur. Events in a number of these sa-
gas mirror campaigns in which Haraldr took part, particularly in Apulia and Sicily. Apulia,
that region in southeastern Italy which borders on the Adriatic, became a Byzantine province
in the sixth century and, apart from intermittent Arab domination (including a short period in
the eleventh century), remained in Byzantine possession until Robert Guiscard of Normandy
set up the duchy of Apulia in 1059. Sicily, likewise, see-sawed between Byzantine and Arab
rule in the tenth and eleventh centuries, until, after an unsuccessful Byzantine attempt at re-
conquest in the 1030s, Robert Guiscard and his brother, Roger, established the Norman king-
dom of Sicily.
As related in Morkinskinna, Heimskringla, and Orkneyinga saga, the travels and cam-
paigns of Haraldr harri, along with those of Sigurur Magnsson and Rgnvald Kali, pro-
vide the model for the itineraries of many riddarasgur heroes who journey to Constantin-
ople: a trail of plunder-rich encounters with Saracen pirates off the coast of Moorish Spain
and in the Mediterranean, service with the emperor, forays from Constantinople against Sara-
cens, visits to Asia Minor and the Holy Land, and a component of amorous adventure. Har-
aldr harri served with the Varangians in the Aegean and in Sicily, and made a pilgrimage
to Jerusalem. Morkinskinna (chs. 12, 13) reports rumours that the formidable empress Zoe,
wife of Michael IV, wanted him for herself and accused him of having designs on her niece.
Enroute to Jersualem, Sigurur Magnsson went first to England, then to Galicia, Lisbon, and
through the Straits of Gibraltar. He harried Muslims in Spain, fought pirates off the coast of
Spain and the island of Formentara; visited Roger II in Sicily, landed in Acre and was lavishly
welcomed in Jerusalem by King Baldwin, with whom he joined in a successful skirmish
against heathens in Syria, and then travelled to Constantinople (Morkinskinna: chs. 6163).
Rgnvald Kali (Orkneyinga saga: chs. 8689) went to France and had a liaison with a lady-
in-waiting in Narbonne, then to Galicia and through the Straits of Gibraltar. He demolished a
Saracen dromond in the Mediterranean, then proceeded to Crete, Acre, Jerusalem, Constan-
tinople, and home via Bulgaria, Apulia, and Denmark. Penitence and pilgrimage tend not to

93
be the primary objectives in Icelandic accounts of visitors to Byzantium and the Holy Land.
As Sverrir Jakobsson points out (2008: 180), the magnificent receptions by Byzantine emper-
ors for Sigurr Jrsalafari Magnsson and Rgnvald of Orkney reported in Morkinskinna,
Hemiskringla, and Orkneyinga saga take precedence over any religious considerations. Simi-
larly in the Islendingasgur, Icelanders who go to Byzantium, such as Bolli Bollason in
Laxdla saga, do so primarily for the acquisition of personal honour and material gain.
Rhetorically, the topos of translatio studii adds a further dimension to riddarasgur en-
gagement with Byzantium. According to the preface of Viktors saga ok Blvus, Hkon
Magnsson (12991319) liet venda morgum Riddara sogum j nornu uR girzsku ok fran-
seisku mali (3), and it is as a variation of that topos, not as a statement of fact or fiction
(Amory 1984), that I would read that statement: a declaration of the transfer of narrative au-
thority from Byzantium and France to Norway and Iceland. The topos recurs elsewhere in the
riddarasgur; for example, at the conclusion of Konrs saga keisararsonar, where it is said
that three copies of the story were made by the emperor of Constantinople one for the ruler
of Saxland, one for the king of Denmark, and one for the emperor himself and that the
sagas exemplar was found in a street of unnamed location, which, by implication, is Constan-
tinople (344).
Sometimes, in both history and romance, its a case of translatio studii in the other direc-
tion. Morkinskinna, for example, incorporates northern mythology and legend into the mag-
nificent statuary of the Hippodrome, which itself celebrated the translatio studii et imperii
from Rome to Constantinople (Bassett 1991: 87). Among the sculptural commemoration of
ancient events said in Morkinskinna to be found on its walls are images of the sir, Vol-
sungs, and Gjukings: The walls are decorated with all sorts of ancient events. You can find
the sir, the Volsungs, and Gjukungs fashioned in copper and iron with such great skill that
they seem alive. With this arrangement people have the impression that they are participants
in the games (ch. 62: 324). I agree with Ted Anderssons comment (Morkinskinna: 453n.)
that the reference may indicate the writers familiarity with the notion that Aesir came from
the East, but perhaps we might also read it as a deliberate attempt to embed the North within
the cultural matrix of Byzantium.
Viktors saga ok Blvus concludes in Denmark and turns out, ultimately, to be about the
provenance of a Babylonian sword and halberd which come into the posession of a Danish
king, and which, according to the closing lines of the saga, are the weapons which slay two
men in an incident related at the beginning of the seventeenth-century Hrmundar saga
Gripssonar. In something of a similar translatio from the classical world to the North, Kiria-
lax saga is ultimately configured as an an ancestral northern narrative, inasmuch as it breaks
off in the very act of representing itself as the precursor to a narrative about kings and heroes
in the northern part of the world: Ok n skrifa ek af eira atfer eigi fleira at sinne ok vikjum
sgunni annan sta til eira knga ok kappa, sem bygu norrlfy heimsins ok vi hljta at
koma essu sgu [] (Kirialax saga: 101).
Among the sagas considered in this paper Konrs saga keisarasonar, Sigrgars saga ok
Valbrands, Jarlmanns saga ok Hermanns, Kirialax saga, Nitida saga, Vilhjlms saga sjs
Byzantine emperors are revered figures, but often past their prime. In Konrs saga keisara-
sonar, the emperor is rkastur [] llum heiminum (280), but his combat skills are rusty
([k]onungr hafi lngugi burtrei framda: 337), and his prospective son-in-law betters him
in the tilting contest which he (the emperor) intitiates. In Kirialax saga, Lotharius, the stlk-
onungr of Constantinople (the customary Icelandic term for the Eastern emperor), is said by
his prospective father-in-law to be elderly and therefore to present an opportunity for the man
who marries his daughter to become his successor. When, in Vilhjlms saga sjs, a hostile
force from Ermland (probably Armenia) demands that Kirialax, emperor of Byzantium sur-
render himself and his daughter, on the condition that his life will be spared on account of his

94
advanced age (77), they are amazed when he leads a spirited counter attack (7779). In Sigr-
gars saga ok Valbrands the emperor Adrianus, an honoured but unassertive figure, is point-
edly given precedence in the sagas three formal banquet scenes: in England; in Villus-
vnaland probably the kingdom of Nubia, where Christians raised pigs in the Middle Ages
(Nelson 1998), and which was strongly influenced by Byzantine culture and in Hungary.
Adrianuss men are, moreover, models of chivalric propriety. When they are defeated in a
tournament by English knights, the Byzantine knight Vgbaldr courteously hands over the
prize of 500 marks of gold and, with equal courtesy, the English prince Sigrgarr immediately
gives it to the poor (Sigrgars saga ok Valbrands: 120). Nor is Byzantium itself always with-
out challenge to its supremacy or immune from the threat of humiliation. In Nitida saga Prin-
cess Nitida, meykongr of France, refuses the suit of the Byzantine emperors son, Ingi, be-
cause, she says, Byzantium cannot compare with France in wealth and importance: ier hafit
eingvan rikdom til motz vit mig. Hafa og litit lond ydar ad yda vit Frackland jd goda (Nitida
saga: 1011).
Although France, andvegi heimsins (the worlds high-seat), explicitly displaces Byzan-
tium as the centre of the world at the beginning of Nitida saga (3), the balance of world power
in that romance is ultimately recalibrated through the intervention of a ruler from further East,
Lifornius of India. There are resonances here of the mythical Christian ruler, Prester John,
allegedly the author of the sensational Letter (ca. 1165) addressed to Manuel I, which prom-
ised aid to Byzantium against the Turks. (A reference in the Konungs Skuggsj [13] indicates
that the work was known in Norway and Iceland.) The new world order is mapped out in the
political power grid constructed by Lifornius at the sagas conclusion: on his marriage to
Nitida, Lifornius becomes co-ruler of France; Ingi of Byzantium marries Liforniuss sister;
Ingis sister, Listaln, is married to Nitidas foster-brother Hlskjldr, the heir to Apulia, and
Lifornius presents the couple with a third of India. Byzantium thus gains an alliance with In-
dia, and, through the marriage of Listaln, regains a dynastic link to Apulia.
There may be resonances in Jarlmanns saga ok Hermanns saga of eleventh-century Arab-
Byzantine power struggles in Apulia, when Jarlmann and Hermann of Swabia successfully
defend Byzantium against the combined heathen forces of Apulia and Serkland (Land of
the Saracens). Interestingly the peerlessness of Jarlmann and Hermann is geographically
measured in relation to Byzantium: fanzt eingi fyrer nordan Gricklandz haf s er eim
vri iafn ad fridleika ok jrottum (5) (there was no one to be found north of Greece who
was their equal in handsomeness and accomplishments).1 Hermann seeks in marriage the
Byzantine princess, Rkilt, a woman of great learning, powers of healing, and piety. Ermanus
of Apulia, a rival contender for Rkilts hand, who boasts of having Blland (Ethiopia), Bul-
garia, and Scythia in his power, threatens to attack the city with an overwhelming and mon-
strous force and to bring certain death to the emperor and utter humiliation to the Byzantines,
if his suit is rejected. The emperors neck is almost broken in the battle which follows, but
eventually Jarlmann visits the same fate upon Ermanus and all the heathens are killed.
In what may be a reminder of the significant Anglo-Saxon presence in the Varangian
Guard after 1066 (Blndal 1972: 1412), Constantinople is championed by English-led forces
in Vilhjlms saga sjs, the story of the lifelong alliance between Vilhjlmr, son of King
Rkarr of England, and Reginbald, son of the Byzantine emperor Kirialax (said also to be
known as Michael [Vilhjlms saga sjs: 28]). Noteworthy in particular is Vilhjlmrs con-
cern for the territorial integrity of the Eastern empire: Reginbald offers him a kingdom but
Vilhjlmr refuses it because he does not want to diminish Reginbalds territory (129). When
Reginbald crowns him King of Babylon, Vilhjlmr, in formal acknowledgment of the sym-

1
As is the beauty of the princess Potentiana in Saulus saga ok Nikanors: a ma so af henni segia at fyrir nordan
Gricklandz haf fddizt eigi fridare kona enn etta blomstur (7).

95
bolic overlordship of Constantinople, takes a splendid sword kisses it, and hands it to Regin-
bald.
Kirialax saga is principally the story of Kirialax, a prince from Thessaly who becomes
emperor of Byzantium. The saga operates within a multiplicity of chronologies, which draw
extensively and sometimes incoherently upon a vast array of learned sources (Cook 1985:
30326). Some events in the narrative are said to be contemporaneous with historically
documented attacks on Rome (and France and Germany) by Goths, Huns, and Vandals in the
fourth and fifth centuries, and by the usurper Eugenius during the reign of the emperor Theo-
dosius (34695); others are contextualized within the pseudo-history of King Arthurs con-
quest of northern Europe. A long sequence in Sicily has echoes of the Ostrogothic conquest of
Sicily under Theodoric in the late fifth century, but it may also evoke the attempted recon-
quest of Sicily from Arab rule in the mid-eleventh, that campaign with which Morkinskina
and Heimskringla credit Haraldr harri with a prominent role. The framework of history is
further extended to ancient and biblical history in the sagas accounts of Kirialaxs visits to
the ruins of Troy and to Jerusalem.
Kirialax saga is a story of Byzantium ascendant. Rome is attacked and threatened from
beginning to end: by Goths, Huns, and Vandals; by insurgents from North Africa; by the
usurper Eugenius; and by King Arthur. A running historical commentary on Roman fortunes
is linked to various episodes within the saga. Egias, son of King Dagnus of Syria, for exam-
ple, is said to have fought alongside the (historical) emperor Valentinian in Mauretania
against potential attackers of Rome, while Theodosius stays behind to guard the city from
assaults by men from the northern alps (11). After his defeat in Sicily, a viking named Eugen-
ius is said to have gone North, assembled forces from Swabia and Holstein (Svafa and Holl-
zetu landi), and on the model of the historical Eugenius gained power over Rome until
his defeat by Theodosius (62). Resonances of the historical emperor Zeno (ca. 425491), who
made the Germanic chieftain, Oadacer, patricius of Italy and later indirectly engineered his
killing by Theodoric, surface in the aftermath of the sagas Sicilian campaign, when the em-
peror Zeno leaves Rome in charge of an unnamed patricius and goes to austur-veg to deal
with hostility and unrest. Towards the end of Kirialax saga, Romanus, a Roman knight and
lifelong companion of Kirialax, returns to defend his patrimony because Arthur of Britain has
subjugated the northern part of the world. War rages everywhere, as the saga comments, with
spurious invocation of the Imago mundi of Honorius Augustodunensis (Cook 1985: 306) and
a work which may, as Kristian Klund suggests (Kirialax saga: 13n), be the Cronica Martin-
iana A eim timum var micill okyrrleikr vida um verolldina, eptir vi sem segir Imgo
mundi ok su bok, er het cronikamericion (Kirialax saga: 13) but Constantinople itself re-
mains undisturbed.
Given the reference to Arthurs conquering of the norlfu heimsins (89), with specific
mention of Italy ( essum time herjai Artus kngr af Bretland Italiam), the lack of any
mention in Kirialax saga of a Western emperor is puzzling. There is, however, some confu-
sion between the keisari Leo and the stlkonungr Lotharius, and an implication that Leo is
overlord of Lotharius (Cook 1985: 322n): at the wedding feast of Kirialax, Leo sits on his
right and Lotharius on his left; Lotharius addresses Leo as herra keisara (Kirialax saga: 88);
Leo crowns Kirialax and makes him overlord of Greece and its seven subordinate kingdoms.
Leo and his predecessor, Zeno, are not identified as Western emperors, but Leos place of
residence appears to be somewhere other than Constantinople, since Lotharius offers him
hospitality on his return from austur-veg.
Brings saga concludes by representing itself as an extended exemplum of how wrongdo-
ing will be avenged and righteousness rewarded (123). When Baeringr, a German knight
wrongfully deprived of his patrimony by the treacherous Heinrekr and brought up at the court
of Rkarr of England, defends Constantinople from Saracen attack, the grateful Emperor

96
Emanuel offers him half of Greece and the hand of his sister, Vindemia. The sexually irre-
sistible Bringr subsequently has a dream in which an angel tells him that Lucinia, daughter
of Lucius, keisari of Rome, and other women will do their best to tempt him, but that he must
remain faithful to Vindemia. Vilfriar, daughter of Pippin, king of France, goes through the
motions of taking the veil in order to avoid marriage to Emanuel and preserve her availability
for the handsome hero. The frustrated Lucina falsely accuses him of rape, and Lucius has him
thrown into a waterfall, from which he is rescued by an angel. Mortified and terrified of
Bringrs wrath when he discovers his error, Lucius shuts himself up in Florence and, when
he finally emerges, surrenders himself and his lands to Bringr. A council of kings and bish-
ops meets on the Feast of the Assumption and declares that the treacherous Lucina has lost
her fathers kingdom: Lucinie [] tapai riki fodr sins (121). In the final washup, Bringr
directs Lucinia to marry the widowed Rikarr of England, with Lombardy as her dowry, and
Vilfriar to wed the Greek emperor; he himself marries the steadfast Vindemia, according to
the laws of God and man, in a splendid ceremony (12223). Lucius dies shortly after, and
Bringr, having killed Heinrekr and regained his patrimony, is crowned emperor of Rome.
The virtuous Vindemia has comprehensively eclipsed the unprincipled daughters of the rulers
of France and Rome, and Constantinople is confirmed as the untarnished capital of Christen-
dom.
Some riddarasgur convey a sense of the geography and landmarks of Miklagarr. There
are references to the opening of the chain across Stlpasund (the Golden Horn) in Jarlmanns
saga ok Hermanns (17) and Vilhjlms saga sjs (36). Kirialax saga refers in some detail to
the Hagia Sophia (which had been ransacked and desecrated in 1204) and the imperial palace.
The long description of the latter, which derives from Karlamagns saga (Cook 1985: 306),
comes after earlier accounts in Kirialax saga of Jerusalem and the ruins of Troy and might, in
accordance with the views of a number of medieval historians (Alexander 1962: 346; Ball
2001: 445; Carile 2006), be said to offer tacit acknowledgment of Constantinople as the sym-
bolic successor to both cities. That many of the walls of Troy remain intact, a detail not pre-
sent in the sagas sources for this episode Alexanders saga and Trjumanna saga (Cook
1985: 306, 31317) serves further as a poignant reminder for us (though perhaps not for the
saga-writer) that the walls of Constantinople are said to have terrified the army of the Fourth
Crusade. The glowing splendour of the imperial palace, with its throne of fire-red gold and
dazzling pillars (Kirialax saga: 8687), mirrors the magnificent stone pillars and the gold
cross studded with jewels in Jerusalems Church of the Holy Sepulchre, as described earlier in
the saga (Kirialax saga: 6465).
The riddarasgur, then, hum with resonances of the Norths historical association with
Byzantium, and the East-West dynamic of Christendom is grounded not in confrontation and
conquest but in deference and defence. There is a rhetoric of cultural connection, too, in the
threads of translatio studii woven by the riddarasgur between the North and Byzantium.
Just as if we accept Hengs argument Geoffrey of Monmouth replaced his sixth-century
Romans with twelfth-century Byzantines, so Icelandic saga-writers of the fourteenth and fif-
teenth centuries looked back to Byzantium through an eleventh-century field of vision (Sver-
rir Jakobsson 2005: 3578) which acknowledged that the Great City was vulnerable to attack
but unchallenged in its moral and spiritual authority.

Bibliography
Alexander, Paul, 1962: The Strength of Empire and Capital as Seen Through Byzantine Eyes. Specu-
lum 1962. Pp. 33957.
Amory, Frederic, 1984: Things Greek and the Riddarasgur. Speculum 59. Pp. 50923.
Angold, Michael, 1997: The Byzantine Empire, 10251204. Second Edition. London and New York.

97
Angold, Michael, 2003: Knowledge of Byzantine History in the West: the Norman Historians (Elev-
enth and Twelfth Centures). In: Proceedings of the Battle Conference 2002. Anglo-Norman Studies
25. Woodbridge. Pp. 3031.
Ball, Warwick, 2001: Rome in the East: the transformation of an empire. London.
Bassett, Sarah, 1991: The Antiquities in the Hippodrome of Constantinople. Dumbarton Oaks Papers
45. Pp. 8796.
Bassett, Sarah, 2005: The Urban Image of Late Antique Constantinople. Cambridge.
Brings saga, 1884: In: Fornsgur Surlanda. Ed. by Gustaf Cederschild. Lund.
Blndal, Sigfs, 1978: The Varangians of Byzantium. Cambridge.
Carile, Maria Cristian, 2006: Constantinople and the Heavenly Jerusalem? In: 21st International Con-
gress of Byzantine Studies. London. www.byzantinecongress.org.uk/paper/VII/ VII.2_Carile.pdf.
Clari, Robert de, 1991: La Conqute de Constantinople. Trans. by Alexandre Micha. Paris.
Cook, Robert, 1985: Kirialax Saga: A Bookish Romance. In: Les sagas des chevaliers. Ed. by Rgis
Boyer. Pp. 30326.
Eiriks saga vifrla, 1983: Ed. by Helle Jensen. Copenhagen.
Heimskringla, 1951: Ed. by Bjarni Aalbjarnarson. slenzk Fornrit 27. 3 vols. Reykjavk.
Heng, Geraldine, 2003: Empire of Magic. Medieval Romance and the Politics of Cultural Fantasy.
New York.
Jarlmanns saga ok Hermanns, 1963: In: Late Medieval Icelandic Romances III. Ed. by Agnete Loth.
Copenhagen. Pp. 366.
Kinoshita, Sharon, 1996: The Poetics of Translatio: French-Byzantine Relations in Chrtien de
Troyess Cligs. Exemplaria 8. Pp. 31554.
Kirialax saga, 1917: Ed. by Kr. Klund. Samfund til Udgivelse af Gammel Nordisk Litteratur.
Copenhagen.
Konras saga keisararsonar, 1954: In: Riddarasgur III. Ed. by Bjarni Vilhjlmsson. Reykjavk. Pp.
34344.
Konungs Skuggsi, 1983: Ed. by Ludvig Holm-Olsen. Oslo.
Macrides, Ruth, 1992: Dynastic Marriages and political kinship. In Byzantine Diplomacy. Ed. by
Jonathan Shepard and Simon Franklin. Pp. 26380.
Morkinskinna, 2000: The Earliest Icelandic Chronicle of the Norwegian Kings (10301157). Trans. by
Theodore M. Andersson and Kari Ellen Gade. Ithaca and London.
Nelson, Sarah M.,1998: Ancestors for the Pigs: Pigs in Prehistory. Philadelphia.
Nicol, Donald M., 1993: The last centuries of Byzantium, 12611453. Second edition. Cambridge.
Nitida saga, 1965: In: Late Medieval Icelandic Romances V. Ed. by Agnete Loth. Copenhagen. Pp. 3
37.
Orkneyinga saga, 1965: Ed. by Finnbogi Gumundsson. slenzk Fornrit 34. Reykjavk.
Saulus saga ok Nikanors, 1963: In: Late Medieval Icelandic Romances II. Ed, by Agnete Loth. Co-
penhagen. Pp. 391
Sigrgars saga ok Valbrands, 1965: In: Late Medieval Icelandic Romances V. Ed. by Agnete Loth.
Copenhagen. Pp. 11194.
Sverrir Jakobsson, 2005: Vi og verldin. Heimsmynd slendinga 11001400. Reykjavk.
Sverrir Jakobsson, 2008: The Schism that never was: Old Norse views on Byzantium and Russia.
Byzantinoslavica 12. Pp.17388.
Viktors saga ok Blvus, 1962: In: Late Medieval Icelandic Romances I. Ed. by Agnete Loth, Copen-
hagen. Pp. 350.
Vilhjlms saga sjs, 1964: In: Late Medieval Icelandic Romances IV. Ed. by Agnete Loth.
Copenhagen. Pp. 3136.
Villehardouin, Geoffroi de, 1938: Ed. and trans. Edmond Faral. La Conqute de Constantinople. Vol
1. Paris.
Wilcox, Rebecca, 2004: Romancing the East: Greeks and Saracens in Guy of Warwick. In: Pulp Fic-
tions of Medieval England: Essays in Popular Romance. Ed. by Nicola McDonald. Manchester. Pp.
21740.

98
 The World West of Iceland in Medieval
Icelandic Oral Tradition
Eleanor Rosamund Barraclough, Department of Anglo-Saxon Norse and Celtic,
University of Cambridge, England

Introduction
The abstract for this paper drew attention to the unusual way in which the slendingasgur
depict the world west of Iceland (with particular emphasis on Greenland), and the implica-
tions of this representation for the construction of a world view of the geographical sphere
as it might have been viewed by medieval Norse society.1 The Greenland of the sagas was a
unique and at times strange place, lying somewhere on the boundary between the known, fa-
miliar Norse world, and an unfamiliar, exotic sphere beyond. An analysis of any saga is en-
hanced by a consideration of its oral dimensions, but the issue is especially important in the
case of texts with Greenlandic episodes, where the provenance and nature of these underlying
oral elements is particularly difficult to ascertain.2 Yet the degree of consistency throughout
various sagas portrayals of the region suggests some stability within the oral traditions con-
nected with the country. This is particularly the case with regard to the Vnland sagas (Eirks
saga and Grnlendinga saga), where their similarities led early scholarship to favour a liter-
ary relationship between the two, although more recent research has concluded that they are
unrelated literary texts with common oral elements (lafur Halldrsson 1978:36971,450).
By examining the strands of oral traditions and common literary themes that reoccur through-
out the sagas, this paper will attempt to construct a mental map of this geographical and so-
cial sphere as it might have appeared to medieval Icelandic society. It will examine the place
of Greenland in the Norse world view, considering why the sagas set in the region tend to
focus on the more negative aspects of landscape and life in the country.3

Greenland in the slendingasgur

1
The term world view is defined by Sverrir Jakobsson (2007:22) as conscious and subconscious ideas about
the world and its inhabitants, including the self, in a historical and geographical perspective. It is also an integral
and inseparable part of the general discourse of a period. It characterises groups social or cultural rather than
individuals.
2
The literary and oral background of texts concerned with Greenland is a matter of debate, made more complex
by the diverse genres and postulated dates of composition for each saga. There is no direct evidence for the pro-
duction of Norse manuscripts in Greenland, although there was probably some form of writing at the Episcopal
see at Garar. However, given the marginal situation of the Norse settlements in Greenland, only the wealthiest
chieftains and the bishop could have afforded to be patrons of literature, and if they were, one might expect them
to trumpet the fact, not pass by it in silence. Consequently, depending on the character of the saga in question, it
may be that the narratives were entirely fictional with no link to Greenlandic society (particularly in fantastical
texts such as Jkuls ttr and Gunnars saga, which focus on supernatural trollish communities). Alternatively, it
is possible that there was a Greenlandic eyewitness informing the writer (for instance, in Grnlendinga ttr
Greenlandic places and landscapes are described accurately, detailing a realistic Greenlandic society and its legal
procedures). Another possibility is that certain saga authors could have been Greenlandic themselves, writing in
either Greenland or Iceland. Additional texts can also be tenuously linked to a Greenlandic literary culture, such
as the two eddic poems Atlakvia and Atlaml, which are given the debateable epithets in grnlenzca and in
grnlenzco in Codex Regius (GKS 2367 4to).
3
The term comes from Gsli Sigurssons work (2004:300) on the mental map of the world west of Iceland. He
develops his hypothesis in terms of geographical orientation, but in the current context the model can be ex-
tended to include abstract ideas concerning the place that regions would have occupied in the world view of
collective medieval Icelandic society.

99
A number of broad literary patterns and themes remain relatively consistent across the corpus
of texts featuring Greenland.4 The physical landscape features prominently, emphasising the
difficulty of settling a harsh terrain. Greenlands saga topography can be envisaged as a series
of horizontal layers, which become increasingly impenetrable as the landscape rises towards
the glaciers and mountains. The lower layers begin in the sea, where inhospitable, dangerous
waters give way to the mutable, intermediate medium of the gravel shoreline. This in turn
rises to a narrow habitable shelf, consisting of fertile pockets of land by the fjords and below
the cliffs. Further into the country, this strip of land becomes impassable rock with the (fre-
quently supernatural) wilderness located beyond. In the upper layers of the terrain, vast, unin-
habitable glaciers and ice shelves loom up, dominating the landscape and curtailing explora-
tion with their solid impenetrability. In terms of other themes associated with Greenland, the
countrys wilderness is a key feature, often complete with supernatural inhabitants and grisly
happenings. Outlaws and social outcasts frequent the shores, either arriving from the lands
from which they have been exiled, or banished at Greenlands own ing. Indeed, Eirkr raui,
Greenlands primary landnmsmar, arrives under a cloud of killings and exile from Iceland,
having already left Norway for similar reasons. Within the community, humans become sick,
livestock die and famine is an ever-present danger, whilst on the countrys storm-battered
coast, boats are shipwrecked and lives are lost.5
A broad pattern of deictic orientation emerges from the sagas concerning Greenland, in
which the more easterly part of the Norse world (particularly Iceland, and to a lesser extent,
Norway) is the conceptual and geographical locus of the texts and the nucleus of Norse social
identification.6 In terms of the mental map that can be constructed of the sphere, an east-to-
west geographical axis emerges, moving from cultural familiarity to exotic western wilder-
nesses. This is true externally, in terms of the relationship between the various lands of the
North Atlantic (as a rule, Iceland is more stable than Greenland, which in turn is more famil-
iar than Vnland, which is less strange than lands such as Einftingaland and Hvtraman-
naland as mentioned in Eirks saga). It is also the case internally, within the two key commu-
nities of Greenlandic society. Generally identified in terms of their relative longitudinal loca-
tions as the Eastern and Western Settlements, the former is often indicated more familiarly by
name (Brattahl or Eirksfjrr), whilst the latter is referred to in more abstract terms as the
vestr-bygg or the vestr bygg (Western Settlement and uninhabited west). This differ-
ence in nomenclature is reflected in the characters of the sites themselves, for the Eastern Set-

4
These are primarily Eirks saga raua, Grnlendinga saga, Fstbrra saga, Krka-Refs saga, Flamanna
saga, Brar saga Snfellsness, Eyrbyggja saga, Jkuls ttr Basonar, Gunnars saga Keldugnpsffls,
Auunar ttr vestfirska and Grnlendinga ttr.
5
There are of course exceptions to the thematic patterns, such as Grnlendinga ttr, which concerns a realistic
Greenlandic community and its attempts to assert national identity against a background of religious tension and
struggles with Norwegian incomers. Yet even here, sickness and death awaits Sigurrs hunting expedition in the
wilderness (a typical Greenlandic motif), where they discover the plague-ridden remains of a lost crew. Further-
more, the Western Settlement is a hostile place in comparison to the east, used as a base by the Norwegians dur-
ing their legal feud with the Greenlanders. In other texts, although the eponymous hero of Refs saga is forced
into exile in an exceptionally bountiful wilderness, the fact remains that he is there because he has been out-
lawed, a common feature of many Greenlandic episodes.
6
Deixis is a linguistic term referring to temporal and spatial co-ordinates in language. The intrinsic orientation
of an object is by definition independent of the position of the speaker and dependent on static observation. In
contrast, the deictic orientation of objects cannot be fixed without the speaker, and is dependent on their dynamic
involvement. Deictic orientation occurs whenever a linguistic sign receives part of its meaning from an extra-
linguistic context (see Andersen 1985). This concept can be incorporated into the methodology of research into
the perception of the familiar Norse sphere and the significance of landscape in sagas concerning Greenland. By
examining the deictic locus of identification through the texts topographical references, it is possible to identify
thematic preoccupations, boundaries of cultural familiarity and attitudes towards the exotic and the unknown in
the Norse world view.

100
tlement is described as a welcoming community, as in Eiriks saga, when orbjrn reaches
Brattahl, Eirkr tekr vel vi honum, me blu, ok kva at vel, er hann var ar kominn (F
4:209). In comparison, the Western Settlement is presented with a gloomy sense of alienation
bordering on the sinister, and although elements such as paganism, plague and supernatural
activity also exist in the Eastern Settlement, in the west they are conveyed with dramatic liter-
ary patterning as the defining features of the community. For instance, in Grnlendinga saga,
following his death, rsteinn Eirkssons prophecy concerning Gurrs future back east
(munu it fara af Grnlandi til Nregs ok aan til slands ok gera b slandi F 4:260)
and her bright descendants (bjart ok gtt, stt ok ilmat vel F 4:260) contrasts starkly
with the gloomy, claustrophobic darkness of the farmhouse in the dead of winter, where the
stagnant community seems to be internally rotting away in the grip of the epidemic.
In part, the literary descriptions must reflect some degree of the geographical and meteoro-
logical reality, with the cultural memory and realistic oral traditions concerning the regions
weather and living conditions being transferred in some form into the written texts that
emerged.7 However, despite the fact that there must have been famines, shipwrecks, bad
weather and plagues in similar landscapes such as Iceland, such adverse features do not define
the country as a whole. In his discussion of why natural phenomena such as volcanoes do not
feature in the slendingasgur despite their presence in the Icelandic landscape, Oren Falk
(2006:232) notes:

The slendingasgur are tight-lipped in general about all kinds of natural calamity. Few wild-
fires or famines ravage the countryside in saga Iceland; harsh winters and disease seldom deci-
mate the population; landslides and floods are mercifully rare; and ravenous polar bears [] are
almost unheard of.

Therefore if, as is generally held, the sagas chiefly attend to meteorological extremes and
natural phenomena for literary effects such as metaphor and mood-setting (see Falk 2006:233,
Ogilvie 2006), then what does this say about the place of Greenland in the Norse world view,
where such descriptions are so prominent?

Test case: landing on the shores of new lands

In order to answer this question, a test case will now be made of a particular aspect of the
landscape, in order to highlight the unique place of Greenland in Norse oral traditions. The
depiction of the landings and subsequent landnm (land-taking) of incomers to Greenland
and Iceland will be compared, which will show the different ways in which the voyagers in-
teract with the two new landscapes.
Beginning with the arrival in Greenland in Grnlendinga saga, from the time when Bjarni
sets out from Iceland, there is a sense of going beyond the controllable and established world
of Icelandic society when Bjarni says vitrlig mun ykkja vr fer, ar sem engi vr hefir
komit Grnlandshaf (F 4:246). The familiar world of Iceland retreats, and the unfamiliar
seascape is threatening and difficult to navigate (alda eir n haf, egar eir vru bnir, ok
sigldu rj daga, ar til er landit var vatnat, en tk af byrina, ok lagi norrnur ok o-
kur, ok vissu eir eigi, hvert at eir fru, ok skipti at mrgum dgrum F 4:246). Similarly,
in Eirks saga they leave Iceland in good weather, yet san ltu eir haf, ok er eir vru
hafi, tk af byri (F 4:205). The subsequent voyage is grisly, for once the wind has dropped,
fengu eir hafvillur, ok frsk eim greitt um sumarit. v nst kom stt li eira (F
4:205). Later, bad weather in this stretch of water is responsible for the discovery of Vnland
(ltr Leifr haf ok er lengi ti ok hitti lnd au F 4:205).

7
Cultural memory is the interplay between the literary inventiveness of saga texts and their ability to reflect
and play a role in the broader social and historical issues of the day (see Glauser 2007).

101
 Beyond the Vnland sagas, the picture of Greenlands physical landscape remains largely
consistent, with journeys to Greenland characterised by storms and shipwrecks. Refs saga
links the theme specifically to the sighting of Greenland (eim ferst vel, ar til er eir f sn
af Grnlandi, ok sian velkir lengi ok hefr norr me landinu F 14:131), while in
Fstbrra saga, Skfrs journey out to Greenland contrasts with his journey to Norway (on
the way out, skip velkir ti lengi; f eir ver str (F 6:223), whilst on the way back, eir
f ga byri; fersk eim vel, taka Nreg(F 6:257)).
In the supernatural world of Jkuls ttr, the description of the sea journey is protracted,
emphasising an otherworldly disorientation, the long time they are lost at sea, and the fero-
cious shipwreck:

Gaf eim ltt byri, og rak fyrir eim myrkr og hafvillur, sv eir vru ti allt sumari; en er
hausta tk, geri storma me miklum hrum og frustum, sv sldi hvern dropa, er inn kom.
[] um sir rak skipi a skerjaklasa miklum me boafllum strum. (F 14:47)

The word hafvillur (witless-at-sea) is significant, implying that the Greenlandic ocean is not
only physically dangerous, but can also affect the sailors minds.8 Furthermore, as the rain-
drops turning to ice (sv sldi hvern dropa), the weather becomes the landscape, and the
storm solidifies into the frost and snow that covers much of the country. Consequently, an
event (the storm) is transformed into a topographical situation (the icy landscape) in its mete-
orological hostility, driving the sailors forward onto the skerries and inhospitable shoreline of
the country.
Such descriptions of the approach to Greenland are compounded by the inclusion of men-
acing supernatural and pagan elements within the story. During the voyage to Greenland in
Flamanna saga, rr appears to orgils, threatening shipwrecks if the company refuse to
believe in him. When this comes to pass, the description of the shipwreck on the Greenlandic
coast focuses on the little ship washed up below the glaciers, compounding the continuing
sense of mans insignificance and vulnerability in the face of the a hostile landscape: eir
brutu skipit undir Grnlandsjklum vk nkkurri vi sandml. Tk skipit sundr efra rmi
(F 13:282). T here is no sense of an external agency being responsible for the breaking ship
(such as storms, skerries or humans), with primacy and power instead given to the solid sheet
of ice that dominates the landscape. In these descriptions the sailors seem to be repulsed by
the land itself, for however good the journey is up to that point, it is hard to control the ap-
proach once they sight the coast.
By contrast, in Iceland the immigrants have much more control as they near the coast, re-
flecting the more powerful way in which they are able to interact with the landscape of their
new home. To some extent, the seascape is still marginal, and as Margaret Clunies Ross notes
(1998:130, with reference to Gsli Plsson 1990):

This privileging of the idea of land taking as a means of humanising the environment had its an-
tithesis in the relative neglect of matters to do with the waters and their inhabitants which were
placed in a special, somewhat marginal category associated with anomaly and uncertainty.

Consequently, as with the seas around Greenland, there are rough passages at sea as they
reach their destination. However, the descriptions are perfunctory and serve little function in

8
The adjective also appears in the journey to Greenland in Eirks saga raua (see above). The word is rare (I
have found it only in Laxdla saga in the description of lfr pis voyage to Ireland and in Finnboga saga,
when Finnbogi is shipwrecked in the far north of Norway) and although not confined to descriptions of the tur-
bulent journey to Greenland, every time it occurs it is in the context of journeys to the inhospitable northerly or
westerly outer reaches of the Norse world.

102
comparison to the place of shipwrecks in the plots of many Greenlandic episodes; the storm
blows down and they land without further ado. For instance, in Egils saga:

Er eir vru komnir vi sland, sigldu eir sunnan at landi; eir sigldu vestr fyrir landit, v
at eir hfu at spurt, at Inglfr hafi sr ar bsta tekit; en er eir kmu fyrir Reykjanes ok
eir s firinum upp lka, stefna eir inn fjrinn bum skipunum. Ver geri hvasst ok
vta mikil ok oka; skilusk skipin. Sigldu eir inn eptir Borgarfiri, til ess er raut sker
ll; kstuu akkerum, til ess er ver lgi ok ljst geri [] fluttu eir kistuna nes at, er
ar var, settu hana ar nir ok hlu at grjti. (F 2:712)

In this instance, the sailors are able to circumnavigate the country freely and weather storms
without serious repercussions, and when they decide to come ashore, the landscape opens up
to receive them so that they might enter it through easily accessible fjords. Once they have
landed, the travellers are able to manoeuvre freely in order to familiarise themselves with the
topography and bring it within their control, their exploration sweeping up across the plains
and into the mountains. This stands in sharp contrast to Greenlands series of topographic
layers, which become increasingly impenetrable as they rise from the coast up towards the
sterile glaciers.
Just as the approach to Iceland is a controlled and manageable affair, once they have
reached the land, the colonisers must take control over their new country. As with Greenland,
supernatural elements play a role, but whilst in that setting they highlight the powerlessness of
the incomers to Greenland, here they are used as a tool of power by Icelandic settlers. This
interaction between the human and the divine take several forms, including benevolent rela-
tionships with Icelands landvttir (spirit-beings who live in the land and safeguard it), pro-
tective affinities with particular gods carried out with the emigrants and transferred to the new
land, and the use of fatalistic determinants such as high-seat pillars to decide on the location
of the new farmsteads (see Clunies Ross 1998:12257).
The approach to Iceland in Eyrbyggja saga encapsulates the importance of supernatural
forces in enabling the seafarers to take control of their approach to Iceland and their settle-
ment of the land:

rlfr kastai fyrir bor ndvegisslum snum, eim er stait hfu hofinu; ar var rr
skorinn annarri. Hann mlti sv fyrir, at hann skyldi ar byggja slandi, sem rr lti r
land koma. En egar r hf fr skipinu, sveif eim til ins vestra fjararins, ok tti eim fara
eigi vnum seinna. Eptir at kom hafgula; sigldu eir vestr fyrir Snfellsnes ok inn
fjrinn. (F 4:78)

As with texts such as Egils saga and Landnmabk, Eyrbyggja saga employs the motif of the
high-seat pillars, carved with rr and cast overboard, in order to create a sense of supernatu-
ral interaction with the meteorological conditions, working in concert to welcome them and to
create an effortless entry into the country. The word sveif (swept) amplifies the sense of
swift movement, emphasised by the information that the ship is moving faster than expected.
As with the description of Skalla-Grmrs landnm in Egils saga, the sailors are not curtailed
by a solid block of land in front of them, but are able to sail freely around the coast, arching
around cape Reykjanes and the headland of Snfellsnes, and propelled into the fjord by a
hafgula (sea breeze) that springs up to speed them on their way.

Conclusion
By means of conclusion, it is worth briefly considering the reasons for the way in which
Greenland is depicted in the sagas, with the prevalence of certain negative characteristics. The

103
question is complex, and a number of factors are likely to be involved, ranging from geo-
graphical reality to changing trends in literary genres.
Firstly, there is the aforementioned issue of Greenlands conceptual location in terms of the
deictic orientation of the Norse world, perched on the edge whilst Iceland was in the middle.
Strange and uncanny events are much more likely to happen in distant, more exotic countries,
since, as Falk states, [un]natural calamities in the sagas tend to occur in far-off lands, where
the boundaries between the real and the fantastic are more porous anyhow (Falk 2006:232).
Additionally, the literary depiction of Greenland is likely to be based on its geographical
reality to some extent, for if Greenland was a difficult land to inhabit, it is not surprising that
it would be represented in the sagas as a place of shipwrecks, storms and plagues. For in-
stance, the differences between the characters of the Eastern and Western settlements can be
explained in part by the different physical conditions in the sites, since despite the nomencla-
ture, the Western Settlement was actually 300 miles further north than the Eastern Settlement.
Exposed to the inhospitable West Greenlandic current, the region was considerably colder and
wetter than its southern counterpart, with a significantly reduced summer growing season (see
Diamond, 2005:215). With a blend of literary patterning and geographical reality which hints
at the nature of the defining cultural memories associated with the region, the real-life loca-
tion and climate of the Western Settlement partly explains why it was characterised as a hos-
tile place of sickness and eerie events. Famine and plague may have also occurred in the east,
but in the west these characteristics were key, embellished with paranormal incidents in order
to generate a darker, more supernaturally inclined world.
Moreover, if the sagas are ranked roughly according to age (for discussion of this problem-
atic issue see rnlfur Thrsson 1990, Degnbol et al. 1989), the increasingly fantastical nature
of Greenland as we move through the centuries suggests that the depiction of the country has
also been influenced by changing fashions in saga genres over the years (the classically per-
ceived pattern being a shift from the socially realistic genre of the slendingasgur to the
more fantastical and continentally influenced fornaldarsgur and riddarasgur). The sagas
traditionally identified as older provide the most realistic and socially detailed accounts of the
country (Grnlendinga ttr, Grnlendinga saga, Fstbrra saga and Eyrbyggja saga),
whilst the later texts have a tendency to use broader brushstrokes, exaggerating the thematic
motifs associated with the region (Flamanna saga, Brar saga, Jkuls ttr and Gunnars
saga Keldugnpsffls). Earlier tendencies to describe shipwrecks, harsh living conditions and
supernatural elements become crystallised as the focus of later stories, with the importance
accorded to the human population dwindling until it disappears altogether to be replaced by a
trollish society in sagas such as Jkuls ttr and Gunnars saga. However, this may not only
be a result of changing literary fashions. It may also be that because the themes are largely
negative, as time passed the literature also reflected the increasingly precarious nature of life
in Greenlands deteriorating physical climate (see Diamond 2005), with an ever more pessi-
mistic cultural memory underpinning the sagas. It might also mirror the diminishing links
between Greenland and the rest of Europe, for as the trading patterns changed and the sailing
routes to the country were slowly abandoned, the position of Greenland in the Norse world
view must have altered significantly (see Olfur Halldrsson 1993:241). This would explain
the fact that in the later texts, human society disappears to be replaced by monsters and giants,
for with less contact between Greenland and the rest of the world (particularly Iceland where
the sagas are likely to have been recorded), there were fewer oral traditions and information
about Greenlandic society emanating from the region.
Finally, there is the question of why, as demonstrated by the test case of the landnm as it
is presented in both Iceland and Greenland, the less positive aspects of the Icelandic landscape
are not the defining features of this country. This can be explained with reference to Jesse
Byocks assertion that over centuries, [the sagas] helped an immigrant people form a coher-

104
ent sense of who they were (2004:303). Central to this impulse were the traditions associated
with the landnm and genealogies, reflecting the Icelanders aspirations to establish their nas-
cent nation within a larger European framework (see Clunies Ross 1993:3756). Within this
context, there would have been little reason for negative descriptions of the countrys land-
scape, particularly during descriptions of the landnm, since such historicising tendencies
were a critical means of defining and legitimising separate Icelandic identity as the countrys
independence became increasingly threatened in the international political arena. In contrast,
the presentation of Greenland in the sagas creates the impression that while it was not a
wholly alien land, it did lie upon the margins between the familiar Norse world and an unsta-
ble, unknowable sphere beyond. Underlying the unsettled nature of the cultural memory
preserved in the sagas, the oral traditions associated with the region were likely to have
stemmed in part from the anxieties and dangers that would have concerned the Norse settlers.
Consequently, in its literary representation, the land on the edge of the world was transformed
into an unpredictable, shadowy place of shipwrecks, plagues, and supernatural happenings.

Bibliography
F = slenzk fornrit. Reykjavk
Andersen, Stephan and Keenan, Edward 1985: Deixis. In: Language, Typology and Language De-
scription, vol. 3 (Cambridge). Pp. 259308
Byock, Jesse 2004: Social Memory and the Sagas: The Case of Egils saga. In: Scandinavian Studies
76. Pp. 299316
Clunies Ross, Margaret 1993: The Development of Old Norse Textual Worlds: Genealogical Structure
as a Principle of Literary Organization in Early Iceland. In: Journal of English and German Philol-
ogy 92. Pp. 37285
Clunies Ross, Margaret (1998): Prolonged Echoes: Old Norse Myths in Medieval Northern Society,
vol. 2: The Reception of Norse Myths in Medieval Iceland (Odense)
Degnbol, Helle & al, ed. 1989: Ordbog over det norrne prosasprog / A Dictionary of Old Norse
Prose. Registre / Indices. Den arnamagnanske kommission (Copenhagen)
Diamond, Jared 2005: Collapse: How Societies Choose to Fail or Succeed (New York, NY)
Falk, Oren 2006: Fragments of Fourteenth-century Icelandic Folklore. In: Pre-print papers of The Fan-
tastic in Old Norse / Icelandic Literature: Thirteenth International Saga Conference, Durham and
York, 6th12th August 2006, vol. 1. Ed. by J. McKinnell & al. (Durham). Pp. 231 40
Gsli Plsson 1990: The Idea of Fish: Land and Sea in the Icelandic World View. In: Signifying Ani-
mals: Human Meaning in the Natural World. Ed. By R. Willis (London). Pp. 11933
Gsli Sigursson 2004: The Medieval Icelandic Saga and Oral Tradition: A Discourse on Method.
Trans. By N. Jones (Cambridge, MA)
Glauser, Jrg 2007: The Speaking Bodies of Saga Texts. Trans. By K. Heslop. In: Learning and Un-
derstanding in the Old Norse World: Essays in Honour of Margaret Clunies Ross. Ed. by J. Quinn
& al., Medieval Texts and Cultures of Northern Europe 18 (Turnhout). Pp. 1326
Olfur Halldrsson 1993: Greenland, Norse. In: Medieval Scandinavia. An Encyclopedia. Ed. by P.
Pulsiano & al. (New York and London). P. 241
Ogilvie, Astrid, and Gsli Plsson 2006: Weather and Witchcraft in the Sagas of Icelanders. In: Pre-
print papers of The Fantastic in Old Norse / Icelandic Literature: Thirteenth International Saga
Conference, Durham and York, 6th12th August 2006, vol. 2. Ed. by J. McKinnell & al. (Durham).
Pp. 73441
lafur Halldrsson 1978: Grnland mialdaritum (Reykjavk)
Sverrir Jakobsson 2007: Hauksbk and the Construction of and Icelandic World View. In: Saga-Book
of the Viking Society 31. Pp. 2238
rnlfur Thrsson 1990: Leitin a landinu fagra: Hugleiing um rannsknir slenskum fornbk-
menntum. In: Skldskaparml 1. Pp. 2853

105
 What do the norns actually do?
Karen Bek-Pedersen, Scandinavian Studies, University of Aberdeen, Scotland
The norns (Old Norse nornir) are a group of female supernatural beings who, in Old Norse
tradition, somehow or other represent fate. As with many other beings from this tradition, the
norns are known to us predominantly through the literary sources.
They are not exactly prominent figures; literary references to these beings are rather few,
while references to them from outwith the literature are incredibly few. However, although
the sources convey relatively little information about them, it is not unfair to say that certain
ideas constitute what one might call our general or even stereotypical knowledge about the
norns. In particular three specific ideas appear to be prominent: 1) that there are three norns;
2) that they are called Urr, Verandi and Skuld; 3) that they represent Past, Present and Fu-
ture.
These facts have been reiterated in so many scholarly references to the norns that they
may even be taken to be common knowledge and, therefore, these ideas merit a bit of close
attention.

The Number Three

Concerning the idea that there are three norns, it is true that norns have a strong tendency to
occur in the plural as a group but instances of singular norns also exist.1 In cases where
there are clearly more than one, the number of norns in the group is rarely specified;2 three
seems a good suggestion, and this is the enumeration we encounter when there is one, but
there is more to be said about it than that.
Vlusp 20 gives three names and Gylfaginning 15 (quite possibly echoing Vlusp) does
the same, though Gylfaginning, perhaps in an attempt to amalgamate contradicting traditions,
goes on to state that there are more than three norns. Here, Gylfaginning cites Ffnisml 13,
which refers to a three-part division of the collective group of norns but not to three indi-
viduals. Ffnisml 13 quite specifically says that some norns are of this kind, some of that
kind and some of that kind, thus bringing the total number, as it were, to more than three
otherwise it would presumably have said one of this and that kind, not some.
The intention is not to discard the idea of the three norns altogether, but simply to say that
it is not the full picture. There are three norns in Vlusp and (at least some of the time) in
Gylfaginning. But there are also other sources that refer to these beings and they do not all
give the same information.

The Names
The idea that the norns carry the names of Urr, Verandi and Skuld obviously feeds off the
notion of there being three of them. Two sources mention these names, namely Vlusp 20
and Gylfaginning 15.
As a trinity, the names do not occur outwith these two texts, although both Urr and Skuld
occur elsewhere but never the two of them together. Verandi occurs nowhere else, and it
has been suggested that her name may have been invented in order to fill in some sort of per-

1
Reginsml 2; Kveldlfrs lausavsa in Egils saga Skallagrmssonar 24; Egills lausavsa in Egils saga
Skallagrmssonar 56; lafs drpa Tryggvasonar 18.
2
It is, in fact, statistically very unusual that a specific number is given; this happens in Vlusp 20 (though this
text does not use the term norns in its description; it simply calls them meyiar), in Norna-Gests ttr and in
Gylfaginning 15 although Gylfaginning 15 also says that there are more norns than the three named ones.

106
ceived gap between the other two. The figure behind the name may still be ancient, even if the
name is not.

Past, Present and Future

The idea that the names as well as the figures hiding behind them represent past, present and
future deserves some attention here because it is in certain ways problematic.
The first problem is that, while the meaning of the name Verandi is in all sorts of ways
close to the meaning of present, Urr means fate, not past, and likewise, Skuld means
something along the lines of debt. Admittedly, the noun skuld has a fairly wide semantic
range, but that it should encompass the concept of future seems to be stretching it a bit far. I
am not aware that skuld is ever used in that sense, nor have I come across any occurrences of
urr used with the meaning past.
However, it is possible that it has been the intention of the Vlusp-poet to create an addi-
tional layer of meaning to the names Urr and Skuld by inserting Verandi in order to lend
the three figures a collective, time-related aspect. This recasting of the names produces a con-
nection that can but must not necessarily be made, namely the temporal understanding of
Urr and Skuld as, respectively, past and future, without erasing the, so to speak, original
meanings of fate and debt.
The second problem of regarding the norns as representatives for time is that fate and time
are not at all the same. Time is concerned about when things happen, and such chronological
concern seems to be quite different from what is meant by fate. Fate stands outside of time,
because it regards the future much as we regard the past; yet, fate is experienced over time, so
if time does not exist, one cannot experience fate. Therefore, we need time in order to have
fate. But this does not mean that they are the same thing. Fate is much more concerned about
what happens. It is not really concerned about when or why something is going to happen,
only about what will happen and the fact that it will happen.3
The recasting of fate in the chronologically orientated guise of time occurs only in Vlusp
and, insofar as it appears to rely on this poem, the recasting can be said to occur in Gylfagin-
ning, too.
The norns, however, are not the exclusive property of these two texts; we are allowed to
also consider what is said about them elsewhere.
What, then, is actually said about the norns? Several things, is the answer, and this paper
will not be able to go into detail with each individual reference. It will focus instead on what
appear to be the two most common ideas about the norns their connection to honour and
their connection to law.

Norns and Honour

The norns occur almost exclusively in contexts involving legendary human characters, not the
Old Norse gods. Vlusp apart, the mythological poems of the Edda do not mention the norns
whereas the heroic poems account for almost half of the total number of references. Also
skaldic poems account for a substantial number of references to the norns, and it seems note-
worthy that a relatively large number of the total references place the norns in the context of
heroic action. By this is meant the type of action that makes a hero or heroine truly heroic,
situations where the protagonists prove their heroic character by acting in ways that accord
with the high standards of the strict code of honour instead of succumb to the pressure that
they find themselves under. In other words, there seems to be a tendency to make reference to
the norns exactly in circumstances that will define a person as truly heroic or as not heroic.

3
Winterbourne (2004:1518); Bek-Pedersen (forthcoming)

107
 Brynhildr, in Sigurarkvia 57, blames the norns for the difficult situation she is in, being
married to Gunnarr instead of Sigurr, as well as for the emotional turmoil this has thrown her
into. She identifies the norns as the ones who are to blame, but this does not stop her from
exacting her revenge on human beings and she proceeds to urge Gunnarr successfully to
kill Sigurr. What motivates Brynhildr appears to be her sense of honour, the fact that she has
been made to break her promise to marry Sigurr, and she embarks on a horrific revenge ex-
pedition, spreading death all around her, but all the while feeling that she is forced to do these
things and does not have a choice. Her strong sense of honour is her choice, brought about, as
she sees it, by the norns.
Helgi, in Helgakvia Hundingsbana nnor 26, on returning to tell his valkyrja-lover Sigrn
of the outcome of the battle he has fought, finds himself having to communicate a tricky piece
of news. He says to her that: erat r at llo[] gefi not all is as you would have it be-
cause, although he has slain Hbroddr, whom Sigrn was expected to marry but decidedly
did not want, he has also killed her father and one of her brothers. The situation is not entirely
unlike that of Brynhildr with honour and love crossing each other in such a way that people
get caught in between the two and Helgi says that: nkkvi nornir valda the norns decided
some of this. His message seems to be that obeying the rules of honour and, with that, the
decisions of the norns is only what is expected of a hero, no matter the emotional cost in-
volved.
Angantr, in Hervarar saga ok Heireks 14, makes the same type of reference to the norns
when he expresses deep regret for having slain his brother Hlr in a dispute over who should
inherit from their father. Hlr demands half of the inheritance, and Angantr initially makes
him what he thinks is a very decent offer. But when Hlr hears himself referred to as ambt-
tarsonr son of a slave woman he regards this as an attack on his honour and sees armed re-
taliation as the only solution to the ensuing conflict. With this, the brothers end up on oppos-
ing sides and one of them kills the other, saying that evil is the judgement of the norns.
While the understanding appears to be that fate was what got between the two brothers, it
seems just as much to be questions of honour that separate them.
Gurn, in Gurnarhvt 13, seems to be thinking along the same lines. After she has suc-
cessfully seen her sons off, sending them to their almost certain death in avenging their sister,
she breaks into a long list of woes, describing how she, on the one hand, feels forced into car-
rying out horrible acts of revenge for the sake of honour and, on the other hand, feels tremen-
dous grief even as she does these things. She is caught in a tragic combination of what is nec-
essary in order to maintain honour and the inhumanity of doing this. For this, she is, as she
puts it, furious with the norns grm vark nornom, emphasising once more that honour and
fate interlink closely.
A happier take on the same situation comes from Hamisml 30 where the sons of Gurn,
Hamir and Srli, have managed to kill Irmunrekkr, but are themselves about to be slain by
the overpowering force of Irmunrekkrs men. They have upheld their own and their familys
honour and this is what is important to them, even at the cost of losing their own lives they
seem almost happy with the outcome, as if contemplating a deed well done, in spite of the fact
that: kveld lifir mar ekki eptir kvi norna no man lives out the evening after the norns give
their verdict, as it is phrased. Again, the norns appear to be involved in a game of honour.
The norns, then, have a strong tendency to be associated with situations where the heavy
demands of upholding ones honour and emotional stability cross each other. That is, when
the figures whom we encounter in Old Norse legends find themselves in situations where their
sense of honour requires them to act in ways that would otherwise be considered unaccept-
able, they often invoke the norns. They do not step down or shy away from what they feel
obliged to do, no matter the fact that certain death is frequently the outcome; instead, they
refer to this as fate and proceed to take the action deemed necessary.

108
 Norns and Law
Such are the contextual settings in which the norns are typically mentioned, and the great
concern that Old Norse heroes and heroines show for their reputation thus appears to interact
with their ideas about fate and destiny. Furthermore, some of their attitude to the obligations
imposed on them by honour and by fate is appears to be reflected in the vocabulary used to
describe the ways in which the norns exercise their influence.
The most common metaphor employed in such instances is a legal metaphor: norna dmr
judgment of the norns turns up in several texts, Ffnisml 11, Ynglingatal 24 and Hervarar
saga 14, as a phrase for death or dying, with the rather similar kvir norna verdict of the
norns being employed in Hamdisml 30. Not unlike such quasi-legal terminology are the
phrases: kve ek nkkvi nornir valda I say that the norns decided some of this in Helgakvia
Hundingsbana nnor 26 and Torf-Einarrs wording: rtt skiptu v nornir the norns settled it
correctly, which he uses in a description of how he avenged his father. Certainly the wording
of Vlusp 20: r lg lgo they laid down laws clearly draws on an image involving law
to describe how the norns operate.
It is important to note that there is no direct linkage between norns and the law as this op-
erated in human society. Instead, the key to the legal metaphor characterizing the norns may
be that law (court cases, juridical counselling and legal disputes) is not what the norns actu-
ally do, but that what they do is considered to be similar to this, only on a different level.
Underlying this legal metaphor seems to be a way of looking at the concept of fate as
though it were akin to some kind of law that it was definite and unavoidable, but also that it
was there in order to help maintain society and uphold a balance between various sections of
society.

Conclusions
As mentioned, it is not possible to discuss each individual reference to the norns in a space as
short as this. Instead, the present paper has taken a more generalising approach but the gen-
eralisations have been made on the basis of what are, statistically speaking, the notions that
are most often linked to the norns in Old Norse tradition.
The three points mentioned at the start: 1) that there are three norns; 2) that they are named
Urr, Verandi and Skuld; 3) that they represent Past, Present and Future, do not actually re-
flect the ideas that are most commonly presented in Old Norse tradition as it has come down
to us. Not that these three points are therefore invalid as such, but when it is clear that the ma-
jority of references to the norns are not at all concerned with these things, perhaps we should
reconsider how representative they really are.

Bibliography
Bek-Pedersen, Karen, (forthcoming) Fate and Weaving Justification of a metaphor.
Bek-Pedersen, Karen: Nornir in Old Norse Mythology. PhD thesis, University of Edinburgh. 2007
(unpublished)
Edda. Die Lieder des Codex Regius nebst verwandten Denkmlern. ed. Gustav Neckel. 1936. Heidel-
berg.
Egils saga Skallagrmssonar slenzk Fornrit 2. ed. Sigurur Nordal. 1933. Reykjavk: Hi slenzka
fornritaflag.
Hervarar saga ok Heireks konungs. ed. Jn Helgason, 1924. Kbenhavn: Jrgensen.
Snorri Sturluson Edda: Prologue and Gylfaginning. tr. Anthony Faulkes. 1987. Oxford: Clarendon
Press.
Winterbourne, Anthony, 2004: When the Norns have Spoken. Fate and Time in Germanic Paganism.
Madison-Teaneck: Fairleigh Dickinson University Press.

109
 smund austrvega: The Faroese Oral Tradition on smund
and its Relation to the Icelandic Saga
Chiara Benati, Universit di Genova, Italy
The story of smund kappabani (the Champion-Killer) seems to have been quite successful
in the Faroe Islands. The character of smund Faroese smundur appears in five Faroese
heroic ballads: Grms rma (CCF nr. 52), Heljars kvi (CCF nr. 63), Frgvin Olrina (CCF
nr. 81), Sniolvs kvi (CCF nr. 91) and Torbjrn Bekil (CCF nr. 98), four of which Grms
rma, Heljars kvi, Snjolvs kvi and Torbjrn Bekil preserved in more than one version.
In addition to these, smund the Champion-Killer is also mentioned in other texts, such as
Triks kongs rma (CCF nr. 97), where he is presented as one of Dietrichs warriors, or the
so-called Dvrgamoy ballads (CCF nr. 6, 7, 8, 9), a large group of texts dealing with all the
three thematic cores which, in Faroese oral tradition, are connected with the character of
Sigurd: the Nibelung cycle, the Dietrich epic and the smund tradition itself.

The Faroese Ballads on smund

Grms rma
This ballad, preserved in the CCF in two different versions A and B , narrates the adven-
ture of Grmur, son of Hildibrand, against smund, a family of giants and, finally, Sigurd. At
the beginning of the text, Grmur expresses his wish to sail in search for adventure, he has a
new ship built and eventually leaves. His voyage leads him to a quiet beach where he finds
the terrible (illur) smund, who challenges him:

smundur vi sn skjldin fra:

Grmur, eg bji tr t at stra!

Fyrsta sting, i Grmur legi,

smund burt r sali hevi.

Annan stingin legi t,

svori hinum av hondum br.

So kannst t tn hestin venda,

sla annans svr av hendi. (CCF 52A: st. 811)

They start duelling: with the first blow Grmur makes smund fall from his horse, while with
the second one he also loses his sword. In version A smunds last action in the ballad is rep-
resented by the words he addresses to his opponent: You can turn your horse and knock
someone elses sword from their hand!. In B smund is described while, after the duel, he is
carving evil (ramar) runes in a grassy garden.
After having met smund, Grmur continues his voyage in search of adventure. A storm
pushes his ship towards the shore of a land where he finds a giant. Grmur goes ashore and,
taking the sword in his hand, enters the cave where the giant lives with his family, beheads the
old giant and, after stealing gold and wealth, sails home. In the meantime the giants son
comes home and discovers his fathers corpse. Once he finds out who the murderer is, he
takes an iron bar on his shoulder and starts looking for Grmur, who, by that time, has reached
the hall of his residence. While hes there drinking both mead and wine, the young giant en-
ters the hall claiming revenge for his fathers death. The two start fighting and Grmur cuts his
opponent in two pieces ( lutir tv).

110
 On a sunny day, early in the morning, Grmur rides towards Hildarfjall (B: Lindarfjall),
where he meets Sigurd Sigmundarson. The two knights fight and Sigurd has the better of his
opponent. According to Grmur, this could happen only with the help of magic and runes. A
ends with this comment, while B narrates the feast celebrating their reconciliation.

2. Heljars kvi
Heljars kvi narrates smunds quest for Heljars beautiful daughter, Silri. In the CCF the
text is preserved in two versions, A and B. As soon as he hears of the existence of this
maiden, smund has his horse saddled to ride to Heljars. He reaches the palace gate, which
is protected by some white bears, kills the animals with his sword and pronounces some
magic words (rnir) to put to sleep the snake further protecting the threshold. In this way, he
is able to enter the hall. He immediately declares his wish to marry Heljars daughter. The
landlord wants Silri herself to decide about her future, since shes not easy to rule. After three
days in the hall, smund hasnt been able to see the girl yet. He, therefore, decides to change
strategy and ask for the help of other warriors. Hildibrand, his son Grmur and Virgar the
Strong accept to come to the palace, but none of them succeeds in seeing the maiden. Then
smund invites Sigurd, who following Nornagestrs advice brings along his good sword and
helmet. On his way towards Heljars, Sigurd meets an old man, who offers him a new sword
able to cut the hardest stone. He also tells him how to face both the white bears at the gate and
the snake. In this way, he reaches the hall where the other warriors sit around the table. After
five days, Silri finally appears in the room. smund makes his marriage proposal, but she
strongly refuses. He then suggests organizing a tournament: Virgar fights against Grmur and
Sigurd against smund. Being in trouble against Sigurd and fearing Virgar and Grmur might
attack him as well, smund finds a diversion heading towards the house of a dwarf living in
the neighborhood. They attack, defeat him and steal his gold. When the warriors are satisfied
with the battle, they all come back home:

rla um morgunin, slin skin,

t fru kempur hvr til sn.

Onga jomfr smundur vann,

hann helt so aftur Suurland. (CCF 63A: st. 103104)

And smund returns home alone, without any maiden.

3. Frgvin Olrina
In this ballad, too, the quest for a beautiful girl is narrated. In a castle on a mountain live two
maidens, Ingibjrg and Olrina. Once Grmur hears that, he rides towards the castle, where
Ingibjrg yields to him. In the meantime, Virgar Valintsson sends a messenger to Olrina. The
messenger reaches the castle and delivers the letter containing Virgars marriage proposal.
Olrina refuses, saying that Virgar has already hundreds of maidens in his castle to have fun
with and doesnt need her. In the following fight Geyti, Virgars messenger, confronts the
maidens father and defeats both him and two of his men. He, then, takes all the gold and sil-
ver he can find and leaves the castle, where the two girls sit alone. Hearing that, Sniolvur de-
cides to follow and kill him. Once Geyti is defeated, Sniolvur rides to the castle.
Thinking the castle and the girl are unprotected, smund wants to take advantage of the
situation and find a new mistress there:

smundur kom so sla degi

vi stt bgvi svr:

111
 N skal eg mr frillu kjsa,
deyur er hvuleysur herur. (CCF 81: st. 49)

Actually, the castle is not unprotected, but rather full of spells and magical ties, which Sniol-
vur uses to bind smund:

Borgin var innan vi mentir full,

leinkjur, lnur og garn,
Sniolvur spenti streingir upp,
smund fastan i jarn. (CCF 81: st. 50)

Bound to an iron bar, smund asks Sniolvur for mercy, is freed and can return to Selgjaland.
In the meantime Geyti, wounded, reaches his master Virgar and reports both on his mission
to Olrinas and on his fight against Sniolvur. Virgar decides, therefore, to go personally to the
castle. Olrina meets him in front of the door and challenges him saying nobody in the world
dares fight against Sniolvur. Hearing these words, Virgar cannot but invite Sniolvur to duel.
Sniolvur is defeated, Virgar escorts the sad Olrina to the castle and proclaims peace for both
farmers and criminals. The ballad ends by saying that Olrina finally entered a nunnery and
that she and Virgar are both saints in Heaven.

4. Torbjrn Bekil
This ballad, preserved in six different versions, narrates the adventure of smund against the
troll Torbjrn Bekil to avenge the damage he has caused to Halgas farm. Unable to kill the
troll by herself, Halga goes to smund and offers to marry him:

Eg havi fari um sland alt

kristi li at kanna,
va man mtt lti fara,
sjlv bii eg mr mann. (CCF 98A: st. 47)

She has travelled all over Iceland to meet Christian people and is looking for a husband, but
she wont sleep with smund before he crosses his sword with Bekil:

[] Hvar er smundur, sonur tn,

eg geri ta ei at loyna?
Hann kemur ei song vi mr
fyrr enn odd vi Bekil royna. (CCF 98A: st. 50)

smunds mother gives him coat, sword and armour. So equipped he is ready for the fight, he
goes to Halgas and accepts her proposal. In the meantime, Torbjrn sends a messenger to
Halgas. When he comes back, he reports having seen a tall man kneeling in front of her.
Wondering who this mysterious man could be, Torbjrn suggests he could be an ashman
(skudlgur)1 and asks fifteen of his men to go against him. They attack smund without
success and are all killed. Seeing that his warriors dont come back, Torbjrn gathers his fam-
1
As pointed out by Conroy (1978: 41), the original protagonist of this ballad couldnt possibly be the same s-
mund we meet in the Icelandic saga and in the saga-related Faroese ballads of the smund cycle. Assuming the
existence of two different characters named smund smund illi or ungi and smund kellingarson it is
possible to explain how a robber and rapist like smund illi could be asked for help against a troll by a maiden.
The nickname kellingarson, son of a witch, or of a female troll (Poulsen & al. 1998: 576), with which smund
is usually referred to, originally indicated only the positively connotated smund we find in Torbjrn Bekil, but
after the confusion of the two characters it became common in the ballads dealing with smund illis adventures
as well.

112
ily to ask for help and advice. The giant Rani, his relative (frndi), offers to fight against s-
mund with his right hand bound. They duel bitterly until smund is able to cut Ranis stom-
ach down to the navel. The same destiny is shared by Gyrilin and Atli, who wanted to
avenge the death of their relative. Finally Torbjrn himself challenges smund: He is killed
and, after him, his mother and his sisters as well.
Tired and wounded smund can go back to Halga:

Ta var smundur kellingarson,

fellur upp sni kn,
mean hann ta vna vv
til ektar festi sr. (CCF 98A: st. 183)

They get married and live happily ever after:

Drukki var teirra brleyp,

ktt var teirra lv,
fru so bi eina song,
smundur og hans vv. (CCF 98A: st. 184)

5. Sniolvs kvi
According to de Boor (1920: 214), this text represents the oldest and possibly the original
Faroese version of smunds story. Certainly this ballad and, in particular, one of the twelve
versions preserved in the CCF, B (447 stanzas divided in seven tttir) which I will analyze
in this study constitutes the longest and most complete witness of the reception of the s-
mund matter in the Faroe Islands.
The first part of version B Rana tttur tells the story of Hildibrands Brautwerbung.
Sitting in armour on his golden chair, Hildibrand asks his men if any of them knows a maiden
deserving to become his wife:

Hildibrand setst gyltan stl,

klddur brynju bl:
Hvar vita tit so vna jomfr,
mr er smi at f? (CCF 91B: st. 4)

One of his men his messenger starts speaking of the daughter of lav of Uppland, the
beautiful Silkieik, whose face shines like the brightest spring sun:

[] Hun ber ikki bleika br

undir snum gula hri,
heldur enn tann fagrasta summarsl,
i fagurt skn um vri. (CCF 91B: st. 8)

After hearing these words, Hildibrand leaves immediately for Uppland. In the meantime Rani
is heading to lavs to conquer Silkieik. Once he is there, Rani asks the girl to follow him to
sansland. Silkieik replies that she is already betrothed to another man, whose name she re-
fuses to tell. Instead of naming him, she calls for her brother, Sniolvur, who challenges Rani
to a duel.
Before the fight takes place, another knight in blue armour is seen riding towards the cas-
tle: Hildibrand. He enters the hall, goes to Silkieik and asks her to follow him to Selgjaland.
Her answer is positive: this is the knight she is betrothed to and whom she loves. In the duel
which follows Hildibrand defeats and kills Rani. Not knowing which of the two opponents

113
has died, Silkieik sends her brother Sniolvur to the battlefield to check it out. Hildibrand
doesnt know he is his future brother-in-law and attacks him. Sniolvur is quick enough to find
a shelter, so that Hildibrand gives up the fight and reaches Silkieik. They get officially en-
gaged and move to Brandavk.
At Christmas Hildibrand and Silkieik invite some nuns who predict that their son a brave
warrior who will hardly find his equal will fall under his fathers sword. Hearing this proph-
ecy, both Hildibrand and his wife are deeply shaken: while Silkieik wants Hildibrand to de-
stroy his sword, he prefers to sink it in the sea, so that nobody can find it:

Hildibrand sigldi for Heljar norur,

ta var mest av st,
hann tk sn gylta, bitra brand
og varpar havi t. (CCF 91B: st. 86)

In Sniolvs tttur Sniolvurs Brautwerbung is narrated. Wishing to find a girl deserving to

marry him, Sniolvur asks his mother for advice. When she suggests he should take a maiden
from Uppland, he answers that none of them can sleep in his arms and that he wants to con-
quer the daughter of the duke of Brunsvk. He has a new ship built and sails towards the
dukes land. Seeing him coming, the duke sends Sigurd to the beach to kill him. As soon as he
sets foot on land, Sniolvur expresses his wish to conquer the dukes daughter, Adalls. The
girl enters the hall and falls immediately in love with Sniolvur. She, therefore, accepts to fol-
low him to Uppland to marry him. After their wedding has been sumptuously celebrated, one
night Adalls wakes up from a strange dream: her husband was fighting against a knight who
eventually cut off his head. His name was smund.
The third part of the ballad, Golmars tttur, focuses on smunds search for Hildibrands
sword sunk in the sea. Having been told of the existence of this extraordinary weapon,
smund leaves for Gantarvk. There he meets duke Golmar who asks him what the reason for
his journey is. smund replies that he wants to seduce the beautiful Ingibjrg, Golmar tries to
resist, but is taken away by force, while smund obtains the object of his desire. The day after
he forces Golmar to accompany him and to show him the very spot where Hildibrand sank his
weapon. He dives repeatedly and finally finds it. With his new sword in hand smund kills
Golmar and returns home together with Ingibjrg.
Hildibrands tttur echoes the description of the fight between smund and Hildibrand at
the end of the smundar saga kappabana. The duels outcome is, however, different, since
the battle doesnt end with Hildibrands death, but with smund returning home naked after
Hildibrand has cut in two pieces his armour:

Hildibrand gav so strt eitt hgg

av so miklum mi,
klvur brynju av smundi,
hann nakin eftir st.

[] smundur snist haan burt

bi vi st og sorg,
glaur snist Hildibrandur
aftur sna borg. (CCF 91B: st. 267, 270)

In Virgars tttur, smund fights against Virgar Valintsson. After forging a new silver armour,
smund sends a messenger to Virgar to challenge him to a duel. Before answering, Virgar
listens to the girls living in his castle, who foresee his defeat on that very day. Sure that no-
one can fairly hurt him while on Skemming, Virgar leaves the hall and reaches the battle-

114
field. He is definitely stronger, but smund knows magic and, in this way, is able to hit his
opponent between the ribs and the shoulders. It will take a while before hes healed and can
return to his golden castle, where he probably still is.
The same narrative scheme of the challenge (German Herausforderungsschema) is present
in the sixth part of the ballad, smundar tttur, where smund encounters Sniolvur. This time
the challenge to the strong warrior is determined by smunds wish to seduce his wife,
Adalls. Sniolvur is killed in the duel and smund rides with his victims head to the latters
castle. Seeing him riding towards her, Adalls understands immediately that this knight is not
her beloved husband. When smund tells her he wants to seduce her, she repulses him, say-
ing she wont have any other man after Sniolvur and adds that smund could defeat him only
by employing magic. He, then, shows her Sniolvurs head, her belt goes into pieces and her
heart is broken.
Grms tttur begins with smund out at sea asking if any warrior is still alive. From that
moment on his name is changed into kappabani.2 Hearing of the existence of Grmur, a war-
rior against whom nobody dares fight, he sends him a messenger to invite him to fight. The
news that smund has already killed Sniolvur scares Grmur, who doesnt want to encounter
a warrior using witchcraft. He, therefore, offers him the armour of fifty warriors, but refuses
to duel with him. smund goes, then, to Oddur the Strong and tells him that one particular
warrior in the wood doesnt want to encounter him. Both Oddur himself and his relative vint
offer to go and fight against Grmur, who eventually defeats and kills them. Since Grmur still
refuses to duel with him, smund goes to Hildibrands, where Silkieik is telling her husband
what she has dreamt: he was fighting with his own son, unseated him and cut off his head.
Hildibrand reassures her saying that his sword is lying in the deep of the sea. When smund
enters his house and explains to him that he cannot defeat a dangerous warrior because he
refuses to encounter him personally, Hildibrand offers to fight against him if he can borrow
smunds sword. Grmur and Hildibrand meet on the battlefield and duel until the
unsuspecting father cuts his son in two pieces. Wishing to know who this valiant opponent is,
Hildibrand asks for his name and discovers that he is his own son. He throws away the sword
cursing the stomach and bones that have picked it up from the sea. The ballad ends with this
remark:

Satt er ta, i tala er,

so er greint fr,
eingin ger at fortvinast,
hvat nornur leggja . (CCF 91B: st. 447)

No-one can change what the Norns have devised.

Icelandic and Faroese smund

As appears from the account given above, the Faroese ballads on smund preserve a version
of smunds story diverging in many respects from the Icelandic smundar saga kappabana.
These divergences concern both the plot and the characterization of the protagonist, smund.
In the first chapter of the Icelandic saga the genealogy of the two protagonists, Hildebrand
and smund, is presented together with the story of the two swords forged by king Bulis
guests Olus and Alus, one of which will be fatal to Hildebrand in the dramatic climax of the

2
This nickname only occurs in another Faroese ballad, Dvrgamoyggin fagra or Dvrgamoy II (CCF nr. 7),
where stanza 53 of version B and C says: Ta er Sjurur Sigmundarson, / hann situr baki Grana: / smundur
ber eitt heiursnavn, / teir kalla hann kappabana. Elsewhere, smund is usually referred to as kellingarson.

115
narrative. None of the Faroese ballads reports this antecedent: in Sniolvs kvi the first time
we meet the sword it is already in Hildibrands possession and we only discover it has some
peculiarity when the three nuns who were probably originally the three Norns foresee it
will cause Grmurs death.
This is, in fact, another striking difference between the saga and Sniolvs kvi: in the
Faroese ballad the dramatic climax is not represented by a fratricide, but by a paternal filicide,
with an unaware Hildibrand killing his own son, Grmur. The family drama of a father fight-
ing against his offspring and finally killing him is reminiscent of the Old High German
Hildebrandslied and is hinted at in both the stanzas inserted at the end of the saga and known
as Hildibrands Death Song:

Stendr mr at hfi hlf in brotna,

eru ar taldir tigir ins tta
manna eirra, er ek at mori var.
Liggr ar inn svsi sonr at hfi,
eptirerfingi, er ek eiga gat,
viljandi aldrs synjaak (Detter 1981: 99).

and in the passage of the saga, where Hildebrands fury is described: En vanstilli essu er
hnum var a s hann son snn ok drap hann egar (Detter 1891: 98). According to
Halvorsen (1951: 15), the author of the written version of the saga derived this piece of
information from the death song, without properly understanding which episode was alluded
to.
The tragic epilogue of the saga is announced in the first chapter by Olus, the sword maker,
who curses king Buli saying this will cause the death of his grandson: Hann segir: jrngott
er sver, enda munu n nckur forfll liggja til hamingjubrotz, vat at mun vera at bana
inum gfgustum brrum, dttursonum num (Detter 1891: 82). In the Faroese ballad
Grmurs death is first prophesied by the three nuns invited by Hildibrand at Christmas and
then recalled by Silkieiks premonitory dream.3
Another fundamental difference between the two texts is represented by the connotation of
the protagonist. In fact, if in the smundar saga all actions and military enterprises carried out
by smund find their explanation both in the political logic of territorial expansion and in the
wish to protect and avenge his own or his allies sovereignty, in Sniolvs kvi smund is de-
picted as a robber and rapist who chooses his victims arbitrarily, following a sort of animal
instinct. Even his repeated successes on the battlefield cannot be fully ascribed to his skill and
valour, but rather to the use of sorcery he had probably learnt from his mother, a notorious
witch, as it is frequently hinted at:

Sniolvur var mn murbrir,

hans lki kann ikki finnast,
ta var alt vi illgerningar,
hann mundi sigur vinna.

Hann hevur tt sr mur ta,

ein er verst land,
hon hevur manga raska kempur

3
In version A of Sniolvs kvi only the premonitory dream is present and Hildibrand himself has it: Hildibrand
vaknar mijari ntt, / hann sigur sn dreym so brtt: / Undarligt hevur fyri meg borist / alla hesa ntt. //
Hildibrand so til ora tk, / lt sr vkja: / Mr tkti, sum mitt ga svr / var komi fr havsins dpi. //
Mr tkti, eg rei grnum vlli / vi so lti tr, / har kom Grmur, sonur mn, / eg hgg hans hvur fr. (CCF
91A: st. 158160).

116
 lagt for eiturgrand. [] (CCF 91B: st. 354355)

smunds familiarity with magic is also attested in other Faroese ballads, such as Grms rma
B, where he is depicted while carving some runes or Heljars kvi, where he pronounces
some magic words to put to sleep the snake protecting the threshold of Heljars palace. Arbi-
trariness of military action and brutal attitude towards women are distinctive features of s-
mund in all Faroese ballads, except Torbjrn Bekil. Here he has the positive connotation of
the hero fighting to protect Halga who has been attacked by the terrible troll. The scene of
Halga going to smunds and asking him to avenge her fathers death parallels in the saga
the sister of the dukes of Saxony complaining about Hildibrands tyranny, so that smund
moves against him to protect her and her country.
Similarly, the stanzas narrating how Torbjrn Bekil sends his warriors to fight against s-
mund who reproaches him with instigating other warriors to fight, instead of encountering
him personally echo this passage of the saga: mlti smundr: fyrir hv hleypir Hildi-
brandr t mnnum snum, en sitr heima sjlfr ok etr mik smmenni? (Detter 1891: 97).
However, only part of the narrative material employed in the Faroese smund ballads
finds a correspondence within the smundar saga kappabana. Apart from the above-
mentioned parallels between the saga and Torbjrn Bekil, Grms rma, Heljars kvi, Frgvin
Olrina and Torbjrn Bekil itself narrate a series of adventures involving smund as well as
other famous warriors, such as Sigurd or Virgar Valintsson and having no connection with
the events portrayed in the Icelandic saga. The same can be said for the Dvrgamoy ballads or
for Triks kongs rma. Even in Sniolvs kvi only three tttir (Rana tttur, Golmars tttur
and Grms tttur) correspond to the saga, while all other parts are but the obsessive and for-
mulaic repetition of the same narrative scheme resulting from the combination of the Braut-
werbungs- and the Herausforderungsschema: wishing to seduce the one or the other beautiful
girl, smund challenges their guardians who are usually stronger than he is, but are humili-
ated when not put to the sword. Only once, against Hildibrand, is he defeated and left naked
with his armour cut into pieces, but this humiliation simply represents one of the motivating
forces of smunds later military enterprises. On the whole, Sniolvs kvi moves from the
antecedent constituted by Hildibrands successful quest for Silkieik and the prophecy about
their sons destiny and, in a sort of spiral movement determined by the incremental repetition
of the duel sequences, culminates in the filicide committed by Hildibrand. The continuous
repetition of both narrative sequences (not only attempted seduction, challenge and fights, but
also premonitory dreams) and poetic formulae helps giving unity and cohesion to the ballad,
which despite various ellipses and inconsistencies, appears quite well-structured in compari-
son to other Faroese kvi.

Concluding remarks
In this study I have analyzed the reception of the smund story in the Faroe Islands in order
to get the most accurate possible image of the development of both the characters and the plot
of the Icelandic smundar saga kappabana on its way east.
On the basis of the results of this analysis, it is possible to agree with de Boor (1920: 214
and following) and exclude that Sniolvs kvi derives from the saga in the form we know it.
On the other hand, the remark on the unavoidability of the fate devised by the Norns in the
last stanza of the ballad and its correspondence with some of the Latin lines inserted by Saxo
in his account of Hildigerus death4 dont seem to be sufficient for assuming Sniolvs kvi
derives directly from Saxos Gesta Danorum, since allusions to the Norns are not infrequent
4
Sed quaecunque ligat Parcarum praescius ordo, / Quaecunque arcanum superae rationis adumbrat, / Seu quae
fatorum serie praevisa tenentur, / Nulla caducarum rerum conversion tollet. (Olrik Rder 1931: 204).

117
in Faroese oral poetry (Halvorsen 1951: 17). In addition to this, the confusion between nornir
Norns and nunnur nuns, which has evidently taken place in Rana tttur, indicates, in my
opinion, that the final reference to the Norns is perceived as purely formulaic and is in no way
put into relation with the events previously narrated.
As suggested by Halvorsen (1951: 50), Sniolvs kvi could be derived from another form
of smunds narrative material. In this version the character of smund has a strongly nega-
tive connotation: hes a robber and a rapist, who doesnt fight fairly, but achieves his victories
with the help of witchcraft. The transformation of his mother into a witch or female troll (kel-
ling) must have taken place once Faroese oral tradition had lost the consciousness of Hildi-
brand and smund being half-brothers. After this detail had been lost, there was no more mo-
tivation for the dramatic climax, since the death of either opponent wouldnt have constituted
a family tragedy. At this point, Sniolvs kvi was probably newly contaminated with the old-
est epic nucleus on Hildibrand and family tragedy was reintroduced in the tradition in the
form of paternal filicide.
Over the centuries in which the ballad survived in a purely oral dimension, this narrative
core was expanded through the insertion of new adventures, roughly corresponding to the
various tttir composing the ballad. These were concluded in themselves and could probably
be sung and danced to separately. Such a complex narrative was very likely to entertain
Faroese people for several kvldsetur (Wylie 1987: 43 and following) in a row, not too dis-
similarly from todays TV-dramas and soap-operas. These expansions some of which, as
Hildardalstr, appear to be quite late are often originated by smunds sexual desire and
contribute to increase, especially in a serialized performance, the tension towards the dramatic
epilogue of the story. Some of these additions are completely new compositions, while some
are the result of the incorporation of characters and events from originally separate traditions
into this particularly successful cycle (Conroy 1978: 38 and following).
smunds attraction for women plays a fundamental role in the other Faroese ballads
where he appears and which, apart from the case of Torbjrn Bekil, dont show any corre-
spondence with the Icelandic saga. In the majority of these texts, smund is simply a warrior
(usually on the quest for a beautiful girl), who often shares his adventures with other famous
heroes, such as Sigurd, Virgar Valintsson or Dietrich of Bern.
Since both the Icelandic saga and the Faroese ballads had been transmitted orally for centu-
ries before being fixed in the form which has come down to us, it is impossible to reconstruct
the exact course followed by the smund story on its way east, towards the Faroes. However,
I have tried to demonstrate, that once it had reached the Islands, the smund story, far from
having a linear development, was repeatedly altered as a result of the loss of original details,
of the insertion of new narrative material or of the contamination with other heroic traditions.

Bibliography
de Boor, Helmut. 1920: Die frischen Dvrgamoylieder. In: Arkiv fr nordisk filologi 37. Pp. 207
299.
Conroy, Patricia. 1978: Sniolvs kvi: The Growth of a Ballad Cycle. In: Frskaparrit 26. Pp. 3353.
Fornaldar sgur Nordrlanda eptir gmlum Handritum. Utg. av C. C. Rafn. 18291830. Vol. 2. Pp.
337354.
Froya kvi: Corpus Carminum Froensium. A S. Grundtvig, J. Bloch comparatum. 19512003.
Kbenhavn.
Froysk orabk. Utg. av J. H. Poulsen & al. 1998. Trshavn.
Halvorsen, E. F., 1951: On the Sources of the smundar saga kappabana. Oslo. (Studia Norvegica 5.)
Saxonis Gesta Danorum. Utg. av J. Olrik., H. Rder. 19311957. Kbenhavn.
Wylie, Jonathan, 1987: The Faroe Islands: Interpretations of history. Lexington.
Zwei fornaldarsgur (Hrlfssaga Gautrekssonar und smundarsaga kappabana) nach Cod. Holm. 7,
4to. Hrsg. von F. Detter. 1891. Halle a. S.

118
 The Other and the Noble Heathen: Ambiguous
Representations of Grettir and Finnbogi
Lisa Bennett, Flinders University, South Australia
In Chapter 38 of Grettis saga smundarsonar, there is a significant episode in which Grettir
begrudgingly agrees to swim an icy channel and retrieve fire for his freezing shipmates.1

Grettir burst into the house, unaware who was inside. By the time he reached land his cowl was
frozen stiff, and he looked frighteningly huge, like a troll. The people inside were startled and
took him to be an evil creature. They struck at him with everything they could lay their hands
on. A great scuffle ensued, and Grettir warded off the blows with his arms. Some of the men
struck him with blazing logs and the fire spread all over the house. Then he managed to leave
with the fire and returned to his companions. They lavished praise on his exploit and his brav-
ery, and said no man was a match for him. The night passed, and the crew felt they had been
saved when they had the fire. The weather was fine the next morning, and the merchants woke
up early and made ready to sail away, saying that they would go and find the people who had
made the fire, to find out who they were. They unmoored the ship and sailed across the channel,
but instead of finding the hut they saw a great pile of ashes with human bones inside, and felt
certain that it must have burned down along with everyone inside it. They asked Grettir whether
he had caused this mishap, and called it a pernicious crime. Grettir said what he had suspected
had come true, that they would reward him badly for fetching the fire, and said it was a bad
thing to help dishonourable men. Grettir suffered greatly for the incident, because wherever the
merchants went they said that he had burnt those men in their house. [] [Grettir] became so
despised that no one wanted to have anything to do with him. (CS II:111)

Scholars approach this fire-fetching episode from several different angles. Much criticism
focuses on Grettirs swimming prowess, here and elsewhere in Grettis saga, as a parallel to
Beowulfs swimming contest with Breca (Puhvel 1971:2778; Jorgensen 1978:556; Went-
ersdorf 1975:1467). Other studies view Grettirs encounters with supernatural beings, par-
ticularly Glmr, as the cause of his persistent ill fortune and eventual downfall (Cook
1989:239; Poole 2004:6). From this perspective, Glmrs curse henceforth outlawry and
killings will fall to your lot, and most of your deeds will bring you misfortune and improvi-
dence (CS II:107) can be held accountable for Grettirs killing of rir and his family as
described above. Finally, and most importantly for this discussion, there is the scholarly opin-
ion that the burning-in is an accident resulting from Grettirs ill fortune. Scholars defend
Grettirs actions by saying he performs a good deed in fetching fire for his companions; he is
generally seen as behaving heroically in this passage (Hume 1974:476; Pencak 1995:8;
Bragg 2004:2467; Poole 2004:15; Hawes 2008:31, 36) or it is assumed that his failure to
undergo the ordeal condemns him to outlawry (Hamer 2008).
However, if Grettir is to be considered a hero then he is a hero out of his time, as Kathryn
Hume argues: Grettirs stormy relations with society gain immeasurably in importance when
viewed not just as the result of personal quarrelsomeness, but as reflecting a clash between
two sets of values [] both of which have merit but which cannot really coexist (Hume
1974: 4856). In other words, Grettirs heroic aspirations are modelled on pagan virtues that
have no place in post-Conversion Iceland. More to the point, although I largely agree with
Humes stance, in this paper I would like to argue that Grettir is remembered in a negative
1
Unless noted otherwise, all English translations of Grettis saga smundarsonar are Bernard Scudders from the
Complete Sagas of Icelanders series; hereafter cited (CS II:pp.). All English translations of Finnboga saga
ramma are John Kennedys from the same series; hereafter cited (CS III:pp.).

119
light in this saga, not only because he demonstrates outdated heroic qualities such as a pro-
pensity for fighting to resolve disputes instead of using cunning or for battling with vikings
(Hume 1974; Hume 1980:11; Bennett 2008) but because, even though he is one of the only
central characters in the slendingasgur to live his entire life in the Christian period, Grettir
is guilty of committing a burning-in. Grettirs intentions in fetching the fire are irrelevant
here; what is important is the implication that Grettir has performed a deed that, in Christian
terms, is no longer acceptable. Although many scholars describe this burning-in as an acci-
dent, and indeed the passage can be interpreted in this way, we nevertheless get the impres-
sion that Grettir is guilty of committing the pernicious crime his companions accuse him of
because of his ambiguous characterisation.
As I have argued elsewhere, the burning-in motif, which depicts the incendiary killing of
victims trapped within their own homes, frequently appears in both pre- and post-conversion
periods in the Sagas of Icelanders and the Sturlunga saga compilation (Bennett 2007). The
treatment of this motif seems to reflect how the saga compilers wanted to perceive their an-
cestors, rather than how historical events might actually have unfolded. For instance, the pat-
tern of burnings seems to indicate that remote pagan characters were able to commit burn-
ings of epic proportions without incurring any serious repercussions in the saga narratives. By
contrast, late pagans, whose stories appear in the period immediately preceding the year 1000,
were frequently portrayed as innately anticipating Icelands acceptance of Christianity
(Lnnroth 1969; Strmbck 1975:23; North 1991; Jochens 1999a:621; Tulinius 2000:253),
even though such behaviour and such supernatural awareness of Icelands conversion is his-
torically illogical. We find, therefore, that many planned burnings are thwarted in sagas de-
picting late tenth century events, which suggests the societys growing reluctance to accept
burnings even though major characters still propose them. Moreover, once the storylines
move beyond the year 1000, only two successful burnings occur in the Sagas of Icelanders:
Grettirs accident and the burning of Njl. Several important figures in the Sturlunga saga
compilation propose burning-in as the solution to ongoing feuds, but when these burnings are
not thwarted, the victims are few and the perpetrators are considered rather despicable charac-
ters. In this way, although such perceptions may be historically inaccurate, the impression we
get is that through the recurrent use of the burning-in motif, saga writers were able to incorpo-
rate their pagan ancestors into their Christian world without condemning them wholly for be-
haviour that is seen in Christian terms as completely unacceptable by the same token, they
were able to paint Christian burners as solitary, pagan, bestial Others (cf. Aalto 2006), as is
the case with Grettir.
Comparing Grettirs burning-in episode to an almost identical sequence of events in
Finnboga saga ramma which was written in the early fourteenth century, well before Gret-
tis sagas composition (B II 1993:42; Hawes 2008:25; cf. Hamer 2008:21) sheds much
light on this argument. Like Grettir, Finnbogi is caught out at sea in inclement early winter
weather. His ship is dashed to pieces on the rocks but because Finnbogi is depicted as a for-
tunate character in his saga, he survives the crash and makes his way to shore. Parallels be-
tween Finnbogi and Grettirs situations are impossible to miss at this point in the narrative:

It was then dark night and neither frost nor wind was in short supply. All his clothes froze on his
body, and the snow storm was intense. [] When he had been walking for a while he noticed
the smell of a fire, and a little after came to a large and impressive farm [] where he heard
many people inside. They were sitting by fires. (CS III:231)

Both characters set out by themselves to seek fire and, as William Ian Miller argues, Soli-
tariness was always ground for suspicion. Being alone gave one the option of holding ones
own counsel and thus the option to be a thief or a murderer, a secret killer [] Only in the

120
rarest of circumstances in the sagas does a man of good character and intention go somewhere
alone (1990:102). In this episode, Finnbogi should be as open to suspicion of solitariness as
Grettir is in his saga, yet the events in Finnboga saga unfold in a very different way.
In Finnboga saga we are presented with one of the rare circumstances where a man of
good character and intention goes somewhere alone. The promise of a warm fire lures
Finnbogi to a farmhouse filled with people, much as it does Grettir. However, considering it is
night time and he, like Grettir, is an enormous man so apprehensions of a veritable frost
giant would be fully understandable as Poole observes (2004:7) Finnbogis behaviour is
beyond reproach. He openly approaches the farmhouse, and knocks at the door three times:

He knocked at the door, and a man spoke, asking one of the servants to answer the door. They
replied that they did not care though the hammering went on all night. Finnbogi knocked a sec-
ond time, more loudly. The man asked them to open the door. They replied that they would not
do so, even if a troll beat on it every night. Finnbogi struck a third time, so hard that everyone
was startled. (CS III:231)

The farmer disparages his servants objections and answers the door himself, at which point
Finnbogi introduces himself as the son of Asbjrn and as an Icelander (CS III:231). Despite
the possibility that a troll might be at their door, the significance of which will be discussed
further below, the farms inhabitants are only startled by Finnbogi because of the strength and
volume of his knocking. All of the elements for disaster are present in this episode: it is night;
Finnbogi is alone and thus open to all of the negative connotations solitariness carries; there is
a house full of people a roaring fire ready to scorch them all. Yet, despite the potential for a
burning-in to occur in Finnboga saga, nothing of the sort happens. Instead, Finnbogi acts rea-
sonably and respectfully, just as one would expect a Christian character to behave regardless
of the fact that Finnbogi has yet to be converted to Christianity. As a result, he is invited in
from the cold and Everyones attitude towards him was very cheerful (CS III:232).
By comparison, when Grettir approaches a farmhouse in almost identical circumstances,
we get the impression that he is behaving as a thief or a secret killer. He is enlisted to seize
fire for his companions, but ends up committing a burning-in; even worse, when he is directly
asked about the burning, Grettir refuses to take responsibility for his actions and publish his
crime. In his analysis of the laws referring to theft in Grgs, Theodore M. Andersson points
out the difference between a rnsmar robber and a jfr thief (1984:497). Robbery in-
volves the open seizure of property, and results in full outlawry if the perpetrator is prose-
cuted; whereas a thief, who is also condemned to full outlawry if discovered, disgracefully
commits his crime in secret. Similarly, the penalty for murder, and it is murder if a man
hides [a killing] or conceals the corpse or does not admit it (Gr I 1980:146), is outlawry. Al-
though the punishment for all of these offences is the same, there is a significant social and
legal stigma attached to the act of thievery and secret killing, as Andersson notes: If a man
took something by force and used it openly, his conduct was less reprehensible in the eyes of
the law than if he took it in secret and continued to hide it (1984:497). In this episode, Grettir
may be seen as seizing rirs fire, since his companions encourage him to do so, but he
steals rirs life when he does not confess about the burning-in: in other words, he commits
murder. In addition, Grettir is aligned thematically with his pagan great-grandfather, nundr
trftr, who commits a mass burning-in in the opening section of Grettis saga (CS II:56;
Hume 1974:479; Hawes 2008:223). Thus, Grettirs behaviour in his approach to the farm-
house, in his being involved in a mass burning, and in his failure to publish his crime puts
him on the same semantic level as thieves, murderers and pagan mass-burners, all of which
highlights his non-Christian tendencies.

121
 Unlike Finnbogi, who introduces himself and makes his lineage clearly known, Grettir is
cast as an Other in this fire-fetching episode. The only declaration we get of Grettirs iden-
tity in this sequence is that he is a troll, and he certainly behaves like one. The word troll is
also used when Finnbogi knocks at the farmhouse door: the servants say that they will not
answer it even if a troll beat on it every night. But since Finnbogi acts like a reputable man,
most notably by speaking instead of barging in unannounced, it soon becomes obvious that
we are not intended to equate him with trollishness. A similar example, as Richard North ar-
gues, occurs when Egill is about to give himself up to Eirkr Blox in Egils saga.

A servant goes in and announces him:

mar er hr kominn ti fyrir durum, segir hann, mikill sem troll.
Theres a man come here standing at the door outside, he says, as big as a troll. (F II 178,
ch.59) With proverbial humour Egill is made a living example of the expression troll fyrir du-
rum, a liability, and yet in the weakening of his stature from urs [ogre] to troll, the author
also seems to made [sic] him more human than his father. (1991:148)

Like Egill, Finnbogi goes through the right channels: he approaches the door and is human-
ised as a result, despite his potentially frightening appearance. When Barr, the owner of the
farm, realises who Finnbogis father is he says, I expect to find the wolf where I see his ears
(CS III:232); yet while the words troll and wolf are used around Finnbogi, there is no indi-
cation that these terms refer to him as a character. There is potential for Finnbogi to be inter-
preted as a troll or Other in this episode, but his behaviour prevents us from seeing him as
such.
By contrast, rir and his family believe Grettir is a troll fyrir durum (literally, a troll, as
well as the proverbial liability), which explains why they react so violently against his intru-
sion. There seems to have been a tradition about Grettirs trollishness that predates Grettis
saga. For instance, in the opening passage of Fstbrra saga, written in the late thirteenth
century (B II 1993:42), we learn:

He was an outlaw at the time, and wherever he went he managed to have people give him what
he wanted. However, what Grettir called gifts would not have been regarded as such, or so read-
ily given away, had people not felt that they had a troll on their doorstep. It was this that eventu-
ally led to the farmers gathering their forces, capturing Grettir, condemning him to death, and
building a gallows on which they intended to hang him. (CS II:32930)

Elsewhere in his own saga, Grettir is referred to, or treated as, as a troll (Hawes 2008:36).
Moreover, he is consistently associated with the marginal elements of the Icelandic world in
which trolls belong. First and foremost, he becomes a permanent outlaw one of the tilegu-
menn (out-lying men) after he commits the burning-in. Kirsten Hastrup explains, [t]his
category differed from the category of skgarmenn (outlaws, forest-men), in that it was []
labelled mythological [] [it] was used as a designation of any who left ordinary human
company [] Later, in Iceland as well as Norway, the notion attained a more wicked mean-
ing, including outlaws and non-humans (1985:142). In addition, Grettir adopts the name
Gestr for some of his (mis)adventures, which means stranger as much as it can mean
guest (Heinrichs 1994:501) and his most powerful confrontations are against supernatural
or mythological creatures. Hastrup adds, it required an outlaw to defeat supernatural beings;
only an outlaw could meet them on equal terms. To fight on common ground, the fighters had
to inhabit the same space. Outlaws and supernatural beings were co-inhabitants of the wild,
and in this sense they were allies against society (1985:153). Thus, although Grettir earns the
reputation as a great ghost-buster in his saga, the impressive battles he fights with trolls,
giants, and the undead simultaneously diminish his honour and underscore his exclusion from

122
Christian society. The impression we get is that Grettir has more success fighting these crea-
tures than fighting other men because, as a Christian who behaves like a pagan (and who, be-
cause of Glmrs curse, has the ability to see what no human wants to see: the supernatural
creatures that haunt the dark (Hawes 2008:20)), Grettir does not inhabit the same space as
other people. He is not worthy to fight with Christian men, so he must find opponents who are
on an equal par with him.
Furthermore, as Sverrir Jakobsson points out, If being a stranger did not automatically
make one a marginal person, some habits of strangers might have worked towards their mar-
ginalization. [] It was, for instance, common to describe marginal figures in terms of besti-
ality (2007:152). It is significant that Grettir seems to be metaphorically characterised as
hamrammr a person who could change shape, usually into an animal, while retaining his
human identity (Hastrup 1985:153) which, I would argue, is suggested by his trollish char-
acteristics in the fire-fetching episode. Perhaps more striking, though, is the sequence of
events in the bear-slaying episode in which Grettir is semantically linked to the realm of ber-
serkir and bears (Heinrichs 1994:55; Miller 1990:208; cf. Hawes 2008:2930). Simply put,
berserkr means bear-shirt (Hastrup 1985:153). Jens Peter Schjdt notes, the traditional way
of looking at berserkir [was] as warriors who in some way were associated with bears []
being a warrior of this special kind demanded that they were strong and savage like bears
(2007:145). Once again, a comparison between Finnbogi and Grettirs association with bears
is illuminating. I would like to argue that Grettirs behaviour when he confronts a bear not
only foreshadows his reprehensible actions in the fire-fetching episode, but it situates him as
an Other by implying that he is a bear himself. By contrast, Finnbogis upstanding treatment
of the bear he fights (in almost identical circumstances, as is the case with the fire-fetching
episode) firmly places him in the world of honourable Christian men.
In both sagas, a vicious bear awakes from his winter hibernation and proceeds to slaughter
livestock and wreak havoc on the farms in the district (CS II:835; CS III:2324). Also in
both sagas, prominent farmers enlist Grettir and Finnbogi to help hunt down the bear. In Gret-
tis saga, a boastful character named Bjrn (another bear) attempts, and fails, to kill the
beast; while in Finnboga saga, Brr, the farmer who Finnbogi opted not to burn-in in the
previous chapter, demonstrates the qualities of a noble heathen: he chooses Christian brains
over pagan brawn by calling together an assembly, at which he has the bear legally outlawed,
which gives people an officially authorised reason to pursue it. Teams of warriors intend to
fight the bear in both sagas, yet Grettir and Finnbogi each end up confronting the animal at
night, by themselves. While his companions are asleep, Finnbogi takes his weapons and sets
out to find the bears den; when he finds it, he once more demonstrates that he is a rational,
honourable, and even-tempered man. He tries to reason with the bear before doing anything
rash:

Stand up bear, said Finnbogi, and attack me. That would be more worthwhile than lying on
this sheeps carcass.
The bear sat up, looked at him, and flopped down.
Finnbogi said, If you think Im over-armed against you, Ill remedy that.
He took off his helmet, laid down his shield, and said, Stand up now, if you dare.
The bear stood up, shook his head and lay down again.
I understand, said Finnbogi, you want us to be on equal terms.
He threw away his sword and said, It will be as you wish. Stand up now if you have the sort of
heart one would expect, rather than that of the most cowardly of all beasts. [] They fought
for a long time [] but it ended up with Finnbogi forcing the bear onto his back and breaking
asunder his spine. (CS III:231)

123
Finnbogi stresses the importance of being on equal terms with his foe, but there is no indica-
tion that he becomes a bear to achieve this equality. If anything, the bear in Finnboga saga
is metaphorically lifted up to Finnbogis level and is humanised: this exchange outlines how
one would expect a person to reason with another person, not the way a beast would challenge
another beast. Thus, as is the case in the fire-fetching episode, Finnbogi is positioned near the
realm of the marginal he is associated with words like troll and he places himself on equal
terms with a bear but he consistently acts like a noble (almost-Christian) human.
Grettir, on the other hand, is symbolically linked to the berserkir when he sets out wearing
a shaggy fur cloak, which his companions throw into the bears den (cf. Hume 1974:481).
When they depart, Grettir turns back alone, affording himself the opportunity to take his own
bad counsel, much as he does in the fire-fetching scene. The bear lashes out when he sees
Grettir; Grettir instantly unsheathes his sword and chops off the creatures paw. There is no
discussion here, only action: Grettir said later that holding off the bear was his greatest feat
of strength (CS II:84), a comment which emphasises his habit of relying on actions and bes-
tial strength rather than on human reason. Grettir takes every advantage he can, succumbing
to the visceral mindset of survival of the fittest, and he is not above using his weapon if it
means he will win. He returns home wearing his tattered cloak, carrying the bears paw as
proof of his victory, and boasting about his achievement in verse. In this episode, Grettir
metaphorically proves that he is stronger than two bears, Bjrn and the animal whose life he
has taken, which reinforces his symbolic status as an Other as a bear who enjoys fighting
with other bears. Conversely, Finnbogi does not even want to take credit for his deed: He
arranged [the bears dead body] so that things looked as they were when he had arrived, took
his weapons and went back to the farm. He was very stiff and lay down on his bed, pretending
he had been asleep (CS III:233). Given the unmistakable parallels and significant diver-
gences between these two bear fights, once cannot help but think that the author of Grettis
saga was familiar with Finnboga saga and that he adapted crucial scenes from it for thematic
purposes (cf. Hume 1974). If this is the case, then the author of Grettis saga ultimately
chooses to differentiate his hero from Finnbogi, in that Grettirs animal qualities are empha-
sised rather than his human ones.
Grettir is not the only saga character to be described as monstrous for narrative purposes.
We need only consult Egils saga, with its generations of dark, half-troll, wolfish shape-
shifters, for a prime example of such Othering (Jochens 1999b:88; North 1991:14755).
Evil trolls, like Kolbjrn and his crew in Brar saga, also represent nature and are repeat-
edly likened to animals (Jakobsson 1998:66); but benevolent trolls and giants, such as Brr
Snfellsss, are not always Othered in this way. It seems reasonable to suggest that saga
authors did not have the same expectations of non-human characters as they did of human
protagonists (Jakobsson 1998:54). Therefore, Grettirs ambiguous characterisation, in which
he is simultaneously cast as human and Other, is significant.
While Janice Hawes states, It is Grettirs contradictory personality above all else that
places him in liminal space and threatens to place him completely outside the human sphere
(2008:32), I would argue that this personality is manifested in Grettirs actions and these
actions determine how readers see him. Thus, in situations where we might expect to see bad
behaviour, such as when he is out alone at night bearing a striking resemblance to a troll,
Grettir behaves like the beast he resembles. The same circumstances are not problematic for
Finnbogi, however, because he does not behave badly. There is no overt indication that
Finnbogi chooses not to commit a burning-in at the farmhouse even though all of the ingredi-
ents are there for him to do so; but his characterisation as a noble heathen especially when
it is seen in light of the pattern of thwarted burnings in the pre-year 1000 period in the other
Sagas of Icelanders means that he simply will not do it. For most of his saga, Finnbogi does
not consciously follow Christian tenets (and he could not logically have been expected to) yet

124
he is portrayed as doing so nonetheless. His noble behaviour seems to be rewarded when, at
the end of Finnboga saga, Finnbogi is one of the first people to convert to Christianity (CS
III:257).
By contrast, Grettir is consistently represented as a bestial or monstrous (Hawes 2008)
character throughout his saga. These negative characteristics are exaggerated when Grettir
confronts the bear; but it is when he causes and then conceals a burning-in in the Christian
period that we get the ultimate proof of his Otherness. Unlike Finnbogi, Grettir is a Chris-
tian from the start; and though Hamer argues Grettis saga is more than a morality tale (19
20), we cannot help but speculate that the saga author was remembering Grettir in a negative
light for instructive purposes, particularly when we situate Grettirs crime amongst the many
other instances of burning-in in the Sagas of Icelanders.

Bibliography
Aalto, Sirpa, 2006. Categorizing Otherness in Heimskringla, Preprint papers of the 13th Interna-
tional Saga Conference, Durham and York, 6th12th August. Vol. I. Ed. by J. McKinnell, D.
Ashurst & D. Kick. Available online: http://www.dur.ac.uk/medieval.www/sagaconf/aalto.htm
rmann Jakobsson, 1998. History of the Trolls? Brar Saga as an Historical Narrative, Saga Book
25. Pp. 5371.
Andersson, Theodore M., 1984. The Thief in Beowulf, Speculum 59. Pp. 493508.
Bennett, Lisa, 2008. Conquering Viking Enemies in Icelandic Cultural Memory, in Vikings and
Their Enemies. Ed. by K. Burge. Melbourne: Vikings Research Network. Pp. 7081
2007. The most important of events: the burning-in motif as a site of cultural memory in
Icelandic sagas. Journal of the Australian Early Medieval Association 3. Pp. 6986.
Bragg, Lois, 2004. Oedipus Borealis: The Aberrant Body in Old Icelandic Myth and Saga. Madison:
Fairleigh Dickinson University Press.
Cook, Robert, 1989. Reading for Character in Grettis saga, in Sagas of the Icelanders: A Book of
Essays. Ed. by J. Tucker. New York: Garland. Pp. 226240.
Grgs / Laws of Early Iceland I, 1980. Trans. by A. Dennis, P. Foote & R. Perkins. Winnipeg: Uni-
versity of Manitoba Press.
Hamer, Andrew, 2008. Grettis saga and the iudicium dei, in Northern voices: essays on Old Ger-
manic and related topics, offered to Professor Tette Hofstra. Ed. by K. Dekker, A. MacDonald and
H. Niebaum. Leuven; Dudley, Mass.: Peeters. Pre-print available online at:
http://www.liv.ac.uk/english/pdf/Grettis_saga_and_the_iudicium_dei1.pdf. Pp.124.
Hastrup, Kirsten, 1985. Culture and History in Medieval Iceland: An anthropological analysis of struc-
ture and change. Oxford: Clarendon Press.
Hawes, Janice, 2008. The Monstrosity of Heroism: Grettis smundarson as an Outsider, Scandina-
vian Studies 80. Pp. 1950.
Heinrichs, Anne, 1994. The Search for Identity: A Problem after the Conversion, Alvssml, 3. Pp.
4362.
Hume, Kathryn, 1974. The Thematic Design of Grettis Saga, Journal of English and Germanic Phi-
lology 73. Pp. 469486.
1980. From Saga to Romance: The Use of Monsters in Old Norse Literature, Studies in Phi-
lology 77. Pp. 125.
slensk Bkmenntasaga II. 1993. Ed. by Bvar Gumundsson & al. Reykjavk: Ml og Menning.
Jochens, Jenny, 1999a. Late and Peaceful: Icelands Conversion Through Arbitration in 1000,
Speculum 74. Pp. 621655.
1999b. Race and Ethnicity in the Old Norse World, Viator 30. 79104.
Jorgensen, Peter A., 1978. Beowulfs Swimming Contest with Breca: Old Norse Parallels, Folklore
89. Pp. 5259.
Lnnroth, Lars, 1969. The Noble Heathen: a Theme in the Sagas, Scandinavian Studies 41. Pp. 129.
Miller, William Ian, 1990. Bloodtaking and Peacemaking: Feud, Law, and Society in Saga Iceland.
Chicago & London: University of Chicago Press.
North, Richard, 1991. Pagan Words and Christian Meanings. Amsterdam: Rodopi (Costerus New Se-
ries 81).

125
Pencak, William, 1995. The Conflict of Law and Justice in the Icelandic Sagas. Amsterdam & Atlanta:
Rodopi. (Value Inquiry Book Series 21)
Poole, Russell, 2004. Myth, Psychology, and Society in Grettis saga, Alvssml 11. Pp. 316.
Puhvel, Martin, 1971. The Swimming Prowess of Beowulf, Folklore 82. Pp. 276280.
Schjdt, Jens Peter, 2007. inn, Warriors, and Death, in Learning and Understanding in the Old
Norse World: Essays in Honour of Margaret Clunies Ross. Ed. by J. Quinn, K. Heslop & T. Wills.
Turnhout: Brepols. (Medieval Texts and Cultures of Northern Europe 18). Pp. 137151.
Strmbck, Dag, 1975. The Conversion of Iceland: A Survey. Trans. by. Peter Foote. London: Viking
Society for Northern Research.
Sverrir Jakobsson, 2007. Strangers in Icelandic Society 11001400, Viking and Medieval Scandina-
via 3. Pp. 141157.
Tulinius, Torfi H., 2000. The Matter of the North: fiction and uncertain identities in thirteenth-century
Iceland, in Old Icelandic Literature and Society. Ed. by M. Clunies Ross. Cambridge: Cambridge
University Press. (Cambridge Studies in Medieval Literature). Pp. 242265.
Wentersdorf, Karl P., 1975. Beowulfs Adventure with Breca, Studies in Philology 72. Pp. 140166.

126
 Mikill vinr rs
Eyrbyggja saga och namngivningen i runinskrifter
Maths Bertell, History of religions dept., Stockholm university, Sweden

Inledning
Denna underskning r sprungen ur tanken att det frkristna namnskicket p de i de flesta fall
kristna runstenarna ger en bild av det hedniska samhllet. Uppland ska ju enligt traditionen ha
varit svrkristnat och man skulle drfr kunna tnka sig att namnskicket skulle kunna besitta
en seghet som ger en lngsam frndring. Adam av Bremens Uppsalaskildring frn 1075 ger
intryck av en sdan seghet och ven Upplandslagen frn c:a 1300 ppekar att ingen skall
blota t avgudar, och ingen skall tro p lund eller p stenar (Adam av Bremen [1075] 1984;
Svenska Landskapslagar, Upplandslagen 1979:12). Eftersom inget sdant nmns i Sderman-
nalagen skulle man ju kunna tnka sig att de frkristna inslagen i det upplndska namnskicket
skulle vara strre n i Sdermanland.

Forskningen om frleder i personnamn

Personnamnen p Rkstenen, mjligen 12 stycken, ger oss en inblick i 800-talets namnskick.
Stenens inskrift kallar skguden rr, men intressant nog r inget av de namn som kan uttol-
kas ur den omdiskuterade texten sammansatt med frleden or- vilket man kanske skulle ha
kunnat vnta sig (Lindquist 1947:9). I Assar Janzns underskning av Landnmabks
namnskatt kan han konstatera att de tvledade namnen med r- som frled r de mest fre-
kommande, hos bde mans- och kvinnonamn. Frleden r ocks typiskt nordisk. De vanligas-
te mansnamnen r orsteinn (83 st), rr (73 st), rir (56 st), orkell (56 st), orbjrn (53
st), orgeirr (51 st), rarinn (45 st), orgrmr (40 st) och de vanligaste kvinnonamnen r
rr (57 st), orgerr (43 st), ords (37 st), Helga (36 st), runn (34 st), ra (22 st),
Yngvildr (19 st), Hallveig (19 st), Valgerr (18 st), orbjrg (17 st), Vigds (16 st), orkatla
(15 st) och Jrunn (15 st) (Janzn 1947:27 f). Vi har en tydlig bild av namnskicket i Norge
vid tiden fr landnamet p Island. Utvecklingen i Norge och p Island gr sedan isr vad det
gller flera av namnen: en del frsvinner i Norge men lever vidare p Island och tvrt om.
Senare, under medeltiden, har namnskicket ndrat skepnad: mnga ldre namn lever kvar men
mnga med hednisk bakgrund har ftt kliva t sidan fr kristna namn. or-namnen har mist
sin ledande stllning i biskop Eysteins jordebok, ven knd som Rde Bog, frn 12001300-
talets biskopsdme fr Oslo med omnejd. Det vanligaste namnet hr r det kristna namnet Jn
(Janzn 1947a:29).
Det vikingatida namngivandet tycks ha fljt vissa principer, gemensamma fr hela Norden.
Frldrarnas syfte med sina telningars namn tycks ha varit nrmast magiska och med en fr-
hoppning om att kunna pverka sina barns framtida karaktr, efter tanken att mnniskan r
vad hon kallas (se Janzn 1947a:31 m litt). Givetvis ville man ge dem en s frdelaktig start i
livet som mjligt. I namngivande kan vi ocks spra en tydlig genusordning: de karaktrsdrag
som ansgs manliga inympades med manliga attribut som strid, rikedom, makt etc medans
flickornas framgng tycks ligga i vrden som sknhet och hjlpsamhet, men ven valkyrjed-
rag uppmuntras (Janzn 1947a:31). Janzn menar att namngivandet r i hgsta grad medvetet,
dvs den som dps till orgeirr blir skyddad av skguden, Eirkr den framfr andra mktige
ska vara mer gynnad n andra. Detta ska dock ha frndrats genom den allt mer populra va-
riationsprincipen som gav ett ondligt antal varianter av tvledade namn. Denna frndring av
namndelarnas genomskinlighet ska ha skett i redan i ldre germansk tid, ven om en med-
vetenhet kring vissa namns betydelse levt kvar lngt fram i tiden. Variationsmjligheterna gav

127
dock mjlighet till hgst orimliga eller betydelselsa kombinationer, rnulfr (rn varg),
Hallsteinn (flat sten sten). Drfr kan man inte lgga alltfr stor vikt vid enstaka namns be-
tydelse (Janzn 1947a:32; Janzn 1947b: 238). I Landnmabk kan dock Janzn konstatera att
mnga namnled gr i arv i namngivningen, bde fr- och efterleder. Ibland hoppar dock upp-
kallningsprincipen ver en generation. Vanligast r att faderns namn teranvnds men det
frekommer ven att moderns namnleder vandrar vidare (Janzn 1947a:33). Av Janzns reso-
nemang kring medvetenhet om namnens betydelse, tycks just de teofora namnen vara sdana
som anvnts i ett aktivt namngivande i Landnmabk.
Den alliteration som knnetecknar germansk namngivning under folkvandringstid har inte
ftt spridning i det nordiska namnskicket (Janzn 1947a:34 f; Janzn 1947b:237). Variations-
principen och uppkallelsen har antagligen varit rdande i samhllets hgsta skick och som
sdan blivit normgivande fr samhllet som helhet (Janzn 1947:36). Det r ocks mjligt att
namn i olika tider varit p modet, t ex genom kunganamn och andra i tiden vlknda karakt-
rer och d orsakat en stor tillfllig frekvens (Janzn 1947:38).
Wessn vill se s- och s-frlederna som varianter p or-frleden och att namngivandet
ska ses som tminstone delvis tnktes ge braren beskydd (Janzn 1947:65, 96). I samma
anda skulle frleden Gu- med variationen Go- tnkts ge braren gudarnas beskydd, om n
utan att peka ut ngon srskild gud. Dock anses Ragn- av regin gudar vara s pass frblek-
nad att man inte kan rkna med ett medvetet namngivande under vikingatid (Janzn 1947:87,
96).
or-, r- med olika vokalvarianter r den vanligaste frleden i Landnmabk. Anled-
ningen till detta r de breda folklagrens gud i vstra Norden under vikingatiden. Inom kunga-
tter lyser dock or-namnen med sin frnvaro. Frleden existerar i princip bara i Norden. Inte
ens kristendomens genomslag dmpar till en brjan namnets popularitet (Janzn 1947:94).
Frleden finns i en mngd variationer men ven former som rr (urspr. rr) och rir
(urspr. r-vr), samt lfr (urspr. rulfr) rknas till namngruppen. Ocks smeknamnet Tos-
ti av orsteinn hr dit (Janzn 1947a:94 f). Vi kan ocks se en genusstruktur som terspeglas
i frlederna. Mannens territorium r vapnen och frleder som Brand-, Bryn-, Geir-, Grm-,
Hjalm-, Ketil-, Odd-, Skjald- och mjligen Jr-, samt ven frleder som Sig-, Vg- och Gunn-
pekar i den riktningen. Men ovanstende kan ocks peka i ett valkyrieideal fr sm flickor.
Som tnkt hrskarinna i det egna hemmet kunde ocks, enligt Janzn, frleder som Rann- och
Sal- tnkas peka i en sdan riktning, med en idealiserad bild av kvinnan som husfru (Janzn
1947a:96 f; Janzn 1947b:256, 260).
I det fornsvenska materialet terfinns benmningar eller namn p gudar och andra mytolo-
giska vsen i en stor del av namnen. Syftet med dessa frleder r, precis som i de vstnordiska
exemplen, att ge individerna gudarnas uppmrksamhet (Janzn 1947b:256 f).

Syfte
I Eyrbyggja saga ges vi ett mjligt scenario kring frkristna gudar och namngivning. Reflek-
teras sagans tradition i det bevarade runstensmaterialet? I sagan berttas om den Torstillvnde
Hrlfr Mostrarskegg som var s hngiven sin gud att han kallades rlfr. I sagan kan vi lsa:

Hrlfr var hfingi mikill ok inn mesti rausnarmar; hann varveitti ar eyunni rshof ok var
mikill vinr rs, ok af v var hann rlfr kallar. (Eyrbyggja saga 1935:6)

rlfrs fixering vid skguden begrnsar sig inte bara till honom sjlv. Platsen fr hans nya
hem och ven hans son fr frleden r-. Intressant nog dper sedermera sonen sin son till
rgrimr.

128
 au rlfr ok Unnr ttu son, er Steinn ht. enna svein gaf rlfr r, vin snum, ok kallai
hann rstein, ok var essi sveinn allbrgrr. (Eyrbyggja saga 1935: 12 f)

En sumar at, er orsteinn var hlfrtgr, fddi ra sveinbarn, ok var Grmr nefndr, er vatni
var ausinn; ann svein gaf orsteinn r ok kva vera skyldu hofgoa ok kallar hann orgrm.
(Eyrbyggja saga 1935: 19)

orsteinns hustru heter ra, men det r orsteinn som ger barnet till rr och ger honom
hans framtida roll. orgrmr blir sedan hofgoi nr han har ldern inne och fr ven han en
son, som dock fds efter hans dd. Sonen fr ocks namnet orgrmr och efter honom brise-
rar anvndandet av frleden i slkten fullstndigt (Eyrbyggja saga 1935:20). Detta namngi-
vande skulle ju kunna vara en litterr knorr och ett stt att visa hur fanatiska de hedniska fr-
fderna var, men skulle ocks kunna terspegla en ldre namntradition. Vi vet ocks att strre
kungatter rknade sig som hrstammade frn de strsta gudarna rr, inn och Freyr. Men
hrskarskiktet har alltid haft en normerande inverkan p befolkningen och en inte alltfr djrv
tanke vore att den aristokratin p samma stt tecknat sin egen slkts ursprung eller tminstone
tillhrighet till ngon av de stora gudarna och att detta skulle kunna haft utslag i namnskicket.
Fr att riktigt pressa tanken kan vi ju ocks tnka oss att en sdan tillgivenhet ocks skulle
kunna haft sitt inflytande p ortnamnen. Enligt Eyrbyggjasagans tv exempel ges namn an-
tingen av frldrar eller av omgivningen. I sagan r det faderns frtjnade frled som frs
vidare till sonen och sonsonen etc. Namnmaterialet i runinskrifterna tillter endast en under-
skning av den frsta prinicipen, dr frldrarnas namndelar ges vidare till nsta generation.
De vikingatida runinskrifterna r ju nstan uteslutande kristna och ngon grad av hedniskhet
kan inte frvntas av generationerna i 1000-talets runinskrifter. Dremot kan en viss seghet i
namnskicket vara mjlig dr en tidigare tradition baserad p ett frkristet namngivande kan ha
drjt sig kvar.

Frgestllningar
Finns det en liknande namngivningsmall i upplndska och sdermanlndska runstenar? Kan
vi upptcka flera teofora frleder eller efterleder inom samma familj, exempelvis en fader vid
namn Frejsten och en son som heter Torsten? Gr det att se en regionalitet i Tornamnen ge-
nom runinskrifternas spridning i Uppland och Sdermanland?

Underskning
Jag har valt ut de inskrifter som har tv eller flera Tornamn, oavsett om de hr till samma ge-
neration eller inte. Sedan har jag analyserat varje inskrift och valt ut de inskrifter som kan
tnkas ha ett Eyrbyggjamnster. Runristaren Torbjrns skalds namn har inte rknats med i
underskningen och inte heller de andra runristare som har namn som brjar p r- eller har
namn avledda drav. Underskningsomrde r Sdermanland och Uppland.
I Uppland finns enligt Jan Owes Svensk runristningsfrteckning 1474 stycken och i Sam-
nordisk runtextdatabas terfinner vi 1392 av dem. I Sdermanland har vi totalt 471, respekti-
ve 361 av dessa i Samnordisk runtextdatabas. Owes sammanstllning bygger p de stora run-
verken i Norden, i vrt fall Sveriges runinskrifter. Uppdateringar av Samnordisk runtextdata-
bas genom nya fynd och ny belysande forskning gr ju att listan pverkas i frga om antal
runinskrifter (Owe 2002:3 f).

Islndskt sagamnster i namngivningen?

Det finns i materialet ett litet antal stenar som uppvisar vad man skulle kunna tolka som ett
Eyrbyggja-mnster i namngivningen, dvs tv generationer med anknytning till namnet

129
rr, antingen som frled eller som avledning. I Uppland r dessa U 418, U481, U 510, U
838, U 878 och U 1034. U 418 r en tydligt kristen inskrift med bnen Gu hjalpi nd hans.
U 878 r intressant eftersom orgerr reser stenen efter sin man och orunn efter sin far. Om
vi antar att orgerr r mor till runnr ser vi att frleden r- d har gtt i arv mellan kvin-
nor. En av Tors stora uppgifter var just att beskydda Asgrds och Midgrds kvinnor, vilket tar
sig uttryck i myterna, men ven i det vikingatida gravskicket, ngot som jag diskuterat i ett
tidigare sammanhang (Bertell 2003:193 ff). Inskriften r dock den enda i sitt slag av de som
jag underskt. Den liknar dock U 104, dr Ingiora och hennes sner orsteinn och rir om-
nmns. Fadern heter Sveinn. P U 1034 har tv av tre sner ftt namn med frleden r-:
orbjrn och orsteinn, sner till orfastr. P vriga stenar med fler n ett r-namn r fr-
hllandena inte riktig lika tydliga. P U 144, U 151, U 176, U 628, U 653 och U Fv1992;157
har tv av brderna r-namn men inte fadern eller modern (i de fall dr hon nmns). P U
180 och U 275 har tv av brderna r-namn men ingen ur den ldre generationen nmns,
stenarna r resta av brder ver broder. Familjerna skulle allts kunna ha en namntradition
kring r-namnen men detta avsljas inte i inskrifterna.

Tor och Frej

Tv inskrifter r srskilt intressanta ur ett annat frkristet perspektiv: U 275 och S 232. P U
275 har tv av brderna r-namn medan den tredje brodern heter Freysteinn. P S 232 har
Freysteinn och gull rest stenen ver sin far orbjrn och rir ver sin bror. P den frsta
stenen har vi en brdraskara med frkristna gudanamn i frlederna och p den andra tycks en
r-generation fljts av en Frey-generation.
P de srmlndska stenarna terfinns ett Eyrbyggja-mnster p S 84 och S 233. S 84
har en bn Gu hjalpi nd orbjarnar. Mjligen r ocks S 61 en Eyrbyggja-inskrift d den
omnmner en man orsteinn som rest stenen ver orbjrn. Inskriften fortstter sedan med att
nmna att Slvi och Smir rest stenen ver sin bror. Vi kan drfr anta att orsteinn r orb-
jrns far, eftersom det antagandet skulle flja det gngse mnstret att den ldre generationen
nmns frst och drefter syskon. Slktingar brukar nmnas fre vnner, varfr vi kan anta att
relationen mellan mnnen har varit av slktkaraktr. Eftersom de inte var brder, som de v-
riga i inskriften terstr alternativet fader. Jmfr med exempelvis U 878 och U 151. Det r
uppenbart att namnleder gr i arv i bde Uppland och Sdermanland i ngon mening men att
detta gller inte endast frlederna utan ven efterlederna. Se till exempel S 229.
I inskrifterna S 54, S 229, S 336 och S 360 nmns endast brder med r-namn och
ingen ur en ldre generation br frleden. Man kan drfr inte tala om Eyrbyggja-namn i des-
sa fall. P S 336 nmns inte den ldre generationen alls, men intressant nog br alla tre br-
der namn vars frled har en frkristen bakgrund: orgsl, sgautr och orgautr. Trots detta
avslutas inskriften med bnen Hjalpi Gu nd, vilket tyder p att namnen r neutraliserade i
religis mening.

U 418 (P3) rlinghundra hrad, Norrsunda socken

()(u)[rfas]tr * auk| |kitilui * au * litu * stain [*] rito [*] ifti[R] * urstain * faur sin *
ku hialbi hont * hons
orfastr ok Ketilv au ltu stein rtta eptir orstein, fur sinn. Gu hjalpi nd hans.
Stenen r smyckad med ett kors och inskriften avslutas med en bn.

U 481 (P4) Lnghundra hrad, Lagga socken

orkis[l] -uk orstin auk uibiarn auk olifr --tu raisa stin eftiR orbi-
arn faur sin
orgsl ok orsteinn ok Vbjrn ok leifr [l]tu reisa stein eptir orbjrn, fur sinn.

130
 Stenen r smyckad med ett kors, men saknar bn.

U 510 (P4) Lnghundra hrad, Krsta socken

frystin orbiurn fasti uiniutr ulfr kuntiarfr aiR brur raistu stin :
at * orstin : faur sin *
Freysteinn, orbjrn, Fasti, Vnjtr, Ulfr, Gunndjarfr, eir brr reistu stein at orstein,
fur sinn.
Stenen r smyckad med ett kors, men saknar bn.

U 838 (P3) Lagunda hrad, Nystra socken

ufr * auk * orfatr * air * litu ' raisa ' stan * at * orborn * faur * sen * koan **
hir maa ' stanta ' stain ' ner ' brautu ' auk ' (k)ilauk ' ri * kirua ' merki ' [at] (b)(o)a---
* sen :
lfr(?) ok orfastr eir ltu reisa stein at orbjrn, fur sinn gan. Hr mun standa ste-
inn nr brautu ok Gillaug r gera merki at b[nda] sinn.
Stenen saknar kristna markrer.

U 878 (P3) Hagunda hrad, Hagby socken

[iokir l]it raisa st[ain iftr olaf bunta sin auk fastlauk auk orun
iftiR faur sin ]
orgerr(?) lt reisa stein eptir laf, bnda sinn, ok Fastlaug ok runnr eptir fur sinn.
Saknar uppgifter om kors. Stenen terstr endast som fragment.

U 1034 (P5) Norunda hrad, Tensta socken

orbia(r)n ' auk ' orstain ' uk ' styrbiarn ' litu raisa stain ' eftiR ' orfast ' faur sin
ybir risti
orbjrn ok orsteinn ok Styrbjrn ltu reisa stein eptir orfast, fur sinn. pir risti.
Stenen saknar kristna markrer.

S 84 $ (KB) Vsterrekarne hrad, Tumbo socken

a...R ...et * raisa * stain * at * orbiorn * borour * sin * sun * orstainR * i skytiki *
ku * hiolbi * ant * * orbiornaR *
... [l]t reisa stein at orbjrn, brur sinn, son orsteins Skyttingi. Gu hjalpi nd orb-
jarnar.
Stenen r smyckad med ett kors.

S 233 $ (FP) Sotholms hrad, Sorunda socken

kun[i au]k orfastr raistu stain at ori fa-... ... [amut]i (h)iuk
Gunni ok orfastr reistu stein at ri, f[ur sinn]. mundi hj.
Stenen r utsmyckad med ett kors, men saknar bn.

Tv av de upplndska stenarna saknar kristna markrer, de vriga r med all tydlighet kristna.

Gr det att se en regionalitet i Tornamnen?

I Uppland har 13,6 % av inskrifterna inslag av r-namn, vilket r frre n i Sdermanland
dr siffran r 15,5 %. I Uppland r det intressant att lgga mrke till att vissa omrden, dr det
nd finns relativt mnga inskrifter som t ex i Bro hrad med 24 inskrifter, helt saknas inslag
av r-namn, eller Lyhundra hrad dr vi hittar hela 50 inskrifter men endast en med r-
namn. Allra flest, procentuellt sett, hittar vi i rbyhus hrad (3 av 8 dvs 37,5 %), Simtuna

131
hrad (4 av 14 dvs 28,6 %) och Hagunda hrad (11 av 47 dvs 26,2 %). Det finns allts ingen
jmn frdelning av namnen ver landskapet som helhet.

I Sdermanland finns en liknande bild. Intressantast r Daga hrad med sina 19 inskrifter,
men som helt saknar inslag av r-namn. ven Eskilstuna och Strngns med 7 inskrifter var
saknar r-namn. Ttast r Svartlsa hrad (10 av 34 dvs 29,4 %), knebo hrad (6 av 20 dvs
30 %) och Jnkers hrad (3 av 11 dvs 27, 3 %). I kerbo hrad r hela 50 % av inskrifterna
r-inskrifter men det stora procenttalet beror p att kerbo endast har 2 inskrifter och en av
dem har ett r-namn. Antalet inskrifter varierar allts kraftigt mellan olika hrader, men vi
ska ocks tnka p den stora procentuella utfall som enskilda belgg ger i hrader med f in-
skrifter. Istllet r hrader med mnga inskrifter, men f eller inget belgg fr r-namn mer
intressanta. Detta tyder p att en viss regionalitet kan sknjas och att namnet trots sin popula-
ritet konkurreras ut av andra frleder.

132
 Frndring ver tid
Jag anvnder mig av Anne-Sofie Grslunds datering av runinskrifter som baserar sig p in-
skrifternas ornamentering och utformningen av rundjuren, srskilt deras huvuden. Grslund
delar in de runristningar som har ett rundjur med huvudet i profil i 5 perioder, samt en period
dr rundjurets huvud avbildats uppifrn:

Stilgrupp Frslag till absolut datering (Grslund 1998:86)

P1 ca 10101040
P2 ca 10201050
P3 ca 10501080
P4 ca 10701100
P5 ca 11001130
FP samtidig med P1 och P2, ca 10101050

Grslund har anvnt sig av uppgifter av genealogisk karaktr och uppgifter om Ingvarstget
fr sin datering. Hon har ocks utnyttjat arkeologiskt material med sker datering (Grslund
1992:195 ff). En del inskrifter faller utanfr indelningen ovan. De har istllet delats in i grup-
perna korsbandsstenar (KB) och raka runinskrifter (RAK). Korsbandsstenarna r ristade s att
slingornas ndar utgr ett fundament fr ett kors. Dessa anses oftast hra till det ldre skiktet
av runinskrifter. Grslunds datering r inte oproblematisk och har ocks kritiserats, nu senast
av Magnus Kllstrm (Kllstrm 2007:64 f).
Jag anvnder mig av dels den tidigare databasen Mlsten och den nyare Samnordisk run-
textdatabas.
Intressant nog ligger tyngdpunkten p r-inskrifterna i de tidigaste perioderna i Sder-
manland men i Uppland dremot ligger de i de senare perioderna. Om man till tidig period
rknar perioderna 1 och 2, samt FP och RAK har hela 53,6 % r-namn, mot endast 1,4 pro-
cent r period 4 och 5. I Uppland r bilden annorlunda: perioderna 1 och 2, FP och RAK har
sammanlagt endast 15 %, medan 23 % faller under period 4 och 5.
Mikromiljer: I Uppland har 28 av 141 stenar med r-namn ven kristna bner, dvs 19,9
%. I Sdermanland r siffran lgre: 10 av 67 stenar dvs 14,9 %.
Ortnamn: Finns det ett samband mellan Tor-namn p runstenar och teofora ortnamn inne-
hllande frleden Tor-? Sockennamn? Kan omrden med mnga ortnamn p Tor- tyda p
gammal Torsbygd, Tor-namn p runstenar och Tor-namn i sockennamn?

133
 Det finns lite som tyder p ett samband mellan teofora ortnamn och namnen p stenarna.
Enligt Per Vikstrands Gudarnas platser finns ett mindre antal teofora ortnamn i Uppland som
kan betraktas som skra och de flesta av dessa upptrder i hrader dr vi inte har ngra r-
namn alls. Torstuna i Torstuna socken i Torstuna hrad och Torsvi i Torsvi socken i Trgds
hrad r exempel p ortnamn som tagits upp i socken- och hradsnamnen, vilket borde indike-
ra att namnen betecknat centrala platser redan i frkristen tid. Per Vikstrand diskuterar mj-
ligheten till en koppling mellan tingsplatser och ortnamn med r- i frleden men tycker sig
inte kunna se ngon tydlighet i materialet (Vikstrand 2001: 146, 162 ff).

Slutsatser
De f exempel vi har p inskrifter med tv generationer med r-namn, visar inte p ngon
namntradition med som en stark stllning i omrdet. Inte heller kan vi se ngon koppling mel-
lan ortnamnen som indikation om starka r-omrden i personnamnskicket.
Vid tiden fr resandet av runstenar i Svealand tycks namnen fr de frkristna gudarna som
del i personnamn helt tappat sin laddning, precis som de teofora ortnamnen. Frleden r-
frkommer i de mest kristna sammanhang, med kors och med bner i samma inskrifter. De
skillnader som nd finns ver tid och rum i underskningen mste ses som popularitetsvgor
som antagligen liknar dem som finns i dag. De teofora ortnamnens laddning tycks ocks ha
gtt ur tmligen fort, eftersom vi har flera exempel p sockennamn och kyrkor som vertagits
frn frkristen tid, exempelvis Odensala i Uppland (Othinsharg 1293). Det gr inte heller att
tolka namnen som ngot vittne frn en segdragen kristning av Uppland. Om namnen hade haft
sin laddning kvar lngre i Uppland n i Sdermanland, dr namnen nra nog frsvinner i peri-
od 4 och 5, borde tyngdpunkten inte ligga s sent i dateringen. Istllet borde vi kunna se en
jmnare frdelning av namnen ver alla perioder. Dock verkar namnelementen ha haft regi-
onalt starka omrden, stillvida att det helt saknas i vissa hrader i bde Sdermanland och
Uppland, trots rikt runinskriftsbestnd, men r kraftigt representerat i andra omrden, procen-
tuellt sett.
Namnelementet traderas p fler stt n i sagan ur ett genusperspektiv. Mn rver frleden
av sin far, men ven av sin mor. Dttrar rver p samma stt frleden fdernet och p mder-
net.
Underskningen leder vidare till nya frgestllningar. Det tycks som om inget av de beva-
rade namnen frn urnordisk tid har r- eller motsvarande som frled. r det ett sent inslag i
namnskicket och i sdana fall varfr?

Litteraturlista
Bertell, M., 2003: Tor och den nordiska skan. Frestllningar kring vrldsaxeln. Stockholms univer-
sitet: Stockholm.
Eyrbyggja saga. slenzk fornrit 4. 1935. Hi slenzka fornritaflag: Reykjavk.
Grslund, A-S., 1992: Runstenar om ornamentik och datering II. Tor 24.
Grslund, A-S., 1998: Ornamentiken som dateringsgrund fr Upplands runstenar. A. Dybdahl & J. R.
Hagland (red.): Innskrifter of datering. Dating inscriptions. Tapir: Trondheim.
Janzn, A., 1947a: De fornvstnordiska personnamnen. Nordisk kultur 7. Albert Bonniers frlag:
Stockholm.
Janzn, A., 1947b: De fornsvenska personnamnen. Nordisk kultur 7. Albert Bonniers frlag: Stock-
holm.
Kllstrm, M., 2007: Mstare och minnesmrken. Studier kring vikingatida runristare och skriftmilj-
er i Norden. Stockholms universitet: Stockholm.
Lindquist, I., 1947: versikt ver de ldsta skandinaviska personnamnen, med huvudvikten p de
urnordiska. Nordisk kultur 7. Albert Bonniers frlag: Stockholm.
Mlsten, The Department of Archaeology and Ancient History, Uppsala university.
http://www.arkeologi.uu.se/ark/projects/malsten/default.asp

134
Owe, J., 2002: Svensk runristningsfrteckning. http://hem.bredband.net/b688266/runologi.html
Samnordisk runtextdatabas, Institutionen fr nordiska sprk, Uppsala universitet. http://www.nordiska.
uu.se/forskning/projekt/rundata
Vikstrand, P., 2001: Gudarnas platser. Frkristna sakrala ortnamn i Mlarlandskapen. Kungl. Gustav
Adolfsakademien: Uppsala.

135
 The Good, the Bad and the Devil!

On rewriting a Religious Motif in some Virgin Martyr Legends

Kjersti Bruvoll, Dept. of Scandinavian studies and comparative literature, NTNU, Norway
Do not have any other gods before me. The first of the Ten Commandments is well known
to most of us. In Exodus 20:4 it is followed by the statement: You shall not make for your-
self an idol The Christian condemnation of idols is a central point in early Christian theol-
ogy and missionary activity, and thus also in several of the early Saints Legends. Idols are
presented as an example of the many traps the Devil has devised to lure humankind away
from God, and to worship them therefore leads to damnation. These are simple religious
truths well suited to missionary activity and to early Christian didactic writings. One of the
most important dichotomies in the Virgin Martyr Legends is therefore the fight between Good
and Evil, represented on the one side by the Holy Trinity and on the other by the Devil and his
servants and idols. However, the presentations of opposed groups differ between the extant
Old Icelandic versions of the legends, and the Latin versions that they have been shown to be
based on, and at times quite substantially. Here, I will primarily examine the alterations in the
Old Icelandic versions of the Legends in the presentations of Evil, the Devil and his other
worldly and human servants; that is the consequences of the changes made in the translation
process and, perhaps even more important, later rewritings of the Virgin Martyr Legends.
Many of the quite extensive changes in the Old Icelandic versions of Virgin Martyr Leg-
ends are to be found in manuscript AM 233a fol. In this manuscript we see a series of amplifi-
cations and interpolations, from single words to larger extracts of text, that are nowhere to be
found in the respective Latin versions of the same legends or, for the most part, not even in
other manuscripts or versions that contain the Old Icelandic Legends. Many of these transpo-
sitions and transformations have therefore presumably been made in the transmission process
rather than in the translation process. The main focus will therefore be on this manuscript,
which today contains 29 leaves that have originally belonged either to one single or two dif-
ferent manuscripts. The Legends of Virgin Martyrs are all to be found in a part that has cer-
tainly originally belonged to one single manuscript of large format.1 The legends of Virgin
Martyrs preserved here are the Legends of St. Agatha, St. Margaret, St. Katherine and the
three sisters Fides, Spes and Caritas. The legend of St. Agnes was also originally been part of
this manuscript, but today all but a very small fragment of it has been lost.
Some of the dissimilarities between the Latin versions and the Old Icelandic translations of
them are naturally due to misinterpretation or to faults or unconscious choices of wording in
the transcription process. This being said, however, there is also a clear tendency in the later
reworking of the texts that they involve conscious changes such as amplifications, reductions,
rewordings or omissions; changes made either to adapt the texts to a vernacular public or po-
etics or to clarify ambiguous parts or elaborate on motifs, phenomena and so on. When it
comes to the motif of the Devil and his idols, another tendency is that a translator or a later
editor adds information about the idols that are not originally mentioned or thematised in the
Latin version. This is evident in a longer interpolation in the Legend of St. Agatha in AM
233a fol, but it is also manifest in several of the shorter and less evident amplifications in this
and other legends in both the same and other manuscripts.

1
I Helgafellsbkur fornar, (1966:3038) lafur Halldrsson writes that the second hand in AM 233a fol. from
the part of the manuscript containing the Virgin Martyr legends is probably the same hand that has copied
several manuscripts in the monastery at Helgafell.

136
 The Idols
A central motif in hagiographic literature is the futility of worshipping the idols. In the Latin
legend of St. Agatha, in the version BHL 133, her main message in the discussion between
herself and her persecutor on this topic is that the idols are bad role models, since they are in
fact demons that wish to alienate humankind from God. People who worship them can also
become like them or meet the same fate as them, since they become tainted by their evil. A
prayer to the idols is therefore futile and stupid. In AM 233a fol we can, however, see a fur-
ther elaboration of this motif through a series of amplifications and amendments, the most
important of which is a quite extensive interpolation following a speech from Agatha in which
she states the stupidity of appealing to stones for help in the Old Icelandic manuscript am-
plified to stocks and stones. Some of the aspects of and points made about the idols are less
important, merely implicit or even not touched upon at all in the original version, as such; the
interpolation in AM 233a fol clarifies, elaborates on or introduces new aspects when it comes
to the idols:

[] ea huerr mun suo farlgr allr skynsem. sa er del uet aa skapara snum. ok kennr stt
los. at hann mun ganga myrkra gtu. sva afskaplga. at hann mun dyrdka greylg go n. er
gor ero manna hndum eptr hnum uerstum monnum. er uert hafa hemnum. er aa allan htt
voro fullr af dfullgum golldrum. ok grnngum. ok ar me af allzkonar fullfna. ok
hordmum. morum ok odaauerkum. er aufundn munu uera era lfssgum. mean au
voro her hem snum bauluuum lkomum. En eptr era skemdarfult lf foro au tl helvts
ok brenna ar ok frosa me dflnum. ok hans rum. ok allr er er au dyrka utan enda. en
au mega huark barga ser ne orum helldr voro au eptr dauann blotut ok mgnut af un-
dum mnnum. ok dfuls krapt tl fyrrdmngar elfrar salfum em ok llvm em er em
treysta. ea hyggr u hnn grmm gus oun. at ek muna sakr red nnar ea fyrr ognr
psla nna ea fyrr nckurskonar blml orda nna gefa upp ast ok traust drottns mns
Jesu Krst. er bd er gu ok mar. [My transcription]

This interpolation adds what for this legend are some new, but generally well-known theo-
logical concepts and truths about the idols. For example, the fact that the idols are made by
human hands is stressed, and we are told that they are sculpted after the worst men that have
ever existed in this world; criminals, murderers, adulterers and so on, men that after their
deaths went straight to Hell where they are to burn and freeze in eternity. Life and espe-
cially punishment after death is in this way further elaborated upon in comparison with the
original. Also emphasised is the fact that anyone who worships the idols will suffer the same
fate as them. In fact, idolatry leads to damnation both for the idols themselves and for their
worshippers. Another perspective added to the text with this interpolation is the insistence on
Christ as both human and God. This is for different reasons also a very interesting addition,
but not one that will be touched upon here.
Some of the above-mentioned aspects of idols and idolatry are further elaborations of al-
ready treated, or at least implicitly mentioned, characteristics and qualities of idols and eternal
punishment; others are new to the text, while not to the genre. Most of what is mentioned here
is information about the idols which is known from other texts of the same genre. The insis-
tence on idols as made by human hands is an important focus in other Virgin Martyr legends.
In the beginning of the prologue of the Latin BHL5303d version of the legend of St. Margaret
for example, humanitys foolishness in worshipping the idols is linked to the raging madness
of the devil. At the same time, the uselessness of the deaf, dumb and blind idols, made by
human hands, is stressed:

137
 Adhuc tamen obtinebat insaniae diaboli rabies homines, et idola surda et muta et caeca, manu
hominum facta adorabant, quae nec illis proderant, nec sibi.2

The foolishness of this behaviour is further emphasised by the comparison of it to the glories
one can expect if one submits oneself to the power of God. The reason why the the idols are
not of benefit to humankind but rather the contrary, lies in their nature as human-made
symbols. This is an argument that is also well known in the Bible. In the Book of Psalms
113.12163 we read:

The idols of the Gentiles are silver and gold, the works of the hands of men. They have mouths
and speak not: they have eyes and see not. They have ears and hear not: they have noses and
smell not. They have hands and feel not: they have feet and walk not: neither shall they cry out
through their throat. Let them that make them become like unto them: and all such as trust in
them.4

This Biblical quotation is rendered in its entirety in the legend of St. Barbara (cf. Unger 1877:
155, line 1723), and in the legend of St. Margaret the characteristics of blindness, deafness,
dumbness and lameness are on several occasions linked to the idols, together with a stress on
the fact that they are made by human hands and as such useless. In the legend of St. Agatha
however, this fact is only implicitly understood and not stressed in any way. The legend men-
tions that the pagans have made the idols themselves using copper, marble and plaster and
covering them with gold. It also mentions that it is futile and insane to call upon them for help
as they are made of stocks and stones. But the deficiency of the idols is not further elaborated
on. It is however an important feature in some of the other Virgin Martyr legends, in addition
to the legends of Margaret and Barbara. St. Katherine states that the idols can neither help
themselves or others: Nam dii nec sibi nec aliis prodesse prosunt5, and St. Cecilia insists,
talking to her persecutor, that it marvels her how he does not see that figures made of stone,
metals or tree cannot be gods. In a further elaboration on this subject she ridicules the idols
and those who believe in them by, among other things, describing how they let spiders spin
webs or birds build nests or shit all over them without interfering, and she stresses how stupid
it is to believe that something like that can be a god. She also compares the idols to dead peo-
ple, adding that they are inferior even to them; they are proved to be less than the dead since
when humans lived they were able to see, hear, walk, talk, caress and smell, while the idols
can do none of these things and never could nor ever will. Cecilia later also asks her persecu-
tor, Almachius, to place his hands onto the idols and in this way experience that they are
merely made of stone, since he will not believe what his own eyes ought to tell him about
them; thus being the laughing-stock of the entire people.
It is not only in the Old Icelandic translation of the legend of St. Agatha in AM 233a fol
that the idols as made by human hands motif is stressed more than it is in the Latin version
it is based on. The portrayal of this motif is also marked by amplification in the legend of St.
Katherine, but here traceable mainly in the other manuscripts containing this version of the
legend; Stock.perg.fol.2 and AM 429 12mo.6 Also, the interpolations in connection with this
motif are primarily suited to clarify or embellish certain aspects in its presentation; aspects
2
Translation: As yet, however, the raging of the madness of the devil held people in its grip, and they wor-
shipped deaf, dumb and blind idols, fashioned by human hand, which were of benefit neither to them nor to
themselves. (cf. Clayton and Magennis 2006: 1945)
3
In numbering here and other places in the text refers to that in the Latin Vulgate Bible.
4
This and later Biblical quotations in English are from the Douay-Rheims Bible. We can also find almost the
same text in Psalm 134.1518. The same idea is also expressed in for example Isaiah 31.7 and Isaiah 37.19.
5
Translation (mine): But the Gods can neither be of help to themselves nor others.
6
PS: Lacuna in AM 233a fol.

138
that either are not mentioned at all in the Latin text or the other Old Icelandic manuscripts, or
that are merely suggested or mentioned in passing. In the opening passages of the Legend,
where we learn that humans are the eager slaves of demons or depictions of idols because
they do not yet know about the true God, these manuscripts have an interpolation that states
that the idols are made by human hands:[] ok kollodv at gud er salfr hfdu er snum
hondvm smdat.7 This is a fact that is merely implied in the Latin version of the legend,
BHL1661b.8 St. Katherine speaks ironically about how her persecutor, Maxentius, admires a
temple made by humans, but the false gods themselves or their graven images as made by
human hands are not mentioned. Is seems that the translator here, or maybe more likely a later
editor, has felt the need to add this otherwise well known fact about them. We can also find
this fact stated in other amplifications and interpolations in the Old Icelandic text in AM 233a
fol. When a group of onlookers after being converted by Katherine curse the idols and every-
one that worships them, they add in this manuscript only that the idols are: handa uerk
manna. ok oll lknesk era. [].9

The Devil and Mister Kvintian10

The second of the more important additions in the above-mentioned longer interpolation in
the legend of St. Agatha states that the idols are filled with devilish sorcery and actions, and
thus also every kind of immorality, indecency, adultery, committing homicides and other evil
deeds, and that after their deaths they went to Hell where thy will burn and freeze together
with the devil and his servants and all people who worship them in all eternity. The fact that
the idols are images of evil people or false gods is thematised in the Latin legend, they are
among other things named as the pagan gods Jupiter and Venus (Odin and Freya in the Old
Icelandic texts), but here it is further embellished upon. A central feature in the legends of
Virgin Saints is in fact the thought that the idols themselves are devils or demons that wish to
remove humankind from God. Since the false gods thus represent evil beings, it is only suit-
able that their worshippers suffer their same faith and conditions; they become like unto
them, as Psalm 113.16 states; their evil rubs off on those who create them and those who
worship them alike. Thus it is only fair that your conditions are the same as those of your
gods and that a person is judged according to the rank of his or her gods. In the legend of
Agatha she explains to Kvintian that: []vt et vos possitis in deorum vestrorum numero
computari11 (ASS Feb I: 616), further she uses the comparison with the roman gods Jupiter
and Venus to insult Kvintian and his wife; or at least he so interprets it. She tells him: Sit
talis vxor tua, qualis dea tua Venus fuit; et tu sis talis, qualis Iupiter deus tuus extitit.12 Kvin-
tian naturally reacts to this insult; a reaction that Agatha turns around and uses against him;
why should it be an insult to be compared with ones own gods if those are true gods? The
gods that a person worships should stand as examples in this persons life. If the worshipper
himself is insulted when a person compares his lives to the gods, then that has to be the ulti-
mate proof that these false gods are not to be trusted. This is ultimately an idea derived from
the Bible. In Matthew 10.25 we read: It is enough for the disciple that he be as his master,

7
The text follows the manuscript Stock.perg.fol.2. AM 429 12mo reads: [] ok kollodv at gud uera er salfr
er hfdu smdat snum hondvm. Lacuna in AM 233a fol.
8
According to Bjrni Olafsson (1972:53) this version, first printed in Bronzini (1960), is closest to the Icelandic
text, not BHL1659 as previously thought. BHL1659 has however some readings that are closer to the Icelandic
texts.
9
My transcription. This addition is not to be found in Stock.perg.fol.2. Lacuna in AM 429 12mo.
10
Kvintian in AM 233a fol, Quintian(us) in the Latin version and in the other Icelandic manuscripts.
11
Translation: (mine) So that you can be judged according to the rank of your gods.
12
Translation (mine): I wish that your wife would be like your goddess Venus, and you like your god Jupiter.

139
and the servant as his lord. If they have called the good man of the house Beelzebub, how
much more them of his household?
In the Latin text Kvintian is presented as both an idolator and a servant of sin. Version 1 of
this legend, found in AM 429 12mo and Stock.perg.fol.2 presents him as one who sacrifices
to the devil, leading to a simplification of the concept of the enemy. In version 2 of the Ice-
landic legend, found in AM 233a fol, the consul (jarl in the Old Icelandic text) is presented as
simply a grand idolater; a blotmar mikill, and not the servant of sin. This leads to less em-
phasis on Kvintian as a servant of sin in this opening passage in both Old Icelandic versions.
The difference seems however quite incidental in both versions. In AM 233a fol similar ex-
pressions are also omitted elsewhere, for example when St. Agatha abuses Kvintian calling
him: Tu minister Satana; this is rendered merely by u in AM 233a fol, again leaving us
with a weakened emphasis on Kvintian as the servant of evil. On other occasions, though, the
statement that Kvintian is the servant of sin is added where it is not stated in the Latin legend
or in version 1 of the Old Icelandic legend; a version that except for the opening passage men-
tioned above follows the Latin rather closely in this respect.
The Latin legend of Agatha is quite clear when it comes to the punishment Kvintian can
expect after death if he does not turn his back on the dark path. St. Agatha tells him that he
should regret his delusions so that he can be saved from eternal torment in the afterlife; tor-
ments that by far exceed the torment that he threatens to wreak upon Agatha. She also tells
him that if he does not abandon the false gods and instead adores the true God, the Creator of
all things, he will be subjected to harsh punishments and eternal flames. Even the physician
who visits Agatha in the dungeon, and who later reveals himself as Gods apostle, predicts
that Kvintians soul will suffer eternal pains. These predictions and threats are faithfully ren-
dered in the Old Icelandic texts, although in some instances we see a clarification of what the
consequences of Kvintians actions might be; for example when Agatha tells Kvintian that:
Omnia verba tua fatua et vana sunt et iniqua, prcepta tua arem ipsum maculant. Vnde miser
et sine sensu et sine intellectu es.13 This is in the Old Icelandic version 1 rendered by: oll
ero ord n tom ok raung bodord n. vat au saurga k salfan ok gera k vtlausan., and
in version 2 by: Aull ero bodord n ok or nyt. ok rangr domar nr. vat er saurga k
ok draga tl heluts. So while St. Agatha both in the Latin and in the first Old Icelandic ver-
sion presents the consequences of Kvintians words and judgements as condemning him to
misery and stupidity, according to the version in AM 233a fol. they actually condemn him to
misery and drag him to Hell.
In this way both the Latin and the Old Icelandic versions of the legend portray the pun-
ishment for idolatry and renouncing God as everlasting torments in the eternal flames of Hell.
In the longer interpolation in AM 233a fol. we see however that Hell is not necessarily only a
place of fire and flames, but that it can also be freezing; perhaps this is a natural expansion of
the concept of Hell in a Nordic country. Further we see an expansion of the description of the
people that the images of the idols were once modelled upon; the worst people that have ever
existed, adulterers, murderers and so on; a colourful description, but not necessarily compati-
ble with the presentation of the idols as based upon false gods that we find elsewhere in the
legend.
A further amendment when it comes to Kvintians expected punishment is evident in the
closing passages of the legend in AM 233a fol., an amendment that instead of clarifying or
narrowing the text, actually opens it up to new interpretation; a rather uncommon move for
the text. The passage in question recounts how Kvintian, after the death of St. Agatha, wants
to get hold of her familys assets. Both the Latin and Old Icelandic versions narrate how while

13
Translation (mine): all your words are foolish, useless and evil, your commands pollutes the air itself. Be-
cause of this you are miserable and devoid of wits and intelligence.

140
on the journey to put his plan into effect, he is killed by two horses while crossing the river in
a boat; they kick him to death and into the river, and he is never found. He has to suffer the
fate of being denied a proper burial. In all but the AM 233a fol. version of the legend, this
passage opens with the narrators statement that Kvintian is actually struck by the revenge of
God here. AM 233a fol omits this rather clear and obvious statement, and instead closes the
passage with a quite colourful description of what happens when Kvintian falls into the river:

[] ok heyru menn na nr. dfullga blstran me p ok gny. a er lk arls sokk ndr. ok

tok uatnt at uella sem katl vr vm rar stundr dags. [My transcription]

Here too it is quite evident where Kvintian ends up after his demise, but the passage depicts
this much more elegantly and somewhat less explicitly than in the other versions. Also the
description of how, after this incident, the water in this spot begins to boil as if in a kettle
three times a day; inevitably calls to mind the Icelandic heitur pottur.
In the AM 233a fol. version of this legend we see that idols through interpolations in the
text are depicted both as representations of false gods and of evil humans, two distinctive fea-
tures characterising the idols that are not usually present in one and the same legend. They are
however both well known qualities used to describe the idols in other texts of the genre. The
different interpolations and other amendments also lead to the inconsistency that the idols are
described both as without autonomous powers, since they are made by human hands and
without the power to intervene on behalf of their worshippers or themselves, and at the same
time as representations of evil powers in society and as the snares of the Devil; two qualities
that are not necessarily compatible. Are the idols useless and helpless or do are they affect the
lives of those that worship them, driving them to punishment in Hell? The presentation of this
motif thus becomes blurred and less focussed in this version, and additional information
might appear incidental and without a clear direction. At the same time additions and emenda-
tions, at least some of the more fortunate ones, broaden the readers perception of the plot by
providing greater and more colourful detail.
We can see some of the same effects as described in the AM 233a fol version of the legend
of Agatha also when it comes to the emendations in the legend of St. Katherine in the same
manuscript. The fact that the idols, named as Jupiter, Odin or Tor, in fact are demons that
have in store the expected and inevitable fate that they are fated to burn in the eternal flames
of Hell together with the people who worship them, is a characteristic of the idolatry motif not
stressed in the Latin Vulgate version of the legend of St. Katherine. The motif does however
surface in the Old Icelandic version of the legend that is extant in AM 233a fol, and, as men-
tioned above, it is well known from other Virgin Martyr Legends, like for example the legend
of St. Agatha. As is the case in the legend of St. Agatha, this manuscript adds aspects not pre-
sent in the Latin legend through amplification and longer interpolations; after the emperor has
condemned Katherine to death, he disputes with his queen who protests fiercely against this
monstrous act. This leads the emperor to ask his queen if she has abandoned faith in the gods,
and she answers him that she now believes in the one true God; the Holy Trinity. In AM 233a
fol. she further elaborates on this:

orr ok onn ok allr gudar ok go henna manna ero dflar ok engu nyt nema tl ess at
brenna elfum heluts elld. ok allr er med em er au dyrka. en au megu huark ser gott
gera ne rum. [My transcription]

This interpolation might be understood as both a reference to other texts of the genre, and also
an attempt to clarify why it is so crucial that people abandon the false gods. In this way the

141
Old Icelandic editor of this manuscript or one of its models interprets the legend for his pub-
lic.

Demonic possession
In the legend of St. Margaret the idol motif is not embellished in AM 233a fol in the same
way as in the legends of Katherine and Agatha. In fact the strategy here seems to be quite the
contrary; omission and contraction. As mentioned earlier, the narrator in the Latin legend
opens the prologue where the idols are characterized as deaf, dumb and blind, and made by
human hands. The foolishness of worshipping them is contrasted with the near-infinite
possibilities awaiting if one subjects oneself to the power of God; a contrast that is repeated
throughout the whole legend.
The entire prologue is omitted in version 2 of the Old Icelandic legend extant in for
example AM 429 12mo, while it is rendered in version 1 extant for example in AM 233a fol
and AM 235 fol. In version 1, however, the contrast between good and evil depicted and
stressed in the latin version is downplayed; the narrator simply states that retaining faith was
difficult at the time of the events narrated in the legend, because there were more people
opposing it than standing behind it. The idols are also portrayed as made by human hands
later on in the narrative, in one of Margarets many replies to Olibrius, her persecutor. This is a
rather extensive reply that in the Old Iclendic versions is, as the only part of this reply,
omitted (AM 429 12mo) or severely condensed (AM 233a fol and AM 325 fol); to the extent
that Margaret simply refuses to sacrifice to the idols, and neither praises the Lord nor scorns
the idols. Also in other instances where the shortcomings of the devil and his idols are
contrasted to the glories of God and Heaven, the contrast is downplayed or even erased in this
Old Icelandic version; often leaving only Margarets refusal to sacrifice. The result is, since
the praising of the Lord is omitted in these passages, a stronger emphasis on the refusal to
submit to the worshipping of idols, at the same time as certain characteristics of the idols, as
for example the fact that they are made by human hands and are thus useless, are downplayed
to the limit of nearly being completely erased.
As mentioned above, defects such as deafness, dumbness, blindness and paralysis are often
linked to the idols in this legend, a connection that is only preserved once throughout the
entire text of the Old Icleandic version 1 (AM 233a fol and 235 fol). This insistance in the
Latin legend is however not accidental. A related motif is in fact a connection between these
same defects among humans and demonic possession; an idea that is also present in the Bible,
where the miraculous healings by Christ are often regular exorcisms.14 This idea is also
expressed in the legend of St. Margaret; for example when in the prologue we are told that
many people have been cured by Christ in the past, or when we are told that many people are
cured in the prescence of Margarets relics. In the discussion between Margaret and the black
demon, the demon also states that he in fact struggles with the righteous while they sleep and
among other things blinds them and confuses their senses. Also in Margarets last prayer these
same defects are mentioned, now in relation to different defects and conditions that infants
should be protected from when they are born; in the latin legend these are paralysis, dumbness
and blindness, and in some manuscripts also deafness. Different Old Icelandic versions men-
tion different conditions in relation to this, and most often fewer conditions than are men-
tioned in the Latin version, the exception being version 3.15
According to Wendy R. Larsson (2002: 26) it is not by chance that the Latin tradition
states that the children should be protected against paralysis, blindness and dumbness (and in
some versions also deafness) during birth, since these defects were often regarded as signs of
14
See for example Mathew 15:2228, Mark 9:1729, Luke 13:1113
15
Lacuna in AM 233a fol.

142
demonic possession in the Middle Ages, an idea expressed in the aforementioned Psalm
113.1216. These characteristics were also often linked to the pagan gods in hagiographical
legends; in the legend of Margaret they are used to describe the pagan gods that her father and
Olibrius sacrifice to. This connection is not made in the Old Norse versions of the legend
however, since both the number of defects that the children are to be protected against and the
number of characteristics linked to the pagan gods is reduced. In addition they are not the
same:

Version the children the idols

Latin: BHL5303d: paralysis, dumbness, blindness (and deafness) deaf, dumb and blind
AM 235 fol. death and paralysis deaf and dumb
AM 429 12mo blindness and paralysis deaf and blind
Version 3: blindness, paralysis, deafness, dumbness and foolery deaf and dumb

While the language in the Latin tradition thus reflects the liturgy, most of the Old Norse ver-
sions lose this aspect of the text. We can also see a similar suppression of this idea in other
parts of the Old Norse texts. The comment in the prologue that states that Christ cured the
blind and deaf is omitted, and the specific conditions we are told that people are cured of
when they come in contact with the saints relics are not mentioned (version 2) or are strongly
reduced (version 1); while the Latin legend mentions that the infirmi, caeci, claudi, surdi,
debiles16 are cured, the Icelandic version 1 only mentions the curing of the deaf and blind,
while version 2 simply refers to sick people. This suggests that the allusions to the Biblical
passages of interest and the theological idea behind them are overlooked either by the transla-
tor or later editor(s) of the Old Icelandic versions.

Bibliography
Acta S. Agath. In Acta Sanctorum, February I: pp. 615618, Ed. by Societ des Bollandistes
Assmann, B, 1889: Angelschsishe Homilien und Heilegenleben. Bibliothek der Angelschsischen
Prosa 3, Kassel.
Bjarni lafsson (forthcoming): Katrnar saga. Reykjavk
Bronzini, Giovanni B, 1960: La leggenda di S. Caterina DAlessandria. Passioni greche e latine. Ro-
ma. Accedemia nazionale dei lincei.
Bruvoll, Kjersti 2000: Sagaen om Margareta fra latin til norrnt. Oversettelsesstrategier satt under
lupen. Hovedfagsoppgave ved NTNU. Trondheim.
Clayton, Mary and Magennis, Hugh, 2006: The Old English Lives of St. Margaret. Cambridge Uni-
versity Press.
Larson, Wendy R., 2002: The role of patronage and audience in the cults of Sts Margaret and Marina
of Antioch. In Gender and holiness: men, women, and saints in late medieval Europe. Ed. by S.J.E.
Riches and S. Salih. London.
lafur Halldrsson, 1966: Helgafellsbkur fornar. Heimspekideild Hskla slands, Reykjavk.
Rasmussen, Peter, 1977: Tekstforholdene i Margrtar saga. Reykjavk.
Unger, C.R., 1877: Heilagra manna sgur. Volume 1. Christiania, Bentzen.

16
Translation (mine): infirm, blind, paralysed, deaf, disabled.

143
 Negotiations of Space and Gender in Brennu-Njls Saga
Katrina Burge, Monbulk, Victoria, Australia
In this paper I shall argue that mapping the social spaces represented in the Icelandic sagas
produces an accurate representation of hierarchies, values and gender associations within the
text. The purpose of my investigation is to demonstrate the effectiveness of mapping personal
space, and to use this mapping as a means to determine the extent to which gender is a key
determinant of spatial location, range and influence, and to identify other factors which influ-
ence the demarcation of space.
The sagas are reflections and representations of aspects of Icelandic culture as it existed
over the centuries, from the tenth and eleventh century setting of the sagas, up to and beyond
the thirteenth and fourteenth century time of composition. Actual events depicted should be
seen as selections from a range of cultural memories rather than documented historical facts.
With caution, these representations can also be extrapolated onto the physical layout of Ice-
land, with, of course, the usual caveats about the conflation of the saga world and historical
reality.
This study focuses on Brennu-Njls Saga, and specifically on the farmstead of Ber-
grshvll, which is the physical centre of the narrative, not only as the residence of its
eponymous hero, but as the radius from which most action emanates. Many major characters
eventually find their way there, as friends, enemies and horrified coroners, and many of the
most prominent spend part of their lives there. Before its untimely Burning, Bergrshvll is a
rich and lively space, a frequent forum for relationships and activities and a working farm in
which the business of spreading manure and searching for sheep is interwoven with feasts and
feuds.
Spatial arrangement can be reconstructed from written material, and analysed to identify
cultural classifications represented by exclusions and proximities. Space should be understood
as emotional and social, as well as geographic. It is not random and it is not a simple equation
in which function is added to topography and technology to produce an inevitable spatial re-
sult. The social significance of spaces indicates the hierarchies, associations and exclusions
that constitute a society. Henri Lefebvre identifies the dynamic relationship between space
and society:

The spatial practice of a society secretes that societys space; it propounds and presupposes it, in
a dialectical interaction; it produces it slowly and surely as it masters and appropriates it. From
the analytic standpoint, the spatial practice of a society is revealed through the deciphering of its
space. (Lefebvre 1991: 38)

A space can be coded as female or male by, for example, its decorations, typical activity or
usual inhabitants; and changes to these factors can change the gender coding of a space. Space
is plastic and can shift or solidify depending on circumstances. How a society organises its
activities spatially and how it relates different spatial contexts to each other indicate critical
cultural concerns. It is always a two-way process: although a culture might prescribe certain
physical characteristics to related sites, those sites then shape and affect everything that takes
place within them.
Spaces transmit meaning to activities and performances located within them. As Gaston
Bachelard has pointed out, memories are really fixed in space, not in time (Bachelard 1994:
9). A space retains vestiges of previous activities that prevent it from ever being neutral or
valueless again and which can be attributed to other events that take place in it.

144
 The spatial arrangement of commonwealth Iceland enforces the centrality of the farmstead
as a social site. In Iceland the entire population lived on isolated farmsteads, united by bonds
of kinship and work. With no permanent markets or meeting places and no constant public,
male spaces in the saga-world, public spaces were created only at specific locations at certain
times, such as the annual Alingi, or the more frequent local assemblies. The farmstead there-
fore becomes the arena for numerous negotiations of personal and corporate space, status and
influence. Each farmstead was an independent and geographically discrete unit, so complex
and significant that each forms what Kirsten Hastrup has termed a complete social universe
(Hastrup 1990: 275).
The farmstead is the location of the dominant Icelandic social grouping, the household. For
all but the low-status itinerant members of society, the household is the pre-eminent mode of
affiliation and it is also a fundamental component of the governance of the society. Through-
out the sagas, a persons identity is linked to her or his position within the household, and to
the relative social position of that household.
The Bergrshvll of Njls Saga is a home, a farm and a social grouping. The holdings of
Bergrshvll encompass the eponymous farmstead, woodlands at Rauaskrir, a subsidiary
farm at rlfsfell and, for the duration of the Alingi each year, Njlls b at ingvellir.
There is no direct correlation between physical and social distance. Rauaskriur, though only
a couple of kilometres from Bergrshvll, is beyond social restraint and influence, whereas
the booths at the Alingi many miles away mirror the spatial practice of the farmsite.
In Bergrshvll and its environs, the processes of daily life in Iceland are mapped out,
from the sleeping arrangements of the people to the clearing of the forests. We see where
women work, where men go, and how physically close and conceptually distant the space of
the sexes can be. Gender is not rigidly embodied in the world of the Icelandic sagas, as noted
by Carol Clover in her pivotal discussion of the application of the one-gender model to Ice-
landic society (Clover, 1993). This situation has implications for the significance of masculine
activity in domestic locations, as well as for the nature of feminine participation in activities
extending beyond the domestic unit. As genders are seen as points on a continuum rather than
a fixed oppositions, it follows that behaviour must constantly reinforce gender distinctions.
The spatial congruence between feminine and masculine realms renders these distinctions
potentially fluid.
The population of Bergrshvll is not fully countable from the text, though thirty-one
residents, sixteen male and fifteen female, are identified by name or position. These thirty-one
are not all in residence at the same time, indeed they are not all alive at the same time, and
their numbers were presumably augmented by others who worked on the farm but never con-
tribute sufficiently to dramatic action to get their names mentioned in the saga.
Relatively few of the residents control significant social space. Njll and his wife Ber-
gra, Skarpheinn, Grmr and Helgi Njlsson, and Kri the son-in-law are all prominent. The
main power brokers are Njll and Skarpheinn, with Kri also occupying a responsible posi-
tion. Njlls other sons form a fraternal block with Skarpheinn but they are not independ-
ently influential. Bergra is the only female who wields significant power within the house-
hold, and it is noticeable that her control increases when Njll is away at the Alingi, when
she directs the feud between herself and Hallgerr that sees the death of numerous farmhands
on both sides.
Careful and intricate spatial patterning is evident at those locations that feature in the narra-
tive, making the few social arenas the sites of multiple expressions of position, identity, gen-
der and authority. The most significant sites are the farmhouse, its immediate surroundings of
farmyard and homefield, and the wild spaces outside direct household control.
Early Iceland is not a leisured society and most of the time, where people are is linked to
the work they are doing. Of course, the nature of that work often determines or influences

145
where it takes place. According to the law code, an innan stokks/tan stokks (that is, inside the
threshold/outside the threshold) division of labour operated (Finsen 1879: 173174), with
men working in the fields tending the stock and crops, and women carrying out activities,
usually related to textiles or food, that are centred on the farmhouse. Jenny Jochens has dem-
onstrated that the work practices presented in the slendingasgur do not always follow a tidy
dichotomous model in their spatial and gender associations (Jochens 1995: 114140). Very
little productive or domestic work features in the Njls Saga, so it is difficult to see a clear
spatial demarcation between masculine and feminine work spaces. Only a few work practices
are described: Bergra serves some meals (34, 127); Svartr cuts wood (36) and Atli burns
charcoal (37); the Njlssons work on their weapons (44); the rlfsfell shepherd finds a lost
sheep (69); and the farm hands cart dung to the tn at Bergrshvll (44). The main distinc-
tion between work location is not gender, it is status. For members of the family, male or fe-
male, work takes place indoors, while the hired hands are only depicted outside.
The farmhouse itself is an intensively negotiated site in which the proximity of women and
men, of high and low status residents, and of different generations creates overlapping and at
times conflicting spatial maps. The biggest room in an Icelandic farmhouse is usually the
skli, or hall. At Bergrshvll the skli is the first room encountered upon entering the house
(129: 328), and there is also a stofa, the room in which meals are taken (127:324).
The stofa appears to be the arena for most interpersonal negotiations. It is mentioned by
name only once, when the family are seated at the table as Njll describes a doom-vision of its
walls running with blood (127:324). It seems likely that other meals take place there, includ-
ing the feast attended by Gunnarr and Hallgerr (36) and the numerous feasts held by the
Njlssons for Hskuldr rinsson (97:248, 109:276). Meals are consumed at the benches on
removable tables or boards that are placed in front of the benches. We are told that, after
meals, bor vru ofan tekin (the tables were taken away, 127:324), an indication of how the
limited interior space is adapted to different uses.
At the head of the room is a raised platform (pallr or verpallr) with benches. The pallr at
Bergrshvll features only at the vetrgri attended by the newly-married Gunnarr and Hall-
gerr (35:91). Trouble arises between Bergra and Hallgerr when Bergra orders Hall-
gerr to surrender her position on the pallr in favour of rhalla sgrmsdttir, wife of Helgi
Njlsson:

Mlti Bergra til Hallgerar: skalt oka fyrir konu essi. Hallgerr mlti: Hvergi mun
ek oka, v at engi hornkerling vil ek vera. Ek skal hr ra, sagi Bergra. Sian settisk
rhalla nir.

Bergra said to Hallgerr: You will move away for this woman. Hallgerr said: Not for
anyone will I move, like an outcast old woman. I am in control here, says Bergra. Then
rhalla sat down.

Seating arrangements are of critical importance in the sagas because, as William Miller has
noted, they provided one of the few occasions in the culture where relative ranking was
clearly visible (Miller 1990: 30). While men have many ways of displaying their social posi-
tion, for women these seating arrangements are critical because the feast is one of their very
few occasions when they are assembled together. The physical placement of Hallgerr and
rhalla is a matter concerned with female markers of honour (Larrington 1991: 15), and the
ranking they display there will codify social positions amongst themselves until the next
comparable occasion.
Immediately outside the house, the farmyard begins. The farmyard is a busy space. Al-
though it would have been the site of intense farming activity, in the narrative it is generally

146
social rather than productive processes that fill the space. The yard is the specific setting for
numerous incidents involving farm residents and outsiders, men and women, high and low
status people.
The farmyard is frequently the space in which the people of Bergrshvll receive strang-
ers and guests. Bergra stands outside warily watching as a strange man rides up on a dark
horse, and she finds out his business before engaging him as a farmhand (36). On several of
Gunnarrs many visits, both solo and accompanied, Njll meets and welcomes him in the yard
(e.g. 21, 35), where a situational space is created for the discussion of his affairs.
The farmyard is a point of exit from Bergrshvll as well as entry. This transition is par-
ticularly well-drawn in incidents featuring Njll and his sons. Twice Njll is woken when the
younger men set off into the night on a killing mission, and on both occasions he follows
them into the yard (44, 92). The conversations that take place there mark a spatial and social
shift for the sons, who are not fully within the boundaries of their home but neither have they
fully passed into the amorphous, wild region beyond. Njlls conversation and presence pro-
vide a conduit between the social world and the spaces beyond, which legitimizes the Njls-
sons subsequent actions even when they move beyond the social realm to carry out their re-
venge slayings.
The space of the farmyard is not obviously associated with any hierarchies of gender, gen-
eration or status. From Sunn, the old foster-mother who foresees the Burning from the pile
of chickweed (124), to Njll himself and assorted visitors, it features a wider range of per-
formers than any other part of the farm. It is an egalitarian and undifferentiated space which
does not appear to be partitioned in accordance with household hierarchy.
As the point at which Bergrshvll meets the outside world, the farmyard serves to em-
phasise the membership of Bergrshvll as an inclusive rather than differentiated classifica-
tion. Internal divisions are sublimated into the overall identity of the household, so that the
stratification is between residents and outsiders rather than between different classes of resi-
dent. The farmyard is inevitably a liminal area and considerable effort goes into maintaining
its boundaries. Once people are within the garr, they are in a position to jeopardise the
farms security, and this vulnerability is carefully monitored. When, at the Burning, the men
of Bergrshvll retreat into the house at Njlls bidding (128), they effectively cede control
to the Burners, with famously dire consequences. This demonstrates the importance of the
farmyard as a defensive zone, and illustrates the danger that ensues when this defence is
breached.
Outside the farmyard the conceptual wild begins to encroach. Travel creates the spaces in
which gender distinctions are most apparent. Both the range and the frequency of travel are
much greater for men than for women, and their travels are also distinguished in the way in
which they are described in the text.
The most mobile Bergrshvll resident is indubitably Njll, who goes repeatedly from
home to assembly to neighbours farm, as well as making trips to his own subsidiary farm of
rlfsfell. Njll is usually accompanied on this travels by his sons, who also make several,
usually violent, journeys of their own. While Njll and his sons often move beyond the farm-
site, other men at Bergrshvll are more restricted. Only Svartr (36), Atli (38) and rr (42)
are depicted outside the farm compound, and their location is subject to the personal control
of others.
Most women are restricted in the movements into the wild. The stronger a womans posi-
tion within a household, the less mobility she has. Bergras elevated status sees her almost
completely immured within the walls of Bergrshvll. Hrn, Njlls sometime mistress, is
of land-owning rank but without the secure social position that comes from formal marriage.
Twice at times of crisis she visits Bergrshvll, firstly after the killing of her son, Hskuldr
Njlsson (98), and secondly to tell Njll about the Sigfssons plot against him (124). The

147
only women in Njls Saga who travel extensively are vagrant women who wander throughout
the region seeking hospitality at various farms. They are never known by name and their lack
of fixed abode is noteworthy enough for them to be called farandkona (44, travelling
women) a term which locates them by not locating them, although the term snauar konur
(92, beggarwomen). None of these women are actually shown in transit. They are here; they
decide to go there; they are there, with no sense of the spatial transition and personal activity
involved.
In comparison, descriptions of mens travels usually mention the process involved: they go
home, they ride, they go out, they meet people. This makes men seem dynamic while women
are static. Men of status usually move en masse to assemblies, to a horse fight, and to other
farms, but women do not usually travel as a cohort. Though women might appear in two dis-
tant locations, they are never shown in the transitional areas between them. This reiterates the
notion that men can legitimately move beyond the boundaries of the farm into the conceptu-
ally wild spaces between, but women are culturally invisible during their sojourns there.
Nonetheless, both Hrn and the farandkona negotiate their way with ease through the wild
spaces of the society, while men are often at risk there. The terrain beyond the farm yard pro-
vides the space where men to carry out the violent negotiations of honour and status that are
integral to social order.
Definitions of public spheres have emphasized the idea that the most significant effect of
the creation of a distinct, separate and usually masculine realm for high-status negotiations is
that it excludes many people, especially women, from the privileged knowledge that is created
and shared within that space (Spain 1992:3). Certainly this is the case with the Alingi, where
women are permitted to be present but their right to act in this arena was greatly restricted in
comparison to men. The gendering of access is necessarily different at Bergrshvll. At the
farmstead, people are shown constantly trying to construct exclusive spaces in which they can
exchange information with select interlocutors, and Bergra is frequently included in these
conversations. At Njll and Gunnarr repeatedly go aside to talk, and the Njlssons go apart
from their parents when they want to plot with Mrr Valgarsson. Bergra leaves the room
when the men have some disturbing news to mull over but storms back in to reclaim her place
within the discussion, and private conversations between her and her husband are implied
though not shown. This domestic space achieves a quasi-public status through the absence of
public space, and Bergras prominence here shows that, along with her menfolk, a women
with sufficient status, and the right personality, can be active in the public arena. The spatial
dichotomy of public and domestic domains is completely dislodged by the social structure of
medieval Iceland, which developed without towns, villages or courts.
To conclude, space at Bergrshvll and throughout the society is not mapped in accor-
dance with a binary gender construct, even though the significance of gender as a means of
ordering society is demonstrated by processes as diverse as nomenclature, work divisions and
personal grooming. While gender distinctions are certainly apparent in spatial range, as men
are represented as more extensive travellers than women, throughout and beyond the country,
there is considerable overlap in the spaces accessible to both genders.
The farmsite, and especially the inner compound of farmhouse and yard, consist of heavily
negotiated areas. These spaces manifest several overlapping, and at times conflicting, maps of
association, exclusion and gender. Classification by gender does not fully explain personal
variance in the occupation of space, as household position is also an important variable in
constructing spatial practice. Those people, particularly Bergra, Njll and Skarpheinn,
who hold important positions in the household hierarchy, occupy a wider range of spaces and,
predictably, display greater authority over others within those spaces, than residents lower in
the hierarchy. Hierarchical positions affect the amount of control an individual has over her or
his own space. High-ranking residents are able to shape the personal space of their subordi-

148
nates, even to the extent of locating those subordinate in regions which categories such as
gender prevent the status-holder from occupying.
Gender is a powerful element in constructing social identity with its concomitant spatial
expression, and the successful operation of gender classifications is fundamental to Icelandic
society. However the spatial layout of the society is inimical to rigid gender boundaries and
its atomisation into individual farmsteads conflates public and domestic performance sites.
This has profound implications for the negotiation of gender. The farmhouse, the centre of
domestic space, is situated as a prominent arena for social action. The multiple spatial values
which are thereby inscribed on the farmhouse leads to the increased proximity of women,
whether as audience or performers, in prestige-related activities from which, were it not for
the insubstantiality of the nominally public spaces, they would otherwise be excluded. Space
in the saga world is a fluid entity that can bind and connect as well as separate.
Perhaps the most interesting and unusual spatial feature of early Iceland is the merging of
appropriate sites for female and male performance. The slendingasgur, and Njls Saga
more than most, are populated by men sensitive to the slightest hint of being anything other
than uncompromisingly masculine. The disputes they create over their own claims to manli-
ness are the most consequential in the saga, which leads one to conclude that the concept of
masculinity is destabilised by the temporal and spatial impermanence of the separate public
male sphere. Men become vulnerable to imputations of effeminacy because they lack constant
recourse to a separate forum in which to reiterate their gender away from and against feminine
influence. The hyper-defensive masculinity that characterises the saga results from the com-
bination of physical proximity and conceptual inequality between genders. Men seek to define
and defend themselves against implications of womanliness, while womens status is en-
hanced by their socially-sanctioned performances in quasi-public spaces, as they appropriate
the positions and activities that were originally devised to demonstrate and reinforce the pre-
scribed gender hierarchy.

Bibliography
Allen, Richard, 1971. Fire and Iron: Critical Approaches to Njls Saga. Pittsburgh: University of Pitts-
burgh Press.
Ardener, Shirley, ed., 1981. Women and Space: Ground Rules and Social Maps. London: Croom
Helm.
Bachelard, Gaston, 1994. The Poetics of Space. Translated by Maria Jolas. Boston: Beacon Press.
Bauman, Richard, 1986. Performance and Honor in Thirteenth Century Iceland. Journal of Ameri-
can Folklore 99 : pp. 121150.
Borovsky, Zoe, 1999. Never in Public: Women and Performance in Old Norse Literature. In: Journal
of American Folklore 112 : pp. 639.
Bourdieu, Pierre, 1990. The Logic of Practice. Translated by Richard Nice. Stanford: Stanford Univer-
sity Press.
Clover, Carol, 1986. Maiden Warriors and Other Sons. In: Journal of English and Germanic Philol-
ogy 85 : pp. 3449.
, 1993. Regardless of Sex: Men, Women and Power in Early Northern Europe. In: Speculum 68 :
pp. 363387.
Dronk, Ursula, 1981. The Role of Sexual Themes in Njls Saga. London.
Eldjrn, Kristjn, 1971. Now the Whole House Began to Burn: An Account of the Excavations at
Bergthorshvoll. In: Atlantica and Iceland Review 9 : pp. 4146.
Finsen, V, ed., 1852, reprint 1974. Grgs: Islndernes Lovbog Fristatens Tid. Odense.
Gurevich, Aron, 1992. Historical Anthropology of the Middle Ages. Ed. by J. Howlett. Cambridge.
Hastrup, Kirsten, 1985. Culture and History in Medieval Iceland: An Anthropological Analysis of
Structure and Change. Oxford.
, 1990. Island of Anthropology: Studies in Past and Present Iceland. Odense.
Jochens, Jenny, 1995. Women in Old Norse Society. Ithaca.

149
Jhannesson, Jn, 1974. A History of the Old Icelandic Commonwealth: Islendinga Saga. Translated
by Harldur Bessason. Manitoba.
Larrington, Carolynne, 1991. What Does Woman Want? Maer & Munr in Skirnsml. In: The
Audience of the Sagas. Proceedings of the Eighth International Saga Conference. Gothenburg. pp.
918.
Lefebvre, Henri, 1991. The Production of Space. Translated by Donaldson Nicholson-Smith. Oxford.
MacCormack, C. P. and M. Strathern, eds., 1980. Nature, Culture and Gender. Cambridge.
Miller, William Ian, 1990. Bloodtaking and Peacemaking: Feud, Law and Society in Saga Iceland.
Chicago.
Nicholson, Linda, 1986. Gender and History: The Limits of Social History in the Age of the Family.
New York.
Spain, Daphne, 1992. Gendered Spaces. Chapel Hill.
Sveinsson, E. ., ed., 1954. Brennu-Njls Saga. Reykjavk.

150
 The Secret Lives of Lawspeakers: the portrayal
of lgsgumenn in the slendingasgur
Hannah Burrows, University of Sydney, Australia
It is an axiom that in Commonwealth-period Iceland the lawspeaker was a unique figure. He
was the only elected official in a society that had no overarching ruler or state system of gov-
ernance, and stood at the head of the legal community in a society whose literature is com-
monly held to attest to a cultural predisposition for law (Miller 1990:224). Valuable work
has been done to establish who the lawspeakers were, in terms of names, dates, and social
connections (Jn Sigursson 1886, Gsli Sigursson 2004), but the slendingasgur have
barely been utilised as sources justifiably in this context, given the contested status of these
texts as reliable historical documents, and the caution which needs be exercised in using them
as such.
The slendingasgur convey other kinds of information than historical fact, however: cul-
tural memories, social traditions and institutions, and thirteenth-century attitudes to the past
(Whaley 2000). Little has so far been written about popular perceptions of the office of
lawspeaker, social memories of individuals, or the portrayal of lawspeakers in literary texts,
and this paper will address these issues. First, the evidence available is detailed for each
lawspeaker in turn, followed by a more general consideration of the way the lawspeaker func-
tions in the world of the slendingasgur. Table 1 below gives an overview of the appearances
of lawspeakers in the slendingasgur.1

Table 1. Lawspeakers in the slendingasgur

lfljtr Hrafn rarinn orkell orgeirr Grmr Skapti Steinn
Br g r
Drop g
Egil r g X g
Eyrb
Finn X
Fl X
Fljt r
Grett L L L
Gull h
Gunnl X
Har L
Lax g
Ljs X X
Njl g X L
Reyk X L
V-L
orstst
orstux h
lk X
Total 2 2 3 2 7 1 8 1

1
Only lawspeakers who feature in the slendingasgur are included. For the evidence for the others, see Jn
Sigursson 1886; Gsli Sigursson 2004.

151
Table 1 cont.
orkell T Kolbeinn Sighvatr Finnr Snorri Sturla Total
Br 2
Drop 1
Egil 4
Eyrb g 1
Finn 1
Fl 1
Fljt 1
Grett r 4
Gull 1
Gunnl g 2
Har 1
Lax g 2
Ljs r 2
Njl 4
Reyk 1
V-L 1
orstst g g g 3
orstux 1
lk 1
Total 1 1 1 1 3 1

Key
h = historical context
L = primarily legal role
r = referred to in passing
g = named in genealogy; no role in saga
X = role in action of saga
Bold type denotes use of the title of lawspeaker; italics denote no mention of title.

The lawspeakers
lfljtr
lfljtr is conventionally considered the first lawspeaker, but while memorising and reciting
the law at the first Aling would seem to qualify him for the title, he is remembered in the
slendingasgur only as having brought law to Iceland and only in two rather minor places,
Gull-ris saga (where he is invoked in passing), and orsteins ttr uxafts (which gives
basic details of his settlement and bringing of the law). Despite lfljtrs role in Icelandic
history, he is too early for the Saga Age proper, accounting for his minimal role in the
slendingasgur.

Hrafn Hngsson (c.930c.949)

Hrafn is credited in Egils saga with the distinction of being fyrstr lgsgumar slandi (F
2:59) the first lawspeaker in Iceland (and gfgastr sona Hngs (ibid. 60) noblest of
Hngrs sons). This does not afford him any widespread fame in the slendingasgur, how-
ever, being mentioned elsewhere only in a Njls saga genealogy. Like lfljtr, Hrafns term
falls earlier than the period covered in detail by the slendingasgur.

rarinn leifsson (c.95069)

rarinn is listed among the sons of leifr hjalti in Egils saga, and identified with the nick-
name Ragabrir Ragis brother in Njls saga and in a genealogy in Laxdla saga. In all
three sagas it is noted that he held office. His nickname also used in slendingabk and

152
Landnmabk is firmly associated with his traditional identity; it seems his credentials as
lawspeaker were not sufficient to make him memorable on his own merits, and Ragi appears
to have been the more interesting member of the family. Unfortunately, no extant source gives
further details of Ragis exploits, though he is described as vgamar mikill (F 12:41) a great
warrior in Njls saga.
Njla describes rarinn favourably as strvitr mar (ibid.) a greatly wise man and intro-
duces a third brother, Glmr, who survives four chapters as the second of Hallgerrs ill-fated
husbands. Baldur Hafsta (2001:334) notes that Glmr is not attested elsewhere, and sug-
gests that Njla makes repeated efforts to connect invented characters with historical figures.
It is worthy of note, then, that he may have seized the opportunity to invent an leifsson to
bring a lawspeaker into the saga.

orkell orsteinsson (97084)

orkell mni moon is another early lawspeaker whose nickname appears to be firmly estab-
lished in tradition and a noble heathen tradition in which his wisdom and foresight are
brought to the fore. As well as in slendingabk and Landnamabk, the nickname is used in
the two slendingasgur in which he appears: Grettis saga and Harar saga. In both he is
identified as lawspeaker and plays a cameo role as a legal expert. He is, however, anachronis-
tic in them, his term of office being later than the events described. In Grettis saga he is said
to have established the shore-rights law, an attribution not made in any other source. In
Harar saga he brings about a successful settlement to a case, and is described as bi vitr ok
ggjarn and kunnigr at allri rttvsi (F 13:267) both wise and benevolent; knowledge-
able in all matters of justice.

orgeirr orkelsson (9851001)

orgeirr is one of the most frequently-mentioned lawspeakers in the slendingasgur, appear-
ing in seven in all with a significant part in three. He is the only lawspeaker who can be said
to have a central role in an slendingasaga: Ljsvetninga saga although he appears only in
the first four chapters, he is head of the Ljsvetningar and the saga continues with tales of his
descendants.
There are two striking aspects of orgeirrs portrayal in the slendingasgur. The first is
that despite being a major figure in Reykdla saga, Finnboga saga ramma and Ljsvetninga
saga, he is nowhere identified as lawspeaker in the former two, nor in the last-named in the
section of the saga in which he actually plays a role. His place in slendingasgur tradition is
as the wealthy and powerful goi of Ljsavatn; this appears to be far more significant than his
sixteen-year term of office as lawspeaker. orgeirr is involved in legal cases and asked for
advice on legal issues, but this was a duty of any goi his part in the legal action of the sa-
gas is that of a powerful and influential man, not a legal expert. Even where he is named only
in genealogies, when we might expect the lawspeakership to be used to highlight the illustri-
ousness of the family, it is not mentioned.
The second striking feature about orgeirrs appearances in the slendingasgur is also
something which seems to be omitted, compared to what we know from elsewhere. In slend-
ingabk and Kristni saga it is orgeirr who goes under the cloak and decides it should be law
that everyone in Iceland be Christian, with the (one would think) memorable words es vr
sltum sundr lgin[]vr monum slta ok friinn (F 1:17) if we break asunder the law, we
will also break the peace. Yet among the slendingasgur it is only Njla that makes any
reference to this event, in an account ultimately indebted to Aris. orgeirr is commemorated
in these texts first and foremost as Ljsvetningagoi, and it is his actions in this role that are
deemed sguligt.

153
Grmr Svertingsson (100203)
Grmr held office for just two summers, and appears among the slendingasgur only in Egils
saga, where he is introduced with reference to his more famous nephew and successor, Skapti
roddsson. It is presented merely as an afterthought that Grmr var ok lgsgumar (F
2:241) Grmr was also lawspeaker (though his term comes after the events of the saga).
Grmr marries Egills niece and foster-daughter, rds, and Egill lives with them at Mosfell
in his later years. Otherwise there is little to distinguish Grmr, though he is ascribed the con-
ventional qualities of being auigr ok ttstrr (ibid.) wealthy and of good family.

Skapti roddsson (100430)2

Skapti is the best-known lawspeaker to the slendingasgur, appearing in eight. Like orgeirr,
Skapti was also a goi; but his lawspeakership is central to his identity. He is explicitly said to
be lawspeaker in five sagas and appears solely in connection with legal matters in a further
two. This is likely due to his twenty-six-year term of office, meaning he was lawspeaker for a
substantial part of his political life, during a substantial part of the Saga Age.
Landnmabk bestows the byname Lg-Skapti and slendingabk portrays a wise, firm
leader, suggesting a respected place in tradition in which Skaptis accomplishments were well
known. However, his portrayal across the slendingasgur varies. His most favourable depic-
tion is in Grettis saga, where he is never mentioned without his title and exhibits ideal if con-
ventional lawspeaker qualities:3 manna vitrastr ok heilrr (F 7:108) wisest of men and of
good counsel.
Demonstrating an admirable sense of fairness, Skapti declares he will not outlaw Grettir in
his absence. However, Grettir is found guilty and exiled regardless, a view of the
lawspeakers power somewhat different to that suggested by Aris reference to Skaptis rkr
[] ok landstjrn (F 1:19) power and governance. Nonetheless, Skapti remains on Gret-
tirs side, advising him during his exile. Lest this aiding of a convicted felon be taken as dis-
regard for the law, however, his help is limited to the strictly legal: en me v at ek skal heita
lgmar landinu, stendr mr eigi at taka vi tlegarmnnum ok brjta sv lgin (F
7:178) because I am called lawspeaker [lit. lawman] in this land, it is not fitting for me to
take in outlaws, and thus break the law.
Elsewhere in the sagas in which Skapti figures his depiction is at best neutral. He plays a
small role in Gunnlaugs saga (as a kinsman of Hrafns, he argues that Gunnlaugrs betrothal
to Gurn is invalid), Valla-Ljts saga (supporting Ljtr in legal matters, though without ex-
plicit identification as lawspeaker), and Flamanna saga (giving legal advice, but again with-
out reference to his title). There is no need to develop his character in these sagas: he is
merely a token legal expert. Njls saga, singularly, dramatises the establishment of the Fifth
Court; but unfortunately for Skapti, preservation of legal history comes second to require-
ments of plot and character development: Skaptis association with the event, evidenced in
slendingabk, is preserved in the saga, but he is relegated to a passive role while Njll ap-
pears the greater legal expert, masterminding the plan.
Skaptis expertise is acknowledged in Njla: he and his father are described as lgmenn
miklir (F 12:141) great legal experts, and he is consulted at several points during the pre-
liminaries to the Burning suit. However, he is insulted by Skarpheinn in a tantalising refer-
ence to a bizarre-sounding episode in his past:

2
rmann Jakobsson 1996 makes some similar observations in using Skapti as a case study on the methods of the
saga writers. This article came to my attention after the present paper had been written.
3
He is more usually referred to as lgmar than lgsgumar here an anachronistic usage of the post-
Commonwealth title owing to Grettis sagas date of composition, probably c.131020.

154
 heitir Skapti roddsson, en fyrr kallair ik Burstakoll, er hafir drepit Ketil r
Eldu; gerir r koll ok bart tjru hfu r. San keyptir at rlum at rsta upp
jararmen, ok skreitt ar undir um nttina. San frt til rlfs Loptssonar Eyrum, ok
tk hann vi r ok bar ik t mjlsekkum snum (F 12:2989).

You are called Skapti roddsson, but previously you called yourself Burstakoll [Smeared-
Head], when you had killed Ketill of Elda; you then shaved your hair and smeared tar on your
head. Then you paid some slaves to raise up a strip of turf and you crept under it for the night.
Then you went to rlfr Loptsson at Eyrar, and he took you in and carried you out in his meal-
sacks.

Sadly, no other source elucidates this incident. Some of the taunts directed at other figures in
this scene appear to allude to events known in other sagas (Cook 2001:331, nn 3, 5), suggest-
ing they could be generally-known; the anecdote about Skapti is specific enough to suggest a
basis in tradition, at least, if not truth. Perhaps this was the sort of behaviour Icelanders chose
to forget about their lawspeakers; hence what sounds like rather a good story does not survive
elsewhere.
Skaptis reputation suffers a further blow at the battle at the Aling. Discovering that his
son is involved, Skapti intervenes to try to stop the fighting, but is skewered through both legs
by a spear, necessitating the indignity of being dragged away. The literal crippling of the
lawspeaker may represent the failure of the law to control disputes and its helplessness against
the escalation of violence. To make matters worse, Kri and Snorri goi both immortalise the
incident in mocking skaldic verse, and Skapti is never compensated for the wounding a fur-
ther insult to his honour.
In lkofra ttr Skapti is one of six goar (all known from other sagas) whose jointly-
owned woodland is accidentally burnt down by the eponymous lkofri. Skapti is identified as
lawspeaker on first mention and prepares the case, albeit v hann sat nst (F 11:85) be-
cause he lived nearest, rather than because of his legal expertise. As far as can be deduced
from Grgs, the goar are within their rights to prosecute (and Skapti himself claims
mlaefni vr eru brn ok g (ibid.:89) the grounds of our lawsuit are just and good),4 but
much is made of the accidental nature of the damage and of the greed of the goar. Eventu-
ally, lkofri enlists the help of orsteinn Su-Hallsson and his brother-in-law Broddi Bjar-
narson, upon whose advice he persuades the goar to accept a settlement. He tricks them into
allowing orsteinn and Broddi to announce the terms: an insultingly low amount, followed by
Broddi verbally abusing the goar. Skapti is accused of composing a love-poem for the wife
of a kinsman, an offence punishable by full outlawry and again something that, if true, Ice-
landers may have wished to suppress about their lawspeaker.
Like Bandamanna saga, for which lkofra ttr has been suggested as inspiration, it is
likely that the ttr is a fictional satire on thirteenth-century chieftains. The named Saga-Age
goar are probably not personally the subjects of attack, although it is possible that the insults
directed at them in the flyting scene have a basis in tradition.5 Skapti may represent a real
thirteenth-century lawspeaker, but more likely is that his lawspeakership is incidental to the
ttr, where he does not actually manipulate the law. If an attack on the office was intended,
it would surely have been made more obvious. I think, rather, that Skapti was chosen because
he was a well-known Saga-Age figure, and that he is named with his title because the
lawspeakership was inextricably part of his identity.

4
There are no extant laws covering the exact situation, but see Gg Ib:94 and Gg II:463 (on burning grass); Gg
Ib:137 (on handling fire); Gg Ia:166 (on accidental damage).
5
For evidence that Skapti composed poetry, see below.

155
Steinn orgestsson (103133)
Steinn is the third lawspeaker to feature in Grettis saga, putting in two appearances in connec-
tion with legal matters. Like the other lawspeakers in the saga, he is said to be a vitr mar (F
7:244) wise man and is sympathetic to Grettir. He declares that the maximum period of out-
lawry should be twenty winters, and that sorcerers should be outlawed, legislation not attrib-
uted to him elsewhere.

orkell Tjrvason (103453)

Virtually nothing is known about orkell, despite his having held office for twenty summers.
Hence is it not certain that the orkell Tjrvason making a lone appearance (without the title)
in Ljsvetninga saga is the same man as the lawspeaker, although it seems likely: he is said to
be a grandson of orgeirr, and father of one Hrlfr, holder of the Ljsavatn goor. Gsli
Sigursson (2004:71) thinks it is highly surprising, that if this orkell really had held the
distinguished position of lawspeaker, that Ljsvetninga saga should make no mention of this
fact, but owns that cultivation of the law seems to have run in families. Since Ljsvetninga
saga makes no particular effort to associate orgeirr himself with the title of lawspeaker, it is
perhaps not so surprising that this lone mention of orkell is silent as to his position.

Gellir Blverksson (105462)

Gellir is not mentioned in the slendingasgur, although an Eyjlfr Blverksson features in
Njla with a genealogy which by comparison to Landnmabk would make him Gellirs half-
brother. Eyjlfr is said to be inn rii mestr lgmar slandi (F 12:363) the third greatest
lawyer in Iceland, but acts on Flosis behalf in the burning suit and is killed in the battle at
the Aling. It may seem surprising that the author of Njla passes up this golden opportunity
to refer to another lawspeaker, Gellir; but intriguingly, Eyjlfr is not attested in other sources.
This of course does not mean he did not exist; it is entirely likely that two brothers could be
trained as legal experts, and given Eyjlfrs negative portrayal in Njla, it could be that he
was conveniently forgotten by the compilers of Landnmabk. The author of Njla may also
have kept silent on Eyjlfrs kinship to Gellir so as not to tarnish the latters reputation (al-
though a negative comment in Eyjlfrs introduction to the saga attacks his family, and as
Skapti proves, the author is not averse to mocking lawspeakers). However, it could also be
that the author, clearly knowledgeable in legal history, invented the character not unlike
Glmr leifsson and gave him a genuine legal pedigree to satisfy a quest for authenticity, or
as an in-joke directed at the more esoterically-minded members of his audience. This suggests
that Gellir and his lawspeakership may have been better known than is indicated by his infre-
quent and sketchy appearances in the extant sources.

Kolbeinn Flosason (106671)

Kolbeinn is another figure whose historical identity is unclear. However, a Kolbeinn
lgsgumar is mentioned in orsteins ttr stangarhggs as having married the granddaugh-
ter of one of the protagonists.

Finnr Hallsson (113945)

A priest named Finnr Hallsson appears in the same genealogy in orsteins ttr as Kolbeinn
Flosason; Snorri Sturluson is also named. The title is not given here, however, and the genea-
logical details differ from those in Landnmabk; consequently it is not certain that the refer-
ence is to the same Finnr Hallsson prestr who was lawspeaker (Gsli Sigursson 2004:80 n.
27).

Snorri Sturluson (121518; 122231)

156
Snorri is mentioned in genealogies in three slendingasgur, but never specifically denoted
lawspeaker. It can be assumed that Snorris general renown, rather than his lawspeakership in
particular, is the reason for his inclusion.

Sturla rarson (1251)

Sturla is referred to as a source on three occasions in Grettis saga, twice with the title lg-
mar. He is treated with respect as a legal authority, and, like the sagas other lawspeakers, is
a Grettir fan: Hefir Sturla lgmar sv sagt, at engi sekr mar ykkir honum jafnmikill fyrir
sr hafa verit sem Grettir inn sterki (F 7:289) Sturla lgmar has so said that no outlawed
man seems to him to have been equally as great as Grettir the strong.

Conclusions
There are surprisingly few references to lawspeakers in the slendingasgur, especially in
light of the iconic role that is often assumed for them (e.g. Hastrup 1990:74). The lawspeaker
may have been unique, but this was apparently not in itself enough to make him a hero of
popular tradition. This suggests that by the thirteenth century, either anecdotes about individ-
ual postholders had been largely forgotten, or the office and its holders were not considered
especially saga-worthy material.
There do not appear to have been many strong traditions about the personality of individual
lawspeakers. There are a few, possibly spurious, associations of lawspeakers with particular
laws, but this is not a frequent feature of the slendingasgur.6 While the bynames of orkell
mni and rarinn Ragabrir are clearly part of their traditional identity, this seems to be
more habitual association than a trigger for anecdotes. It should not be forgotten, though, that
the slendingasgur are limited in range to a particular time period, and that two lawspeakers
orgeirr and Skapti held office for a combined total of forty-four summers at the heart of
this period. The potential for appearances by other lawspeakers is thus restricted, and indeed it
is these two, of all the lawspeakers, who figure most often. It is also these two who have any
sort of saga personality but this is not necessarily consistent across the sagas they appear in.
One can conclude from the slendingasgur that orgeirr was a powerful goi though,
perhaps excepting Njla, one would struggle to note his lawspeakership and that Skapti,
uniquely, was a well-known lawspeaker. Beyond this it is difficult to construct a clear identity
for either. The picture of the wise orgeirr and his fundamental role in the conversion, painted
in slendingabk and so familiar to the modern scholar, stands out among the slendingasgur
when drawn upon in Njla. To the sagas, orgeirr was goi at Ljsavatni. Despite the impor-
tance of the conversion in a number of sagas, it is not their place to document its history. It is
not that orgeirrs role is misremembered; rather, on the whole, it is just not detailed. Yet it is
difficult to reconcile what appear to be two very different traditions circulating about orgeirr,
and while I do not wish to draw any too-clearly delineated distinctions between audiences,
perhaps one tradition Conversion-orgeirr was the preserve of a learned and/or ecclesias-
tical community; the other goi-orgeirr that of popular folk-tale and secular concerns.
There was evidently widespread knowledge of Skaptis name in the thirteenth century (and
beyond), and his lawspeakership was clearly a fundamental part of his traditional identity. The
saga authors, however, apparently felt free to manipulate his character or draw selectively on
tradition to suit their needs, and holding the post did not guarantee him respect. It is possible,
though, that some of the less favourable traditions about him were suppressed because of his

6
Some sagas discuss the adoption of new laws but do not attribute them to any particular figure, e.g. Eyrb ch.
38, Gunnl ch. 11. Four clauses in the Konungsbk manuscript of Grgs name lawspeakers as their originators
(Gg Ia:1223, 184; Ib:23, 147), while slendingabk suggests a degree of legislative autonomy for lawspeakers.
Again, however, such instances are few and far-between (Burrows 2007:99102).

157
position. Hints at these incidents suggest that there were more memories of Skapti in circula-
tion than have survived. If we take into account the evidence of Skldskaparml, which cites a
skaldic helmingr attributed to him (he is also listed in Skldatal) (see Burrows forthcoming),
there is almost enough material for an immanent saga of Skapti roddsson and what with
his holding the lfus goor, outlawing important chieftains, establishing the Fifth Court,
having a skaldic career and an affair with the wife of a kinsman, and being smuggled into
Iceland covered in tar and flour, it would be interesting indeed.7 Nonetheless, he remains by
some distance the only lawspeaker about whom this can be said.
When lawspeakers are mentioned in the slendingasgur, then, they tend to have one of
two main functions:
1 A number of sagas refer to lawspeakers in a genealogical context, even if they play no
other role. In these cases, the lawspeaker is usually identified with his title, the exceptions
being orgeirr Ljsvetningagoi, a well-known saga figure for whom the lawspeakership is a
minor part of his identity; and Snorri Sturluson, who, likewise, was known for reasons other
than his time in office.
The saga genealogy was not merely a faithful account of all the members of a family, but a
way of linking them with the beginnings of Icelandic society and with important ancestors or
descendants. It also provided an opportunity for descendants of settlers or saga personages to
demonstrate their lineages (Clunies Ross 1993). It seems a reasonable assumption that a
lawspeaker in the family would be worth drawing attention to, and on the face of it, the genea-
logical appearances of titled lawspeakers in the slendingasgur bears this theory out.
However, the inference can be taken only so far. Although the number of lawspeakers re-
ferred to in this way is not insignificant, neither is it considerable. While mention of a post-
holder usually comes with explicit use of the title, it is more common that the lawspeaker
does not feature at all.
2. Lawspeakers are introduced to a saga to perform a legal role. This accounts for the ma-
jority of appearances of lawspeakers in the sagas. Their roles in these cases are usually brief;
they are token figures without any distinctly-drawn character portraits, though the traits of
wisdom, knowledge and good intentions are often highlighted and seem to be stock
lawspeaker characteristics perhaps an affirmation that, at least in general, the office carried
positive connotations. It is also worthy of note in this regard that although individual
lawspeakers are not beyond being criticised, there are no occurrences in the slendingasgur
in which a lawspeaker is shown to exploit his position for his own benefit.
Though individual lawspeakers do not have starring roles, lawspeakers in general are
drawn upon effectively by some individual sagas. Njls saga and Grettis saga each refer to
four different lawspeakers, more than any other, and it is interesting that these are both among
the later, post-Commonwealth sagas perhaps these authors felt freer in their use of the now-
defunct office than did those for whom it was still current. In Grettis saga, lawspeakers are
wise and respected, and all come out in support of Grettir, demonstrating the unjustness, even
unlawfulness, of his outlawry. While the letter of the law demands his exile, those most
learned in its spirit disagree but are increasingly powerless in a changing society in which less
noble factors are ever more prominent. In Njla, however, the lawspeakers are on the whole
either incidental to the action (Hrafn, rarinn) or are portrayed with scorn (Skapti). Both
sagas demonstrate the decreasing power of the lawspeaker and the offices ultimate futility,
but Grettis saga with a fond respect, Njla with a cynical criticism.
Overall, then, lawspeakers are not a major part of the subject matter of the slendin-
gasgur, and when they do feature they are more often types than personalities. But the

7
The concept of the immanent saga was first posited by Carol J. Clover (1986:1039). For a recent case study
see Gsli Sigursson (2007).

158
slendingasgur do not often provide superfluous detail for its own sake (Heinrichs
1976:142). Mention of a lawspeakers title in genres not limited to the slendingasgur
tends to be made only when it is relevant to the plot.8 Similarly, bringing a lawspeaker into a
saga could be misleading unless he was to perform a specific legal function in it. Lawspeakers
come into the sagas in an advice-giving capacity, but because postholders had no judicial au-
thority and could not influence the outcome of lawsuits, they were not major players at least
in their official role in the conflicts that the sagas commemorate.

Bibliography
rmann Jakobsson, 1996: Skapti roddsson og sagnaritun mildum. rnesingur 4, 21733.
Baldur Hafsta, 2001: Egils saga, Njls saga and the shadow of Landnma: The Work Methods of
the Saga Writers. In: Sagnaheimur: Studies in Honor of Hermann Plsson on his 80th Birthday, 26
May 2001. Ed. by sds Egilsdttir & R. Simek. Vienna. Pp. 2137.
Burrows, Hannah, 2007, Literary-Legal Relations in Commonwealth-Period Iceland. Unpubl. Ph.D.
thesis, University of York.
, forthcoming: Rhyme and Reason: Lawspeaker-Poets in Medieval Iceland. Scandinavian
Studies.
Clover, Carol J., 1986: The Long Prose Form. Arkiv fr nordisk filologi 101, 1039.
Clunies Ross, Margaret, 1993: The Development of Old Norse Textual Worlds: Genealogical Struc-
ture as a Principle of Literary Organisation in Early Iceland. Journal of English and Germanic
Philology 92, 37285.
Cook, R. (trans.), 2001: Njals saga. London.
Heinrichs, Anne, 1976: Intertexture and its Functions in Early Written Sagas: A Stylistic Observa-
tion of Heiarvga saga, Reykdla saga and the Legendary Olafssaga. Scandinavian Studies 48,
12745.
Gsli Sigursson, 2004. The Medieval Icelandic Saga and Oral Tradition: A Discourse on Method.
Trans. by N. Jones. Cambridge, MA.
, 2007. The Immanent Saga of Gumundr rki. Trans. by N. Jones. In: Learning and Under-
standing in the Old Norse World: Essays in Honour of Margaret Clunies Ross. Ed by J. Quinn, K.
Heslop & T. Wills. Turnhout. Pp. 20118.
Gg Ia, Ib: Islndernes Lovbog i Fristatens Tid, udgivet efter det Kongelige Bibliotheks Haandskrift.
Ed. by Vilhjlmur Finsen, 1852. 2 parts. Copenhagen.
Gg II: Grgs: efter det Arnamagnanske Haandskrift Nr. 334 fol., Staarhlsbk. Ed. by Vilhjlmur
Finsen, 1897. Copenhagen.
Hastrup, Kirsten, 1990: Island of Anthropology: Studies in Past and Present Iceland. Odense.
F 1: slendingabok Landnmabk. Ed. by Jakob Benediktsson. 1968. Reykjavk.
F 2: Egils saga. Ed. by Sigurur Nordal. 1933. Reykjavk.
F 7: Grettis saga smundarsonar. Ed. by Guni Jnsson. 1936. Reykjavk.
F 11: Austfiringa sgur. Ed. by Jn Jhannesson. 1950. Reykjavk.
F 12: Brennu-Njls saga. Ed. by Einar l. Sveinsson. 1954. Reykjavk.
F 13: Harar saga. Ed. by rhallur Vilmundarson & Bjarni Vilhjlmsson. 1991. Reykjavk.
Jn Sigursson, 1886: Lgsgumanna tal og Lgmanna slandi, me skrngargreinum og fylgisk-
jlum. Safn til sgu slands og zlenzkra bkmenta a fornu og nju 2, 1250.
Miller, William Ian, 1990: Bloodtaking and Peacemaking: Feud, Law, and Society in Saga Iceland.
Chicago.
Whaley, Diana, 2000: A Useful Past: Historical Writing in Medieval Iceland. In: Old Icelandic Lit-
erature and Society. Ed. by M. Clunies Ross. Cambridge. Pp. 161202.

8
I examine the appearances of lawspeakers in other genres of medieval Icelandic writing in Burrows 2007, ch. 2.

159
 Vatnsdla saga and Onomastics:
the case of Ingimundr orsteinsson
Jrg Bschgens, Bonn University, Germany
Vatnsdla saga is not exactly the most cherished text within the corpus of Icelandic family
sagas. One could, however, argue that it is the most orthodox to our nomenclature since it
concentrates not on a single character or a narrow selection of heroes, but tells us in a more or
less unfocussed manner a story about the establishment of Vatnsdal chieftainship, beginning
in the days predating the Norwegian exodus and ending in the first half of the 11th century.
Nevertheless, it seems to lack a greater theme or a structural unity apart from the family
ties that bind the different narrative units together. At least on the surface, it is lacking a dis-
tinguishable climax or central event that could give the mere accumulation of often underde-
veloped feud patterns some kind of meaning or coherent structure.
According to Theodore Anderssons and comments from other scholars on saga rhetorics,
such a climactic event1 should have been made identifiable by the writer through staging, a
slowing down of the narrative pace, a heightening of several levels of detail and a broadening
of dialogue (Andersson 1967:55). When he analyzed Vatnsdla saga, he was indeed able to
identify a staged event: the death of Ingimundr orsteinsson. But since Ingimundr dies half-
way through the story, and the storyline that is connected with his death ends with the killing
of his murderer Hrolleifr and his trollish mother Ljt a few pages later, Andersson concluded:

The only dramatic piece is the death of Ingimundr, prepared as it is by the unfolding friction be-
tween Hrolleifr and his neighbors, but it is impossible to regard this as the pivot of the saga.
Hrolleifr tends rather to be absorbed into the series of ill-fated sorcerers and scoundrels eradi-
cated by several generations of beneficent Vatnsdlir. (Andersson 1967:221)

While I share Anderssons opinion that the death of Ingimundr is clearly marked by staging, I
think that there is another event in this text highlighted by this method, namely the death of
Jkull Ingimundarson the Elder. It happens right at the beginning of the saga and is thus
hardly describable as a climactic event. But regardless of how we classify these two events,
they are rhetorically clearly distinct from the rest of the narration and should thus be worth a
closer look.
What drew my attention to the Jkull episode, apart from its narrative quality, was the per-
sonal name Jkull, which catches ones eye because it is clearly readable as an (Old-) Ice-
landic expression for ice, glacier and also serves as name for a being in Hversu Nregr
Byggist as well as the older and somehow related Fundinn Nregr, a descendant of primor-
dial Fornjtr and himself the father of king Snr, which might suggest that he was conceived
as some kind of giant. Beside the names occurrence in Hversu Nregr Byggist, which I
would like to come back to later in this paper, the name Jkull is featured in a handful of For-
naldarsgur2. A scene in Sturlaugs saga starfsama shows that the name Jkull was used for an
onomastic pun. When the character Jkull asks who is willing to fight him, the similar illus-
triously named Frosti replies: Mun eigi at makligast, at ek gangi r i moti, v at frostit
herir jkulinn? (Sturlaugs saga:127). When we reach the horizon of the Sguld, the name
occurs for a very limited number of persons apart from the two Jkuls in Vatnsdla saga3.
1
As Lars Lnnroth has shown (Lnnroth 1976:94), every incident of some importance for the main plot within
saga narratives is marked by staging, but when a climax is reached, the staging becomes more intense.
2
Cf. FAS4, Vol. IV, pp. 388
3
Characters with the name Jkull do also appear in Vglundar saga and Gunnars saga Keldugnpsffls. In these
two texts they are rather evil characters with a violent temper. Jkull Ingimundarson the Younger is also the

160
One is Jkull Brarson, who is described in Grettis saga as: mikill mar ok sterkr ok in
mesti ofsmar, hann var siglingamar ok mjk dll, en mikillhfr mar. (Grettis
saga:117) As Jkuls ttr Brarsonar tells us, he is later killed by Saint lf. The other char-
acter is Jkull Bason, the son of the giantess Frr, as Kjalnesinga saga tells us. He kills his
father by accident and travels to the borders of the world to fight trolls in Jkuls ttr Baso-
nar. This shows that the name Jkull could mark in a literary work a character with aspects
associated with the world of giants and similar inhabitants of the tgar sphere which mani-
fest itself either directly through a biological link or through a rather unsocial and violent
temper.
It is thus not much of a surprise that our two Jkuls in Vatnsdla saga could also accu-
rately be described as rather physical characters with a tendency for violent behavior. Jkull
the Elder is a robber, while Jkull the Younger is contrasted through his violence against his
mildly mannered brother orsteinn, who is the brain in nearly all of their operations4. At least
some of these Jkuls represent the Grettir type5 of the variety of Old-Icelandic heroes, and
one could suggest that this kind of hero is already marked through a certain type of personal
name6. It is hardly surprising that a medieval writer, who lived in an age that was obsessed
with etymologies, (Haubrichs 1975) and language in general (Clunies Ross 1987:30f), would
use such a technique, and that an audience, even if it does not have a formal education, could
get at least parts of the message. Onomastic wordplays were also not uncommon in skaldic
poetry (cf. Mundal 2004) and even in saga writing.
I think it is a common observation among saga readers that many characters that are mar-
ginal both in regard to the story and socially bear names that consist of just one element, are
either readable or of foreign origin, and have a derogatory meaning that is in line with the role
they play or the social stratum they represent. Typical examples of such names are frequent
thrall names like Svartr and Kolr. In Vatnsdla saga we encounter Ljt and Ljtr, Hrolleifrs
companions in mischief, and both characters without any official genealogical links. Both
the masculine and the feminine form of the name seem to have been common as real personal
names, but the direct combination of these two names in a part of the plot where the epitome
of Vatnsdal chieftainship is murdered, together with Ljts association with heathen worship
and black magic, strongly suggest that we again have a case where personal name and type of
character form a unity.

opponent of Finnbogi in Finnboga saga ramma.

4
While this is a constant theme in the interaction of the two brothers, who seem to form a kind of symbiotic
unity, this concept is nicely prepared in the naming scene: Sj sveinn hefir hyggiligt augnabrag, ok skal eigi
seilask til nafns; hann skal heita orstein, ok mun ek ess vilnask, at hamingja mun fylgja. Sj sveinn var
snimma vnn og grviligr, stilltr vel, orvss, langsr, vinfastr ok hfsmar um alla hluti. Son ttu au annan;
sj var ok borinn at fer snum, ok skyldi hann ra fyrir nafni; hann leit ok mlti: essi sveinn er
allmikilfengligr ok hefir hvassar sjnir; hann mun vera, ef hann lifir, ok eigi margra maki ok eigi mikill skap-
deildarmar, en tryggr vinum ok frndum ok mun vera mikill kappi, ef ek s nkkut til, mun eigi naur at min-
nask Jkuls frnda vrs, sem fair minn ba mik, ok skal hann heita Jkull. (V. s.:37)
5
Lnnroth 1976:62; perhaps one should modify Lnnroths type slightly to fit Jkull orsteinsson. He does not
suffer a terrible death or is forced to live in outlawry, but it is actually his brother who prevents him from this
kind of fate. The relationship between the two brothers is aptly illustrated in a short dialogue in chapter 34:
mlti orsteinn: Hver er n ragr n? Jkull svarar: etta veit ek ik eigi fyrr grt hafa, at leita ra
undir mik; mun hr ok til ltils at sj, ef eira arf vi, en verr mr eigi til essa rftt; vit skulum fara til
Undunfells, ok skal rir fara me okkr, brir okkar. This kind of balance between these two types of charac-
ter is a constant theme throughout the saga. It reaches its peak in the conflict between ttar Grmstungum and
the Vatnsdlir, where orsteinn is forced to use a legal trick to prevent his son Inglfr and his brother Jkull
from starting an unjust bloodfeud. (cf c. 37)
6
I think that it might not be purely due to chance, that darker heroes like Grettir, (Vga-)Glmr, (Vga-)Styrr,
Egill, Skarp-Heinn and Jkull bear names that point toward the direction of danger or general violence.

161
 While it is surely a complicated question how far an uneducated audience could follow a
literary expert, I think that my thesis regarding the name Jkull is not all that problematic be-
cause it is a noun of everyday use and thus clearly understandable. Even if there existed a
person with the name of Jkull in the genealogy of the Hofverjar, which, of course, cannot be
repudiated, it seems to be clear that name and character form a unity and become a literary,
fictionalized figure. In his article about The Significance of Names in Old English Literature,
Fred Robinson takes up an old discussion in his discipline concerning the interaction between
name and character in the case of King Heremod in Beowulf and the problem of him being a
figure of Danish historical tradition. In response to a point made by Frederick Klaeber in
Beowulf and the Fight at Finnsburg, Robinson writes after his analysis of both hagiographic
and heroic poetry:

Despite this aptness of the name, however, Klaeber dismisses Mllenhoffs theory7 with the
objection, But later studies have shown him [Heremod] to be a definite figure in Danish his-
torical-legendary tradition. But is the implied premise valid? Must we exclude the possibility
that even unalterable names inherited from tradition can bear pregnant meanings when the poet
wishes them to? (Robinson 1968:5151)

He ends his article with the name Hygelac/Hugleikr and a summarization of both Old English
and Old Norse scholarship about this name. The consensus appears to be that both the Beo-
wulf poet, Saxo Grammmaticus and Snorri etymologized the name and arranged their literary
realization according to their respective etymology (Robinson 1968: 5257).
A similarly interesting observation was made by Paul Beekman Taylor regarding the con-
ception of Kri and Flosi in Njls saga. He sees a link between the name Flosi, from flsa
(to run precipitously designates irresponsible, deceitful, fickle, Beekman Taylor
1998:146) and the transformation of the otherwise quite honorable character into a vicious
avenger through Hildigunns charge (Beekman Taylor 1998:146). Kris revenge and recon-
ciliation with Flosi is in two instances directly influenced by the wind (Beekman Taylor
1998:146), which is remarkable since he is the namesake of another Kri, the son of Fornjtr,
who is the mythological embodiment of the wind (Clunies Ross 1983:57).
A rather puzzling example of the poetics of personal names can be found in Eyrbyggja
saga. The writer, who has a tendency to explain each and everything, tells us that our well
known saga heroes (Vga) Styrr and Snorri goi did not receive the names under which they
rose to literary fame at birth. They first had other names but were renamed at a later stage in
their lives because of their temper (Eyrbyggja saga: pp.2021). That Snorri goi was first
called orgrmr and later received the name Snorri because of his temper is also told in Gsla
saga Srssonar (p. 57).
The poetical use of names in Old Norse prose narrative certainly is a field that would de-
serve further research, but I think that these examples alone support Robinsons comment on
the creative possibilities of a medieval writer or poet.
But let us briefly return to the name of Jkull.
Since the event of Jkulls death through the hand of orsteinn Ketilsson is a staged narra-
tive unit, we should take a look at what happens in this scene. orsteinn is characterized both
by the author directly and his father (V.s.:3f) as a physically rather insignificant figure, but
after his father has egged him on he decides to set out for the yet anonymous robber who
threatens the community. When he finally enters Jkulls house, we witness a rare instance of
inner dialogue during which orsteinn struggles with his conscience about the necessity of
7
Mllenhoff (Mllenhoff Karl: Beovulf: Untersuchungen ber das angelschsische Epos und die lteste Ge-
schichte der germanischen Seevlker. Berlin 1889, p.51.) translates the name Heremod as kriegerischer Mut
and concludes that name and character fit together remarkably well.

162
killing Jkull8. Jkull Ingimundarson the Elder should be more to the liking of Ketill raumr,
orsteins father, because he is the prototype of Fornld ethics, as he himself declares: eptir
htti rkra manna sona aflaa ek mr fjr, tt heldr vri frekliga att ort (V.s.:9). Jkull
shows mercy towards orsteinn because he sees his extraordinary moral qualities9. The con-
stant mirroring of superficially opposed virtues running through the whole story of the Ingi-
mundarsons is already present in the story of the killing of Jkull the Elder through orstein
the Elder, and that is the reason why the writer put so much detail into this passage. In this
scene the foundation is laid for understanding the interaction between orsteinn and Jkull
Ingimundarson.
While I think that the connection between name and character type in the case of Jkull is
quite transparent, I would like to spin the thread a little further. As the title of this paper al-
ready indicates, I would like to take a closer look at the figure of Ingimundr orsteinsson, the
most impressive representative of Vatnsdal chieftainship.
The name Ingimundr itself is perfectly regular and joins the wide variety of -mundr names
that can be found in the corpus of Old Norse literature. For a name that is so stereotypically
Germanic, it is somewhat odd that it appears only very rarely both in texts that are concerned
with the Fornld and those that play in the Sguld. It becomes more common in the post
conversion age up to the late Middle Ages (Lind 190515:637). While it is of course nearly
impossible to make a suggestion about the frequency of names in a basically illiterate age, the
earliest more or less historical bearer of this name I was able discover is the Norwegian viking
leader (H)Ingamund, who, according to the Fragmentary Annals of Ireland, raided Chester
and established a settlement on Angelsey at the beginning of the 10th century (Fragmentary
Annals:169). From Uppland and Sdermanland we have a variety of runestones which men-
tion people named Ingimundr from around the middle of the 11th century10.
In Old Icelandic literary texts that place their matter into the pre-Christian time, the only
two Ingimunds I was able discover apart from Vatnsdla saga were a character in Hversu
Nregr Byggist, Ingimundr Aalfsson (Flateyarbk I 1860:24), and Ingimundr Hafrsson in the
longer version of orvalds ttr vfrla; that the writer again used an otherwise rare name
that also appears in Hversu Nregr Byggist is an interesting feature that might serve to shed
some light upon the sagas function.
The longer A redaction of orvalds ttr vfrla, which is supposed to already have been
included in Gunnlaugr Leifssons Latin lfs saga Tryggvasonar (Rafnsson 2005:114), tells
us the story of five year old Ingimundr Hafrsson, who seeks out bishop Frirekr to receive
baptism. While Vatnsdla saga most likely integrated parts of orvalds ttr into its story,
the writer perhaps also drew some inspiration from this marginal character. When we look at
Ingimundr orsteinsson we see a man who is deeply rooted within the lifestyle of the heroic
Fornld but who points at the same time toward a yet distant future. He has a natural piety
that shows itself most obviously in his death, when he warns his slayer so that he can dodge
the revenge of Ingimundrs sons, a deed that orsteinn Ingimundarson characterizes as
ggirnd (V.s.:62). He is undoubtedly an anima naturaliter christiana (Kristjnsson
1994:242) but through his death he is transformed into a martyr. While Lars Lnnroth ob-
served that the noble heathen retains characteristics of the heroic age, especially the duty of
revenge (Lnnroth 1969:15), Ingimundr even stands above this legally supported act of just
violence. The hagiographic undertones in the narration of Ingimundrs death are further un-

8
Slkt (his fathers hvt and the troubles of his community) hvatti orstein fram, ok leitai hann sr fris, at
hann mtti einn hefna margra vanrttis, en ru lagi tti honum skai mikill um manninn. (V.s.:8)
9
mr segir sv hugr um, at munir gfumar vera (V.s.:9)
10
S 194, S 10, U 72, U 296, U 495, U 388 $, U 808, U 826, U 898, U 922 $, U 1090 g, see
http://www.nordiska.uu.se/forskn/samnord.htm

163
derlined by the use of curses directed against Hrolleifr: heljarmar (pp 53, 59), mannfjndi
(pp 56, 60), manndjfull (p 59), fjndinn (p 59) etc.
This depiction of Ingimundr as a saintlike figure is quite remarkable for someone who ear-
lier on seemed to have a special relationship to the pagan god Freyr, but this relation is a
rather odd one. Young Ingimundr is a down-to-earth viking who shares the stereotypical aver-
sion of the noble heathen against agents of the supernatural (V.s.:29). After a talk with his
protg Haraldr hrfagri he has to accept that Freyr wants him to leave Norway for recently
settled Iceland. Ingimundrs settlement is heavily loaded with imagery connected to Yngvi-
Freyr (cf. Meulengracht Srensen 1992), and some form of godly protection is not merely
indicated, but obvious through the supernatural guidance to Vatnsdal, a place that is separated
from its surrounding by its bursting plentifulness (V.s.: 4043), where Ingimundr establishes
a community of extraordinary tranquility (V.s.:47). The first thing Ingimundr does when he
constructs his homestead is erecting a large temple. This is again a rather odd behavior for a
noble heathen, since the world of heathen worship should be a suspicious one for him. While
Vatnsdla saga has a rich stock of all sorts of sorcery and heathen practices, Ingimundrs
temple only serves to trick the Norwegian Hrafn into giving his sword, which will become the
emblem of Vatnsdal chieftainship, to Ingimundr as compensation for violating the sanctity of
the building.
How can we summarize the religious ideas surrounding Ingimundr? The saga leaves no
doubt that the divine intervention is not an illusion but an observable fact, the protagonists
interpret this divine force as a form that is known to them as Freyr, the aristocratic (Motz
1996:13) sentinel over r ok frir (Snorraedda:29). Freyr might have been less problematic as
a prefiguration of the Christian god than giant slaying rr or the sorcerer inn. At some
point Ingimundr seems to have realized that Freyr is just a shell for the real divine guide of
his fortunes, the one who created the sun and the whole world (V.s.:62). Because of this
knowledge he is able to mock the old gods by using their temple for a sham.
Besides this heavenly king, Ingimundr also has a close relation to Haraldr hrfagri, who is
conceived as a just ruler and the greatest king of pre-Christian Scandinavia (V.s.:35). Earlier
on in this paper, I called Ingimundr King Haraldrs protg, and I think this description fits
their relationship quite well. Haraldr provides Ingimundr with rich gifts, arranges his mar-
riage, makes an exception to his anti-witchcraft policy so that Freyrs will can be revealed,
encourages Ingimundr to settle in Iceland and supplies him with timber to build a representa-
tive farm. Ingimundr, on the other hand, is obedient to Haraldr (V.s.:34) as he is (subcon-
sciously) obedient to Christian principles. While King Haraldr rewards him with gifts and his
friendship, God rewards Ingimundr with a rich and prosperous life and thefor us today
rather dubiousgift of martyrdom.
If we divide the name Ingimundr into its two elements ingi and mund(r), we get two nouns
that were understandable by a contemporary audience.
While ingi might have been a bit outdated, it was still in use in 13th century skaldic poetry,
as a stanza by Sturla rarson shows, in which he calls a Scottish nobleman j inga
(Skjaldedigtning I A:221). Far more interesting for our purpose is a Lausavsa ascribed to
lfr Haraldsson in the Legendary lfs saga hins helga in which he substitutes the name
Ingibjrg by gramr ok brattir hamrar (lfs saga hins helga:134).
More problematic is the element mund(r), not because of its meaning or its use in the 13th
century, but because I was not able to discover any poem or stanza where it is used in the con-
text of an onomastic wordplay. Nevertheless, as a mostly poetically used synonym for
hnd/armr (Lexicon Poeticum:413), mund is quite common. Perhaps it is not too far out to
assume that the name Ingimundr could express a special relationship between the name bearer
and earthly and heavenly kingship, perhaps with the meaning protected by/in the hands of
the lord/king. Such a meaning would perfectly go along with both Ingimundr Hafrsson and

164
Ingimundr orsteinsson, unfortunately not with Ingimundr jarl, from whom Ingimundr or-
steinsson inherits the name. It is tempting to read the name Ingimundr as a type-name, like the
name Jkull, since its meaning fits to the characters roll perfectly. He is the foundation for
the Hofverjar clan and Vatnsdal chieftainship, and through his death, he transformed to an
Icelandic version of innocently slain martyrking11.
I do not think that Vatnsdla saga is primarily concerned with religious matters or written
from a clerical perspective12, but the development of religious ideas or knowledge is clearly
employed by the writer as a structural device. orstein Ketilsson is reluctant to kill Jkull and
ends his life with the words: uni ek v bezt vi vi mna, at ek hefi verit engi gangsmar
vi menn (V.s.:32). Ingimundr orsteinsson gets rewarded for his ggirnd by a yet un-
known creator and orkell krafla, who is saved through an act of mercy by orsteinn and
rir Ingimundarson, replies to bishop Frirekr that he does not want to have any other faith
than eir orsteinn Ingimundarson hfu ok rir fstri minn; eir tru ann, er slina
hefir skapat ok llum hlutum rr. Byskup svarar: smu tr boa ek me eiri grein, at
tra einn gu fur, son ok helgan anda (V.s.:125). When the time is right, and the com-
munity of secular chieftains accepts and understands the necessity of the new faith, orkell
gets baptized, er kristni var lgtekin slandi (V.s.:126). Within this successive devel-
opment of a religious idea, Ingimundr orsteinsson bridges the historical and ideological gap
between the heroic age and the age of settlement. Through his character, the theme of being
chosen, both by the king and by god, enters the saga, and completes the figure of the ideal late
13th century chieftain, whose secular ideals are discussed and set in relation to each other in
two generations of orsteins and Jkuls.
It is not exactly a new idea that Vatnsdla saga discusses the properties an ideal chieftain
should have and might have been conceived as a kind of mirror for its contemporaries.
Sigurur Gumundsson called it hfdingja skuggsj (V.s.:XXXII), while Einar lfur Sveins-
son shared his, perhaps tongue in cheek, classification and said that its development should be
seen against the background of the changing political landscape of late 13th century north-
western Iceland (V.s.: XXVIIXXXII). I hope that my paper made clear that I, at least par-
tially, share their view about this rather unusual saga. I think that my analysis of the Jkull
figure supports the view that different aspects or qualities of chieftainship, and perhaps human
nature in general, are set against each other in the form of the central characters that exem-
plify both the positive and the negative traits of the respective quality, while characters like
Ingimundr orsteinsson and orkell krafla represent the ideal synthesis of boldness and re-
straint, muscle, brains and ethical behavior (cf. V.s.: XXXI). What distinguishes Ingimundr
from his sons and orkell, who all are men with extraordinary strong luck and without doubt
11
For the importance of the several types of martyr kings in both Anglo-Saxon and Scandinavian nation building
processes cf. Erich Hoffmann: Die heiligen Knige bei den Angelsachsen und den skandinavischen Vlkern.
Knigsheiliger und Knigshaus. (=Quellen und Forschungen zur Geschichte Schleswig-Holsteins, Vol. 69).
Neumnster 1975
12
I think that this is made clear in the dialogue between orkell krafla and bishop Frirekr, which I already par-
tially cited above. orkell takes baptism when the Icelandic legislature, consisting of secular chieftains, agrees to
accept it. It is tempting to see this attitude as a literary reflection of the struggle between sgrmr orsteinsson
and bishop rni orlksson that was only solved on a personal level when sgrmr died in 1285 and rni lifted
his excommunication. As Elizabeth Ashman Rowe has shown, there is a close connection between orsteins
saga Vkingssonar and the family of sgrmr (Elizabeth Ashman Rowe, 2004: Absent Mother sand the Sons of
Fornjtr: Late-Thirteenth-Century Monarchist Ideology in orsteins saga Vkingssonar. In: Mediaeval Scandi-
navia 14. Pp. 133160, cf. pp 154), Vatnsdla saga is like orsteins saga part of the Gautland Cycle that Lee
M. Hollander identified (Lee M. Hollander: The Gautland Cycle of Sagas. I. The Source of the Polyphemos Epi-
sode of the Hrlfssaga Gautrekssonar. II. Evidences of the Cycle. In: JEGP 11, 1912. Pp. 6181 & 209217),
therefore it is in my eyes not too far fetched that the struggle between secular and clerical powers at the end of
the 13th century in both Norway and Iceland also influenced the theme of Vatnsdla saga. That does not mean,
however, that I subscribe to the long outdated and never convincing concept of the two cultures.

165
noble heathens, is that he appears to be a chosen figure, one of these rare characters with
whom the Christian god communicates on a personal level. He becomes the spiritual founda-
tion for Hofverjar chieftainship. This is clearly shown by the narrative and indicated by his
name. Wherever the author got his names from, he was able to give them a deeper meaning
within his story.

Bibliography
Andersson, Theodore M., 1967: The Icelandic family saga. An analytic reading. Cambridge MA.
(Harvard Studies in Comparative Literature 28).
Beekman Taylor, Paul, 1998: Sharing Story: Medieval Norse-English Literary Relations. New York.
(AMS Studies in the Middle Ages 25).
Clunies Ross, Margret, 1983: Snorri Sturlusons Use of the Norse Origin-Legend of the Sons of Forn-
jtr in his Edda. In: ANF 98. Pp. 4766.
, 1987: Skldskaparml. Snorri Sturlusons ars poetica and medieval theories of language. Odense.
(The Viking Collection 4).
Eyrbyggja saga. 1935. Ed. by Einar lfur Sveinsson & Matthas rarson (slenzk Fornrit 4).
Flateyarbk I. 1860. Ed. by Gubrandur Vigfsson & Carl Unger. Christiania.
Fragmentary Annals of Ireland. 1978. Ed. & Transl. by Joan N. Radner. Dublin.
Gsla saga Srssonar. 1958. In: Vestfiringa Sgur. Ed. by Bjrn K. rlfsson. (slenzk Fornrt 6).
Grettis saga smundarsonar. 1939. Ed. by Guni Jnsson. (slenzk Fornrit 7).
Haubrichs, Wolfgang, 1975: Veriloquium Nominis. Zur Namensexegese im frhen Mittelalter. Nebst
einer Hypothese ber die Identitt des Heliand-Autors. In: H. Fromm et.al. (Eds.): Verbum et
Signum. Beitrge zur medivistischen Bedeutungsforschung. FS Friedrich Ohly. Vol. I. Mnchen.
Pp 231266.
Kristjnsson, Jnas, 1994: Eddas und Sagas. Die mittelalterliche Literatur Islands. Hamburg.
Lind, Erik H., 190515: Norsk-Islndska Dopnamn ock fingerade Namn frn Medeltiden. Uppsala &
Berlin.
Lnnroth, Lars, 1969: The Noble Heathen: A Theme in the Sagas. In: SS 41,1. Pp. 129.
, 1976: Njls Saga. A Critical Introduction. Berkeley.
Meulengracht Srensen, Preben, 1992: Freyr in islndingesagaerne. In: G. Fellows-Jensen & Bente
Holmberg (Eds.): Sakrale Navn. Uppsala, 1992 (=Norna rapporter, 48) Pp. 5575.
Motz, Lotte, 1996: The King, The Champion and the Sorcerer. A Study in Germanic Myth. Wien
(SMS 3).
Mundal, Else, 2004: Nomen est Omen. Umgang und Spiel mit Namen in der altnordischen Skalden-
dichtung. In: Astrid van Nahl et al. (Eds.): Namenwelten. Ort- und Personennamen in historischer
Sicht. Berlin. (RGA ErgB 44). Pp. 56576.
lfs saga hins Helga. Die Legendarische Saga ber Olaf den Heiligen (Hs. Delgard. Saml. nr.8II).
1982. Ed. & Transl. by Anne Heinrichs et al. Heidelberg.
Rafnsson, Sveinbjrn, 2005: lafs sgur Tryggvasonar. Um gerir eirra, heimildir og hfunda. Reyk-
javk
Robinson, Fred C., 1968: The Significance of Names in Old English Literature. In: Anglia 68,1. Pp.
1451.
Sturlaugs saga starfsama. 41950. In: Fornaldarsgur Norurlanda. Vol. 2. Ed. Guni Jnsson. Reyk-
javk.
orvalds ttr vfrla. 2003. In: Biskupa Sgur I. Ed. by sdis Egilsdottir (slenzk Fornrit 15).

166
 Sagas and Archaeology in the Mosfell Valley, Iceland
Jesse L. Byock, Scandinavian Section and Cotsen Institute of Archaeology, UCLA, USA
The relationship between sagas and modern archaeology is just beginning. This paper dis-
cusses the nature of the relationship in light of findings of the Mosfell Archaeological Project
(MAP). In particular, I will discuss the recent excavations in the Mosfell Valley (Mosfellsda-
lur) in Iceland, where we are unearthing a chieftains establishment at Hrsbr including a
longhouse, a church, a graveyard, and a cremation grave and other sites in the Mosfell Val-
ley including a stone ship setting and ships landing. The Mosfell Valley was the home of the
Mosfell chieftains (the Mosfellingar) a family of warriors, farmers, and legal specialists. Fo-
cusing on this glaciated and once wooded valley, our task is to unearth the prehistory and
early history of the Mosfell region. We seek the data to provide an in-depth understanding of
how this countryside or sveit evolved from the earliest Viking Age habitation.
The Mosfell excavation is an interdisciplinary research project employing the tools of ar-
chaeology, history, anthropology, forensics, environmental sciences, and saga studies. The
work is constructing a picture of human habitation and environmental change in the region of
Mosfell (Mosfellssveit). As part of our excavations we are developing a concept of valley-
system archaeology. Mosfellsdalur, the surrounding highlands, and the lowland coastal areas
form a valley system, that is, an interlocking series of natural and man-made components that,
beginning in the ninth-century settlement or landnm period, developed into a functioning
Icelandic community of the Viking Age.
The archaeological work began with surveys and test excavations in the mid 1990s and
major excavations began in 2001. The yearly archaeology, which continues into 2009, has
documented a rich Viking Age and landnm period occupational history.1 The 2001 excava-
tion at Hrsbr revealed the presence of significant remains, including an early church, a sur-
rounding cemetery, and an adjacent burial mound containing remains of human cremation.
The goals of our subsequent field seasons have been to expand the scope of this work, and in
recent years we have excavated a large (28 meters long) and exceptionally well-preserved
early tenth-century eldskli (firehall or longhouse).
Our excavations on the Hrsbr farm focus on four archaeological deposits: Kirkjuhll
(Church Knoll), the hillock just behind the modern farms stable; The tn or hayfield just
north of Kirkjuhll; Hulduhll (Elfin Hill), a hillock located about 60 m west of Kirkjuhll;
and Loddahll, a small knoll at the far north-eastern corner of the home field (tn), the hay
meadow immediately north of Kirkjuhll (see Fig. 1). Elsewhere in the valley we have several
major sites under excavation. This concept of multidisciplinary archaeology, combining
analysis of the cultural and environmental landscapes of a valley including the surrounding
highlands and coast, is particularly well-suited to Viking and North Atlantic archaeology.
From the start we have sought the significant oral memory of the local families. When we
began excavating in the Mosfell Valley in 1995, the knolls at Kirkjuhll and Hulduhll were
used as pasture. Both of these adjacent knolls were covered with grass, and their surfaces
were undisturbed except where the tramplings of cows exposed small patches of earth. The
farmers, lafur Ingimundarson and Andrs lafsson, whose family has lived on the land for
many generations, are extremely knowledgeable about life and the changes in land use in the
Valley.

1
For an overview of the work of the earlier excavations of Mosfell Archaeological Project, see Byock et al.
2005:195218 and Holck 2005:340348.

167
Figure 1. Site plan of the Church Knoll and Tn Excavations at Hrsbr.

No agricultural machinery had ever been used on the knoll because of the reverence attached
to Kirkjuhll in oral memory as the site of an ancient church. To date this remains the case, a
situation that is relatively rare on contemporary Icelandic farms which are highly mechanized.
The same has held true for Hulduhll, with oral stories attaching to it the interdiction that it
was to be left alone because it was inhabited by the hidden people or elves. As it turned out,
both knolls were connected with ancient mortuary rites, Christian and pagan.
Of crucial importance, the archaeology at the Mosfell Valley sites is aided by a wealth of
surviving medieval Icelandic writings, including The Book of Settlements (Landnmabk),
Egils Saga (Egils saga Skallagrmssonar), The Saga of Gunnlaug Serpent Tongue
(Gunnlaugs saga Ormstungu), Hallfreds Saga (Hallfrear saga), The Saga of the People of
Kjalarness (Kjalnesinga saga), The Saga of the People of Floi Bay (Flamanna saga), and
The Short Saga of Orm Storolfsson (Orms ttur Strlfssonar) in Flateyjarbk. These
sources describe sites in the Mosfell Valley and at Leirvogur (Clay Bay), the inlet on the coast
below the mouth of the valley into which the rivers of the valley flow.
If we are to believe the written sources, the Mosfell chieftains loomed large in the Viking
Age history of Icelands western region. The geographical position of their lands and their
area of power allowed the Mosfellingar to monitor and benefit from the travel and trade that
passed through their valley system. Egils Saga tells us about one of these leaders, Grmr
Svertingsson, who lived at Hrsbr. Grmr was the lawspeaker of Iceland from 1002 to1004,
the years immediately following the conversion in the year 1000. Grmr converted and is said
to have built a church at Hrsbr. From the medieval writings, one can piece together consid-
erable information about the Mosfell chieftains. For instance, Gunnlaugs Saga, Hallfreds
Saga, and Egils Saga indicate that the Mosfellingar controlled the Nesses, the region of mod-
ern-day Reykjavk, extending perhaps out to present day Seltjarnarnes. From the Nesses these
chieftains are said to have called up men to support their authority with force.

168
Figure 2. The longhouse at the end of MAPs 2007 excavation. The building seen from the eastern end
is divided into three rooms. The sunken central hall, which had a longfire down the center (and was
still to be excavated at the time of this picture), dates from around the year 900. The benches in the
central eldskli or fire hall can be clearly seen. The room at the far western end had a wooden floor.

The Mosfellingar are also said to have entered into marriage alliances with the goar (chief-
tains) at Borg in Borgarfjrr, the descendents of the landnmsmaur Skallagrmr
Kveldulfsson. Such an alliance, if it did in fact take place, was logical, and it would have
added considerably to the power and authority of both the Mosfellingar and the people at
Borg. The two were close enough to support each other but far enough away not to compete
for thingmen. The scene in Chapter 81 of Egils Saga when Egill comes to the support of his
son Thorsteinn, is one of the great moments in the sagas. When matters of feud and law look
bad for Thorstein, a man, leading a group of warriors, rides into the local assembly in Borgar-
fjrr. This was Egill Skallagrmsson, who had come with eighty men all fully armed as if
ready for battle, a choice company, for Egill had taken with him all the best farmers sons in
the Nesses. (Egils Saga 1976:226.)
Having medieval narrative sources, such as those connected with the Mosfell sites, or
written sources at all, is exceptional in Viking archaeology. Extensive Viking Age sites are
found throughout mainland Scandinavia, the British Isles and northern Europe, but because of
the paucity of written sources, archaeologists, historians, and anthropologists often know little
about the inhabitants, their personal history or specific socio-economic and political relation-
ships. The Viking Age sites in Mosfellssveit are somewhat different. The fact that we use all
the available sources is a distinguishing feature of our archaeology. This much we can say,
that despite all the saga evidence and in the face of the location right on the outskirts of pre-
sent-day Reykjavk, no one had excavated these sites since the twelfth century, when Egils
Saga tells us that the graveyard at Hrsbr was dug when the old conversion-age church was
taken down and a new church built further up the valley (Byock 1993).
Just how to find this graveyard was a question. We tried geophysical tests of the Kirkjuhll
and tn sites but the resulting magnetometer and resistivity maps of these areas did not sug-

169
gest the presence of subterranean architectural features. Nevertheless, we decided it was
worth testing the site because of its place name. Once the excavations began, we soon found
concentrations of burned animal bone and other domestic refuse from a settlement period
(landnm) farm, graves with an east-west orientation indicating the presence of a Christian
cemetery, and finally the foundations of buildings.
Thirteen of the twenty-three skeletal remains excavated at Hrsbr were suitable for analy-
sis, offering considerable evidence about the health status and living conditions of Icelands
early inhabitants (Walker et al. 2004). From the written sources we know that the economic
life of these people centered on a settled pastoral life of stock-raising, coastal fishing, and the
gathering of wild foods in a challenging marginal environment. The skeletons witness a rough
and violent kind of life, with infectious diseases and probable occurrence of tuberculosis.
Traumatic injuries appear to have been common. One person buried in the cemetery is an ap-
parent homicide victim with massive head injuries. Another has a healed leg fracture. In addi-
tion to traumatic injuries, skeletal lesions associated with heavy labor and infectious diseases
are also common in this tenth and eleventh century population.
Several individuals, including an adolescent, show evidence of strenuous physical activity
involving the hands and arms and osteoarthritis is prevalent. One young man from this ceme-
tery is of special interest owing to the presence of lesions associated with a chronic ear infec-
tion that resulted in a brain abscess. Another adolescent male has lesions on the pleural sur-
faces of his ribs. Although other diagnoses are possible, the lesions in both of these cases sug-
gest that tuberculosis was present in the Hrsbr population. Stature comparisons with the
early conversion period burials at Hrsbr and contemporaneous skeletal remains from Nor-
way provide additional data on the living conditions of these people. These data show that
stressful living conditions and heavy labor were common among early Icelanders even at such
a prominent site as Hrsbr.
Archaeology, history, and saga studies are sciences and studies for exploring the past, and
all have their methods and foci. This paper offers insight into archaeological methods and
presents some of the types of data from which saga scholars, historians, and anthropologists
can draw inferences from the archaeology. At its most obvious, we can now draw the
conclusion that the descriptions found in Egils Saga and Gunnlaugs Saga about the
farmstead of the Mosfellingar (Egils saga 1933, chapter 86; Gunnlaugs saga, p. 105) are
reflected in the archaeological finds. We now know much more about the material culture of a
site described in the sagas than was possible within the scope of the traditional analysis of the
written sources.

Figure 3. A man
in his mid-forties
found just east of
the church chan-
cel at Hrsbr.
He died of
wounds. Radio-
carbon dating
places the man in
the later half of
the tenth century
or the early part
of the eleventh.

170
It is hard to imagine it now, especially in light of the rich archaeological finds, but at the start
of our excavations, a many archaeologists, historians, and saga scholars thought it was futile
to consult the family sagas as sources for aiding in locating sites. We were told that everyone
already knew (or was supposed to know) that the slendingasgur were thirteenth-century
fictional literary creations. The question we asked was whether a careful researcher should or
should not use every tool and clue at hand in the process of discovery, especially in light of
the rather clear hint in Egils Saga (1933, Chapter 86) about when, why, and by whom, a
conversion-age church was built at Hrsbr (Gunnlaugs saga ormstungu, p. 105).2

Grmr at Mosfelli3 var skrr, er kristni var lg leidd slandi; hann lt ar kirkju gera. En
at er sgn manna, at rds hafi ltit flytja Egil til kirkju, ok er a til jartegna, at san er
kirkja var gr at Mosfelli, en ofan tekin at Hrsbr s kirkja, er Grmr hafi gera ltt [...].

When Christianity was adopted by law in Iceland, Grmr of Mosfell was baptized and built a
church there. People say that Thrds had Egils bones moved to the church, and this is the evi-
dence. When a church was built at Mosfell, the one Grmr had built at Hrsbr was taken down
[].

Figure 4. A Viking Age ring pin, Western Norse/Celtic style. Such pins were used by men to hold in
place their cloaks. This pin is of iron and the only such iron pin found so far in Iceland. It was found
in the soil lying above the head of the man pictured in Figure 3 above, whose skeleton is drawn (fea-
ture 2) in the 2003 site map below.

2
For the full passage in English, see Egils Saga 1975, pp. 170171.
3
Grmr was the lawspeaker of Iceland from 1002 to 1004. His wife Thrds was the stepdaughter of Egill
Skallagrmsson, who lived with Grmr and Thrds and was said to be buried in the Mosfell Valley. For the
posthumous travels of Egill, see Byock 1995:8287.

171
Figure 5. Beads, some with exotic designs. More than twenty beads were found within the longhouse.
The largest number of such finds within an Icelandic turf house. They offer some indication of the
wealth and high status of the inhabitants of the Hrsbr farmstead.

While we do not by any means believe everything found in the written materials, the sources
concerning Mosfell are often basic and detailed. We have in these writings a core of informa-
tion from a variety of sources about settlers, chieftains, warriors, women, lawgivers, slaves,
laborers, travelers, and merchants passing through Mosfellssveit. Much of this information
speaks to the material and social culture, describing habitation sites, lands, a ships port, buri-
als, social standing, kinship relations, economic arrangements, as well as determinations of
causes and places of conflict. The same can be said for many sagas, and the modern archaeo-
logical as well as anthropological, historical, and literary use of Icelands medieval texts re-
quires a methodology which recognizes both the oral and the written nature of these sources.4
The passages about the Mosfell region are a case in point. As a grouping of sources about a
regional chieftaincy or goor, the passages from different texts have been largely overlooked
by historians and anthropologists. Together the recent archaeological finds by MAP and the
ancient written materials offer a new combination of information about a 250-year period in
the past of an important region from the early 10th to the mid-12th century, a time which spans
the transition from prehistory to history, from paganism to Christianity.
Mosfellssveit encapsulates the major ecologies of Iceland: coastal, riverine, and highland.
Culturally, the region is equally representative. In some ways it was a self-contained social
and economic unit. In other ways, it was connected to the rest of Iceland, not least, through a
network of roads, including an east-west route to the nearby meeting of the yearly Althing.
With its coastal port at Leiruvogur, the region was in commercial and cultural contact with the
larger Scandinavian and European worlds, possibly as far east as Constantinople and perhaps
further to the west.

4
Concerning such a methodology see Byock 2001, pp. 2124 and pp. 149151. See also Byock 1982, which
explores the oral saga in light of narrative technique and the cultural and social backgrounds of a feuding culture.
Distinguishing social memory is also a central issue, Byock 2004, pp. 299316.

172
Figure 6. The conversion-age stave church in early stages of excavation. The church chancel at the
right is excavated. The foundation stones from the later period building overlaying the church nave
were removed in the 2004 excavations. The nave below was undisturbed by the upper agricultural
building. Beneath the church is a older turf building.

The research, in reconstructing the early social history of the Mosfell Valley region, integrates
information on the changing periods of occupation. We excavate individual sites, both secular
and religious, and consider their placement in relationship to one another. We examine the
apportionment of open spaces and the utilization of common lands in the highlands and on the
coast. Written, archaeological, and other scientific information are integrated into this study as
we construct a picture of early life.
The different specialists on the MAP team explore among other subjects the development
of roads and paths, the importance of the ships landing at Leiruvogur, the changes over time
in subsistence strategies, the state of health and disease in the Viking Age and later popula-
tion, developments in building techniques, and the usage of smaller activity areas, such as the
sel, or summer dairy stations. We are asking questions about the production of iron in the
early period5 and finding the locations of burials and early farm sites. In some instances our
task is to find the remains of turf buildings, roads, burials, agricultural enclosures, and port
facilities before they are destroyed by modern construction.

5
The iron artifacts in this late iron age society are numerous, see Zori 2007:3247.

173
Figure 7. Man-made stone settings shaped like ships. These are the first such monuments found in
Iceland.

Figure 8. Architectural renderings of the buildings at Hrsbr in the Mosfell Valley. The church is
approximately twelve meters distance from the longhouse (drawn by Grtar Marksson).

174
The Mosfell Archaeological Project is comprised of an international team and is conducted
under the direction of Prof. Jesse Byock of the University of California at Los Angeles
(UCLA). The field director is Davide Zori (UCLA). The international group works in Iceland
in cooperation with archaeologists from jminjasafn (Icelands National Museum) and with
members of the local Mosfellsbr community as well as with professors and students at the
University of Iceland and other Icelandic researchers. This article is dedicated to the memory
of Phillip Walker, my friend, colleague, and co-director of the Mosfell Archaeological Pro-
ject.

References
Byock, Jesse, 1982: Feud in the Iceland Saga, Berkeley and Los Angeles: University of California
Press.
, 1993: The Skull and Bones in Egils saga: A Viking, A Grave, and Pagets Disease. Viator: Me-
dieval and Renaissance Studies 24:2350. Translated into Icelandic as Hauskpan og beinin
Egils sgu Skrnir 1994 (Vor):73109.
, 1995: Egils Bones: A Viking Warrior and Pagets Disease. Scientific American 272/1 (Janu-
ary):8287. Translated as Die Egil-saga und das Paget-Syndrom, Spectrum der Wissenschaft
(Mrz, 1995); Les os dEgil, hros viking, Pour La Science 209 (Mars, 1995):5258; Le ossa di
Egill, Le Scienze 319 (Marzo, 1995):7479; Koci Egila wiat Nauki (Marzec 1995): 7277;
Archaeology: Annual Edition 96/97. Guilford: Brown and Benchmark Publishers, 1996. pp. 8085.
, 2001: Viking Age Iceland. London and New York: Penguin Books, 2001.
, 2004: Social Memory and the Sagas: The Case of Egils saga. Scandinavian Studies. 76/3: 299
316.
Byock et al. 2005 = Jesse Byock, Phillip Walker, Jon Erlandson, Per Holck, Davide Zori, Magns
Gumundsson, and Mark Tveskov: A Viking-age Valley in Iceland: The Mosfell Archaeological
Project. Medieval Archaeology 49:195218. Translated to Icelandic as: Jesse Byock, Phillip
Walker, Jon Erlandson, Per Holck, Davide Zori, Magns Gumundsson, and Mark Tveskov 2007:
Valdamist Mosfellsdal: Rannsknir fornleifum fr tmum Vkinga a Hrsbr og Mosfelli,
lafa, Rit fornleifafringaflags II, Reykjavk, pp. 84106.
Egils saga. Ed. Sigurur Nordal, slenzk fornrit 2. Reykjavk 1933.
Egils Saga, trans. by Christine Fell. London: Everymans Library, 1975.
Egils Saga. Trans. Hermann Plsson and Paul Edwards. London: Penguin Books, 1976.
Gunnlaugs saga ormstungu Borgfiringa sgur. slenzk fornrit 3. Ed. Sigurur Nordal and Guni
Jnsson. Reykjavk 1938.
Holck, Per, 2005: Egil Skallagrimssons grd og kirke p Island fra utgravningene 20012005.
Michael (Publication Series of The Norwegian Medical Society) 2:340348.
Walker et al. 2004 = Walker, P.L., J. Byock, J.T. Eng1, J.M Erlandson, P. Holck, K. Prizer, M.A.
Tveskov. Bioarchaeological evidence for the health status of an early Icelandic population, Paper
presented at the 73rd meeting of the American Association of Physical Anthropologists, Tampa,
Florida.
Zori, Davide, 2007: Nails, Rivets, and Clench Bolts: A Case for Typological Clarity. Archaeologia
Islandica 6:3247.

175
 An Icelandic Genesis
Betsie A.M. Cleworth, England
In this paper I contend that there is a subtle yet fundamental set of Biblical parallels in the
opening sections of Landnmabk as it appears in the Sturlubk and Hauksbk redactions.1
My particular focus is on the apparent comparability of Flki and Noah, I also compare
Inglfr with Abraham and orkell mni (Inglfrs grandson) with Moses.2 I will first explore
the biblical parallels in greater detail and then move on to their potential implications, func-
tions and the contemporary precedents for them. It is my belief that these parallels were delib-
erately developed to forward religious and socio-political ends and ideals: to suggest a parallel
between the early Icelandic settlers and the Old Testament patriarchs; between the Icelandic
people and the chosen people; and between Iceland and the Promised Land.
The parallel between Hrafna-Flki and Noah is well established in Norse scholarship, but
the origins and nature of that comparison are not as well known.3 According to Ldn Flki was
one of the first explorers from Norway to discover Iceland and reputedly gave it its name.4
While there had been rumours of an uninhabited land to the north no one knew exactly where
it was and so Flki took three ravens with him in order to help him find the way. It is the re-
lease of these birds, his utilisation of them to find land and the sequence of their flights that
creates certain parallels with Noahs sending out flights that creates certain parallels with
Noah sending out of a raven and dove in Gen 8. Admittedly the idea of birds being used in
navigation could have many sources, from Pliny to real life practices.5 In addition the general
use of ravens (and other birds) as messengers and omens is certainly well attested both in
Norse and continental sources.6 On the other hand the sequence of the ravens flights within

1
Hereafter Sturlubk = S, Hauksbk = H, Landnmabk = Ldn. S is thought to be composed by Sturla rar-
son c. 127580, the only surviving manuscript is a seventeenth-century copy of the earlier vellum AM 107 fol. H
survives primarily in AM 105 fol. although fourteen leaves of the text in the original codex (AM 371 4to) sur-
vive. It is thought to have been composed by Haukr Erlendsson c. 13068. These are considered to be the oldest
surviving redactions of the text, although there are differing opinions over whether the Melabk (AM 445b 4to)
fragments derive from an older original; see, Jn Jhannesson (1941:5467, 2216); lsen (1920:283300);
Jnas Kristjnsson (1956). For overview of all the Ldn manuscripts and redactions see Jakob Benediktsson
(2003); Jn Jhannesson (1941).
2
Finnur Jnsson (1900:59(H), 1304(S)). For normalized texts of both redactions see Jakob Benediktsson
(1968:3747).
3
An un-sourced footnote in Herman Plsson and Edwards (1972:17, f. 9) leads after a long paper-chase to Bern-
strm (1964:cl. 170), who describes a critical discussion that he states ultimately ended in acceptance of a prob-
able connection between Ldn ch. 5 and Gen 8. Bernstrm missatributes Fritzner, (188696:47) as the source of
this theory.
4
Jakob Benediktsson (1968: 3639(S,H)).
5
Pliny the Elder Naturalis Historia 6:24; Rackham (193863). For Plinys possible connection to Ldn ch. 5 see
Bernstrm (1964:cl. 170); Jakob Benediktsson (1968:36, f. 5). Hornell (1946) catalogues some interesting exam-
ples of navigation using birds, although this article should perhaps be treated with circumspection.
6
With regards ravens as messengers, assistants or omens in Norse literature the archetypal pair would seem to be
Huginn and Muninn, inns attendants. The direct discussion of there role is minimal and interdependent:
Grmnisml 20, Gylfaginning 38 and Ynglinga saga 7 Neckel (1983:61); Faulkes (1982:323); Bjarni
Aalbjarnarson (194151:189); but there is substantial evidence attesting an association of some kind in the
form of kennings, heiti and (admittedly ambiguous) iconography. For kennings and heiti linking inn and ra-
vens see, Eysteinn Bjrnsson, http://www3.hi.is/~eybjorn/ugm/kennings/allraven.html and
http://www3.hi.is/~eybjorn/ugm/kennings/kappa.html. Ravens do appear as both messengers and omens in
Norse literature outside the context of (direct) divine association, see Boberg (1966:434); and Fritzner
(1891:478). Earlier and contemporary Insular and continental literature do less frequently portary ravens in a
similar fashion. See Colgrave (1940:1003, 2225); Colgrave (1956:11621, 1247); Kalinke (2005: 113, 122
137, 175181). These sacred continental examples possibly stem from the biblical example of Job in I Kings
17:26 whom God sustained in the wilderness by sending ravens to bring him meat.

176
the Ldn passage are similar enough to those of Noahs raven and dove to suggest a deliberate
parallel may have been intended. If we compare Gen 8:613, with chapter 5 of Ldn we can
see these similarities more clearly:

Vulgate Ldn7
S: Flki hafi hrafna rj me sr haf.
H: aan sigldi hann t haf me hrafna rj, er
hann hafi bltat Nregi.

8.6 cumque transissent quadraginta dies aperiens ok er hann lt lausan enn fyrsta,
Noe fenestram arcae quam fecerat dimisit corvum

8.7 qui egrediebatur et revertebatur donec fl s aptr um stafn;

siccarentur aquae super terram
8.8 emisit quoque columbam post eum ut videret si annarr fl lopt upp
iam cessassent aquae super faciem terrae
8.9 quae cum non invenisset ubi requiesceret pes ok aptr til skips;
eius reversa est ad eum in arcam aqua enim erant
super universam terram extenditque manum et ad-
perhensam intulit in arcam

8.10 expectatis autem ultra septem diebus aliis

rursum dimisit columbam ex arca
8.11 at illa venit ad eum ad vesperam portans
ramum olivae virentibus foliis in ore suo intellexit
ergo Noe quod cessassent aquae super terram
8.12 expectavitque nihilominus septem alios dies et enn rii fl fram um stafn
emisit columbam quae non est reversa ultra ad eum
8.13 igitur sescentesimo primo anno primo mense tt, sem eir fundu landit.
prima die mensis inminutae sunt aquae super terram
et aperiens Noe tectum arcae aspexit viditque quod
exsiccata esset superficies terrae

7
Jakob Benediktsson, ed. (1968:36 (S), 37, 39 (H)). With the exception of the opening lines, which I have in-
cluded for context, the S and H redactions contain no substantive differences in this section. For ease of use the
normalised S text is given here. For the variations between the texts see Finnur Jnsson (1900:5(H), 130(S)).

177
Vulgate Ldn
S: Flki had three ravens with him at sea,
H: Then he sailed out to sea with the three ravens
which he had consecrated in Norway.
8.6 And after that forty days were passed, Noe, and when he set loose the first one
opening the window of the ark which he had made,
sent forth a raven
8.7 Which went back and forth until the waters were that flew back from the stern;
dried up upon the earth
8.8 He sent forth also a dove after him, to see if the the second flew up in the air
waters had now ceased upon the face of the earth
8.9 But she, not finding where her foot might rest, and back to the ship;
returned to him into the ark: for the waters were
upon the whole earth: and he put forth his hand, and
caught her, and brought her into the ark
8.10 And having waited yet seven other days, he
again sent forth the dove out of the ark
8.11 And she came to him in the evening, carrying a
bough of an olive tree, with green leaves, in her
mouth. Noe therefore understood that the waters
were ceased upon the earth
8.12 And he stayed yet other seven days: and he the third one flew from the prow
sent forth the dove, which returned not any more
unto him
8.13 Therefore in the six hundredth and first year, in that direction they found the land.
the first month, the first day of the month, the wa-
ters were lessened upon the earth, and Noe opening
the covering of the ark, looked, and saw that the
face of the earth was dried

As the table seeks to demonstrate, the flight of Flkis ravens can be credibly compared with
the flight of Noahs raven and the first and third flight of Noahs dove. In such a comparison
the raven that flies from the stern of Flki ship is like Noahs raven which flies away with
ambiguous hopes of return; and the raven that flies upwards and then returns to the boat is
equivalent to Noahs dove, its first unsuccessful flight and return. The final raven which flies
straight from the prow in the direction of land represents the doves final flight and its depar-
ture from the ark, which indicates the emergence of land.
The elliptical brevity of the Ldn passage, compared to the Biblical section, cuts the action
to the bare minimum to enable comparison, but this brevity may argue a universal referent
which the audience could use to decode these actions. The image of a bird returning with an
olive branch (Gen 8:1011) throws into sharp relief the differences in context and function
between the Genesis and Ldn passages and perhaps has not been paralleled by Ldn for this
reason, despite its iconic status: there are no olive trees in Iceland. Finally, the fact that all of
Flkis birds are ravens whereas all the unambiguously successful flights in Genesis are car-
ried out by a dove is a suggestive discrepancy that I will discuss below.
Apart from Noah and Flki, I propose two further parallels between the characters in Ldn
and the Old Testament: between Inglfr and Abraham and between orkell mni and Moses.
According to both Ldn and slendingabk, Inglfr Arnarson was the first permanent settler in
Iceland,8 Ldn states that he was frgastr allra landnmsmanna, v at hann kom hr at
byggu landi ok byggi fyrstr landit and furthermore that geru at arir landnmsmenn

8
On slendingabks portrayal of Inglfr, see Jakob Benediktsson (1968:5)

178
eptir hans dmum.9 It is arguable that his status as the exemplary settler and thus symbolic
founder of the Icelandic nation makes him a quasi-Abrahamic figure. As Abraham fathered
the twelve tribes of Israel so Inglfrs example gave birth to Iceland.10 Furthermore, Ldn de-
picts Inglfr as both particularly assiduous in performing his pagan sacrificial duties and as
prospering, apparently as a result of this assiduity.11 This is in direct contrast to his blood-
brother Hjrleifr whose ignominious death at the hands of his own slaves Inglfr attributes to
his unwillingness to sacrifice.12 Abraham was so assiduous in his sacrificial duty to God that
he was prepared to kill his own son. 13 Margaret Clunies Ross interprets Ldns description of
Inglfrs sacrificial practices as a ritual transference of luck from old land to new, but sug-
gests that, at least in the context of landnm, it does not matter whether the luck transferred is
pagan or Christian:

[] the advent of Christianity by no means extinguished the land rights and authority of the
Christian descendants of the first settlers. Rather the new religion preserved the authority vested
by the pagan deities in the practitioners and upholders of the old.14

What better way of preserving that authority than casting their pre-Christian forebears as
types of Old Testament patriarchs? If Inglfr was seen by twelfth- and thirteenth-century au-
diences as an Abrahamic figure, then his sacrificial practices can be conceived of as having
both the disassociated historical context and perhaps even the divine sanction of Abrahams
example. Not only were such notable men of the Old Testament known to be guaranteed re-
demption despite not being Christian,15 they had been further redeemed by exegetes who
made the Old Testament foreshadow the New Testament through typology.16 The comparison
between Inglfr and Abraham is more tenuous than that between Flki and Noah, as it is one
of type rather than episodic detail. While this makes the comparison difficult to substantiate it
does not entirely rule it out. Arguably, after being keyed-in to the existence of biblical paral-
lels within the text by the more obvious Flki comparison, the audience would be prepared to
read further, more subtle, parallels in the following sections.
Immediately after the description of Inglfrs ultimate success as the exemplary settler
there is an outline of his lineage through to the time of the redactions writing. Most of his
descendents are merely named but one his grandson, orkell mni is the subject of a more
sustained description. He is said to have been a lawspeaker and though a heathen, he was ex-
ceptionally pure in his person and behaviour.17 Furthermore we are informed that: lt sik bera
slargeisla banastt sinni ok fal sik hendi eim gui, er slina hafi skapat.18 Giving

9
[]the most famous of all the settlers, because he came here when the land was uninhabited and he was the
first man to settle. other settlers came and followed his example. Jakob Benediktsson (1968:46). This is the S
text; H uses auu (desolate) to replace the uncommon compound byggu (uninhabited). Jakob Benediktsson
(1968:47).
10
For Abrahams life see Gen 1125, his naming as father of nations Gen 17:5. For Pauls pivotal description of
Abraham as the father of all believers see Rom 4:1, 13 and Gal. 3:74:22. For a brief overview of Abraham, see
Metzger and Coogan, sv. Abraham.
11
Clunies Ross (1998:1619).
12
Jakob Benediktsson (1968:42, 44).
13
Gen 22.
14
Clunies Ross, (1998:25).
15
See Turville-Petre (1953:1268) on Icelandic versions of the harrowing of hell.
16
Typology was used by Christian exegetes to connect the Old Testament to the new and make it symbolically
foreshadow Christ. See Smalley (1982); Weber (1987).
17
Jakob Benediktsson (1968:46(S), 47(H)). For the life of Moses see Exod 6-Deut 35. For general discussion of
Moses see Metzger and Coogan, sv. Moses, and sv. Law.
18
Jakob Benediktsson (1968:46(S)). In his fatal illness he had himself borne into the suns rays and committed
himself to the hands of that god who had created the sun. There are minor but none substantive differences in

179
reverence to the one who made the sun, particularly in combination with orkells other
qualities, suggests he is a typical example of what Lnnroth terms the noble heathen. One of
the main purposes for the Christian writers to make the anachronistic insertion of such a char-
acter into settlement age narratives is to justify the past and to bring it into concordance with
the values of their own time.19 A Christian audience may have read in orkells death a fur-
ther redeeming narrative element beyond that offered by the noble heathen form, as it is
loosely comparable with the death of Moses as described in Deut. 3334. At the end of his
life Moses is informed by God that despite leading his people well he will not live to set foot
in the Promised Land. He therefore has himself taken to a mountain-top from which he can
look upon it in his last moments:

Ascendit ergo Moyses de campestribus Moab super montem Nebo, in verticem Phasga contra
Jericho : ostenditque ei Dominus omnem terram Galaad usque Dan []

Dixitque Dominus ad eum : Hc est terra, pro qua juravi Abraham, Isaac, et Jacob, dicens :
Semini tuo dabo eam. Vidisti eam oculis tuis, et non transibis ad illam.

Mortuusque est ibi Moyses servus Domini, in terra Moab, jubente Domino

et sepelivit eum in valle terr Moab contra Phogor : et non cognovit homo sepulchrum ejus
usque in prsentem diem.20 (Deut. 34:1, 46)

Both S and H immediately follow the narration of orkells death first with the information
that hafi hann ok lifat sv hreinliga sem eir kristnir menn, er bezt eru siair and then
procede to inform us that [s]on hans var ormr, er var allsherjargoi, er kristni kom
sland.21 This tripartite juxtaposition is full of subtextual pathos: orkell was such a noble
heathen that he was as good as the best of Christians; he died looking toward the Christian
truth but from the outside, without the full knowledge of God or the joy of a Christian Iceland
which his son would be able to enjoy. This is a very similar pathos to that evoked by the death
of Moses who can only look on the Promised Land, which his spiritual children, the Israelites,
will enter. This parallel is perhaps less tenuous than that between Abraham and Inglfr, as the
emotional details of the two scenes are arguably so similar, and yet it must be noted that there
is still a level of interpretation in this parallel that is not required in the Flki-Noah compari-
son. One of the only ways of testing this assumption is to consider the viability of its wider
implications. The perception of Iceland or perhaps more exactly, Christian Iceland, as the
Promised Land is one that may have been extremely appealing and politically rewarding to
twelfth- and thirteenth-century Icelandic audiences.22 If Ldn is a kind of narrative charter es-

the H text, Jakob Benediktsson (1968:47).

19
Lnnroth (1969:28). Lnnroth also suggests that the noble heathen is frequently pictured as a prophet augur-
ing the advent of a new and better faith, Lnnroth (1969:29).
20
Then Moses went up from the plains of Moab upon mount Nebo, to the top of Phasga over against Jericho:
and the Lord shewed him all the land of Galaad as far as Dan[].
And the Lord said to him: This is the land, for which I swore to Abraham, Isaac, and Jacob, saying: I will give it
to thy seed. Thou hast seen it with thy eyes, and shalt not pass over to it.
And Moses the servant of the Lord died there, in the land of Moab, by the commandment of the Lord:
And he buried him in the valley of the land of Moab over against Phogor: and no man hath known of his sepul-
chre until this present day.
21
Jacob Benediktsson (1968:46(S)), he had lived as well as the best behaved of Christian men his son
ormr was the supreme goi when Christianity came to Iceland. H is virtually identical at this point, Jacob
Benediktsson (1968:47).
22
The seemingly complex concept of a place-time is in fact arguably consistent with contemporary Norse per-
ceptions, see Gurevich (1969).

180
tablishing the rights of the twelfth- and thirteenth-century landowners, then paralleling it with
the Promised Land would turn that charter into a divine covenant. And if the intent of the
twelfth- and thirteenth- century authors was to apologetically incorporate their heathen ances-
tors within a Christian schema without completely erasing their cultural past, then suggesting
that they are parallel to Gods chosen people might fulfil this criteria: simultaneously incorpo-
rating them in an internationally recognised paradigm while marking them out as different.
If we now return to Flkis ravens; the use of three ravens instead of a raven and a dove on
a very simplistic level takes the biblical narrative and makes it Norse. Ravens have a negative
or ambiguous connotation in a biblical and exegetical context,23 but in Norse literature they
are a vital component in battle scenes;24 they can be used almost as a cultural emblem,25 and
they carry connections to the pagan past through their Oinic overtones and possible links to
pagan ritual.26 Implicitly the substitution of the dove for ravens may also be in dialogue with
the strong exegetical tradition concerning Noahs raven, utilising its God-denying connota-
tions to comment on and characterise Flki and other early pagans, without completely dis-
missing them. The ravens function and the purpose of his mission in the biblical text is am-
biguous, partly as inherent in the original Hebrew text and partly due to the misconceptions
created by the Old Latin translation.27 For this reason a variety of exegetes tried to explain it
using tropological typology, suggesting that the raven (as implied by the Old Latin transla-
tion) did not return to the ark but instead stayed in the flood-waters feeding off the flesh of the
drowned sinners.28 Within this typological schema, Noah was Jesus, the ark the church, the
dove the good Christian and the raven the apostate or sinner. Hrafna-Flki, while no apostate,
arguably mirrors the exegetes raven in his actions during his attempted settlement. Like the
raven, he is seduced by his greed, which causes his downfall, he spends all of his summer
gorging on the plentiful salmon, so that when winter comes and he has forgotten to make hay
all his animals die, and he is forced to abandon his settlement plans.29

Before I close this discussion I wish to briefly regard the precedent for such biblical compari-
son within the context of Norse literary culture and a wider medieval schema. It has been
widely discussed and generally accepted that historical writing is another kind of fiction, and
that histories from the twelfth to fourteenth centuries were the first intimations of the creation
of national/cultural origin myth.30 Howe (2001:179) has argued that Anglo-Saxon literature
from its very beginnings sought to create such a cultural myth using an intimate blend of bib-
lical parallels and the native (pagan) heritage to cast the Anglo-Saxon people as the chosen
people and to portray their cultural journey as one fated and divinely mandated.31 Rowe
(1998:89) argues that one of the scribes of Flateyjarbk depicts the conversion of western
23
For a medieval exposition of the multiple biblical representations of ravens see Rabanus (PL, 111.2523). For
the exegetical view of Noahs raven, see below.
24
See Jesch (2001:24354; 2002).
25
Lukeman (1958) catalogues all the incidences of the raven banner, in Norse and Insular literature, while the
article contains little analysis it is clear that in almost all incidences the banner denotes Norse warriors, and often
has strong pagan connotations.
26
For a discussion of the ravens possible shamanistic links see, Turville-Petre (1953:61); Lindow (2003:1023,
106); Price (2002:978).
27
See Gatch (1975:37); Murdoch (2003:1168); Moberly (2000).
28
This exegesis was known within a Norse context contemporary to the composition of the S and H redactions,
as Isidore of Sevilles Old Latin influenced interpretation (PL 83.233) as filtered through Comestor (PL
198.1085) appears in Stjorn I (AM 226 fol. and AM 227 fol.), Unger (1862:59). For dating and provenance of
Stjorn I and its use of Comestor see Asts, (1991:738, 1503) Seip (1956:24); Kirby (1986:5156, 723).
29
Jakob Benediktsson (1968:38, 39).
30
For texts concerning the fictional nature of Ldn see, Herman Plsson (1988); Adolf Fririksson and Orri
Vsteinsson (2003); Sveinbjrn Rafnsson (1974). For general texts see, Geary (2002); White (1987).
31
Howe (2001:179).

181
Scandinavia typologically as a re-enactment of world history in parvo. Another example per-
haps even closer to the one posited within this paper can be found in a group of Frisian
chronicles in which the Frisians comparing themselves to the chosen people, inverted the
order of historical events in their history in order to get a closer correspondence with the Old
Testament.32 Several Icelanders, including Haukr Erlendsson, did in fact claim a biblical
heritage, tracing their lineage back to Abraham, Noah and even as far as Adam.33 The apolo-
getic history within the Prologue to Snorra Edda and in Gylfaginning uses euhemerism to
include the pagan deities within a biblical time-frame, and further blurs the lines between pa-
gan myth and Biblical text by conflating comparable incidents such as the Biblical flood and
the mythic drowning of the giants in their own blood.34

Ultimately it is impossible to prove that any of the biblical parallels posited here exist within
Ldn, however there is precedent and the social, religious and political need for such compari-
sons did exist. Personally I find compelling the idea of the Ldn redactors either creating or
working within a subtle biblical schema, that incorporated and enhanced their Norse pagan
heritage, highlighted their unique history and both foreshadowed and celebrated their Chris-
tian present.

Abbreviations
PL Migne ed, 184455, Patrologia Latina.

Bibliography
Adolf Fririksson and Orri Vsteinsson, 2003, Creating a Past: A Historiography of the Settlement of
Iceland, in, Culture Contact, Continuity and Collapse: The archaeology of North Atlantic coloniza-
tion, A.D. 8001800, ed. J. Barrett and J. Tuck, Turnhout (Studies in the Early Middle Ages 5)
Pp.13961.
Asts, R., 1991, An Old Norse Biblical Compilation: Studies in Stjrn, American, New York, NY
(Univerity Studies 7, Theology and Religion 109.)
Bernstrm, J., 1964, Korp, in, Kulturhistoriskt lexikon fr nordisk medeltid frn vikingatid till refor-
mationstid, Malm (9), cl. 16773.
Bjarni Aalbjarnarson, ed., 194151, Heimskringla, Reykjavik (slenzk fornrit 2627.)
Bjrn M. lsen, 1920, Landnamas Oprindelige Disposition (Aarbger for nordisk Oldkyndighed og
Historie.)
Boberg, I. M., 1966, Motif-Index of Early Icelandic Literature, Copenhagen (Biblotheca Arna-
magnana 27.)
Clunies Ross, M., 1998, Land-Taking and Text Making in Medieval Iceland, in, Text and Territory,
Geographical Imagination in the European Middle Ages, ed. S. Tomasch and S. Giles, Philadel-
phia, PA, Pp. 930.
Colgrave, B., ed., 1940, Two Lives of Saint Cuthbert, A Life by an Anonymous Monk of Lindisfarne
and Bedes Prose Life, Cambridge.
Colgrave, B., ed., 1956, Felixs Life of Saint Guthlac, Cambridge.
Dronke, U. and Dronke, P.,1977, The Prologue of the Prose Edda: Explorations of a Latin Back-
ground, in Sjtu Ritgerir: Helgaar Jakobi Benediktssyni, ed. Einar G. Ptursson, Jnas
Kristjnsson, Reykjavik, Pp. 15376.
Eysteinn Bjrnsson, ed., Lexicon of Kennings and Similar Poetic Circumlocutions: The Domain of
Battle, http://www3.hi.is/~eybjorn/ugm/kennings/kennings.html.
Faulkes, A., 19789, Descent from Gods (Mediaeval Scandinavia 11), Pp. 92125.
Faulkes, A., ed., 1982, Edda, Prologue and Gylfaginning,Oxford.

32
Smalley (1982:xxvii)
33
Faulkes (19789:1056).
34
Faulkes (1982:36, 11); Dronke and Dronke (1977); Faulkes (1983).

182
Faulkes, A, ed., 1983, Pagan Sympathy: Attitudes to Heathendom in the Prologue to Snorra Edda,
Edda: A Collection of Essays, ed. R. Glendinning and Haraldur Bessason, Winnipeg, MB, Pp. 283
314.
Finnur Jnsson, ed., 1900, Landnmabk IIII, Hauksbk, Sturlubk, Melabk, Copenhagen (Det
kongelige nordiske oldskrift-selskab.)
Fritzner, J., 188696, Ordbog over det gamle norske Sprog 13, New York, NY; London.
Gatch, M. McC., 1975, Noahs Raven in Genesis A and the Illustrated Old English Hexateuch, (Gesta
14. 2), Pp. 315.
Geary, P., 2002, The Myth of Nations, The Medieval Origins of Europe, Princeton, NJ; Oxford.
Gurevich, A., 1969, Space and Time in the Weltmodell of the Old Scandinavian Peoples (Medieval
Scandinavia 2), Pp. 4253.
Herman Plsson and Edwards, P., trans., 1972, The Book of Settlements: Landnmabk, Winnipeg,
MB (University of Manitoba Icelandic Studies I.)
Hermann Plsson, 1988, A Foundation Myth in Landnmabk, (Mediaeval Scandinavia 12), Pp. 24
28.
Hornell, J., 1946, The Role of Birds in Early Navigation (Antiquity 20:79), Pp. 1429.
Howe, N., 2001, Migration and Mythmaking in Anglo-Saxon England, Notre Dame, IN.
Jakob Benediktsson, ed., 1968, slendingabk; Landnmabk, Reykjavik (slenzk fornrit 1.12.)
Jakob Benediktsson, 2003, Landnmabk, in, Medieval Scandinavia: An Encyclopaedia, ed. P. Pul-
siano, New York, NY, Pp. 3734.
Jesch, J., 2001, Ships and Men in the Late Viking Age: The Vocabulary of Runic Inscriptions and
Skaldic Verse, Woodbridge.
Jesch, J., 2002, Eagles Ravens and Wolves: Beasts of Battle, Symbols of Victory and Death, in, The
Scandinavians from the Vendel Period to the Tenth Century, ed. J. Jesch, Woodbridge, Pp. 25171.
Jn Jhannesson, 1941, Gerir Landnmabkar, Reykjavik.
Jnas Kristjnsson, ed., 1956, Eyfiringa sgur, Reykjavik (slenzk fornrit 9)
Kalinke, M. E., 2005, St. Oswald of Northumbria : Continental Metamorphoses: with an edition and
translation of svalds saga and Van sunte Oswaldo deme konninghe, Tempe, AZ (Medieval
&Renaissance Texts & Studies ; v. 297.)
Kirby, I. J., 1986, Bible Translation in Old Norse, Geneva (Universit de Lusanne Publications de la
Facult des Lettres 27.)
Lindow, J., 2003, Cultures in Contact, in Old Norse Myths Literature and Society, ed. M. Clunies Ross
(The Viking Collection 14), Pp. 89109.
Lnnroth, L., 1969, The Noble Heathen: A Theme in the Sagas (Scandinavian Studies 41) Pp. 129.
Lukeman, N., 1958, The Raven Banner and the Changing Ravens: a Viking Miracle From Carolingian
Court Poetry to Saga and Arthurian Romance (Classica et Mediaevalia 19), Pp.13151.
Metzger, B. M. and Coogan, M. D., ed., 1993, The Oxford Companion to the Bible, New York, NY;
Oxford.
Moberly, R. W. L., 2000 , Why Did Noah Send out a Raven?, (Vetus Testamentum 50), Pp. 345356.
Murdoch, B., 2003, The Medieval Popular Bible, Expansions of Genesis in the Middle Ages, Cam-
bridge.
Neckel, G., ed., 1983, Edda: die Lieder des Codex Regius nebst verwandten Denkmlern, 5th ed., H.
Kuhn, Heidelberg.
Price, N., 2002, The Viking Way, Religion and War in Late Iron Age Scandinavia, Uppsala (Aun 31.)
Rackham, H., ed., trans., 193863, Pliny, Natural History, With an English Translation, London.
Rowe, E. A., 1998, Cultural Paternity in Flateyjarbk lfs saga Tryggvasonar, Berlin (Alvssmal 8),
Pp. 328.
Seip, D. A., ed., 1956, Stjrn, AM. 227 fol: A Norwegian Version of the Old Testament Transcribed in
Iceland. Copenhagen (Corpus Codicum Islandicorum Medii Aevi 20.)
Smalley, B., 1982, 3rd ed., The Study of the Bible in the Middle Ages, Notre Dame, IN.
Sveinbjrn Rafnsson, 1974, Studier i Landnmabk: Kritiska bidrag till den islndska fristatsidens
historia, Lund (Bibliotheca Historica Lundensis 31.)
Turville-Petre, G., 1953, Origins of Icelandic Literature, Oxford.
Unger, C. R., ed., 1862, Stjorn : Gammelnorsk bibelhistorie fra verdens skabelse til det babyloniske
fangenskab, Christiania.
Weber, G. W., 1987, Intellegere historiam: Typological Perspectives of Nordic Prehistory (in Snorri,
Saxo, Widukind, and Others), in, Tradition og historieskrivning: Kilderne til Nordens ldste

183
 historie, ed. K. Hastrup and P. M. Srensen, Aarhus (Acta Jutlandica 63.2. Humanistisk serie 61),
Pp. 95141.
White, H., 1987, The Content of the Form, Narrative Discourse and Historical Representation, Balti-
more, MD; London.

184
 Poets and Ethnicity
Margaret Clunies Ross, University of Sydney, Australia
The Uppsala University Library manuscript De la Gardie 11 (U) of c. 1300 contains the earli-
est text of Snorri Sturlusons Edda and several other works, among them Skldatal List of
Skalds on fols 22r24r. A printed text of the U version is in SnE 184887 III, 25969. An-
other version of Skldatal is to be found in early modern paper copies of the Heimskringla
manuscript Kringla (K), which was largely burnt in the Copenhagen fire of 1728. rni
Magnssons transcript of K in AM 761 a 4x, a paper anthology of early skaldic poetry of c.
1700, was chosen by the editors of the 184887 edition of SnE (III, 2519) as the basis of
their printed text. The relationship between the two versions, which differ in several signifi-
cant respects, has been studied closely by Gurn Nordal (2001, 12030). She has argued that
the Kringla version was intended as a source list for the writing of kings sagas (2001, 129),
which explains its concentration on poets who composed at the courts of kings and earls in
Scandinavia, while the version in the Codex Upsaliensis is not so closely tied to a Scandina-
vian historical project and expands its scope to include two kings of England, Athelstan and
Ethelred, and their Icelandic poets, Egill Skallagrmsson and Gunnlaugr ormstunga. The U
version also admits important Norwegians chieftains to the list of patrons, while K excludes
any patron who was not a king or jarl. This argument implies that the original function of
Skldatal is more closely conveyed by the K version than the version in the Uppsala manu-
script.
Skldatal is a unique work. It is unique both in its subject-matter and (at least in the Upp-
sala version) its textual arrangement on the manuscript pages. It is a chronologically arranged
list of Scandinavian skalds beginning with the prehistoric Starkar inn gamli the Old
Strlfsson and concluding with poets of the second half of the thirteenth century. Alongside
the names of the poets, which are arranged in three columns from top to bottom of the page in
U, are placed the names of the patrons whom they served. The patrons names are juxtaposed
with their poets names by being written sideways beside them.
Skldatal is anonymous, but it is likely to have been created by someone in Iceland whose
knowledge of Old Norse skaldic poetry was as extensive as that of Snorri Sturluson (perhaps
in its original version it was compiled by Snorri himself if it was produced as an aide-
mmoire for Heimskringla) and who felt moved to compile an economical but expressive re-
cord of the roll-call of skalds and their patrons from prehistoric times to the late thirteenth
century. Its aim appears to have been to capture and record available information about the
aristocratic and royal encomiastic tradition in Norway and other parts of mainland Scandina-
via, from the ninth to the thirteenth centuries: it is a Whos Who of poets and their patrons
from this milieu. However, it excludes poets who did not serve noble or royal patrons, just as,
with a few exceptions, it largely excludes poets who operated outside Norway, in the Orkneys
or the British Isles, for example.1 Thus the majority of poets whose compositions appear only
in sagas of Icelanders or in other saga sub-genres that relate chiefly to Icelandic history do not
appear in Skldatal, nor do those skalds who composed poetry on Christian subjects, unless
those subjects happened to relate to Scandinavian kings, like Einarr Sklasons Geisli Light
beam, in honour of King lfr Haraldsson or Marks Skeggjasons Eirksdrpa, in honour of
the Danish king Eirkr inn gi Sveinsson. Finally, by its very nature, Skldatal excludes all

1
For example, Jarl Rgnvaldr Kali Kolsson of the Orkneys is not mentioned in Skldatal, as he did not compose
for a Scandinavian patron. The skald orkell Skallason, who composed a flokkr in honour of Earl Waltheof of
Northumbria, does not appear on the list either.

185
anonymous skaldic poetry, whether or not it was composed in honour of royal or noble pa-
trons.
It may seem curious, in view of Skldatals focus on the relationship between named poets
and their patrons, that the list makes no mention of the ethnicity of the poets who served the
kings and jarls of Norway and other Scandinavian societies, unless that information is con-
veyed indirectly by a nickname or a patronymic. The ethnicity and rank of the patrons is re-
corded or implied, but not that of the skalds. Indeed, the patrons ethnicity and rank are the
factors that determine their places in the various divisions of the list in both versions of
Skldatal, and the poets are grouped accordingly. As the focus of Skldatal is firmly upon the
nexus between skald and patron in a courtly environment, the question of where the skalds
came from, and where they were brought up, may have been considered of secondary impor-
tance in the context in which the list was compiled. However, it could hardly be said that the
matter of a skalds ethnicity was a subject of little interest in Old Norse literature more gener-
ally. On the contrary, several different kinds of prose text, most of them of Icelandic prove-
nance, indicate that a poets ethnic origin and often his family connections were of great im-
portance to the texts authors and therefore presumably to their audiences.
Information about the ethnicity of skalds comes from the following types of sources in the
main: from sagas of the kings of Norway, where the skalds works are quoted as evidence for
events in the lives of Norwegian kings; from ttir often associated with kings sagas, where
the exploits of Icelandic skalds at the Norwegian court are often the ttrs main subject; from
some historical sources such as Landnmabk and Orkneyinga saga; and from sagas of Ice-
landers, especially from the sub-group whose protagonists are skalds, the so-called
skldsgur. As the majority of these works were written by Icelanders, it is not surprising that
these sources show a particular interest in Icelandic skalds and their success abroad, both po-
etic and otherwise.
That same interest in skalds ethnicity is to be found in many modern editions of skaldic
poetry, studies of skalds or skaldic verse and general literary histories. In fact, it is usually
accepted as more or less axiomatic that, after the very early period (the ninth and early tenth
centuries), before the settlement of Iceland or during its early days, when skalds were Norwe-
gian, the art of skaldic poetry was the subject of a successful take-over by Icelanders, who
came to dominate the profession. A typical assessment is that of Whaley (2005, 479): Skaldic
poetry was composed throughout the Scandinavian-speaking world, and by poets of diverse
origin, but mainly Norwegian in the first phase, then Norwegian or Icelandic in the tenth cen-
tury. After c. 1000, most skalds seem to have come from Iceland (especially the west or
northwest) or Orkney, though some Norwegian kings are credited with poetry. The main
purpose of this paper is to test the validity of our common assumption about the changing
ethnicity of skalds from Norwegian to Icelandic, not in order to deny its basic rightness, but to
question whether there are some aspects of the general picture that may need modification in
the light of recent research.
Now that the new edition of the skaldic corpus, Skaldic Poetry of the Scandinavian Middle
Ages (SkP), is firmly underway and has already published two volumes, with more in active
preparation, we are in a very good position to begin to test some of our older assumptions
about a range of skaldic subjects, including the question of the skalds ethnicity. I have util-
ised information in the new editions and on our database in order to approach this question
here. My starting point has been the information about skaldic ethnicity that Finnur Jnsson
included at the head of each named skalds entry in his Den norsk-islandske skjaldedigtning
(Skj), information repeated by E. A. Kock in his Den norsk-islndska skaldediktningen. I have
compared it with the information our editors have been gathering about known details of each
named skalds biography for the skald biography section of each poets oeuvre.

186
 I have confined my analysis to Volumes I to III of the skaldic edition, that is, Poetry from
the Kings Sagas 1: From Mythical Times to c. 1035 (Volume I, forthcoming, edited by Diana
Whaley), Poetry from the Kings Sagas 2: From c. 1035 to c. 1300 (Volume II, 2009, edited
by Kari Ellen Gade) and Poetry from Treatises on Poetics (Volume III, forthcoming, edited
by Edith Marold). Of these three volumes, only Volume II is yet published, but material to-
wards Volumes I and III is available on the skaldic project website
http://skaldic.arts.usyd.edu.au. I have excluded the rest of the skaldic corpus from analysis,
because it is either clearly of Icelandic provenance or it is anonymous, or both. There is of
course some anonymous poetry in Volumes IIII, which cannot be included in the study for
obvious reasons. Leaving aside the anonymous poetry in Volumes IIII, these three volumes
constitute a substantial part of the skaldic corpus by named poets to have survived from the
Scandinavian Middle Ages.
It was Finnur Jnssons practice, when compiling his entries for named poets in Skj A and
B, to place below the skalds name a line about his ethnicity and his floruit, if known, as well
as information about his personal status. For example, of jlfr from Hvinir he wrote
jlfr r Hvini, enn hvinverski, Norsk skjald, 9. rh., while of jlfr rnrsson he
wrote Islandsk skald, d. 1066. He differentiated poets not only on grounds of ethnicity, but
also according to whether he considered them to be professional poets or not. The former are
designated poet, the latter Icelander or Norwegian and so forth, indicating their amateur
status. He gave further information about poets who were also kings, jarls, district chieftains
or lawmen. In some cases, Finnur indicated uncertainty about the information he provided,
usually by means of a question mark, but most often his statements about the poets identities
were unencumbered by doubt, unlike the more judicious commentary in the nineteenth-
century Arnamagnan edition of Skldatal (SnE 184887 III, 287752), which frequently
admits to a lack of knowledge about certain skalds.
I am not the first to suspect that Finnur was often too ready to assign Icelandic ethnicity to
skalds about whom little is known from medieval sources. In a survey of poetry and its impor-
tance in medieval Icelandic culture published in 2000, Kari Ellen Gade questioned whether
[the dominance of Icelandic skalds after the tenth century] is an accurate depiction of the con-
ditions in eleventh- to thirteenth-century Scandinavia, or whether skaldic poetry had become
one of the vehicles by which late twelfth- and early thirteenth-century Icelanders sought to
assert their national uniqueness (2000, 76). Her doubts can now be seen to have been well
founded, as recent research (some of it her own) into the life histories of the skalds who ap-
pear in Volumes IIII of SkP establishes a more accurate picture of what we are entitled to
deduce from available evidence about their ethnicity.

The data
Volume I of the skaldic edition covers court poetry from the known beginnings of the skaldic
art up to about 1035. It spans a period of about 150 years and includes a total of 54 named
poets, as well as some anonymous verse. Of that total of 54 named poets, 22 or just over 40%,
are Norwegian, and there is no doubt about the ethnicity of these individuals. They include
three women of high social status as well as nineteen men, four of them kings. The remainder
are Icelanders (18 or 33%), Orkadians or, in one case each, the Faroese chieftain rndr
Gtu, the Danish king Sveinn tjguskegg and the Jmsvkingr Vagn kason. There are 5 in-
dividuals (9%) whose ethnicity is not clearly defined in the prose sources that record their
poetry, and these are listed in Table 1 below. In all 5 cases Finnur Jnsson classified the poets
as Icelandic.

187
Table 1. Unknown skalds from Skp I. (* denotes mention in Skldatal)
Skald name Comment Date Finnur Jnsson (Skj) Further comment
*Eyjlfr daskld Ethnicity unknown c. 1000 Icelandic poet
*Gizurr svarti gull-Ethnicity dubious d. 1030? Icelandic poet Foster-father of
brrskld Hofgara-Refr?
*Halldrr kristni Ethnicity unknown Early C11th Icelandic poet
Hallar-Steinn Probably Icelandic C12th Icelandic poet See GS 2004, 110
*rr Sreksson Ethnicity unknown C11th Icelandic poet See Biography on
skaldic database
Total 5
Volume II, Volume II, which will be published this year (2009), contains the work of 57
named skalds as well as some anonymous poetry. It continues chronologically from where
Volume I leaves off, at 1035, and continues down to the demise of skaldic court poetry in the
second half of the thirteenth century. The statistics here are interestingly different from those
of Volume I and in part bear out the prevailing hypothesis of increasing Icelandicisation of the
skaldic profession after the early eleventh century. But that is not the whole story, as we shall
see. From the total of 57 named skalds in Volume II, 11 or just over 19% are definitely Nor-
wegian, and five of these, or just under half, are Norwegian kings, while the majority of the
remainder of six are either Norwegian district chieftains or close friends of royalty or the up-
per nobility. Two men are foster-fathers of Norwegian kings. None of this Norwegian group
in Volume II is a professional poet, and in several cases only a single stanza is attributed to
the poet in question. This statistic suggests that after about 1035 in Norway skaldic verse was
considered a largely royal or aristocratic accomplishment for Norwegians, not a professional
skill, and this idea is expressed in the well-known stanza (Hhar Gamv 4II) of Haraldr
harri Sigurarson on his eight accomplishments, which include poetic composition, re-
peated almost verbatim by the Orkney jarl Rgnvaldr Kali Kolsson (Rv Lv 1/58II).

Table 2. Unknown skalds from SkP II. (* denotes mention in Skldatal)

Skald name Comment Date Finnur Jnsson (Skj) Further comment
Bjarni Klfsson Ethnicity unknown C12th Icelandic (?) poet See skald biography
SkP II
*Bjrn krepphendi Ethnicity unknown Early C12th Icelandic poet
*Blakkr Ethnicity unknown C12th Icelandic poet
*Bvarr balti Ethnicity unknown C12th Icelandic poet
*Blverkr Arnrsson Possibly Icelandic C11th Icelandic poet brother of jlfr?
*Grani skld Ethnicity unknown C11th Icelandic poet
*Halldrr skvaldri Ethnicity unknown C12th Icelandic poet See biography SkP II
*Halli stiri Ethnicity unknown C11th Icelandic poet Post-medieval as-
cription
*Hallr Snorrason Ethnicity unknown C12th Icelandic poet
*Illugi bryndlaskld Possibly Icelandic C11th Icelandic poet Suggested by patro-
nymic
*Kolli inn pri Ethnicity unknown C12th Icelandic poet
*Oddr kkinaskld Ethnicity unknown C11th Icelandic poet
*Valgarr Velli Possibly Icelandic C11th Icelandic poet From Vllur, South-
ern Iceland?
*rarinn Skeggjason Possibly Icelandic C11th Icelandic poet ? relative of Marks
*orbjrn skakkaskld Ethnicity unknown C12th Icelandic poet
orbjrn svarti Possibly Icelandic C12th Orcadian (?) poet Orkneyinga saga
only
*orkell hamarskld Possibly Icelandic Early C12th Icelandic poet Could equally be
Norwegian
orkell Skallason Ethnicity unknown C11th Icelandic poet Retainer of Waltheof
of Northumbria

188
*orleikr fagri Possibly Icelandic C11th Icelandic poet Suggested by nick-
name?
Total 19
Compared with the Norwegian group in Volume II, 22 poets or just over 30% are indisputably
Icelandic, and their ethnicity is confirmed by prose sources, usually kings sagas or ttir.
Most of these poets are professionals or semi-professionals in the service of Norwegian kings.
An exception is the group of poets attested from Orkneyinga saga, most of whom are only
known from this saga. In two cases (rmr and Oddi inn litli Glmsson), men that the other
manuscripts say are Icelandic are said in Flateyjarbk to be from the Shetland Islands.
The most interesting statistic in the case of Volume II (see Table 2 above) is provided by
instances in which Finnur Jnsson stated a poets ethnicity to be Icelandic, but our editors
have not been able to find convincing evidence of where he came from. There are almost as
many such cases as there are clear-cut examples of poets who are definitely Icelandic and this
high number (19 out of 57 or 33%) must raise questions about whether the Icelandic domi-
nance of the skaldic profession from the mid-eleventh to the late thirteenth centuries was
really as great as has been assumed up to now. There is also the related question of why the
biographical details of such a high number of skalds of this period are virtually unknown, in
contrast to those of the earlier period.
The situation with the poets named in treatises on poetics (Volume III) is closer in some
respects to that of Volume II than it is to Volume I, yet there are also some significant differ-
ences between this group and those of both the kings sagas volumes. For the most part Vol-
ume III comprises poetry in Snorri Sturlusons Edda, in his nephew lfr rarsons Third
Grammatical Treatise (TGT, c. 1250) and in the mid-fourteenth century Fourth Grammatical
Treatise, as well as additional material recorded in the Codex Wormianus (W) and the Laufs
Edda (LaufE) of Magns lafsson (c. 1610). As with the other volumes, the statistics exclude
anonymous poetry. Seventy-three named poets appear on the skaldic database for Volume III
(see Table 3 below). Of those 73, only three are definitely Norwegian (lvir hnfa, Bragi
Boddason and jlfr of Hvinir, all from the earliest period), a much smaller percentage
(4%) than for either of the kings sagas volumes. There is one definite Orcadian, Rgnvaldr
Kali Kolsson, and one supposed Dane, Starkar. Thirty-one poets or 42.5% are definitely Ice-
landic, while 37 or 51% are either unknown or of dubious ethnicity, the majority of them only
being cited in one or other of the poetic treatises, where for the most part the only information
supplied about them is their names. Even their floruit, in many cases, is debatable and is often
dependent on an editors recognition of the likely period during which the subject-matter they
treat would have been topical. It is also noteworthy that many fewer of the poets in Volume
III appear in Skldatal, indicating that this group is less associated with a court environment
and aristocratic or royal patrons. Probably, if we knew more, many of this unknown cohort
would turn out to be Icelandic, but that would be an educated guess.
The evidence presented here about the skalds whose work illustrates the major Icelandic
treatises on poetics is revealing in the light of Snorri Sturlusons claim to wish to familiarise
young poets with the traditional terms and ancient kennings of the kind of skaldic verse that
the chief poets (hfuskld) were happy to use (SnE 1998 I, 5, ll. 2530). Aside from three
Norwegians, those chief poets are probably almost exclusively Icelandic, and in just over half
the sample from Snorris Edda and the Third and Fourth Grammatical Treatises, are poets
who are virtually unknown and, in some cases, whose works are not attested anywhere else in
the Old Norse textual corpus. To the extent that the poetic treatises are normative and pre-
scriptive,2 they reflect the practices of Icelandic skalds more than any other group. This gen-
2
I do not think that any of them is entirely normative or prescriptive: Snorris agenda, in my opinion (Clunies
Ross 1987), was influenced by the views he expressed in the Preface to the Edda, while both lfr and the au-
thor of the Fourth Grammatical Treatise were keen to demonstrate how skaldic poetry could be explained in

189
eral observation to some extent echoes Gsli Sigurssons finding (2004, 93119), in his study
of lfr rarsons sources in the Third Grammatical Treatise, that a considerable number
of them were local to the north and west of Iceland. If, in addition, we consider that much of
the anonymous verse in both the Third and Fourth Grammatical Treatises is likely to have
been the work of lfr and the unknown author of the latter, then the poetic manuals are even
more firmly grounded in Icelandic poetic practice.

Table 3. Unknown skalds from Skp III. (* denotes mention in Skldatal)

Skald name Comment Date Finnur Jnsson (Skj) Further comment
sgrmr Jnsson Probably Icelandic C13th Icelander In SnE W and LaufE
*Atli ltli Ethnicity unknown C11th Icelandic poet
Bjarni []ason Ethnicity unknown C12th? Icelandic poet
*Eilfr Gornarson Ethnicity unknown c. 1000 Icelandic poet Norw. or Icel.
Eilfr klnasveinn Ethnicity unknown C12th? Icelandic poet
Erringar-Steinn Ethnicity unknown C11th? Icelandic poet
*Eyjlfr daskld Ethnicity unknown c. 1000 Icelandic poet
Eysteinn Valdason Ethnicity unknown C10th? Icelander
Gamli Gnvaarskld Ethnicity unknown C10th? Icelandic poet in C10th
*Gizurr svarti gull-Ethnicity dubious d. 1030? Icelandic poet Foster-father of
brrskld Hofgara-Refr?
Grani skld Ethnicity unknown C11th Icelandic poet
Gulaugr Probably Icelandic C12th Icelandic poet See GS 2004, 109
Hallar-Steinn Probably Icelandic C12th Icelandic poet See GS 2004, 110
*Halldrr skvaldri Ethnicity unknown C12th Icelandic poet See biography SkP II
Hallgrmr Probably Icelandic ? Not in Skj In LaufE
Hallr Snorrason Ethnicity unknown C12th Icelandic poet
*Hallvarr hreksblesi Ethnicity unknown C11th Icelandic poet
*Illugi bryndlaskld Possibly Icelandic C11th Icelandic poet Suggested by patro-
nymic
Leilfr skld Probably Icelandic C11th? Icelander? In TGT, see GS
2004, 109
Ormr barreyja(r)skld Ethnicity unknown C10th or 11th Orcadian poet? Or from Barra, in the
Hebrides
Ormr Steinrsson Ethnicity unknown C11th? Icelandic poet
Skldhelgi rarson Ethnicity unknown C11th? Icelandic poet
Skldrir Ethnicity unknown C12th Icelandic poet
Skraut-Oddr Probably Icelandic C11th? Icelander? In TGT only
Skli orsteinsson Ethnicity unknown C11th Icelandic poet
Snbjrn Ethnicity unknown C11th? Icelander
Steinarr Ethnicity unknown C11th? Icelander?
Steinrr Ethnicity unknown C11th? Icelander?
Styrkrr Oddason Ethnicity unknown C11th? Icelander?
Sveinn Ethnicity unknown C11th? Icelander? Or Greenlander?
rlfr (-valdr) Ethnicity unknown C11th? Icelander?
orbjrn dsarskld Ethnicity unknown C10th Icelander
rr mauraskld Ethnicity unknown C11th? Icelander?
*rr Sreksson Ethnicity unknown C11th Icelandic poet See Biography on
skaldic database
*orkell hamarskld Possibly Icelandic Early C12th Icelandic poet Could equally be
Norwegian
*orleikr fagri Possibly Icelandic C11th Icelandic poet Suggested by nick-
name
*orvaldr blnduskld Ethnicity unknown C12th Icelandic poet
Total 37

terms of Latin rhetoric.

190
 Analysis of the data
The poetic treatises, like Skldatal, do not normally mention a skalds ethnicity and in this
respect they and Skldatal contrast with those kinds of texts that do present biographical in-
formation, like kings sagas in the major compilations, some historical works like Landnm-
bk and Kristni saga, ttir and sagas of Icelanders. For the most part, it is where we are able
to access information about skalds in the non-pedagogical texts that we can determine some-
thing of a poets biography and where he came from. The reason why we have so much fuller
information about the skalds who lived between the beginning of the historical period and c.
1035 is that most of these people are at least mentioned, if not more fully represented, in sagas
and historical works, particularly Landnmabk. The few unknowns in this group (Table 1)
are unknown because no narratives of a personal kind, however small, attach to the mention
of their names and compositions. By contrast, the period between 1035 and c. 1300 is covered
mainly in the historical compilations and ttir and in Skldatal, as it falls outside the
chronological scope of the slendingasgur and somewhat before the contemporary sagas,
and, as we have already seen, the poetical treatises do not mention biographical data about the
poets at all.
To judge by the numbers of poets who composed for the Norwegian kings of the eleventh
and twelfth centuries but are otherwise unknown, as Table 2 shows, there seems to have been
little literary impulse to create even embryonic biographical narratives about them. This may
be partly because literary convention favoured narratives, especially about Icelanders, from
either the earlier or the later period, or it may be because, once these poets became part of the
royal entourage, they were simply swallowed up in the system, as it were, and there was little
to report about them, particularly if they were of low social status, which we know some
skalds were, though that in itself did not always debar them from a life in literature, witness
Sneglu-Halli, who came from a poor family in the north of Iceland.3
There are two other probable factors that must have determined whether skalds composi-
tions and biographical information about them was recorded in writing, and that is whether or
not their royal patrons and the compilers of their biographies considered it important to use
skaldic encomia to support their narratives and give the poets credit for their courtly roles. It
is well known that the major historical compilations vary considerably in the amount of
skaldic poetry they use and in their choices of verse illustration (cf. SkP II, Introduction, lviii-
lxxx). Equally, the kings themselves probably played a part in determining whether the com-
positions and life histories of their skalds were recorded in writing. According to Skldatal,
King Sverrir had either ten (U) or thirteen (K) skalds composing for him, yet little of their
work has survived, they are mostly mere ciphers, and the poetry cited in Sverris saga is not
quoted to corroborate historical events but to ornament the prose, especially within the
speeches attributed to Sverrir himself. To judge by the names of many court poets from the
later twelfth and early thirteenth centuries listed in Skldatal, their compositions did not form
part of any historical narrative and so are lost to us. This applies to poetry composed for
Swedish and Danish rulers and dignitaries, as well as for Norwegian patrons.
Gurn Nordal (2001, 11795; cf. Gade 2000, 856) has demonstrated that an Icelandic
backlash against the apparent loss of status of skalds as authorities and their poems as histori-
cal witnesses in Norway seems to have led to a resurgence of poetic activity of a courtly kind
in Iceland in the early thirteenth century, as members of the major Icelandic families began to
surround themselves with entourages of professional poets. The poets and literary entrepre-
neurs of the Sturlung family (Snorri Sturluson, lfr and Sturla rarson) also recognised
this loss of status and tackled it both by assuming the role of professional poet themselves and
3
The names of some of these unknown poets, like Blakkr and Bvarr balti, lacking patronymics, suggest that
their family connections were unimportant and hence are likely to denote low social status.

191
by producing pedagogical works analysing skaldic poetry (Snorri, lfr) or historical works
in which their own poetry took the place of that of the standard court poets of past times
(Sturla, lfr).4 In this way, they probably sought to regain the status that skalds and skaldic
poetry appear to have lost, at least to some extent, during the twelfth century and the begin-
ning of the thirteenth, probably in part challenged by the newly translated courtly romances,
religious and historical literature from abroad that King Hkon Hkonarson encouraged (cf.
most recently Wanner 2008, 805).
This survey has upheld the standard view of an Icelandic dominance of the composition of
skaldic poetry in Scandinavia after the tenth century, but it has also cast considerable doubt on
its extent or at least upon the certainty of any modern estimation of its extent. A great deal of
information about the identities of many skalds mentioned either in Skldatal or in other
sources has been lost, even when some of their poetry has been preserved, though a great deal
of that has probably been lost too. Additionally, the medieval sources themselves in which
skaldic poetry has been preserved were mostly written by Icelanders, so we cannot forget that
the overall perspective from which we view the skaldic corpus is Icelandic. If we had been
able to see it from a different ethnic perspective, the view may have been different.

Bibliography
Clunies Ross, Margaret. 1987. Skldskaparml. Snorri Sturlusons ars poetic and medieval theories of
language. Odense. (The Viking Collection 4.)
Gade, Kari Ellen, 2000: Poetry and its changing importance in medieval Icelandic culture. In: Old
Icelandic Literature and Society. Ed. by M. Clunies Ross. Cambridge. (Cambridge Studies in Me-
dieval Literature 42) Pp. 6195.
GS = Gsli Sigursson, 2004: The Medieval Icelandic Saga and Oral Tradition. A Discourse on
Method. Cambridge, Mass. and London, England. (Publications of the Milman Parry Collection of
Oral Literature No. 2.)
Gurn Nordal, 2001: Tools of Literacy. The Role of Skaldic Verse in Icelandic Textual Culture of the
Twelfth and Thirteenth Centuries. Toronto, Buffalo and London.
Kock, E. A., ed., 194650: Den norsk-islndska skaldediktningen. 2 vols. Lund.
Skaldic Poetry of the Scandinavian Middle Ages. Ed. by M. Clunies Ross et al. 9 vols. 2007.
Turnhout.
Skj = Finnur Jnsson, ed., 191215: Den norsk-islandske skjaldedigtning. Copenhagen.Vols AI, AII
(tekst efter hndskrifterne), BI, BII (rettet tekst).
SnE 184887 = Edda Snorra Sturlusonar. Edda Snorronis Sturli. Ed. by The Arnamagnan Commis-
sion. 3 vols. Copenhagen.
SnE 1998 = Snorri Sturluson Edda. Skldskaparml. Ed. by A. Faulkes. 2 vols. University College
London.
Wanner, Kevin J., 2008: Snorri Sturluson and the Edda. The Conversion of Cultural Capital in Medie-
val Scandinavia. Toronto, Buffalo and London.
Whaley, Diana, 2005. Skaldic Poetry. In: A Companion to Old Norse-Icelandic Literature and Culture.
Ed. by R. McTurk. Malden, MA, Oxford and Carlton. Pp. 479502.

4
In a recent study, Wanner (2008) has invoked Pierre Bourdieus model of cultural capital in his theory of prac-
tice to account for the details of Snorris life and the creation of his Edda largely as a political manoeuvre for his
personal advancement, but the nature of both the human response and the product (the Edda) is too complex to
be explained only in these terms.

192
 Passing Time and the Past in Grettis Saga smundarsonar
Jamie Cochrane, West Sussex, England
In any narrative text there are a number of relationships concerning time. Firstly there is a
relationship that has been referred to as intrinsic relative chronology (lfia Einarsdttir 1964
also Clover 1982: 120121); that is the chronological relationship between the events de-
scribed, measured either in terms of one another, or against the historical backdrop of the
saga-age. Secondly there is a relationship that I will call narrative pace; that is the relationship
between the words themselves and the events in the story. Thirdly there is a relationship be-
tween the time represented either by an author or a supposed reader or audience1 and the
events described in the text. In this paper I am going to review these relationships in one me-
dieval Icelandic text Grettis saga smundarsonar.
Grettis saga is considered one of the four major slendinga sgur (alongside Njls saga,
Egils saga and Laxdla saga), although it is the youngest among these, dating from around
1310 to 1320 (Cook 1993, 242)2. The events the saga narrates cover the years from 885 to
1047. As lafia Einarsdttir observes in her 1964 study of chronology in the sagas, dates oc-
cur infrequently in saga literature. lafia notes four dates occur in slendingabk, around
which the dating of the story can be ascertained, by counting forwards or backwards. No dates
occur in Grettis saga, but there are a number of prominent historical events historical events
from which the dating of our story can be deduced. The most notable of these are:
1. The battle of Hafrsfjrr, in which nundr trftr takes part, which consolidates Haraldr
hrfagris power in Norway (c.885) (chapter 2).
2. Hkon jarl Eirksson ruler of Norway (1012) (chapter 20).
3. lfr Haraldsson becomes King of Norway (1015) (chapter 37).
4. The deaths of Skapti roddsson and Snorri goi (1030 and 1031 respectively) (chapter
76).
5. The return of Haraldr Sigurarson from Constantinople to share the Norwegian throne
with his half-brother King Magns inn gi and Magns subsequent death (1047) (chapter
90).
Other seemingly incidental mentions of historical events date sections of our story within a
few years, for example the mention of the early Christian missions to Iceland (c. 981985) in
chapter 14 or the mention of orkell mni as lawspeaker (970984) in chapter 12. We cannot
be certain that the original saga audience knew the exact dates of these events (although the
more educated among them may have done), but they would have been relatively familiar
with their approximate relationship with each other and with the audiences own post-saga-
age period. As such, these events form part of the overall fabric of the saga-age that could be
taken as assumed knowledge. The story of Grettir, his conviction, time as an outlaw and even-
tual death are set against a backdrop of historical and easily datable events.

1
I will not tackle the thorny problem of whether the intended recipient of this or indeed any saga was an audi-
ence or reader. There is justification for the use of either term. Grettis saga as it is preserved is a written text and
thus our actual evidence is confined to a readership, but it undoubtedly had an oral background and furthermore
the existence of written text does not preclude ongoing performance either read directly from the text, memo-
rised or extemporised to an audience. For the remainder of the paper I will use the word audience to refer to
the thirteenth century intended recipient of the saga. In the current paper I will not specifically consider the rela-
tionship to the modern reader.
2
Despite the relatively late date for the extant saga, that a tradition existed about Grettir smundarson can be
deduced from mentions of him in a number of earlier sources such as Gsla saga Srssonar (Bjrn K. rlfsson
and Guni Jnsson 1943: 70), Fstbrra saga (Bjrn K. rlfsson and Guni Jnsson 1943: 121122, 148
and 191) and Landnmabk (Jakob Benediktsson 1986: 199, 211, 213 and 280281).

193
 From these events it is possible to count forwards or backwards in the saga according to
the supposed years (winters) passed in the text to build up an overall picture of the internal
chronology of the story. To do this I have taken the online version of the text available at
http://www.snerpa.is/net/isl/grettir.htm (Grettis saga 1997), and divided it into 126 sections of
500 words each.3 At the beginning of each of these arbitrary sections I have tried to estimate
the historical year when the events are occurring and plot them on a graph (figure 1). The
shaded area represents the course of the saga and the solid vertical bars the five points datable
with relative certainty listed above. This is undoubtedly an artificial and somewhat crude
means of approaching the text, but nonetheless offers an insight into the passing of time in the
text. We find a reasonable amount of cohesion and accuracy showing that the saga-writer had
both an awareness of the historical dates and how they related to the specific chronology of
his own story. This intrinsic chronology is far from perfect. There are a few sections where
the saga lacks specificity (for example dvalisk hann ar um hr, Guni Jnsson 1936: 271),
and some where it seems distinctly strained, one of which I will discuss further below.

Plotted in this way the saga forms a flattened S shape. The gradient of the line refers to the
speed at which years pass in the story. This brings me to the second relationship mentioned
above that of pace or rhythm. As one might expect the pace at which these events are nar-
rated is varied according to the level of detail or focus the narrator gives at each point. In the
most basic sense there is a relationship between the words themselves that is the actual
length of time it might take to read each word aloud and the movement of time in the story.
At one extreme hours, days, seasons and indeed whole years might be passed by in a single
sentence. A saga-writer had a number of stock phrases to allow for such progression of time.
Examples of these in Grettis saga include the following:

3
Counting words seems the most consistent and accurate measure of a text (pagination and chapters are both
variable and editorial). For simplicity I have used a computer based text rather than the slenzk fornrit text cited
in the rest of this article. There will be a few discrepancies due to editorial choices between the fornrit and online
versions. Chapter markers and numerals have been counted as words (i.e. 1. kafli = two words).

194
 N lei af nttin. (Guni Jnsson, 1936: 57).

[] ok stu um kyrrt um hr. (Guni Jnsson: 1936, 100)

Such phrases are often characterized by a mention of a particular season or a point in the year:

[] lei sv fram misumar (Guni Jnsson, 1936: 113)

Lei n sv ndverr vetrinn af. (Guni Jnsson, 1936: 159)

Lei sv vetrinn framan til jla, at ekki bar fleira til frsagna. (Guni Jnsson, 1936: 61)

Another means at the writers disposal is to start the description of an event with a mention of
a particular point in time:
Um vrit [] (Guni Jnsson, 1936: 113)

etta sumar it nsta [](Guni Jnsson, 1936: 61)

This has the effect of suspending the characters while the narrator moves time forward to the
next event of significance in his narrative. At the other extreme, events are narrated in actual
time; that is the time it would take to read the text aloud is more or less the same as the time it
would take for the events to actually happen. The clearest example of actual time pace in the
sagas are passages of direct speech. The wealth of direct speech in the slendinga sgur means
that many events are narrated in actual time with a one to one relationship between the words
of the text and the pace of the events. Actual time is the slowest pace routinely found in the
slendinga sgur (which do not have either the proliferation of highly detailed descriptive
passages or emotional responses found in modern novels). I will argue, however, that at par-
ticular points in sagas the pace of the narrative slows beyond this actual time pace. The over-
all pace of the narrative can therefore be measured according to the way in which these de-
tailed passages with direct speech are interspersed with summary passages.
If we return to the graph, this admittedly crude measure nonetheless illustrates the points I
intend to make. The overall direction is upwards, indicating that the vast majority of events in
the saga are narrated in chronological order. In the early chapters, time passes relatively
quickly with only a few events narrated in detail. There are episodes told in detail and with
direct speech, but these are relatively few and several years can be passed quickly in-between
them. As one approaches the climactic portions more of the events are narrated in detail, with
greater use of direct speech and thus a one-to-one temporal to textual relationship and with
less time passing between each episode. As one approaches the end of the saga, the style once
again becomes synoptic and years pass more quickly. The little bumps in the line, show that
although by and large the order is chronological (as indeed is the case for all slendinga
sgur) there are a few areas of narrative overlap.4 For example at the end of chapter 41 (sec-
tion 58 on the graph) Grettir has been given permission to return from Norway to Iceland. The
narrator, however, leaves Grettir poised to embark and returns to Iceland the previous summer
to describe the death of smundr hrulangr and killing of Grettirs brother Atli. Grettir is left
suspended (literally at sea) until the opening of chapter 47. This is a convenient conceit of the

4
Although chronological narration in the slendinga sgur is observed as a general principle (Vsteinn lason
1998, 95100), that is not to say there is no chronological overlap between separate strands within a saga-plot
necessitating the author to backtrack chronologically and fill in the gaps in a particular narrative strand (for dif-
ferent strategies enabling this see Clover 1982, 109135).

195
story ensuring that Grettir is out of the way and unable to intervene against the slaying of his
brother, but it shows the self-awareness of the writer of the intrinsic chronology as he manipu-
lates the various strands of his story.5
This awareness is exhibited in the fact that time and chronology become a theme in the
saga. For example, in chapter 51 the Lawspeaker Skapti roddsson questions which came
first, the killing of Atli or the sentence of outlawry against Grettir. It turns out that the sen-
tencing had occurred first by just a week, thereby invalidating a case against Grettirs family
brought by roddr drpustfr. More generally in the saga the 19 years of Grettirs outlawry
are a recurring theme. It is likely that Grettirs outlawry and the extraordinary length of time
he spent in the wilderness was part of the traditional material handed down to the writer part
of traditional material the writer. Gsla saga Srssonar (assumed to be earlier than Grettis
saga) indicates that Grettir was the only man to have spent more time as an outlaw than Gsli
(Bjrn K. rlfsson and Guni Jnsson 1943: 70). In Grettis saga, however, this length of
time becomes a theme, giving the reader a reference point, both concerning how long since
Grettir sentence and building tension as Grettirs inevitable final stand approaches. This is
perhaps the most problematic use of time in the saga. It proves hard to fit the 19 years exactly
with the external events referred to in the saga. Grettir is made an outlaw in chapter 46 (sec-
tion 62/63 on the graph) which seems to accord to 1016 and killed the winter following the
death of Skapti roddsson, normally assumed to be around 1031, giving a gap of only 15
years (on this discrepancy see Guni Jnsson 1936: lxvi-lxvii). Thus the setting of Grettirs
story against the existing fabric of saga lore has been imperfectly made. It seems possible that
this discrepancy has come from imperfectly reconciled oral traditions one of which had Gret-
tir as an outlaw for exactly 19 years, the other dating his outlawry against historical events.
Nonetheless the 19 years are an important part of the saga, creating the tragic irony of Gret-
tirs life in that he would have become a free-man once again had he lived one winter longer.
Although the saga-writers use of time is imperfect it is self-conscious and relevant to his
story.
The third relationship mentioned above is that between the events described in the text and
the audience. Assuming the early fourteenth century date for the saga is correct, there is a gap
of three centuries between the event described in the text and the original audience. Many
technologies and ideas would have changed or developed in the intervening years. Grettir and
his contemporaries must have been very much part of the history of the first saga audiences;
not distant and obscure like the heroes of the fornaldar sgur, but part of history nonetheless.
The narrator of Grettis saga shows an unusual level of awareness of this historical gulf be-
tween his protagonist and audience. As one can see from the graph the tail end extends right
into the mid-thirteenth century, this is in view of the final chapter of the saga which states the
opinion of the thirteenth century law-speaker, author and power-magnate Sturla rarson (d.
c.1284) on Grettirs life. It is possible that this relates to a text, perhaps an early version of the
saga, written by Sturla about Grettir (Jnas Kristjnsson 1988: 235). It has the effect, how-
ever, of bringing the narrative up to the present, or at least into the life-times of the parents or
grandparents of the original audience.
This is not, however, the first reference to Sturla in the text. He is also mentioned in chap-
ter 69 and particularly strikingly in chapter 49. In one of the sagas best framed set pieces
Grettir attacks his enemy orbjrn xnameginn and his son in a hay field. The scene is set
through a detailed description of the hay-field:

5
On the self-consciousness of the Grettis saga author see de Looze 1991.

196
 ar gengr ein mrr ofan r hlsinum, ok var ar sltta mikil, ok hafi orbjrn ltit sl ar
mikit hey, ok var fullurrt; tlai hann at heim at binda ok sveinninn me honum, en kona
tk rkin (Guni Jnsson 1936: 153)

Grettir too is described in detail. As he approaches, orbjrn and his son discuss how best to
deal with him:

Mar rr ar at okkr, ok skulu vit htta at binda heyit ok vita, hvat hann vill [] mlti
orbjrn: etta er mikill mar, ok eigi kann ek mann velli at sj, ef at er eigi Grettir s-
mundarson, ok mun hann ykkjask eiga rnar sakar vi oss; ok verum vi rskliga ok ltum
engan bilbug okkr sj. Skulu vit fara at me rum, ok mun ek ganga at honum framan ok sj,
hversu til teksk me okkr, v at ek treysti mr vi hvern mann, ef ek einum at mta. En
gakk bak honum ok hgg tveim hndum milli hera honum me xinni; arftu eigi at varask,
at hann geri r mein, san er hann snr baki at r. (Guni Jnsson 1936: 154)

Although this is ostensibly narrated in direct speech and therefore in actual time, it has the
effect of slowing time for the audience. Grettir continues to approach throughout the speech,
but as if in slow motion, while we hear the conversation of his adversaries. It is an example of
a saga motif whereby the audience see the approach of some attackers through the eyes of
their nave victims (a good example of this, which also has the effect of expanding the time
taken of the attackers approach, is where Helgi Harbeinsson discusses in great detail the
clothing and appearance of the attackers approaching him with his shepherd in Laxdla saga,
Einar l. Sveinsson 1934: 186191). This slowing of time in the episode heightens the ten-
sion for the audience by delaying the eventual conflict while constantly pointing forward to it.
We can see the same technique used to tremendous effect in the build up to the Grettirs con-
flict with the ghost Glmr. Through a succession of episodes each of which progress us only a
little further towards the eventual confrontation between Grettir and Glmr, the tension is
built up, leading to the detailed description of Glmr crashing against the roof of the house
and finally stooping inside. I would even argue Grettirs sudden paralysis as he is dragged by
Glmr out into the moonlight is the ultimate example of freezing time in the saga literally
freezing Grettir while Glmr delivers his speech and places his evil curse on Grettir.
If we return to Grettirs conflict with orbjrn xnameginn, we find Grettir finally kills
both men and goes to the farm to announce the killings. As part of this episode, Grettir re-
moves the pin from his spear head under the assumption that he would not want orbjrn to
throw it back at him.6 In fact the plan comes to nothing as without the pin the spearhead falls
off and is lost. The fate of the spear is revealed in the next chapter:

Spjtit at, sem Grettir hafi tnt, fannsk eigi fyrr en eira manna minnum, er n lifa; at spjt
fannsk ofanverum dgum Sturlu lgmanns rarsonar ok eiri mri er orbjrn fell, ok
heitir ar n Spjtsmrr [] (Guni Jnsson 1936: 157)

Within the plot of his story the spearhead is entirely functionless and so its inclusion is spe-
cifically for the purpose of creating a relationship between the now of the story and the
now of its fourteenth century performance. This relationship between the past and present is
played with again and again in the text. Sometimes this is in the differences between the past
and present, for example there are various antiquated customs that the author feels the need to
illucidate for the purpose of his reader, such as the absence of dlur (pumps) from boats and
therefore the need for manual bailing (Guni Jnsson 1936: 55) or specific legal points which

6
There are instances of weapons being returned in this way in sagas, see for example Njls saga (Einar l.
Sveinsson 1954: 403).

197
may not have been familiar to the readers (e.g. Guni Jnsson 1936, 897). Conversely the rela-
tionship between the past and present can be expressed through references to things still exist-
ing in the audiences own time which feature in the story, such as the mention of the place-
name Spjtsmrr in the passage above. Such mentions of places and things still in existence
and presumably familiar to the original audience are by no means uncommon in saga-
literature, but seem particularly plentiful in sections of Grettis saga. For example. according
to the saga stones lifted by Grettir are still visible (e.g. Guni Jnsson 1936: 201), places still
known are named after saga events (e.g. Guni Jnsson 1936: 197 and 243) and even a troll
woman turned to stone which Grettir battled against can still be seen (Guni Jnsson 1936:
213). It is unlikely that every member of the original audience knew each of these places per-
sonally, but by referring to them this way the narrator stresses the relationship between the
time of his story and the present and reminds the audience of the supposed truth behind the
story. Elsewhere such details serve to place the audience right within the text. For example the
detailed description of the interior of the damaged farmhouse before Glmr bursts into the
farm not only further delays his entrance but also may have reminded the audience of the very
building they were currently in one can picture the original audience hearing the story in the
depths of an early fourteenth century Icelandic winter glancing nervously at the roof-beams of
their own farm-house wondering whether Glmr might not visit them too.
This playful relationship between the past and present can be extended to the style of the
saga itself. The many supernatural events in the saga has led to a comparison by some schol-
ars with the fornaldar sgur (see for example Tulinius 2005, 457). The earlier sections bear
this comparison well. The sea-battles described, both the battle of Hafrsfjrr and nundrs
encounters with vikings, fit this epic-heroic model well. Such events were distant both tempo-
rally and spatially from the audiences own lives. The mutilation of nundr at the battle of
Hafrsfjrr and his subsequent nickname name (trftr) are also reminiscent of mutilations
found in the fornaldar sgur (see for example Egils saga einhenda ok smundar berserkja-
bana). Throughout this section (with the exception of the clearly datable battle of Hafrsfjrr)
time is relatively fluid and it is difficult to pin down exact dates for events.
Upon arrival in Iceland, however, the style changes subtly. Although supernatural elements
abound, there is little that is not comparable with supernatural elements in other slendinga
sgur such as Eyrbyggja saga, Laxdla saga or drauma ttir such as Bergba ttr and
Kumlba ttr. Such elements are also found in Icelandic folk-tradition. Indeed I would argue
that, despite its supernatural interest and folk-motifs, the central section of Grettis saga
should be seen very much as an slendinga saga. There is no sense of a quest-narrative as
there is in many fornaldar sgur and Grettir is outlawed by a legal ruling not a supernatural
event (notwithstanding Glmrs curse upon him). He is forced out beyond society into the
wilderness, where strange and wonderful things happen, but there is no sense of the escapism
that run through the fornaldar sgur. Even where fornaldar sgur characters are clearly inde-
finable historical persons, the stories in which they occur exist in a fictional past which is not
clearly defined and while such characters might be datable, the sagas themselves show no
interest in these dates. Grettir in his saga is a real man acting in a time three centuries before
his readers. Although the intrinsic chronology of the central portion of the saga is far from
perfect, it was a chronology of which the saga-writer was thoroughly conscious.
With Grettirs death, the style of the saga changes again. Grettirs brother pursues his killer
ngull to Constantinople and kills him. There then follows a short narrative in which or-

7
See, however, Gunis footnote to this passage, where he questions the saga-writers accuracy on this legal
point. Whether the saga-writer was right, wrong or deliberately falsifying the law at this point need not under-
mine my overall point that a distinction is being drawn in the text between the now of the events and the now
of their narration.

198
steinn falls in love with a married woman named Spes and narrowly avoids being caught on
several occasions by her jealous husband until she finally divorces him. It has been suggested
that motifs in this narrative have links to the Tristram legend (see for example Craigie 1913,
66 and Cook 1993, 242), more generally however, the whole style is much closer to that of a
European romance perhaps even influenced by early fabliaux with its emphasis of cunning
and, in particular, sexual deception. It is not inconceivable that the so called Spesar ttr,
originally had no connection to the oral material surrounding Grettir, but nonetheless for the
writer of the extant version of Grettis saga it was an important final act. Such a writer cannot
have been oblivious to the mismatch in tone between the episodes through the main portion of
the saga and that in the final six chapters and therefore we must assume he had a particular
purpose for such change in style. It seems possible that the episodes of this final act were de-
liberately mimicking what the saga-writer saw as a modern or new style. In this way the style
and tone in which Grettis saga is written reflects the comparative distance of the narrative
from its original audience. The earliest sections, with their reliance on Viking battles resem-
bling those found in the fornaldar saga are deliberately distanced from the audience. The
main body of the text, remains distinct from the audience, but only to the distance of all the
slendinga sgur, with the great heroes living wild lives but within the familiar surrounds of
the Icelandic countryside and with specific reference to things with which that audience will
identify thereby embedding the link between the heroic past and textual present. The final
section plays out in a far off land, in a foreign mode, imported from Europe but rapidly grow-
ing in fashion in saga writing. Therefore, the writers use of time and the past in Grettis saga,
while not perfectly reconcilable with historical events, is nonetheless always thoroughly con-
scious and literary.

Bibliography
Bjrn K. rlfsson and Guni Jnsson, eds, 1943: Vestfiringa sgur. Reykjavk (slenzk fornrit 6).
Clover, Carol J. 1982: The Medieval Saga. Ithaca, New York (Cornell University Press).
Cook, Robert, 1993: Grettis saga. In: Medieval Scandinavia: An Encyclopedia. Ed. Phillip Pulsiano.
Co-ed. Kirsten Wolf. New York (Garland Publishing). Pp. 242242.
Einar l. Sveinsson, ed., 1934: Laxdla saga. Reykjavk (slenzk fornrit 5).
Einar l. Sveinsson, ed., 1954: Brennu-Njls saga. Reykjavk (slenzk fornrit 12).
Einar l. Sveinsson and Matthas rarson, eds, 1935: Eyrbyggja saga. Reykjavk (slenzk fornrit 4).
Fjalldal, Magns, 1998: The Long Arm of Coincidence: The Frustrated Connection between Beowulf
and Grettis saga. University of Toronto press. Toronto, Buffalo, London.
Grettis saga, 1997. Ed. unknown. http://www.snerpa.is/net/isl/grettir.htm (accessed on 22.04.2008).
Guni Jnsson, ed., 1936: Grettis saga. Reykjavk (slenzk fornrit 7.)
Guni Jnsson and Bjarni Viljlmsson, eds, 194344: Fornaldarsgur Norurlanda 13. Reykjavk
(Bkatgfan forni).
Jakob Benediktsson, ed., 1986: slendingabk Landnmabk. Reykjavk (slenzk fornrit 1).
Jnas Kristjnsson. 1988: Eddas and Sagas. Trans. Peter Foote. Reykjavk. (Hi slenzka bkmennta-
flag).
Looze, Laurence de. 1991: The Outlaw Poet, The Poetic Outlaw: Self Consciousness in Grettis saga
smundarsonar. Arkiv fr nordiks filologi 106. Pp. 85103.
ODonoghue, Heather, 2005: Skaldic Verse and the Poetics of Saga Narrative, Oxford and New York
(Oxford University Press).
lafia Einarsdttir, 1964: Studier i kronologisk Metode it tidlig islandsk Historieskrivning. Bibliothe-
ca Historica Lundensis XIII. Stockholm (Natur och Kultur).
Schach, Paul, 1955: The Anticipatory Literary Setting in the Old Icelandic Family Sagas. Scandina-
vian Studies 27. Pp. 113.
Tulinius, Torfi H., 2005: Sagas of Icelandic Prehistory (fornaldarsgur). In: Old Norse-Icelandic Lit-
erature and Culture. Ed. Rory McTurk. Malden MA, Oxford, Victoria (Blackwell). Pp. 447461.
Vsteinn lason, 1998: Dialogues with the Viking Age. Trans. Andrew Wawn. Reykjavk (Heim-
skringla).

199
rhallur Vilmundarson and Bjarni Vilhjlmsson, eds. 1991: Harar saga. Reykjavk (slenzk fornrit
13).

200
 Knsla och oro i Fstbrra saga
Tommy Danielsson, Ludvika, Sweden

I
Som vi alla vet r verklighetens skeenden oerhrt komplicerade affrer, frihetsgraderna r
mnga och vad som ska hnda i nsta sekund inte ltt att frutse, srskilt i avgrande och kri-
tiska lgen av ett frlopp. Eftert kanske man har en viss uppfattning om det intrffade, men
man vet sllan skert varfr det gde rum. Det finns emellertid ett oerhrt effektivt stt att i
efterhand inskrnka frihetsgraderna. Man kan bertta en berttelse om det som hnde. Bert-
tandet i sig r lika komplicerat och oberkneligt som hndelserna sjlva, liksom tillgnelsen
av det berttade, men texten frefaller entydig och kristallklar svl ur konceptionens som
receptionens synvinkel, och det r ltt att bli vertygad om att den intrngande frstelsen gr
att n. Berttandet blir drfr ett oundgngligt stt att hantera komplexiteten i tillvaron, ett stt
dessutom dr emotioner spelar minst lika stor roll som den rationella tanken. Men vad ska
man gra om texten, den frrdiskt enkla och vertygande texten, r det enda man har tillgng
till, och om det man r ute efter r en sedan lnge frgngen komplicerad kommunikations-
process?
Det allra mest basala r att inte betrakta texter som ofrnderliga, statiska objekt som man
kan vrida och vnda p och dra slutsatser kring. Det finns ingen text utanfr medvetandet, det
finns bara textsignaler och affordances som lsarkonstrueras dynamiskt i lsgonblicket ut-
ifrn den dynamiska diskursvrld som medvetandet oupplsligt och dynamiskt ingr i. Ett
frsta steg kan drfr vara att s gott det gr betrakta den egna nutida receptionsprocessen och
frska f en bild av de kraftflt som vi tenderar att konstruera nr vi konfigurerar textens
signaler. Hr mste vi komma ihg att kraftflten bde beror av vr egen erfarenhetsbaserade
uppfattning om vrlden och av textsignalernas urval och fokuseringar och direkta kommenta-
rer. Vi kan inte subtrahera bort inflytandet frn den egna diskursvrlden, men vi kan tmin-
stone ha som medvetet ml att bttre frst de diskursvrldar som den dtida konceptionen
och receptionen frsiggick i.
Det finns all anledning att tro att konceptionens art pverkar texters utformning, men det r
svrare att formulera kriterier och knappast heller nskvrt med tanke p de inblandade dy-
namiska processerna. En frfattare som skriver ner en helt fiktiv berttelse, en frfattare som
bygger framstllningen p egna upplevelser eller andras utsagor, en muntlig berttare som
infr publik framfr en egen berttelse, en muntlig berttare som knner sig reproducera en
berttelse ur en levande tradition det borde finnas identifierbara skillnader hr. I ett par tidi-
gare underskningar har jag diskuterat problemen med att alltfr ensidigt betrakta den is-
lndska sagolitteraturen som en skriftlig kulturyttring, men det r naturligtvis ett vanskligt
fretag att hitta otvetydiga bevis fr muntlighet i de bevarade texterna. Men ven om en skri-
ven saga innebr ngot helt annat n ett muntligt framfrande texten r, som Gsli Sigurs-
son (2004) ppekat, inte oral utan orally derived r det min vertygelse att det kan finnas
spr av muntlighet i det skrivna, och att det beror p vilka perspektiv vi intar om spren blir
synliga eller inte. Ett stt kan vara att frska nrma sig det texterna ser ut att vilja de
affordances som de ser ut att erbjuda och de grundlggande anxieties som de ser ut att ut-
trycka. Det gller helt enkelt att identifiera kraftflten i sagorna, bde hur vi uppfattar dem och
hur texten tycks vilja att de ska uppfattas. Kan man se en genomgende strategi eller finns
det motsgelser, och beror i s fall motsgelserna p en process under lng tid eller p funda-
mentala problem i den omgivande diskursvrlden? Var det de frmedlade knslorna och den
underliggande oron som gjorde berttelsen vrd att bertta?

201
 II
Som bekant handlar Fstbrra saga om de bda fosterbrderna orgeirr Hvarsson och
ormr Kolbrnarskld Bersason, och i ett frsta parti ligger fokus p deras gemensamma
vldsdd p vstra och nordvstra Island. De skils sedan t och det andra partiet berttar om
deras fortsatta den var fr sig. orgeirr drps i en sammandrabbning mellan ett par grupper
sjfarare i ett hamnlge i nordost, och den sista avdelningen har drfr ormr som ensam
huvudperson. Frst tar han p Grnland en gruvlig hmnd p den ene av orgeirrs drpare,
och sedan beger han sig tillbaka till Norge, fljer kung lfr i landsflykten och deltar i slaget
vid Stiklastair, dr bde han och kungen dr.
Redan de frsta raderna slr fast ett makthierarkiskt ntverk frn Gud, via kung lfr helgi
till makthavare i Norge och andra lnder.1 Det var, sgs det, de hvdingar som Gud skattade
hgst som ocks lfr tyckte bst om, och den gode hvdingen i det hr sammanhanget p
Island heter Vermundr. Ntverket r allomfattande eftersom hvdingarna i sin tur hller ord-
ning i den enskilda bygden, och det stlls omedelbart p prov genom att den fredlse Grettir
objuden kommer till trakten. De drabbade bnderna lyckas emellertid vermanna honom och
stter upp en galge. Vermundr r bortrest fr tillfllet, men nr hans kloka hustru orbjrg fr
rtt p vad som r p vg att hnda skyndar hon tillsammans med sina huskarlar till platsen,
frklarar situationens allvar och ser till att Grettir blir befriad. orbjrg framhller dels att
Grettir r av hg tt och skyddad av mktiga frnder, dels att han r vrdefull p grund av
styrka och duglighet. Det har allts sina risker att dda honom, och i extraordinra krislgen
kan sdana som han vara oumbrliga, ven om mest till besvr i vardagen. Nr bnderna i alla
fall hvdar att fredlsa rnare inte br f behlla livet, slutar orbjrg argumentera och hnvi-
sar bara till sin bestmmandertt.
Som vi ser r det redan hr ett stort antal kraftflt som samverkar. Man kan fr det frsta
ana en viss spnning i den vertikala maktordningen mellan hvdinge- och bondenivn, efter-
som bnderna antyder att hvdingarna fr sin vilja igenom oavsett om de har rtt eller fel.
Vidare r frhllandet mellan de bofasta och icke bofasta komplicerat och kan uppenbarligen
vara av bde positiv och negativ art. Det finns ocks slktskapsband som man omjligen kan
bortse frn. Relationen mellan kvinnor och mn ser ut att fungera vl den hr gngen, efter-
som orbjrg r klok och har fr vana att styra bygden i makens frnvaro. Dessutom tillhr
hon en ansedd slkt, men det r ju inte sjlvklart att Vermundr ska uppskatta all hennes fre-
tagsamhet det finns en potential fr konflikt hr. Tjnstefolk r p plats i episoden och lyder
order. Slutligen fr vi en intressant glimt av den laddade typsituation dr faran rent konkret
str fr drren. Mnga skulle inte ha varit s undfallande mot Grettir ef eim sndisk eigi
troll fyrir durum (121).
Textsignalerna styr frsts receptionen inte bara genom urval och perspektivskiften och
kausalitetsbedmningar, utan ocks genom direkta och indirekta vrderingar av det som hn-
der, s ocks i den hr objektiva islndska sagan. Gud tycker bst om de hvdingar som
lfr tycker bst om. Vermundr r ansedd och klok och vnsll, och orbjrg r klok och
hgsinnad. Mnnen r alltid njda med det orbjrg bestmmer. Och det enda som hindrar
hngningen att bli av r att orbjrg kommer ridande. Genom den lilla strofen mot slutet av
episoden fr vi ocks del av Grettirs perspektiv och hans insikt att han verkligen var illa ute.
Till och med berttelsens moral sls fast den visar vilken hederskvinna orbjrg var. Det
finns naturligtvis alternativa stt att sammanfatta skeendet, det ser vi om vi tar i beaktande att
episoden ven r tergiven, i betydligt utfrligare form, i Grettis saga. Dr r det i mngt och

1
Jag utgr frn utgvan i slenzk fornrit 6, dr versionen i Mruvallabk r huvudtext s lnge den finns beva-
rad, och versionen i Hauksbk drefter tar ver. Mrk att versionen i Flateyjarbk inkluderar introduktionen av
Vermundr och orbjrg men saknar Grettirepisoden.

202
mycket andra kraftflt som styr n i Fstbrra saga, frmst genom en tydligare polarisering
mellan en verordnad niv, dr Grettir ingr, och en mer komiskt skildrad bondeniv.
Vad hnder d med kraftflten i resten av sagan? lfrntverket hlls samman av vnska-
pen med de utvalda hvdingarna och genom att islnningar och grnlnningar tillbringar vint-
rar vid den norska hirden. De som lfr bedmer som srskilt dugliga kan bli upptagna i hir-
den och f srskilda befogenheter och anvndas fr att tgrda instabiliteter i ntverket, och
lfr kan ocks p vernaturlig vg gripa in i skeenden. Men faktum r nd att den stabila
lfrvrlden med alla sina ntverk havererar i sagan, i och med att bandet mellan Gud och
kung och hvdingar r brutet efter slaget vid Stiklastair. Bland hvdingarna kan man srskil-
ja de som efterstrvar jmvikt (orgils och Illugi Arason p Reykjahlar och orkell Leifsson
p Grnland), de mer problematiska (Vermundr) och de som str jmvikten (orgrmr trolli).
Frndskapsband gr sig ofta gllande, inte minst vad betrffar det std orgeirr fr frn Reyk-
jahlar, men ocks som krafter utifrn (drpet p orgils Msson och hndelserna p Grn-
land). Vanligt folk representeras framfr allt av skeppsbesttningar, som ger ett hgst nor-
malt intryck, och bnder, som vi redan sett bli ordentligt tillrttavisade (en av de bnder som
enligt versionen i Grettis saga deltog vid hngningsfrsket r i det nrmaste skrckslagen i
Butraldiepisoden). Och det r i allmnhet tufft att befinna sig lgt ner p rangskalan i den hr
vrlden. Man kan bli drpt utan anledning, och ofta ingr man anonymt i grupper av huskarlar
och tjnstefolk, som frlngda armar helt enkelt. I enstaka fall fr man ett namn och en identi-
tet innan man blir ddad, och ngon gng kan man till och med f friheten till fljd av sina
insatser. De halvt utanfrstende vldsmnnen tar upp stor plats i sagan och gr ibland nytta,
genom att ta hand om andra vldsmn som terroriserar sin bygd eller genom att utfra kung-
ens uppdrag. Men de r ocks en plga, och man kan frvnta sig vad som helst av dem.
Mycket handlar dessutom om vldsmnnens relationer inbrdes, i fiendskap eller vnskap,
och om hur uppgrelserna dem emellan kan bli till problem fr samhllet i vrigt.
Kvinnorna spelar en stor, om n sekundr, roll i sagan och r betydligt mer mngfacettera-
de n mnnen. I frvnansvrt f fall r det frga om hustrur som, liksom orbjrg, egentligen
str tillbaka fr mnnen, betydligt oftare om kvinnor som agerar helt i egenskap av mdrar
och husfruar. Lgre p den sociala stegen str de mnga tjnstekvinnorna som ppnar drren i
natten och frmedlar budskap men ocks aktivt kan skydda sina husbnder. Hr finns allts
mnga handlingskraftiga och stridbara kvinnor som hittar metoder att frsvara sig sjlva och
sina egna. Inte minst gller detta ormrs bgge krestor, som verkligen inte r att leka med.
ormr r sledes inte oemottaglig fr kvinnlig fgring, men det frefaller bara vara nr han
r uttrkad som han tar till den typen av tidsfrdriv. orgeirr r mer explicit och menar att det
bara r slseri med krafterna att ligga och krypa p kvinnfolk. Dremot har han ett gott frhl-
lande till sin mor, och nr en kvinna formulerar ett uppdrag r han snabbt p hugget. Flera av
kvinnorna i sagan besitter vernaturliga krafter (Grma p Island och Grma p Grnland,
orbjrg kolbrn, undantagskringen i lfsdalr, ords p Grnland), och det talas ver hu-
vud taget mycket om hedniska frhllanden i sagan men d oftast ur berttandets kristna per-
spektiv. Bortsett frn lfrs ingripanden r det annars bara vid tv tillfllen som det verna-
turliga gr sig gllande (den blodiga processionen i Garpsdalr och pilen som ddar ormr).
Vdrets krafter r ocks ovanligt viktiga i Fstbrra sagas vrld, framfr allt till en brjan,
och utgr ett framtrdande kraftflt med betydelse fr hndelsernas gng.
Men vilken r d meningen med berttelsen, vad r det sagan vill skdliggra och varfr
var den vrd att lyssna till? r det bara frga om en samling lst forbundne fortllinger
som blir sammanhngande frst i och med ormrs hmnduppdrag? (Meulengracht Sren-
sen 1994: 581) eller helt enkelt om en saga med two protagonists instead of one (Andersson
2006: 70). Eller ska vi lita p sagans egna slutord att lkr hr frsgn eiri, er vr kunnum at
segja fr ormi, kappa ins helga lfs konungs (276)?

203
 Jag begrnsar mig hr till ngra enskilda iakttagelser. Fr det frsta r fosterbrderna till
en brjan minst lika besvrliga som Grettir nr han gr livet surt fr bnderna i trakten, men
de hjlper samtidigt en kvinna som trakasseras av vldsmn, och orgeirr hmnas fadern trots
sin ungdom. Samma polaritet fortstter efter orgeirrs fredlshet i och med att han skaffar sig
alltmer anseende utomlands, medan p Island drpsmlen blir mer och mer absurda och futti-
ga. Det r tydligt att tv vrldar av fundamentalt olika karaktr r verlagrade varandra i sa-
gan, och att orgeirr agerar i dem p helt olika stt, som hjlte i den ena och problembarn i
den andra. En annan, hgst lokal, instabilitet, ser ocks ut att organisera helheten. Som fram-
gtt r orgeirr och ormr frst oskiljaktiga men agerar senare var fr sig p egen hand,
och det enda som egentligen hnt r att orgeirr frgat ormr vad han tror skulle hnda om
de prvade krafterna mot varandra. ormr reagerar omedelbart och krver att de ska skiljas
t, och blotta tanken p handgripligheter dem emellan r tydligen frskrckande fr honom.
Att incidenten r central framgr ocks av digressionen om fosterbrderna i lfsdalr, dr
man verkligen ser vad som kan hnda nr vnskap vergr i fiendskap. Kanske r det p
grund av ormrs frnvaro som orgeirr inte alls fungerar p Island, utan bara i den andra
vrld dr kung lfr tar ver fosterbroderns roll. Och fr ormr gr det egentligen inte
mycket bttre. Han blir upprepat rastls och trstar sig med diktning och problematiska kvin-
noaffrer, och inte frrn orgeirr drps fr han en verklig uppgift att ta tag i. Kanske finns
hr ven en frklaring till det oproportionerliga hmndprojektet p Grnland, och till och med
lfr frgar ju ormr varfr han slog ihjl s mnga. Uppenbart r samtidigt att lfr kom
att bli en ersttningsgestalt ven fr honom, och betecknande nog r det en kvinna som fr
lyssna till hans sista hyllningsord om kungen.
Hypotesen r slunda att det finns en verordnad niv som inrangerar de mer lokala och
tillflliga instabiliteterna, och att vi p den nivn hittar ett antal grundlggande spnningsflt.
Det finns en problematik mellan nytta och bekymmer vad gller de asociala vldsmnnen. Det
finns i fenomenet manlig vnskap en inneboende spnning, som tycks svr att hantera och
som kan f vittgende konsekvenser. Och det finns en motsttning mellan lfrs (kristna)
vrld och de vrldar dit hans inflytanden inte nr, och det r drfr svrt att inte tillmta bety-
delse t det faktum att lfrvrlden bryter samman i sagaslutet.

III
Lt mig n en gng understryka att det vi hittills sagt om systemdynamiken i den fiktiva vrl-
den bygger p dynamiska receptionsprocesser dr textsignalerna lsarkonstrueras i en nutida
diskusvrld. Det r sledes inte sjlvklart att den hypotes jag presenterat ovan har det minsta
att gra med hur sagan uppfattades i sin samtid, men det r samtidigt mjligt att man genom
att stlla den nutida receptionen mot konkreta uttalanden och konkreta frhllanden i texten
skulle kunna sga ngot om de obekanta storheterna i ekvationen textsignaler + dtida
diskursvrld = dtida reception. Och om vi frutstter en lng traditionsprocess kan kanske till
och med interna spnningar och paradoxer tolkas som tecken p att texten delvis inte frstr
sin egen vrld. Risken fr bevisfring i cirkel r frsts stor, men samtidigt ofrnkomlig, ef-
tersom det r omjligt att isolera textsignaler frn lsarkonstruktion och eftersom lsaren p-
verkar texten i lika hg grad som texten lsaren. Det som fljer nedan ska drfr betraktas
som ett antal hypotetiska exempel p spnningar (och sammanfall) mellan den nutida recep-
tionen och textens konkreta uttalanden och konkreta frhllanden.
1) Det finns en viss diskrepans mellan presentationerna av orgeirr och hans verkliga upp-
trdande i sagans frsta del. Han r mycket riktigt stor och stark och besvrlig, men det sgs
ocks att han r oblid mot vanligt folk, att han sllan ler och att han inte bryr sig om kvinnor.
Vi fr allts intrycket av en tystlten, brutal vldsman, men det r inte som sdan vi mter
honom i sagan. Han skrder inte alls orden infr vare sig tjnstefolk, faderns drpare eller

204
modern, han pratar vnligt med Sigrflj och antar med entusiasm hennes uppdrag, i Gervi-
dalr frskrar han den uppbragte orkell att ingen skada ska samkas honom, och med orgils
Msson fr han ett lngre samtal. Det r frst senare som han mer lever upp till karakteristi-
ken, och d bara hemma p Island. Snarare r ormr han som bara presenteras med epite-
ten rask och modig den tystltne och reserverade. De f gnger han kommer till tals r han
den minst tillmtesgende av de bda. Han r inte alls pigg p Sigrfljs planer, och vi har
sett hur han reagerar p orgeirrs, eventuellt frflugna, frslag. Inte heller fortsttningsvis
kommer man honom riktigt in p livet, och hans beteende bde p Grnland och i samband
med slaget vid Stiklastair r d och d minst sagt mrkligt.
2) De s kallade digressionerna har diskuterats ingende och en viktig frga har varit om
de r ursprungliga eller inte. Onekligen ger de ett omedelbart intryck av fristende kommen-
tar, och det r nog knappast i en traditionsprocess som de har sin naturliga upprinnelse. Det
myckna talet om karaktrsegenskaper och inre organ tillfr antagligen fga i en muntlig bert-
tarsituation, och exempelvis karaktristiken av orgeirr i samband med att han fr beskedet
om drpet p fadern knns definitivt pklistrad. Att inte ltsas om nyheten r inte automatiskt
samma sak som att aldrig bli vred eller hatisk eller ilsken.
3) Fstbrra saga innehller s mycket som ett tjugotal dramatiserade ankomstscener
med hlsningar, direkt dialog och specificerad kontext (tid, plats). En frklaring skulle kunna
vara att en frfattare anvnder ett typscensberttande hmtat ur traditionen, men tminstone
sex av ankomsterna r mycket speciella. De innehller frmedlare (tjnstekvinnor eller hus-
karlar) och tta perspektivskiften och r i flera fall oproportionerligt lnga i frhllande till
den viktiga hndelse som de oftast resulterar i (fretrdesvis ett drp). Som spnningsskapan-
de grepp borde de fungera bttre i ett muntligt sammanhang p en ensligt belgen grd i en
vrld dr bultandet p drren i natten utgr en central existentiell anxiety n som enskild
lsning eller upplsning p ting eller brllop.
4) Det finns en myckenhet av drastisk komik i sagan, inte minst i samband med digressio-
nerna (vlbekanta r alla de skakande delarna av den skrckslagne Ffl-Egills kropp), men
parodin och satiren fastnar ofta i halsen. Det r fnen och originalet och trlen och den livrd-
de smbonden som frljligas och fiender som har den dliga smaken att bli ddade eller
skadade p ett skrattretande stt. Nu kan man kanske sga att ven ormr dras in i det hr
genom att han s ofta drar sig kroppsliga men, och riktigt roligt r det frsts nr ords p
Grnland s nr lyckas f honom att avslja sitt gmstlle genom att ifrgastta hans manlig-
het.
5) Mrkligt nog gr kvinnorna i stort sett fria frn frljligande. Epiteten r genomgende
upphjande, och inte minst de trollkunniga skildras med stor respekt sagan fastnar ocks
under en lngre sekvens digressivt i den islndska Grmas perspektiv. I den enskilda situatio-
nen r kvinnorna mestadels bde listiga och frnuftiga. De vet vad de vill och driver sin sak,
vare sig det gller att snrja mn eller driva hmndaktioner, och ven om de inte alltid uppnr
sina ml har man en knsla av att de behller initiativet och hela tiden str klara fr nsta drag
i spelet.
6) Mnnen ger ett betydligt oskrare intryck, och det intressanta r att detta inte kommen-
teras som ngot exceptionellt utan tycks tas fr givet. Nr de hrda mnnen mts r labiliteten
stor, och utgngen svr att frutse. Som nr orgeirr mter Gautr eller Butraldi, eller nr han
begr ngot av sina mest meningslsa drp. Som nr ormr mter Gestr eller Loinn eller
bonden i ladan vid Stiklastair. Och som nr orgeirr och ormr gr skilda vgar. Den hr
oskerheten nrmast tematiseras i konflikten mellan fosterbrderna i lfsdalr, Eyjlfr och
orgeirr.
7) Jag har i ett annat sammanhang Danielsson (2009) underskt slutfasen av kontrover-
sen mellan Eyjlfr och orgeirr och ppekat hur mycket det r som hnger i luften och hur
textsignalerna snarast frsvrar frstelsen. Min frsiktiga slutsats r att avsnittet avspeglar en

205
traditionsprocess dr problematiken varit s drabbande att berttandets verkan lg i sjlva
bristen p svar.

IV
Man kan reagera p olika stt infr en berttelse s komplex som Fstbrra saga. Man kan,
som tidigare inom forskningen, tala om ursprungstexter som blivit interpolerade. Eller man
kan, vilket r vanligare i dag, frska frst de bevarade skrivna versionernas enhet utifrn
tnkestt som var rdande under 1200-talet. En mycket vertygande analys i den traditionen
har lagts fram av Meulengracht Srensen (1993). Slutsatsen dr r att vi str infr en frfatta-
re som hgst medvetet bryter mot en fljd sagakonventioner fr att n det vergripande syftet
att only under the kings rule will the actions of the leaders and men under him be in the ser-
vice of the good (410). ormr nr fram till den insikten och kommer drfr till himlen
efter slaget vid Stiklastair, medan orgeirr lever kvar i den hedniska vrlden och bara blir ett
avskyvrt avhugget huvud efter dden och en direkt orsak till att man inte frmr hindra
Eyjlfr och orgeirr frn att dda varandra i Garpsdalr.
Min tanke r att man mste lta bda vgarna frbli ppna. Vi vet inte om det fanns en tra-
dition vrd namnet, men vi kan heller inte vara helt skra p att den inte fanns. Och om den
fanns kommer vi aldrig att kunna sga med skerhet hur den tog sig ut eller om den lmnat
spr i de skrivna versionerna. Men de flesta strvanden att enhetliggra texter r lika spekula-
tiva, och det r ltt att bli frfrd om argumenten r tillrckligt briljanta. Diskussionen br
hllas ppen och mjligheten inte automatiskt avvisas att sagatexter kan innehlla lager av
olika lder och hrkomst, av fusioner och av senare missuppfattningar av tidigare avsikter.
Personligen misstnker jag att det i Fstbrra saga finns spr av en existentiell problematik
som inte lngre var lika aktuell vid nedtecknandet, och som terspeglas bland annat i an-
komstscenerna, i frhllandet till de oregerliga vldsmnnen, i den svrhanterliga vnskapen
mellan mn och kanske i kvinnosynen. Drmed inte sagt att det r ointressant att bli klar ver
hur en skribent p 1200-talet hanterade det traderade stoffet och gjorde det till sitt.

Referenser
Andersson, Theodore, 2006: The Growth of the Medieval Icelandic Sagas. IthacaLondon.
Danielsson, Tommy, 2009: Social eller existentiell oro? Fostbrdradrp i tv islndska sagor. I: Scrip-
ta Islandica 59. Sid. 7391.
Fstbrra saga. Utg. av G. Jnsson. 1943. Reykjavk. (slenzk fornrit 6.)
Meulengracht Srensen, Preben, 1993: On Humour, Heroes, Morality, and Anatomy in Fstbrra
Saga. I: Twenty-eight Papers Presented to Hans Bekker-Nielsen on the Occasion of His Sixtieth
Birthday 28 April 1993. Odense. (NOWELE 21/22.) Sid. 395418.
1994: Mundtlig tradition i Fstbrra saga. I: Sagnaing helga Jnasi Kristjnssyni 10. april
1994. Utg. av G. Sigursson, G. Kvaran & S. Steingrmsson. Reykjavk. Sid. 581591.
Sigursson, Gsli, 2004: The Medieval Icelandic Saga and Oral Tradition. A Discourse on Method.
Cambridge, Mass.

206
 Editing the Fornaldarsgur Norurlanda
Matthew J. Driscoll, Arnamagnan Institute, University of Copenhagen, Denmark
The fornaldarsgur norurlanda (literally ancient sagas of the northern lands, but often re-
ferred to in English as mythical-heroic or legendary sagas) represent one of the major gen-
res of mediaeval Icelandic saga narrative although to what extent they actually do constitute
a genre remains the subject of scholarly debate (see e.g. Quinn et al. 2006). Unlike many of
the standard saga genre designations slendingasgur, konungasgur etc. which actually
are attested in the medieval literature, the term fornaldarsaga is a modern coinage, first used
by Carl Christian Rafn as the title of his three-volume edition Fornaldar Sgur Nordrlanda,
published in Copenhagen in 182930. Although all but one of the sagas included there had
already appeared in print, Rafns edition brought together, for the first time, essentially all the
prose narratives preserved in Old Icelandic dealing with the early history of mainland Scandi-
navia, i.e. before the unification of Norway under Haraldr hrfagri and the settlement of Ice-
land (Rafn 182930: I, v). Rafns edition thus defined the corpus and gave that corpus its
name in accordance with that definition.
In their present form, the fornaldarsgur are thought to date predominantly from the 14th
and 15th centuries, and are thus regarded as one of the younger genres of saga literature. Most
of them have at least some basis in significantly older tradition, however, and it has been cus-
tomary to distinguish between them on the basis of their relationship to that tradition. Thus
while works such as Vlsunga saga and Hervarar saga ok Heireks, which are demonstrably
related to and/or derived from ancient Germanic poetry, have long been accorded a measure
of scholarly respect, others, such as Egils saga einhenda ok smundar berserkjabana and Bsa
saga, with their fondness for the fabulous, stock characters, lengthy battle scenes and so on,
have often been dismissed as historically unreliable and of scant artistic merit. It was, how-
ever, perhaps not surprisingly, these same sagas which were generally the most popular, as
attested by the very large number of manuscripts in which they are preserved.
The importance of the fornaldarsgur is many-fold. They are, to begin with, a valuable
source of information on the history at least the legendary if not the actual of early Scan-
dinavia. Fornaldarsaga-like narratives were used as a source by Saxo in his Gesta Danorum,
as he himself acknowledges, and the sagas were combed for information about the early histo-
ries of the kingdoms of Denmark and, not least, Sweden, by 17th- and 18th-century scholars:
in fact, the first saga texts ever to be printed in the original were fornaldarsgur, published in
Uppsala in the second half of the 17th century.
The influence of the fornaldarsgur is also to be found in other literary works. Almost all
of them were turned into the lengthy Icelandic metrical romances known as rmur, generally
more than once, and many also formed the basis for ballads in Norway, Denmark, Sweden
and the Faeroe Islands (Mitchell 2003). They have also served as a source of inspiration for
more serious writers. Johannes Ewalds Rolf Krage: et Srgespil (1770) and Adam Oehlen-
schlgers Helge: et Digt (1814) were both based on Hrlfs saga kraka, the former via Saxo,
the latter directly, while Esaias Tegnrs poem Frithiofs saga (1825), praised by Goethe and
famous throughout 19th-century Europe, was based on Frijfs saga ins frkna. Wagner
drew on Vlsunga saga at least as much as he did on the Nibelungenlied for his Der Ring des
Nibelungen (1876). And while specific models are harder to identify, the influence of the for-
naldarsgur on J. R. R. Tolkiens works, the Star Wars films and on modern fantasy in gen-
eral is also considerable.
Unfortunately, study of the fornaldarsgur has long been hampered by a lack of reliable
editions. Recognising this, the Arnamagnan Commission agreed in 1937 that a new edition
of the complete fornaldarsaga corpus should be among its first priorities. A detailed plan for

207
the work was drawn up and an editor for the project, the Icelandic scholar Einar l. Sveins-
son, was appointed in 1939. The advent of the war prevented the editor from taking up his
duties, however, and the project was abandoned (Driscoll 2008). Although a handful of for-
naldarsgur have subsequently appeared in scholarly editions, it is unfortunately still the case
that the majority of them have yet to be edited properly.
What would happen if this project were to be taken up again today? What would a new
edition of the Fornaldarsgur Norurlanda look like anno 2008?

Defining the corpus

The first question which would need to be asked is quite simply what to include. Assuming
that the fornaldarsgur do indeed constitute a genre, how many sagas are to be ascribed to
that genre?
Rafn included in his edition texts of 31 sagas or 32, if Hversu Noregr bygist and Fund-
inn Noregr, which were placed together by Rafn under the title Fr Fornjti ok hans
ttmnnum, are counted separately three of them in two recensions, in addition to the po-
ems Bjarkaml hin fornu, printed with Hrlfs saga kraka, and Krkuml, printed with
Ragnars saga lobrkar. Among these there are several shorter pieces dealing with Scandina-
vian pre-history, such as Af Upplendingakonungum and Hversu Noregr byggist, which were
for the most part taken out of longer compilations into which they had arguably been inter-
polated such as Hauksbk and Flateyjarbk. Their decidedly non-narrative nature is in sharp
contrast to the sagas proper, however, and the justification for their inclusion could certainly
be questioned. At the same time, there are others, specifically Yngvars saga vfrla, Tka
ttr Tkasonar, Helga ttr rissonar and orsteins ttr bjarmagns, which were not in-
cluded by Rafn but certainly could have been, as they conform to his criteria of time and
place. And there are still others which might also be included, for example sagas like la
flekks saga, Hrings saga ok Tryggva, Sigurar saga fts, Sigrgars saga frkna, Vilmundar
saga viutan and jalar-Jns saga; these are normally classed as romances (riddarasgur), but
while set outside Scandinavia proper, they take place in a Viking, rather than a chivalric, mi-
lieu. There is also the question of ireks saga, seen by some as at least closely related to the
fornaldarsgur (e.g. Schier 1970: 8283). ireks saga is in many ways atypical of Old Norse
works and has generally defied generic categorisation; those wishing to place it among the
fornaldarsgur have presumably done so because it is derived perhaps directly translated
from German sources, rather than British or French.
There is also the question of lost fornaldarsgur (see Mitchell 1991: 185). Some of these
are so completely lost that nothing remains of them at all, such as *Hrks saga svarta, which
is named in Geirmundar ttr heljarskinns but of which nothing survives though it is possi-
ble that the saga referred to by this name is in fact identical with Hlfs saga ok Hlfsrekka
(see Seelow 1981: 15859). There are no such references to *smundar saga flagagfu, but
its existence can be inferred by the fact that it was the basis for a set of rmur, also lost. A
fairly lengthy prose summary survives, Inntak r sgutti af smundi flagagfu, written
down by sr. Eyjlfur Jnsson Vllum around 1700 on the basis of stories told him by his
mother and maternal grandmother, but it is not entirely clear exactly what these stories were
based on, whether the rmur, a written saga or, as seems most likely, both (see Jesch 1982).
What does seem clear is that there once existed a fornaldarsaga-like narrative of which this is
the closest representation we have. As such, one might not unreasonably argue for its inclu-
sion in the corpus.
A number of fornaldarsgur survive only in rmur that were based on them, such as Grms
rmur og Hjlmars, also known as Grimlur. These were printed by Birner, along with prose
translations into Swedish and Latin, in his Nordiska Kmpadater (Birner 1737) the first

208
(secular) rmur to appear in print, and the only rmur, to my knowledge, to appear in Latin
translation. There are other examples of this, and one could argue that, in the absence of the
prose texts on which they were based, all such rmur should also be included in the corpus.
In such cases there often are prose texts as well, but these are secondary, in that they are
prose retellings of the medieval rmur, what Peter Jorgensen has called rmur retreads
(Jorgensen 1990; see also Driscoll 1997: 1213, 194205). There is, in fact, a younger prose
version of Grimlur preserved in AM 601 4to, a manuscript which contains a prose version of
Ormars rmur, which were also based on a lost fornaldarsaga. Here the situation is even more
complicated, as there is also a younger ttur or vintri preserved in AM 119 8vo and some
half-dozen manuscripts in Landsbkasafn; this ttur was then the basis for a younger set of
rmur, composed in 1833 by Sigurur Jnsson Reykjum (Bjrn Karel rlfsson 1934: 336
38 and 41618).
Probably the best known example of the rmur retread phenomenon is Hrmundar saga
Gripssonar, which was one of the sagas included by Rafn in his edition. Although there is
evidence for the existence of a saga by this name in the medieval period the famous wed-
ding feast at Reykjahlar in 1119 (Foote 195357) this saga has not survived, and the text
printed by Rafn is a late 17th-century prose version of the rmur known as Griplur, which
were themselves based on that lost saga (Jesch 1984). A similar case is provided by Haralds
rmur Hringsbana, which are thought to have been composed in the first half of the 15th cen-
tury on the basis of a lost fornaldarsaga (lafur Halldrsson 1973). There is a younger saga,
probably written in the 17th century. This saga was not, according to Bjrn Karel rlfsson
(1934: 4057), based on the rmur, but rather on the older saga. This younger saga was in turn
the basis for two further sets of rmur (Finnur Sigmundsson 1966: I, 2046).
Yet another example is lfhams saga, recently edited in admirable fashion by Aalheiur
Gumundsdttir (2001). The saga exists in three distinct versions, the earliest from the 17th
century, the youngest from the 19th. All derive, directly or indirectly, from lfhams rmur,
also known as Vargstkur, which are thought to have been composed in the beginning of the
15th century though exactly on the basis of what is unclear.
As we have seen, the existence of medieval rmur does not always guarantee that a corre-
sponding prose narrative also existed in written form in the middle ages. There is also a sig-
nificant number of post-medieval fornaldarsgur, works which were certainly written after
the Reformation, generally on the basis of older material, in particular Saxos Gesta Danorum.
There are almost as many sagas of this type as there are proper fornaldarsgur certainly
some 25. While some are only found in one or two manuscripts, others were very popular
indeed. A few even managed to find their way into print, chiefly in cheap, popular editions
from the second half of the 19th century or first decades of the 20th. One such is Sagan af
Starkai Strvirkssyni gamla, which was written by Snorri Bjrnsson (17101803) on the
basis of Saxos Gesta Danorum, Gautreks saga, Heimskringla and the Sgubrot af fornkunun-
gum, with verses in all probability by Gunnar Plsson (171491), of which a popular edition
appeared in Winnipeg in 1911. Though some of these sagas are mentioned in works such as
Margaret Schlauchs ground-breaking study Romance in Iceland (Schlauch 1934), only a
handful have been the subject of detailed scholarly investigation, notably Rosemary Powers
fine article Saxo in Iceland (Power 1984; see also Driscoll 2003). Otherwise, where they are
mentioned at all, they are usually dismissed as spurious, something entirely different from
the fornaldarsgur of the middle ages, certainly nothing to be taken seriously. And yet they
are quite clearly part of the same tradition, a tradition which, arguably, continued unbroken
from the (early) medieval period until the end of the 19th century. For this reason they too,
one could say, deserve inclusion in the corpus.

209
 Editorial principles
The foundation of any scholarly edition is an examination of all the surviving texts, or wit-
nesses as they are known in traditional textual criticism, a thorough interrogation of which
will bring one as close to the original as it is possible to get. Even limiting oneself to the
classic corpus of 36 sagas, viz. the 31 included by Rafn plus Helga ttr rissonar, Yngvars
saga vfrla, Tka ttr Tkasonar, jalar-Jns saga and orsteins ttr bjarmagns, there
are a lot of witnesses to be interrogated: at last count 1542 texts, contained in a total of 779
individual manuscripts (Driscoll & Hufnagel 2009), giving an average of just a fraction under
two texts per manuscript. Of these, just over 100 are defective or one way or another, while
just under 100 are fragments, i.e. where more than half the text is missing. Several contain
only the very beginning of ending of the saga, in some cases obliterated so thoroughly that
nothing can be read. Not infrequently this was done by none other than rni Magnsson him-
self, who split up a number of manuscripts containing more than one saga (and in such cases
always made an exact one trusts, for generally there is now no way of checking copy of
the text he had eradicated). Extracts or excerpts are found in 23 cases, while about 120 are, or
contain alongside the Icelandic text, translations into other languages, predominantly Swedish
and Latin; these have never, to my knowledge, been the subject of scholarly investigation but
are potentially of great interest, if only because some may be translations of manuscripts no
longer extant.
Most of these manuscripts are, or can be, dated and are written by identifiable scribes. The
distribution of manuscripts and texts by century is as follows:

Table 1.Distribution of fornaldarsaga manuscripts and texts by century

Century MSS Texts Texts/MS
XIV 1.0% 1.3% 2.8
XV 3.0% 3.9% 2.5
XVI 0.9% 0.6% 1.8
XVII 28.7% 19.7% 2.1
XVIII 43.0% 41.6% 1.9
XIX 21.9% 0.9% 1.9
XX 1.2% 0.9% 1.3

As is immediately apparent from this table, the vast majority of the extant manuscripts con-
taining texts of fornaldarsgur are from after the Reformation, with nearly half coming from
the 18th century. This pattern of distribution is probably not dissimilar to that of other saga
genres, though in the absence of more large-scale statistical analyses it is difficult to draw any
conclusions with any degree of certainty. One reason for this pattern of distribution, though, is
certainly the great increase in popular literacy in Iceland in the course of the 18th century,
with something like universal literacy being achieved by the end of the century, which led to
an increase in literary activity generally (Loftur Guttormsson 1989; Sigurur Gylfi Magns-
son and Dav lafsson 2002). At the same time, not all the manuscripts included here were
produced in Iceland, many having been copied, usually by Icelandic students, in Denmark or
Sweden for use by Scandinavian antiquarians. Even so, the bulk of fornaldarsaga manuscripts
from the 17th and 18th centuries were produced in Iceland itself, apparently for domestic con-
sumption, something which cannot be entirely unrelated to the interest in this material in the
rest of Scandinavia; it would be nice to know exactly how.
While a very large number of these manuscripts are obviously valueless from a tradi-
tional textual-critical point of view, in that they are and often admit to being copies of
extant manuscripts or, in not a few cases, printed editions, they are certainly not without their
interest. The editorial project envisaged by the Arnamagnan Commission in 1937 involved

210
an examination of all the extant witnesses, principally, in keeping with the precepts of the
nascent Arnamagnan School, with an eye toward identifying the best text, i.e. that which
was as close as possible to the works original form. In the last three decades or so, not least
with the advent of the so-called new philology, there has been less focus on origins and
more on the processes of literary production, dissemination and reception, with the result that
texts which would hitherto have been rejected as unreliable, corrupt and worthless can now be
seen as valuable sources of information on these very processes (Driscoll in spe). Anno 2008,
one would still want to examine the extant texts, but with an eye toward charting the entire
process of transmission and identifying interesting textual manifestations of the works in
question, including, but in no way limited to, those which best represent their oldest identifi-
able forms. One would want to describe and transcribe the individual textual artefacts as care-
fully as possible, but also link them to other artefacts preserving texts of the same (and other)
works. More importantly, one would want to map the relationships between these artefacts
and the people who produced and consumed them, to show how the manuscript matrix
worked. One would then try to present all this material as part of a dynamic, interactive digital
text archive, rather than as static, read-only texts on the page (or screen), though printed texts
for simple reading could easily be generated from the archive on demand. Fortunately, the
technological architecture to do this exists: it is known as Web 2.0. Only in this way, it
seems to me, can we do this vast and utterly fascinating body of material any justice.

Bibliography
Aalheiur Gumundsdttir (ed.), 2001: lfhams saga. Reykjavk.
Bjrn Karel rlfsson, 1934: Rmur fyrir 1600. Reykjavk.
Bjrner, Erik Julius (ed.), 1737: Nordiska kmpadater. Stockholm.
Driscoll, M. J., 1997: The unwashed children of Eve: The production, dissemination and reception of
popular literature in post-Reformation Iceland. London.
Driscoll, M. J., 2003: Fornaldarsgur Norurlanda: The stories that wouldnt die. In:
Fornaldarsagornas struktur och ideologi, ed. rmann Jakobsson, Annette Lassen & Agneta Ney.
Uppsala. pp. 25767.
Driscoll, M. J., 2008: Plans for a new edition of the Fornaldarsgur, anno 1937. In: Fornaldarsagaerne:
Myter og virkelighed, ed. Agneta Ney, rmann Jakobsson & Annette Lassen. Copenhagen. pp. 17
25.
Driscoll, M. J., in spe: The words on the page: Thoughts on philology, old and new. In: Creating the
medieval saga: Versions, variability, and editorial interpretations of Old Norse saga literature, ed.
Judy Quinn & Emily Lethbridge. Odense.
Driscoll, M. J. & Silvia Hufnagel (2009), Fornaldarsgur norurlanda: A bibliography of manuscripts,
editions, translations and secondary literature, http://www.staff.hum.ku.dk/mjd/fornaldarsagas/.
Finnur Sigmundsson, 1966: Rmnatal. Reykjavk.
Foote, Peter, 195357: Sagnaskemtan: Reykjahlar 1119, Saga-Book 14. pp. 22639. Repr. in
Aurvandilst: Norse Studies. Odense, 1984. pp. 6583.
Jesch, Judith, 1982: smundar saga flagagfu. Arv 38. pp. 103131.
Jesch, Judith, 1984: Hrmundr Gripsson revisited. Skandinavistik 14. pp. 89105.
Jorgensen, Peter A., 1990: The neglected genre of rmur-derived prose and post-Reformation Jnatas
saga. Gripla 7. Reykjavk. pp. 187201.
Loftur Guttormsson, 1989: Lsi. In: Munnmenntir og bkmenning, slensk jmenning 6.
Reykjavk. pp. 11944.
Mitchell, Stephen A., 1991: Heroic sagas and ballads. Ithaca, NY.
Mitchell, Stephen A., 2003: The Fornaldarsgur and nordic balladry: The Smsey episode across
genres. In: Fornaldarsagornas struktur och ideologi, ed. rmann Jakobsson, Annette Lassen &
Agneta Ney. Uppsala. pp. 24556.
lafur Halldrsson (ed.), 1973: Haralds rmur Hringsbana. Reykjavk.
Rafn, Carl Christian (ed.), 182930: Fornaldar Sgur Nordrlanda. Copenhagen.
Schier, Kurt, 1970. Sagaliteratur. Stuttgart.
Schlauch, Margaret, 1934: Romance in Iceland. New York.

211
Seelow, Hubert (ed.), 1981: Hlfs saga ok Hlfsrekka. Reykjavk.
Sigurur Gylfi Magnsson & Dav lafsson, 2002: Barefoot historians: Education in Iceland in the
modern period. In: Writing peasants: Studies on peasant literacy in early modern northern Europe,
ed. Klaus-Joachim Lorenzen-Schmidt & Bjrn Poulsen. rhus. pp. 175209.
Power, Rosemary, 1984: Saxo in Iceland, Gripla 6. pp. 24158.
Quinn, Judy et al., 2006: Interrogating Genre in the Fornaldarsgur. Viking and Medieval Scandinavia
2. pp. 27596.

212
 Anatomies off the Map: Secret and distant freaks and the
Authorization of Identity in Medieval Icelandic and Irish Lit-
erature
Amy Eichhorn-Mulligan, Dept. of English, Univ. of Memphis, USA
In 1185 Giraldus Cambrensis wrote:

Just as the countries of the East are remarkable and distinguished for certain prodigies peculiar
and native to themselves, so the boundaries of the West also are made remarkable by their own
wonders of nature. For sometimes tired, as it were, of the true and the serious, she [Mother Na-
ture] draws aside and goes away, and in these remote parts indulges herself in these secret and
distant freaks (OMeara 1982: 31).

Fig. 1. National Library of Ireland, MS. 700, fol. 48r.

213
Giraldus parallels the remarkable wonders or monstrous races for which the East was re-
nowned with figures from the western geographic extremes of Ireland and Iceland. In a map
attributed to Giraldus (Fig. 1) he situates both Ireland (Hybernia) and Iceland (Yslandia) as
outsider lands. England is enclosed in mainland Europes embrace while Iceland, and Ireland
are set beyond those community defining European boundaries (Lavezzo 2006: 689). Ca.
1200, about the same time Giraldus is writing, the Danish chronicler Saxo Grammaticus as-
serts that Iceland is a land very squalid to dwell in, but noteworthy for marvels, both strange
occurrences and objects that pass belief. A spring is there, which by the malignant reek of its
water, destroys the original nature of anything whatsoever[] (Elton 1893: 1011). Saxo
(writing, like Giraldus, in the prestige language of Latin) characterizes Iceland as a strange
land of marvels and such terrifying things as springs that erase identity. Giraldus and Saxos
statements suggest that the marginalized westernmost lands possess dangerously other be-
ings and wonders that rival the more famous Eastern monstrous races.
When we shift westwards to these remote parts, we see that Giraldus and Saxo have a
point: Irish and Icelandic texts are rich in prodigies of nature that rear their ugly heads, but
instead of operating as hideous monsters designed to shed glory on their slayers, the heroes
are the monsters. This may be seen as proving propagandistic claims about Icelandic and Irish
otherness and dangerous subhumanity. What Irish and Icelandic authors do, however, is give
their monstrous heroes articulate, intelligent voices and functions, and endow their abnormal-
ity with an impressively informed logic. Two figures comprise my main examples: the Ice-
landic poet-warrior Egill Skalla-Grmsson, in Egils saga Skalla-Grmssonar (13th c.) and the
Irish hero C Chulainn from the mid-12th c. Book of Leinster Tin B Cuailnge (The Cattle-
Raid of Cooley.) Egils saga and the Tin spend an unusual amount of time on the heroes
ugly appearances and aberrant bodily behavior, and while both characters violate bodily
norms, they are nonetheless celebrated for the ways they advertise and display their dangerous
abnormality. By situating Irish and Icelandic secret and distant freaks like C Chulainn and
Egill within the context of the politically tumultuous 12th and 13th centuries, we see how their
authors take the terms and images of their othering and re-configure them to endow a heros
non-normative body with legitimating power and intellectual cohesion. In other words, nei-
ther the Irish nor Icelanders allow themselves to be disenfranchised by propagandistic rhetoric
like that of Giraldus or Saxo. Rather, they show their mastery of that type of discourse and its
forms, and are able to use that language and body of images as an enabling and empowering
discourse. As Houston Baker argues in the influential Modernism and the Harlem Renais-
sance (1987), this is exactly what early-20th c. African-American writers and performers were
doing when they inverted and re-formed the terms and images of blackface minstrelsy, mas-
tering those forms for their own purposes. Thinking about the theories of mastery of form and
de-formation (or construction of new forms) as articulated by Baker, I argue that C Chulainn
and Egill are deployed to create a specific vision of a hero that, to use Bakers punning lan-
guage, is both gorilla (animal making a show of pounding its chest, bellowing and strutting
around in order to guard his territory) and guerrilla (effective in camouflage, in resisting
easy decoding) C Chulainn and Egill are able to float like a butterfly in order to sting like
a bee, (Baker 1987: 50) and terrify the opponent with their fierce appearances.
Set in a heroic past syncretized with Christ, the Book of Leinster (LL) Tin is nonetheless
from a much later period. Compiled between 1152 and 1161, the manuscript was completed
shortly before the Norman Invasion of Ireland of 1169, but after the groundwork had been
laid and changes had begun to take place. These destabilizing events contributed to the compi-
lation of massive manuscripts (like LL) whose roles were to preserve native traditions as po-
litical and cultural order and literary institutions were being eroded with troubling speed.
Written in the early 13th c., at the end of the Icelandic Commonwealth or Free State period
(9301262/4), Egils saga is also situated in a fraught historical and political context. During

214
the final fifty years of the Commonwealth period, the Norwegian crown exerted a great deal
of pressure on the Icelanders, economic, ecclesiastic and political, and in 12624, twenty
years after the assassination of Snorri Sturluson, the King of Norway succeeded in annexing
Iceland. Written during this tumultuous historical period, and possibly the work of Snorri
himself, while Egils saga is mainly about the Icelander Egill Skalla-Grmsson (born ca. 910),
the saga also spans four generations and provides an extended study of one familys problems
with the ambitious kings of Norway, problems very relevant to 13th c. Icelanders faced with a
Norwegian king who seeks their incorporation into his kingdom. The Mruvallabk text
cited here is, as a mid-14th c. manuscript, furthermore set in the context of Icelands 1380 in-
corporation into the Danish Empire Egils saga and its manuscripts are inextricably bound
up in issues of Icelandic independence (and its loss). Indeed, Egill (like C Chulainn) is mar-
shaled throughout the ages as a symbol of Icelandic pride as an Icelander born in 1924 said,
the sagas and their heroes justified our striving for recognition as a nation (Jn Karl Hel-
gason 2005: 75). Both the Tin and Egils saga are very much linked to Irish and Icelandic
national identity, and are suitable texts for exploring how Irish and Icelandic authors re-
sponded to colonialism and its dehumanizing rhetoric. Indeed, while textual depictions of the
bodies of Egill and C Chulainn do come dangerously close to proving claims about the ex-
treme Western lands being populated with freaks, strategies can also be detected that show
how the Irish and Icelanders shift from being othered on account of their monsters to using
their monstrous heroes to advertise their own power.
Egils saga opens and closes with statements that highlight unusually ugly appearances and
semi-monstrous lineage. The saga begins with Egills grandfather, lfr (nicknamed Kveld-
lfr (Evening-Wolf), son of Bjlfi (Animal-Pelt,) whose mother, Hallbera (Gemstone-
She-Bear,) is daughter of another lfr (Wolf) and sister to a Hallbjrn hlftroll (Gemstone-
Bear half-troll) (Sigurur Nordal 1933: 3). Kveld-lfr (Evening-Wolf) is suspected of be-
ing a shape-changer which, given a pedigree that includes wolves, bears and trolls, is not en-
tirely surprising. The saga furthermore shows Kveld-lfr going into berserk fits or frenzies,
which becomes another family trait, and one which Egills father, and to a lesser extent, Egill
himself, shows evidence of undergoing, this warrior frenzy also enacted by the Irish C Chu-
lainn. Egills grandfather Kveld-lfr marries the daughter of an even more thoroughly berserk
warrior. While connotations of berserks in Norse sources are mostly negative berserks are
typically dishonorable thugs-for-hire, incredibly strong but rather stupid in the case of
Egills family, the saga author combines berserk attributes and dangerous behavior with their
intelligence, political savvy and economic successes as landed farmers, which moves the allu-
sions to berserks and shape-shifters at least partially away from the usual set of damning
stereotypes.
The family genealogy highlights a cross-generational pattern of a handsome, fair and pro-
royal son contrasted by a dark, ugly, and anti-social son. Kveld-lfr and his wife have two
children, the good-looking and well-liked rlfr, and Skalla-Grmr (Bald Grim), a dark
and ugly man [svartr mar ok ljtr,] like his father both in looks and temper (Nordal: 5.)
Later in the saga, Skalla-Grmr and his companions pay an unfriendly visit to the King of
Norway who is told upon their arrival that Some men have arrived outside [] if they
should be called men: they are more like ogres [ursum] in size and appearance than human
beings (Nordal: 63.) Skalla-Grmr marries Bera (She-Bear), and they have two daughters
and two sons, another light-colored, handsome and charismatic rlfr, and Egill, who, the
saga tells us that As he grew up, one could soon see that he would become like his father,
terribly ugly [mjk ljtr] and with black hair [svartr hr] (Nordal: 80.) The saga stresses
the triad of ugliness, martial prowess and intelligence in its last lines: It remained true for a
long time in that family, that the men were strong and great warriors [sterkir ok vgamenn
miklir], and some were highly intelligent. It varied greatly, though, so that in that family were

215
some of the most handsome men to have ever been born in Iceland [] but, most of the
Mramenn were superlatively ugly [ljtastir] [] (Nordal: 299300.) Where most sagas
genealogies focus on wealth, good birth, or intelligence (as this does in part), here we also see
an emphasis on, and perhaps pride in, extreme ugliness.
A willingness to defy authority using the fearsome appearance he has inherited character-
izes Egills actions throughout life. In one scene Egill bodily intimidates the English king
Aalsteinn into paying the honor-price for his brother rlfr:

Egill had strongly marked features [mikilleitr]: a broad forehead, heavy eyebrows, a nose not
long but exceedingly thick, a wide, long beard, a chin as extremely broad as his jaws, a stout
neck and broad-shoulders, more so than other men, [and he was] harsh- and fierce-looking,
when he was angry [harleitr ok grimmligr, er hann var reir]; he was well-proportioned
and taller than other men, and had thick wolf-gray hair, but went bald-headed at an early age; he
sat, as just described, and dragged one eyebrow down to the cheek [ hleypi hann annarri
brninni ofan kinnina], and the other one up to the roots of his hair [en annarri upp hrrtr];
Egill was black-eyed and with crooked brows [] and alternately, he shot his eyebrows down
and up. (Nordal: 1434).

Sufficiently alarmed, Aalsteinn pays up, and Egill composes a poem of thanks heavily reli-
ant on appearance: Knttu hvarms af harmi / hnpgnpur mr drpa,/ n fann ek anns en-
nis/ slttur r rtti []( Nordal: 145). (In bitterness my brows / beetled over my eyes; /
Now my forehead has found one/ To smooth its furrows) (tr. Plsson & Edwards 1976:130.)
The king wants Egill to stay on, yet when Egill refuses the two nonetheless part the best of
friends (Nordal: 145). Egill uses his body to intimidate the king, and they come to an agree-
ment on Egills own terms the saga here, and elsewhere, shows how easy it is for Egill to
gain the respect of important foreign kings, for a mere Icelandic farmer to rub shoulders with,
and largely dictate the terms of engagement to a powerful monarch. His threatening body, and
the way he advertises his badness (Baker: 50) is shown to be persuasive and effective. It is
worth recalling that the good-looking, agreeable brothers die early and not entirely honorably,
the first rlfr killed by the king and the second dying as Aalsteinns mercenary. The saga
shows that ugly, almost inhuman appearance is an advantage, a powerful and persuasive tool
that gives Egill authority and the ability to survive and thrive.
A later episode in which Egills bodily behavior is at stake vividly shows, on one hand,
that Egill cannot be taken anywhere, that hes an ill-bred, backwater lout who belongs on
the margins among secret and distant freaks. On the other hand, with careful reading we can
see this account as cleverly resisting easy dismissal of the Icelanders, and legitimizing Egill
by paralleling him with inn, god of poetry and war. After being misled by the king of
Norways envoys, Egill and his companions arrive, cold and hungry, at the treacherous r-
mrs farmstead. rmr says all he has are curds, and no ale. His defiant wife sends Egill a
message that there is good food and drink, and at Egills protests, rmr begrudgingly has
fine dishes served and strong ale poured. It becomes a heavy-going drinking contest, Egill
eventually drinking on behalf of his over-sated men, until unable to continue, he

stood up, walked across the floor to rmr, put both hands on his shoulders and pressed him
up against the pillar, then heaved up a vomit of massive proportions that gushed all over r-
mrs face, into his eyes, nostrils and mouth, and flooded down his chest so that he was almost
suffocated. When he recovered his breath he spewed up and all of his servants there began to
swear at Egill. What hed just done, they said, made him the lowest of the low, and if he wanted
to vomit he should have gone outside, not made a fool of himself inside the drinking hall [].
Then Egill went back to his seat, sat down and asked for a drink. After that he recited this verse
at the top of his voice: With my spew I swear / Thanks for your sociability!/ We have wit-

216
 nesses that/ I could walk the floor:/ Many a guests gift/ Is even more gushing;/ Now the ale has
ended up/ All over rmr. (Nordal: 2256; tr. Plsson & Edwards: 1878)

We do not need the servants comments to recognize that, despite its fitness as a punishment
for the bad host rmr, this is not a shining moment for the Icelandic hero, almost suffocat-
ing someone with vomit not a noble mode of confrontation. However, a different sense of
Egill and his actions obtains when we read this scene in terms of the mead of poetry myth as
recorded by Snorri Sturluson in Skldskaparml. Since Snorri may also be author of Egils
saga, connections between the two texts are particularly relevant. Snorri explains that the gi-
ant Suttungr set his daughter Gunnl (in a parallel position to rmrs wife and daughter)
the task of guarding the divine poetic mead. In order to win it for the sir, inn sneaks into
Suttungrs mountain stronghold, and Gunnl agrees to reward inn with a draught from
each container of mead for every night he sleeps with her. With a massive thirst he drains
each successive vessel, and

[] turned himself into the form of an eagle and flew vehemently [] when inn came in
over sgarr, then he spat out the mead into the containers [] inn gave Suttungrs mead to
the sir and to those people who know how to make verse. Therefore we call poetry inns
booty and discovery, as well as his drink, his gift and the sirs drink (Faulkes, ed., 1998: 4.38
5.8).

Egill is, of course, cleverly drawn in terms of inn, the divine hero who secures the poetic
drink of the gods. Egill turns a trick and attempted ambush by the kings men into an occa-
sion to showcase the Icelanders superiority in terms of gorilla-esque drinking machismo that
is simultaneously suggestive as a humorous, yet accurate, rewriting of the mead of poetry
account Egill literally delivers the poetic mead to his host, rmr, in Odinic fashion, by
spitting it up. In this situation the guest-host, poet-patron relationship has been poisoned, and
the usual praise poem becomes both a torrent of vomit and a biting skaldic verse. The next
morning, Egill rises with the dawn to kill Armr in revenge, but because the wife and daugh-
ter intervene, Egill agrees to only cut Armrs beard and, more severely, blind him in one
eye, marking him as a sacrifice to one-eyed inn. Egill presents himself as a god of poetry,
as an inn figure, and claims not only literary authority, but the moral high-ground as well,
as one whose grotesque act is far less savage than the Norwegian rmrs murderous viola-
tion of hospitality.
Egills physical appearance (inherited and explicitly his), as well as his bodily actions con-
stitute an advertisement of his dangerous power to a range of non-Icelanders. Additionally,
Egils saga, much of which is set abroad, is also a performance for an Icelandic readership it
is crafted to create a specific vision of an Icelandic hero, to authorize an identity that, citing
Houston again, is both gorilla and guerrilla. As such an aberrant and powerful, secret
and distant freak, Egill Skalla-Grmsson is heroically successful.
Shifting to the parallel Irish figure, C Chulainn strategically deforms his body and its be-
havior, his heros form a highly structured and richly allusive gorilla/guerrilla space. Like
Egill, C Chulainn puts on the mask of the secret and distant freak, though it is by donning
this very mask that C Chulainn, and the scribes who write his body into being, insist on his
superiority as a national defender and hero. When enraged and preparing for conflict, C Chu-
lainns body undergoes an extensive and fantastic contortion:

A crooked bout of destruction [dberge] fell upon his body in the center of his skin. His feet and
his shins and his knees shifted till they were behind him. His heels, his calves and his buttocks
moved till they were in front of him. The sinews of his calves moved around till they were on

217
 the front of his shins, so that each huge, knotted muscle of them was as thick as a warriors
round fist (ORahilly, ed. 1967: ll. 2344 56, tr. my own).

C Chulainns retropedism links him to the family of Eastern prodigies popularly known as
Antipodes or Retropedes, and described by Isidore of Seville as having their soles turned
around behind their legs (Etymologiae XI.iii.24). Pliny states that retropedism enabled great
speed (Historia naturalis VII.ii.30), and one specifically Irish pair allowed movement around
all of Ireland in a single day. These highly useful, almost doglike legs would allow C Chu-
lainn (whose name means Hound of Culann) to speedily move around the country and
guard its borders. The next part of the description also highlights a link to the monstrous
races:

The sinews of his crown were dragged to the hollow of his neck so that each of them was the
size of a one-month-old childs head[] Then he made a red bowl (?) from his face and coun-
tenance. He sucked one of his eyes into his head in such a manner that a wild crane could hardly
have reached in to pluck it out from the back of his skull onto his cheek. The other [eye] sprang
to the outside of his cheek (ORahilly: ll. 2271 76.)

A single-eyed appearance is associated in Ireland with corrguinecht (crane-slaughter,) or

casting the evil-eye while uttering verse satire, sometimes on the battlefield (Kelly 1997:
128), which is closely linked to one-eyed warrior inn, and the eye-brow contortion Egill
effects. Monocular C Chulainn also recalls the Cyclopes, with one eye in the middle of their
foreheads [] (Isidore, Etymologiae X.iii.16) a well-known monstrous race represented, for
instance, in the Icelandic Physiologus MS (Halldr Hermannsson 1938).
The descriptions of contortions resulting in retropedism and cyclopism do similar work.
Giraldus opening quote stressed the orientalizing relationship between the Easternmost and
Westernmost wonders of nature: here the scribes make their hero retropedal and cyclopean
this suggests a link between C Chulainn and the widely attested monstrous races of the
East. To some extent the Irish writers internalize the arguments about their peripherality that
propagandistic accounts suggest. At the same time, they also subvert that otherness by al-
lowing these unusual attributes to transform their defender C Chulainn into an unparalleled
hero what was seen as negative, other and foreign become empowering, speed-enabling, and
evil-eye endowing attributes that make his opponents quiver.
The next stage of the distortion features animalization as C Chulainns mouth is trans-
formed from a civilized organ, the locus of speech, into a grotesquely distended snapping jaw,
and his innards confront us with a proleptic spectacle of being devoured:

His mouth was contorted fearsomely. He dragged back the cheek from the jawbone until his
gullet was revealed. His lungs and his liver came up till they were bounding in his mouth and in
his throat. His upper palate struck a lion-felling blow (?) against the lower palate, and each
stream of fire which washed into his mouth out of his throat was as wide as the skin of a three-
year-old ram. The resounding blows of his heart against his rib-cage sounded like the howling
of a slaughter-hound or like a lion overpowering bears. The torches of the Badb [war-goddess],
poisonous clouds and furious sparks of fire, were seen in the air above his head as the boiling
angry rage rose from him (ORahilly: ll. 2276 85.)

These images provide a direct link to a crucial early scene in C Chulainns life. The boy ar-
rives late to the smith Culanns fort to find that the massive hound has been set outside to pro-
tect the fort and attack all who approach. When the bloodthirsty hound sees the boy it
stretches its gaping mouth back to swallow him whole past the wall of his chest and the
breadth of his throat and the midriff of his breast(ORahilly: ll. 8801.) The boy-hero foils

218
the hounds plans to consume him and tears the animal apart. While a puppy is trained as
watchdog, the boy himself replaces Culanns hound and is consequently renamed C Chu-
lainn (Hound of Culann.)
Returning to the distorted form C Chulainns body takes when warding off a threatening
opponent, we see that his violently pounding palate, gaping mouth and chest copy the chop-
ping jaws of Culanns watchdog, and his beating heart is also likened to the sound of a
hounds baying, which further reinforces the canine nature of his deformity, the animalistic
links having a similarly dehumanizing effect that the animal-heavy pedigree does for Egill.
The distorted and fearsome C Chulainn boils over with fiery energy, flaming torches rising
from his body, and is given a heart that beats with the intensity of a hounds baying or a lions
roar. This Irish Hound of Culann is celebrated as the human watchdog and guardian of Ul-
sters borders, the figure who takes on the mask of the hound when necessary to defend his
people. C Chulainns body is built logically, and the elements of his description encompass a
wide range of attributes featured on both animals and monstrous figures of the East. There is
evident effort invested in projecting and advertising C Chulainns physical aberrance as
powerful and dangerous, as an entry in the Annals of Connacht demonstrates. C Chulainns
aberrance, his gorilla-esque fearsomeness, is seen as so effective that when the Irish Annalists
describe the Irish warrior Aed OConnors 1256 fight against the Anglo-Norman lord Walter
de Burgh, they depict Aed as a latter-day, fire-emitting C Chulainn:

the warriors of the host on that field could not look on the face of the high lord [Aed], for two
great wide-glancing torches were flaming and flashing in his head, so that all feared to speak
with him [] he uttered his high kings war cry and his champions shout in the midst of the
fight (Lydon 1988: 5960.)

C Chulainn and Aed are proclaimed in their abnormality, and through them the monstrous,
flaming body becomes a site of textual prominence, military power, and heroic celebration.
Their very non-normative attributes are here explicitly deployed as the terrifying face that the
native Irish warriors show to Anglo-Norman invaders of Ireland.
Both C Chulainn and Egill have bizarre bodies that effectively terrify and demonstrate
their power. By giving C Chulainns distorted body a narrative logic that could be decoded,
at least by an educated Irish readership, the scribes to some extent systematize and validate
his most monstrous features. Similarly, by depicting some of Egills most outrageous bodily
behavior in terms of inn and important myths about poetry, his body also gains legitimacy.
Despite rhetoric that shows Ireland and Iceland, and their inhabitants, occupying marginal,
almost inhuman territories, the two aberrant heroes Egill and C Chulainn are both endowed
with a kind of sophisticated body logic that makes them forces to be reckoned with, and not
just ugly thugs from out in the European sticks.
It is telling to briefly ponder the afterlives of Egill and C Chulainn as ambiguous heroes
that their respective narrative traditions seek to recuperate. In one tale C Chulainn is raised
from the grave by Saint Patrick himself, repents of his earlier sins, and requests a place in
Heaven, which Patrick actually grants. It is significant that the storytellers go to such lengths
to anachronistically have C Chulainn given a divine blessing by St. Patrick himself. Anxie-
ties about the conflicted form and nature of Egill also emerge at the close of Egils saga, which
records that under the altar were found human bones; they were much bigger than other hu-
man bones. People were certain that, on account of the stories of old men, they must have
been Egills bones. A priest places Egills exceptionally large and heavy skull, entirely
ridged on the outside like a scallop-shell, on the churchyard fence, and to determine the
thickness of the skull, he took a weighty hand-axe and heavily struck the skull to break it, yet

219
 it whitened, and did not become dented or cracked, and one can tell from this that the skull
would not have been easily damaged by blows from small persons while scalp and flesh were
on it. Egills bones were interred down in the outer part of the churchyard at Mosfell (Nordal:
29899).

The bones are relics celebrating the aberrant nature of Egill and his body strangely shaped,
troubling, but impressive and stubbornly refusing defeat or fragmentation. However, while his
bones are allowed into the churchyard, marking an acceptance, they are still put on its edges.
The Tin and Egils saga endow Egill and C Chulainn with authority and power that lasts
even after their deaths. But, there is still hesitation on the part of the scribes and manuscript
copyists who write about Irish and Icelandic heroic non-normativity. The scribe who con-
cludes the Book of Leinster Tin gives us a fascinating look into the power and pull between
celebration and censure. The scribe, in Irish, writes: A blessing on every one who shall faith-
fully memorize the Tin as it is written here and shall not add any other form to it. Just be-
low that though, the same scribe, in the same hand, switches to Latin and records:

But I who have written this story, or rather this fable, give no credence to the various incidents
related in it. For some things in it are the deceptions of demons, others poetic figments; some
are probable, others improbable; while still others are intended for the delectation of foolish
men. (ORahilly: ll. 491925, tr. p. 272.)

The scribe begins with a blessing, in the Irish language, for those who keep this version of the
Tin and its remarkable hero C Chulainn alive and in circulation. The same scribe that en-
ables pride in native literature and characters, and furthermore provides the increasingly
threatened textual means for celebrating these anatomies that are off the map, shifts to the
prestige language of Latin and denounces the lively hero and epic as entertainment for fool-
ish men, a comment that might be read as speaking to the complex psychology of conquest
and colonialism. Despite the scribes caveats, however, it is important to devote attention to
Icelandic and Irish self-authored bodies, often overlooked in discussions about medieval post-
colonialism and alterity, and when encountering characters like Egill and C Chulainn who
demonstrate a mastery of form through deformation, to consider how they operate as re-
sponses to the disempowering takeovers of Ireland and Iceland that began in the 12th and 13th
centuries.

Bibliography:
Elton, Oliver (ed., tr.), 1893: Saxo Grammaticus: Danorum Regum Heroumque Histori. London.
Faulkes, A. (ed.), 1998: Snorri Sturluson: Skldskaparml. London.
Halldr Hermannsson (ed.), 1938: The Icelandic Physiologus. Islandica 27. Ithaca, NY.
Jn Karl Helgason: Continuity? The Icelandic Sagas in Post-Medieval Times. In A Companion to Old
Norse-Icelandic Literature and Culture. Ed. R. McTurk. Malden, MA. Pp. 6481.
Kelly, Fergus, 1997: Early Irish Farming. Dublin.
Lavezzo, Kathy, 2006: Angels on the Edge of the World. Ithaca, NY.
Lydon, James, 1988: Lordship and Crown. In The British Isles, 11001500. Ed. R.R. Davies. Edin-
burgh. Pp. 4863.
OMeara, J. (ed., tr.), 1982. Giraldus Cambrensis: The History and Topography of Ireland. London.
ORahilly, C. (ed.), 1967. Tin B Calnge from the Book of Leinster. Dublin.
Sigurur Nordal (ed.), 1933. Egils saga Skallagrmssonar. slenzk fornrit 2. Reykjavk.

220
 Which came first the smith or the shaman? Vlundarkvia,
craftspeople and central place complexes
Leif Einarson, Department of English, University of Western Ontario, Canada
In his studies of central place complexes in early medieval Scandinavia, Stefan Brink suggests
a model of inter-relations between highly qualified smiths, pagan priests, warriors and royal
or king-like figures (1996: 23942). Aspects of the Old Norse poem Vlundarkvia probably
date back to this early period, and the narrative as it is preserved portrays such an interaction
between a skilled smith on an island workshop and the military power of a royal family in an
aristocratic hall. The role of religious or spiritual features and figures in this poem is, how-
ever, difficult to determine. Studies of the smith and the shaman in the history of world relig-
ions have influenced the interpretation of the figure of the smith in Vlundarkvia: philologi-
cal, literary and archaeological studies alike develop parallels between the figure of the master
smith Vlundr and the figure of the shaman in general, or more specific aspects of inic
shamanism, Norse seir and Smi noaidi (Dronke 1997: 257, 260, 26668, 318; Grimstad
1983: 2014; Hedeager 2002: 9). Some traditions and contexts may suggest that The smith
and the shaman come from the same nest,1 but it is necessary to remove Vlundr from that
nest. In what follows I will outline several key features that preclude Vlundr from being
categorically similar to the shaman. Vlundr is a highly skilled craftsperson of a different eth-
nicity than the royal family he interacts with: the artisanal motifs and inter-group relations
portrayed in Vlundarkvia are parallel to similar motifs and structures in both Old Norse
mythological narratives and archaeological evidence of the role of craftspeople in early me-
dieval Scandinavia.

Plot Summary:
Vlundarkvia survives in the Icelandic Codex Regius manuscript of the Poetic Edda, dated
to 1270. Elements of this narrative also survive in several texts, carvings and runic representa-
tions from Scandinavia and the British Isles, dating back as early as the 7th century (Dronke
1997: 26974; Nedoma 1990: 12939). Vlundarkvia is an interspersed prose and verse nar-
rative about the famously skilled smith Vlundr. He and his two brothers are princes of the
Smi, an indigenous group of people inhabiting areas of northern Norway, Sweden, Finland
and Russia. These three brothers travel on skis, hunt and establish a residence together near a
lake. They meet three swan-maidens from the south, who are weaving fine linens on the
shore. Each swan-maiden marries a brother. The three couples live together for seven winters
before the swan-maidens begin to long and ache for something else: they spend a final eighth
winter together, and in the ninth winter, while the brothers are out hunting, the maidens fly
away. Vlundrs two brothers leave to search for their mates, but Vlundr remains alone,
hunting bears and smithing 700 gold rings. He seems to be anticipating the return of his mate.
Vlundr is now called a prince of elves. The Swedish King Nur discovers Vlundrs abode
and has Vlundr shackled in his sleep and brought to his hall. Anxious about Vlundrs
threatening presence, the queen orders that he be hamstrung and sent to work at an isolated
island workshop. Vlundr sleeplessly makes precious objects with remarkable speed for the
royal family. He has his revenge in two parts. First, Vlundr forges three sets of gruesome
gifts: silver-gilded bowls from the skulls of the kings two sons, jewels from their eyes, and
brooches from their teeth. Second, Vlundr seduces and impregnates Bvildr, the kings
only daughter. With the aid of a magical device (vl) of his own crafting Vlundr lifts himself

1
This is a proverb of the Yakut tribe (Dronke 1997: 257; Eliade 1978: 83).

221
into the sky twice, declaring that his revenge is complete and exactly appropriate to the harms
inflicted upon him.

Three features:
Three key features form the basis for abstract comparisons of Vlundr and studies of shaman-
ism and relations between the smith and the shaman in various traditions around the world.
First, Vlundarkvia demonstrates dualities that are suggestive of interactions between the
Norse and the Smi, particularly the shamans of the Smi. Second, Vlundrs re-forging of
the boys skulls seems magical and evocative. Third, Vlundrs escape flights are suggestive
of supernatural or spiritual transformation. Each of these features, however, demands close
examination in and of itself. With regards to the third feature, which will not be discussed in
detail here, Vlundr is a magical figure, but Norse mythological figures like jazi and Loki
also demonstrate magical flights and transformations that are not necessarily shamanic.
Within the limited scope of this presentation I will focus on the first two features, the dualities
of Vlundarkvia and Vlundrs re-forging of the boys skulls.

1) Dualities and Dichotomies

In her summary of The story according to the Poem Ursula Dronke emphasizes how
Vlundarkvia portrays a duality in human nature (Dronke 1997: 255) between the natural
or mortal and the supernatural: the human passion of the women who marry the brothers is,
for instance, contrasted with the supernatural wings of the valkyries who fly away. According
to this argument, Vlundr is part human and part supernatural demon. The basis for the dual-
isms of Vlundarkvia, however, seems to have less to do with this duality, and more to do
with two other contrasts: 1) the contrast between the Swedish colonial kingship and Vlundrs
Smi and elvish extraction; and 2) the power to create material wealth versus the desire and
power to control material wealth. Both of these contrasts are part of the cultural context for
traders, craftspeople and the central place complex in Migration Period and Viking Age Scan-
dinavia. These contrasts need to be clarified with regard to the interpretation and reception of
Vlundarkvia.
Following the introduction of Nur as a colonial king of the Swedes in the prose pro-
logue of Vlundarkvia, Vlundr is immediately introduced as a Finn or Smi. Norse traders
were familiar with the Smi. The early 13th century Egils saga, for example, portrays a series
of interactions between the Norse and the Smi. In the late 9th century events related in Chap-
ters 10 through 17, rlfr Kveldlfsson and his rivals, the Hildirarsons, go on independent
trips north to collect tribute and trade with the same group of Smi. rlfs dealings go
smoothly, but also with some additional ease and success due to intimidation: this is likely
because he takes more than three times the customary number of armed men (Bjarni 2003:
13). Conversely, the Hildirarsons travel with the customary thirty men and have much less
success in extracting tribute and engaging in profitable trade (Bjarni 2003: 20). If this narra-
tion is a characteristic example, then trading interactions between the Norse and the Smi
were complex, with an established history of customs. Meetings were regular and often
friendly enough, although with a spectrum of possible degrees of alliance, intimidation, hos-
tility and exploitation.
The literature also demonstrates that the Norse were familiar with the reputation of Smi
shamans. The Latin text of the 12th century Historia Norvegiae relates the observations of
some Norwegian merchants who witness Smi shamanism. The writer redacts the commen-
tary of the Norwegian Christian merchants on the intolerable paganism and literally in-
credible devilish superstition of Smi shamanism (Tolley 2006: 12). The passage details a
number of possibly shamanic actions, including supernatural prediction by the use of an un-

222
clean spirit, theft of desirable items over impossible distances, use of a decorated drum, and
transformation into the shape of a water beast. It is potentially suspect that these merchants
just happen to be present during a shamanic sance: Clive Tolley suggests that the most prob-
able explanation may be that the Smi deliberately, but perhaps surreptitiously, arranged the
sance for their Norwegian visitors; the aim would be a demonstration of their superiority in
the field of magic, the only area in which they were recognised as excelling their otherwise
more powerful overlords, with a view to securing a better trading deal (Tolley 2006: 5).2 The
Norse people were familiar with the Smi, and with the reputation of their shamans, and both
groups probably used intimidation and other, perhaps ethnically specific, performances to
secure better trading outcomes.
Vlundrs ethnic and regional otherness as a Smi is represented in a more mythical or
legendary aspect in the verses of the poem, where he is repetitively called a ruler or kinsman
of the elves (Neckel & Kuhn 1983: 118, 122). This elvish nature is first declared just as King
Nur and his queen abusively enslave the smith. The context of Vlundrs antagonistic be-
haviour, coupled with his Smi and elvish extraction, has led to the implication of dark and
demonic supernatural associations from the Christianized discourses surrounding shamanism.
Robert Nedoma, for instance, observes that his elvishness establishes that Vlund is of a
demoniac nature (1990: 138). Ursula Dronke similarly suggests that The poet epitomizes as
elvish the demonic nature of the human smith born in the same nest as the shaman (1997:
2567). It seems to be Snorri who introduces a suggestively Christian dichotomy between
light and dark elves: he assigns the lighter variety to the highest level of the heavens and the
darker variety to the subterranean realm (Faulkes 1988: 19). As both Kaaren Grimstad and
John Lindow point out, literary evidence on distinctions between dwarfs, giants and the light
and dark elves is scant and ambiguous at best (Grimstad 1983:19395; Lindow 2002: 109
10). Lindow emphasizes that the only important figure explicitly assigned to the elves is
Vlund (2002: 110). Grimstad concludes that there was often no clear distinction made be-
tween dark-elves and dwarfs: both lived in the earth, were potentially dangerous to man,
and were superior smiths, skilled in magic (1983: 195). There are, however, no descriptions
of Vlundrs environment in this poem that would suggest an underground situation. Associa-
tions between Welandes smie (Welands Smithy) and the megalithic grave near the Berk-
shire Downs in the Old English charter of 955 obviously offer a different perspective (Kemble
1964: 332; Dronke 1997: 259; Nedoma 1990: 133), as do more generalized associations be-
tween dwarves, smiths and underground workshop environments in mythological and folk-
loric sources (Ellis Davidson 1958; Motz 1977). But it is important to remember that these
associations are not explicitly demonstrated in Vlundarkvia. In fact, Vlundr is described as
having a white neck in the second stanza, which would seem to preclude him from member-
ship in the ranks of the dark elves, who Snorri tells us are darker than pitch, svartari en
bik (Faulkes 1988: 19). This demonization of Vlundr has more in common with the Christian
dichotomies of God and Devil, light and dark3 that distort encounters with elves and Smi
shamanism than it does with understandings of the smith in the poem and in the cultural con-
text of trade with skilled craftspeople in the early Viking Age. If it is from shamanic dis-
courses that Vlundrs demonic associations arise, then it must be pointed out that the cate-
gory of shamanism can prove highly problematic, as Alice Beck Kehoe elucidates (2000:
26, 15, 379, 535). Furthermore, nowhere in the poem does Vlundr wear a cape, cover
himself in a blanket, go into a trance, heal people or carry a staff or a drum: this smith is not
to be confused with a shaman.

2
cf. Tolleys two-volume study on Shamanism in Norse Myth and Magic (2009), which was not yet available at
the time of this writing.
3
cf. Steinsland 2005: 141.

223
 2) The Skulls
Vlundrs re-forging of the boys skulls is also not a sustained shamanic parallel. In her
commentary on Vlundarkvia, Dronke repetitively cites Mircea Eliades The Forge and the
Crucible. One of the key implications made by citing Eliades work is that Vlundrs forging
of silver bowls and jewels from the boys skulls is parallel to shamanic initiation rituals
amongst several Siberian tribes (Dronke 1997: 267). According to Eliades paraphrase (1978:
83), these dream-narratives involve a spirit journey in which the initiate meets a supernatural
smith figure who re-forges the initiates skull using a special anvil, or reassembles the initi-
ates body using iron either in place of bone or as a connective agent between bones. These
narratives involve a consistent sequence of events: 1) spiritual journey to the smith, 2) dis-
memberment, 3) re-integration of the body, sometimes using metal components, 4) spiritual
return journey with confirmed status as shaman.
Vlundrs re-forging of the boys skulls, however, does not operate as a parallel to these
shamanic narratives: this is vengeance, not initiation. I acknowledge the abstract parallels
involving skulls, smithing, metal, magical skills, transformations and powerful figures of
knowledge with different ethnicities or supernatural dwarf/elf extractions. But Nurs sons
are really not in a better position to become shamans because of Vlundrs actions. Nor is
Vlundr acting from a motive that could be confused with wanting to promote these two
young cubs to a powerful spiritual vocation. Nor is Vlundr raising Nurs sons through a
warrior initiation rite, a parallel that Grimstad suggests to the smith Reginn raising the hero
Sigurr (1983: 203). The boys travel to the smithy, where the smith decapitates them. Any
potential for sustained comparative parallels to shamanic or heroic initiation narratives begins
and ends here. The boys heads are re-forged into bowls and jewels, their decapitated bodies
remaining in the waste slag beneath the forge.

3) Textual Parallels
I would like to offer an alternative interpretation of Vlundarkvia, one that is not based upon
placing the smith in the same nest as the shaman. Two brief case studies, one textual and one
archaeological, suggest that the basis for Vlundrs actions is a statement about the relations
between craftspeople and aristocratic power in the central place complex.
Vlundrs gruesome bodily transformation of Nurs sons is part of an un-doing of the
central place complex that Nur and his queen have attempted to establish. Vlundrs ac-
tions here are parodic parallels of the re-forging of Ymirs skull and body into the cosmos by
inn and his two brothers, Vili and V (Faulkes 1988: 913). This is the archetypal Old
Norse myth of creating a central place complex for the gods and humanity, and Vlundrs
revenge echoes both the destructive and murderous as well as the creative and magical aspects
of this action. Comparing Vlundarkvia to this creation story highlights imbalances in power
and tensions between different types of communities, particularly related to the repercussions
of unilateral exploitation of craftspeople. Vlundr and Vlundarkvia are the expression of
complex social networks of ideologies and anxieties.
The Old Norse myth of the creation of the cosmos is recounted most fully in Snorris Edda
and in Grmnisml, and aspects of this narrative also appear in skaldic kennings that arguably
have origins in the 10th or 11th centuries. Ymir is the first Frost-Giant, and the first anthropo-
morphic being: from him are descended all other giants and the gods. inn and his two
brothers kill Ymir and create the cosmos from his body parts: the sky is Ymirs skull, and it is
held up by four dwarfs, the clouds are his brains, the seas and lakes his blood, the earth his
flesh and the mountains his bones, while rocks and scree are made from his teeth and any
bones that were broken. This is how the gods first establish their own central place, and a cen-
tral place for humanity: the respective powerful halls and fertile regions lie at the centre of

224
this construction, surrounded and protected by Ymirs eyelashes. The giants, and some
dwarfs, are generally the foes of the gods and they live on the periphery of the world in
mostly de-centralized pluralized locations (cf. Clunies Ross 1994: 506).
There are at least two key parallels to Vlundarkvia here. First, both are narratives of
magical creative actions complicit with murderous destruction and decapitation: the homolo-
gous imagery of skulls, bowls and the sky is a consistent feature in early skaldic kennings.
One such example is a kenning that refers to the sky as the wide hand-basin of winds, va
munnlaug vinda (Bragi Frag 2; Faulkes 1998: 34).4 This kenning is attributed to the oldest
known skld, Bragi Boddason, who lived during the 9th century and composed verse for sev-
eral Swedish kings. The 10th century skld Arnrr jarlaskld refers to the sky as Ymis hauss
Ymirs skull (Faulkes 1998: 33). Awareness of the Vlundarkvia narrative is also apparent
in several skaldic kennings. In jlfr of Hvinirs early 10th century Haustlng, for instance,
the kenning grjt-Nur (rock-Nur) refers to the giant jazi (Faulkes 1998: 32),5 who is
also known in the same poem as the god of skis, perhaps suggesting similar associations of
itinerancy and Smi hunting techniques as are seen in Vlundarkvia.
Second, both narratives perform specific social structures: the aristocratic power of the
central place repetitively marginalizes and unilaterally abuses the resourceful powers of
skilled craftspeople. The gods establish a central place that, as Margaret Clunies Ross ob-
serves in the first volume of Prolonged Echoes, is based upon a refusal to acknowledge their
giant lineage and also, whenever possible, a unilateral abuse of the resources, skills and magic
that are associated with the marginalized giants: the giants live across the sea on the islands at
the edge of the world, and it is particularly taboo for male giants to breed with female gods.
Anxieties about incursions from the giantlands necessitate rrs constant giant-killing activ-
ity. But interactions between the two groups are necessary and ongoing: the gods needed the
giants, their knowledge, their competence and their powerful objects, just as the giants for
their part desired objects belonging to the gods (Steinsland 2005: 143). This all plays out in
cyclical revenge narratives, which are the contexts of two of the kennings just cited, and fre-
quently revolve around the theft of cultural objects or the abuse of specialized skills. All this
culminates in the apocalyptic downfall of the gods reign and the destruction of the cosmos.
The parallels to Vlundrs infiltration and undoing of Nurs aristocratic central place and
the continuation of his patriline are clear and not at all vague or abstractly comparative: both
the mythological creation narrative and Vlundarkvia demonstrate the destruction and abuse

4
Hinn es varp va
vinda ndurdsar
yfir manna sjt margra
munnlaug fur augum.
(Faulkes 1998: 34)
He who threw into the wide winds basin the ski-goddesss [Skadis] fathers eyes above the dwellings of the
multitude of men (Faulkes 2001: 89).
5
Sr ba sagna hrri
sorgra<n> mey fra
er ellilyf sa,
ttrunnr Hymis, kunni.
Brunnakrs of kom *bekkjar
Brsings goa dsi
girijfr gara
grjt-Naar san.
(Faulkes 1998: 32)
The scion of Hymirs race [giants] instructed the crew-guider, crazy with pain, to bring to him the maid who
knew the sirs old-age cure [Idunn]. The thief of Brisings girdle [Brisingamen] afterwards caused the gods
lady [Idunn] to go into the rock-Niduds [giants] courts to Brunnakrs bench (Faulkes 2001: 87).

225
that results in acting upon the covetous desire to unilaterally control 1) skilled craftspeople of
different ethnic or social extraction and 2) the distribution of valuable goods that define and
maintain social structure and power within early medieval Scandinavia. The covetous, de-
structive and dualistic themes of this poem are mutually shared between the smith and the
king and queen, to the benefit of none of them.

4) Archaeological Parallels
The anxieties and differences that are expressed in the mythological texts and in Vlundark-
via are parallel to the development of communities of itinerant craftspeople that are distinct
from agrarian communities in late Migration Period Scandinavia. Johan Callmers archaeo-
logical study of hus in southern Sweden suggests that there were long-standing workshop
sites throughout Scandinavia that did not have a design like the agrarian complexes that some-
times developed into powerful aristocratic and religious nodes: rather than having an aristo-
cratic hall or large religious space at its centre, the grids at hus are regular, with habitation
plots suitable to families of five to ten people.
Callmer hypothesizes that these locations developed from small temporary sites into larger
communities that were constantly occupied by mostly itinerant craftspeople who formed col-
laborative and mutually supportive communities. All sorts of crafts were practiced at these
locations, especially since close collaboration was necessary in order to make many artefacts.
Some of these craftspeople were largely itinerant, while others were more permanent. Many
were generalists, while some were specialists. There does not seem to have been one particu-
lar figure of the smith or craftsperson, but rather a variety of roles within one developing type
of community. Callmer suggests that

The lifestyle, culture, perhaps also their vernacular set the people active as craftsmen and traders
aside from the inhabitants of the different regions. Frequently the remoteness (in relation to cen-
tral locations in the regions) and the coastal location of the places [like hus] contributed to this
social isolation. Local society of the period had great difficulties in assimilating a population,
which by its habits, doings and for many, by its extraction was alien. Consequently it is most
likely that many of these traders and craftsmen never became part of the local society and then
we must consider the probable issue of the formation of a separate society. We may tend to
imagine these people, on the margin of the majority population, weak and vulnerable and ex-
posed to conditionality. This may be a false picture. They gathered many together [hus could
have hosted 5001000 at its peak size] and they could certainly instantly muster a relatively
large troop of armed men. (Callmer 2002: 155)

Vlundr fits into this community: at the beginning of the narrative he lives near the shore of a
lake in a cooperative family unit of itinerant craftspeople from distinct geographical and eth-
nic origins. Vlundr is later forced to be at the isolated location of the workshop svar st,
at sea-venue, i.e. on a beach (Neckel & Kuhn 119; Sveinbjrn & Finnur 1931: 559). This
location resembles the culturally liminal yet highly practical workshop and market sites on
shorelines or beaches, as well as at sites like the large workshop on the island of Helg, which
may have produced goods for chieftains on site and in a defined region around Lake
Mlaren (Hjrthner-Holdar, Lamm & Magnus, 2002: 169). It is also plausible that a king like
Nur who had enslaved and maimed a renowned smith like Vlundr with his own aristo-
cratic, familial, and foreign connections might not have been without his own fair share of
anxieties about the insurgences of hostile or disruptive traders and craftspeople.
Vlundarkvia evokes these anxieties, as do the interactions between the gods and giants in
the Old Norse mythological corpus. As Clunies Ross points out, the chief method of maintain-
ing the distinctions between the centralized gods and the marginalized giants is through rr

226
and his suggestively smith-like tool, the hammer. Mjllnir is made by a magical smith, and it
is a sacred tool that reinforces the oppositional structures that, as both Clunies Ross and Lin-
dow argue, are the basis for the creation of the cosmos by the sir. Amulets of rrs ham-
mer are an early and persistent feature in the archaeological record. The master smiths, the
vlundar, could help to establish and maintain cultural, aristocratic, spiritual, military and
agrarian distinctions and prowess in the central place complexes they worked within: they
could also threaten to undo them. Vlundarkvia demonstrates these possibilities, as well as
the challenges of sustaining familial and communal structures from both the aristocratic, co-
lonial Swedish perspective and the aristocratic, crafting, hunting, itinerant indigenous Smi
perspective.
Instead of concluding that Vlundr demonstrates shamanic magico-religious power over
fire (Eliade 1978: 7981), or that Vlundarkvia has degenerated from a sacred initiation rite
into a misunderstood narrative about the profane revenge of a dark demonic smith (Dronke
1997: 256257; Nedoma 1990: 138; Grimstad 1983: 204), I prefer to end with an ongoing
appreciation for the complexity of the specific contexts of various smith figures in central
place networks and the persistence of these structures in Old Norse literature. These com-
plexities are brought together in figures like Vlundr, poems like Vlundarkvia, and in the
objects and central place complexes made by skilled craftspeople, like the iron seir-staff
from Klinta with a miniature Trelleborg-style hall on top of it (Price 2002: 184). Smiths made
these structures: they were not alone.

Bibliography
Bjarni Einarsson ed. 2003: Egils saga Skalla-Grmssonar. London: Viking Society for Northern Re-
search.
Bragi Frag 2, 20019: Skaldic Project Academic Body. ed. M. Clunies Ross. 16 Apr. 2009.
<http://www.skaldic.arts.usyd.edu.au/db.php?table=verses&id=1936>
Brink, Stefan, 1996: Political and Social Structures in Early Scandinavia. A Settlement-historical Pre-
study of the Central Place. In: Tor: Journal of Archaeology 28. Pp. 215281.
Callmer, Johan, 2002: North-European trading centres and the Early Medieval craftsman. Craftsmen at
hus, north-eastern Scania, Sweden, ca. AD 750850+. In: Central Places in the Migration and the
Merovingian Periods: papers from the 52nd Sachsensymposium, Lund, August 2001. Acta archae-
ologica Lundensia 39. Uppkrastudier 6. Eds. B. Hrdh and L. Larsson. Stockholm: Almqvist &
Wiskell. Pp.125157.
Cleasby, Richard and Gudbrand Vigfusson, 1874: An Icelandic-English Dictionary. Oxford: Claren-
don Press. 16 Apr. 2009. <http://lexicon.ff.cuni.cz/texts/oi_cleasbyvigfusson_about.html>
Clunies Ross, Margaret, 1994: Prolonged echoes: Old Norse myths in medieval Northern society. Vi-
king Collection 7. Odense: Odense University Press.
Dronke, Ursula, ed. and trans. 1997: Vlundarkvia. In: The Poetic Edda, with translation, introduc-
tion, and commentary, vol. 2 Mythological Poems. Oxford: Clarendon. Pp. 243328.
Eliade, Mircea, 1978: The Forge and The Crucible. 2nd Edn. Trans. S. Corrin. Chicago: University
Press of Chicago.
Ellis Davidson, H. R., 1958: Weland the Smith. Folklore 69.3. Pp. 145159.
Faulkes, A. ed. 1988: Snorri Sturlusson. Edda: Prologue and Gylfaginning. London: Viking Society
for Northern Research.
Faulkes, A. ed. 1998: Edda: Skldskaparml:1. Introduction, Text and Notes. London: Viking Society
for Northern Research.
Faulkes, A. trans. & ed. 2001: Edda. London: J. M. Dent.
Grimstad, Kaaren, 1983: The Revenge of Vlundr. In: Edda: a collection of essays. Eds. R. J. Glend-
inning and Haraldur Bessason. Winnipeg: University of Manitoba Press. Pp. 187209.
Hedeager, Lotte, 2002: Scandinavian Central Places in a cosmological setting. In: Central Places in
the Migration and the Merovingian Periods: papers from the 52nd Sachsensymposium, Lund, Au-
gust 2001. Acta archaeologica Lundensia 39. Uppkrastudier 6. Eds. B. Hrdh and L. Larsson.
Stockholm: Almqvist & Wiskell. Pp. 318.

227
Hjrthner-Holdar, Eva, Kristina Lamm and Bente Magnus, 2002: Metalworking and Central Places.
In: Central Places in the Migration and the Merovingian Periods: papers from the 52nd Sachsen-
symposium, Lund, August 2001. Acta archaeologica Lundensia 39. Uppkrastudier 6. Eds. B.
Hrdh and L. Larsson. Stockholm: Almqvist & Wiskell. Pp. 159183.
Kehoe, Alice, 2000. Shamans and Religion: An Anthropological Exploration in Critical Thinking.
Long Grove: Waveland Press.
Kemble, J. M. ed., 1964: Chartae Anglosaxonicae. In: Codex Diplomaticus Aevi Saxonici 5. Londini:
Sumptibus Societatis.
Lindow, John, 2000: Thors Visit to tgaraloki. In: Oral Tradition. 15.1. Pp. 170186.
Lindow, John. 2002: Elves, In: Norse Mythology: a guide to the gods, heroes, rituals and beliefs. New
York: Oxford University Press. Pp. 109110.
Motz, Lotte, 1977: The Craftsman in the Mound. In: Folklore. 88.1. Pp. 4660.
Neckel, Gustav, rev. Hans Kuhn. 1983. Edda. Die Lieder des Codex Regius nebst verwandten Denk-
mlern. 1. Text. 5th rev. edn. Heidelberg: Carl Winter.
Nedoma, Robert, 1990: The Legend of Wayland in Deor. In: Zeitschrift fr Anglistik und Amerikanis-
tik 38. Pp. 12945.
Price, Neil S., 2002: The Viking Way: Religion and War in Late Iron Age Scandinavia. Uppsala: De-
partment of Archaeology and Ancient History.
Steinsland, Gro, 2005: The Late Iron Age Worldview and the Concept of Utmark. In: Utmark: the
outfield as industry and ideology in the late Iron Age and the Middle Ages. Eds. I. Hom, S. Innsel-
set and I. ye. UBAS international. Bergen: University of Bergen. Pp. 137146.
Sveinbjrn Egilsson and Finnur Jnsson, eds. 1931: Lexicon poeticum antiqu lingu septentrionalis:
ordbog over det norsk-islandske skjaldesprog. 2nd ed. Copenhagen: Mller.
Tolley, Clive, 2006: The Shamanic Seance Presented in the Historia Norwegiae. In: The Thirteenth
International Saga Conference: Durham and York, 612 Aug. 2006.
<http://www.dur.ac.uk/medieval.www/sagaconf/tolley.htm >
Tolley, Clive, 2009: Shamanism in Norse Myth and Magic. 2 vol.

228
Love affairs versus Social Status: A Theme in Kormks saga?
Eln Bra Magnsdttir, University of Bergen, Norway
Kormks saga is considered to be among the earliest sagas of Icelanders, as it is generally
presumed to have been written in the period of 12001220. The dating of the saga has mainly
been based on the absence of textual connection with any other sagas. In addition, the saga is
rather primitive in both composition and style, which could indicate that it belongs to the first
written sagas of Icelanders.
The early dating of Kormks saga has made the question of its origin of considerable im-
portance. For a long time, scholars have believed that the principal source of the saga could be
its many verses (in all 85 verses). Most of the verses (64) are ascribed to Kormkr gmundar-
son, the sagas main character. Most of them have been considered authentic and composed
by Kormkr in tenth century (Einar l. Sveinsson 1939: lxxxlxxxiii). The prose has been
thought to have derived from oral tradition; some part of it may have originally belonged to
the verses although this is not true of other parts, especially those where discrepancies can be
found between verse and prose (Einar l. Sveinsson 1939: xcivxcv).
In his study, Skldasgur (1961), Bjarni Einarsson argued for foreign influences on the
saga, both in terms of verse and prose. He believed that the verses was composed by the saga-
author and that they had been influenced by French troubadours love poetry. Bjarni also main-
tained that the main-theme of the saga, the tragic love between Kormkr and Steingerr, also
had French connections. More specifically, that it was derived from a version of the Tristran
romance which was known in Scandinavia as Tristrams saga in the Norwegian translation of
1226. Bjarnis thesis has not won general acceptance among scholars but it did spark a debate
about the origin of the saga (Andersson 1969: 741, Bjarni Einarsson 1971: 2141,
ODonoghue 1991, Finlay 2001: 23271).
In this paper, I will discuss one of the episodes in Kormks saga, which tells of the love-
affair between Kormkr and Steingerr and the reason why Kormkr didnt want to marry
her. This episode has many parallels with a narrative pattern in the sagas of Icelanders. The
analysis of the theme in this particular episode could suggest that the whole episode is based
on this narrative pattern.

A venja kvmur snar : A Narrative Pattern in the Sagas

In a memorable scene, the author of Kormks saga tells of Kormkrs and Steingerrs first
meeting in Gnpsdalr and how they fell in love with each other. In this scene Kormkr speaks
his first verses about Steingerr and in the first stanza he expresses the feeling that this love
will bring him bad luck (in prose order): N var mr ramma-st mnu jtuns sntar leii;
menrei rttumk risti fyr skmmu; eir ftr fald-Gerar munu vera mr at fri optarr an n
[] (207).1 Thus, from the first moment of their love-affair, fate seems to be against the lov-
ers, and the question remains why fate was against them?
After their first meeting, Kormkr decides to continue visiting Steingerr; this is told by
using a common phrase in the sagas of Icelanders: Eptir etta venr Kormkr gngur snar
Gnpsdal at hitta Steingeri [] (215, italics are mine). In the sagas, the phrase a venja
kvmur snar is more common, in fact gngur in this context, is only used in Kormks saga
and Vatnsdla saga. On the other hand, the phrase a venja kvmur snar occurs, in twelve
sagas of Icelanders (lexis.hi.is; searchwords gngur and komur). In most of these sagas this

1
All citations to Kormks saga are taken from Einar l. Sveinssons edition in slenzk fornrit VIII (1939).

229
phrase appears as a part of a narrative pattern that tells of love-affairs which cause problems
in society; a man venr kvmur snar to a woman, because he wants to have a love-affair with
her, but since he has no plans to marry her, his behaviour is seen as unacceptable. His visits
are therefore not looked upon favourably by the family of the woman because they bring dis-
honour to the woman and her family.
This pattern occurs in all of the skaldsagas, i.e. Kormks saga, Hallfrear saga van-
drasklds, Bjarnar saga Htdlakappa and Gunnlaugs saga ormstungu, but also in eight
others sagas of Icelanders. The difference between the skaldsagas and the other sagas is that
this pattern becomes the main-theme in the former and leads to a love triangle conflict be-
tween the skalds and the husbands of the women. This pattern, a mans love of a married
woman has been designated as the love-motif (Bjarni Einarsson 1961: 4051). In other sagas
of Icelanders, the pattern is not the main-theme, or engender the main conflict, but is only an
issue in one or two episodes (or one to three chapters) in the sagas. The sagas of Icelanders
tells therefore only about the conflict between the visitor of the woman and her father, or her
closest relatives, and ends there.2 This entails that the womans future life is not an issue in
these narratives. To distinguish this narrative pattern from the love-motif, which includes the
love-triangle, we can label it as the venja kvmur-pattern.
The skaldsagas have two variants of the love-motif which explain why the skalds did not
get married to their beloved girlfriends (Bjarni Einarsson 1961: 49). In Kormks saga and
Hallfrear saga the reason is in fact made unclear and therefore never directly mentioned in
these sagas. In the other two skaldsagas, on the other hand, the reason for the love triangle is
very clear; Bjrn and Gunnlaugr were betrayed by their rivals and this is why they did not get
married. In this case, both Kormks saga and Hallfrear saga, share the same narrative pat-
tern as the sagas of Icelanders regarding the part that tells about why the love-affairs caused
problems. As we will get back to later, Kormks saga contains many parallels with the venja
kvmur-pattern as it features in other sagas.
In the sagas of Icelanders the purpose of the visitors is only erotic and therefore not in-
tended to have any social function, including marriage. As a result the behaviour of the pro-
tagonist is seen as unacceptable. But why did the protagonists behave in this way which dis-
honours their girlfriends and brings problem to their families? The sagas do indeed give us
some answers to this question.
Class differences seem to be one of the causes. Besides the skaldsagas, the best-known nar-
rative of the venja kvmur-pattern, is perhaps the one that tells about the love relationship
between Inglfr and Valgerr which figures in Hallfrear saga and Vatnsdla saga. Inglfr
was the son of orsteinn Ingimundarson, chieftain in Vatnsdalr and according to Hallfrear
saga assumed to be mestr mar ar sveitum (F VIII, p. 141). Valgerr belonged to a
lower social class; she was the daughter of ttarr orvaldsson and the sister of Hallfrer
vandraskld who was a farmer but, according to the saga, a rather wealthy one (F VIII,
p. 141). In Hallfrear saga, Inglfr shows no interest in marrying Valgerr even though her
father has offered him to do so. Inglfrs lack of interest is not explained in the saga but the
communication between him and ttarr underlines their different social status; because of his
high status, Inglfr indicates that ttarr is not in a position to interfere in his affairs: Hann
kvazk mundu vera sjlfri fera sinna, hvat sem ttarr segi, lt sv a eins skipaan dalinn,
at hann kvazk engis manns nauarmar vera skulu. (F VIII, p. 143). Inglfrs attitude to-
wards ttarr and his daughter leads to a lawsuit between the two families. At Hnavatnsing,

2
The only exception is perhaps to be found in Eyrbyggja saga (ch. 29, 40 and 47), that tells about the love-affair
between Bjrn Breivkingakappi and urr of Fr. In the beginning urrs brother, Snorri goi, gets in-
volved in this affair but eventually it leads to a conflict between Bjrn and urrs second husband, roddr
skattkaupandi.

230
orsteinn manages to judge single-handedly in the lawsuit and he uses the opportunity to get
rid of ttarr from his neighbourhood: ek geri hlft hundra silfrs til handa ttari, en hann
skal selja jarir snar ok rask brott r essi sveit (F VIII, p. 144).
Vatnsdla contains another version of the story; the lawsuit is described in a different way
and it becomes ttarrs own decision to move from the district (ch. 37). But Inglfrs attitude
towards ttarr is here also characterised by hubris and accordingly the underlying theme in
both versions is the different social status of the participants. After ttarr moves to Norrr-
dalr, Inglfr continues to visit Valgerr, although this is against the whishes of her father. At
that time Inglfr had married Hallds, the daughter of lafr from Haukagil. The saga does not
tell anything about their relationship or marriage, but since her father is introduced as a
wealthy man in Hallfrear saga (F VIII, p. 141), Hallds seems to have been a more appro-
priate wife for the chieftains son.
In Fljtsdla saga class differences is also the reason for why lovers do not get married.
Helgi Droplaugarson has a love-affair with Helga orbjarnardttir who belongs to a lower
class than his. A neighbour of Helga tries to prevent the affair and seeks support from Bersi,
Helgis foster father. Because of the different social status of the lovers, Bersi realises that
Helgi is not going to marry Helga: ok at Helga vri gefin fstra mnum, tti mr ar
fr kona vel gefin hraustum manni. En get ek, at honum yki sr at of lgt fyrir mann-
viringar sakir. (F XI, p. 250). In the end, Bersi manages to get Helgi to break off the rela-
tionship with Helga, a relationship which seems to be the last one: Helgi leitar aldri
konu optar ok ngva ara sv at menn viti. Er at ok alu manna sgn, at Helgi hafi ngva
konu elskat sv at menn viti. (F XI, p. 256).
Fstbrra saga tells of ormr Kolbrnarsklds not very serious love-affairs. ormr
first has an affair with rds, the daughter of Grma in gur. The lovers seem to belong to
the same social class for both are children of rather wealthy farmers, so class-differences is
not an issue here. On the other hand, Grma does not like ormrs visits to her daughter and
offers him to marry her. ormr refuses on the following grounds: eigi er skaplyndi mitt til
ess at kvngask; en vtti ek mr ekki framar en eiga dttur na, en mun at fyrir
farask. (F VI, p. 161). In ormrs case it is obvious that he just wants to have fun. For
instance the saga tells that he was often bored at home, at his fathers farm, and his way of
having fun was to visit the girls. The same observation is also applicable to the relationship
with his second girlfriend, orbjrg kolbrn (ch. 11).
In sagas where class differences is not an issue the reason why men do not want to get mar-
ried is usually not mentioned. Consequently, we can probably assume that protagonists simply
wanted to have fun as in the case of ormr Kolbrnarskld. But in at least two of the sagas,
Ljsvetninga saga (ch. 1) and Vatnsdla (ch. 18), the issue is protecting women from violent
and abusive men. In most of these sagas, those involved belong to the farmer class. Indeed
only two sagas tell about love-affairs within the chieftain class, i.e. Ljsvetninga saga (ch. 5)
and rar saga hreu (ch. 5).3
It is interesting to note, that many of the men involved in the venja kvmur-pattern, are in
fact skalds that belong to the farmer class, i.e. Hallfrer vandraskld, ormr Kolbrnar-
skld and Bjrn Breivkingakappi. Indeed Kormkr is the only one who belongs to the chief-

3
In two sagas, the women involved are widows, but in other sagas they are young and unmarried girls who still
lived under the protection of their father or close relatives. Widows had more rights than unmarried girls
(Meulengracht-Srensen 1995: 3435), but in their private life they do not seem to have had more rights than
unmarried girls; both urr of Fr in Eyrbyggja saga (ch. 29) and shildr in Flamanna saga (ch. 18) are not
allowed to have relationships with men, even though they wanted to, and for the same reasons as young girls
because it brings dishonour to them and their families. In Eyrbyggja, for instance, urrs relationship with
Bjrn Breivkingakappi, becomes so embarassing for her family, that in the end her brother, Snorri goi, insists
that Bjrn should leave the country.

231
tain class. Since skalds have such an important role and are so visible in this narrative pattern,
there is a reason to believe that because of their talent, they had a privileged status in society.
For this reason they were able to act like men of higher social rank. Their privileged status
meant that on occasion they had the opportunity to marry women of higher rank. Hallfrers
refusal to marry Kolfinna, who belonged to the same class as him, seems to reflect his ambi-
tion to marry for higher social status (Torfi H. Tulinius 2001: 201205).
As a result, the venja kvmur-pattern is common in the sagas of Icelanders. It occurs in
twelve sagas, including the skaldsagas, and in some sagas in more than one episode. Since
this pattern is so ubiquitous, we can assume that it derives from oral tradition. The pattern
could in fact reflect a social problem at the oral stage and therefore, in some extent, social
reality regarding relations between the sexes. First, the pattern reveals a class divided society
where men of higher social rank were neither able nor willing to marry beneath them. The
pattern therefore reflects a society where marriage agreements were supposed to either main-
tain or improve ones social status. Secondly, it reflects a social problem in the Saga Age,
where people, belonging to either same or different class, did not follow unwritten rules or
traditions of society about sexual relations.4

Class differences: A Theme in Kormks saga?

The episode in Kormks saga (ch. 46) that tells about the love-affair between Kormkr and
Steingerr has many parallels with the venja kvmur-pattern. Class differences seem to have
been the reason why Kormkr did not wish to marry Steingerr. On the other hand, the author
of the saga does not seem to have wished to emphasise that fact and therefore he blames
Kormkrs refusal of Steingerr solely on rveigs curse (seir). This episode is therefore
very inconsistent, for the authors oral source does not seem to fit with his interpretation of it.
This is perhaps the reason why scholars have not agreed with the authors explanation and
constructed their own interpretations.5
After Kormkrs first meeting with Steingerr in Gnpsdalr, he tells his mother, Dalla, that
he will continue to visit Steingerr. Dalla responds negatively to her sons decision for she
realises that it will cause problems for Steingerrs father: Dalla kva mannamun mikinn ok
eigi vst, at til ynis yri, ef etta vissi orkell Tungu. (215). These words, mannamunr
mikill, indicates that there was a class differences between Kormkr and Steingerr. Although
it was in favour of Steingerr to marry Kormkr, Dalla nevertheless seems to know that her
sons visits will not please the girls father. Her words therefore indicates that she knew that
her son had no intention of marrying Steingerr and that his visits will only cause problems
for Steingerrs family. When orkell is later informed about the situation, he reacts quickly,
and his daughter is forced to move home to Tunga: orkell spyrr n brtt, hvat um er at vera,
ok ykkir sr horfa til viringar ok dttur sinni, ef Kormkr vill etta eigi meir festa; sendir
eptir Steingeri, ok ferr hon heim. (216).

4
Hallfrear saga is the only saga, that deals directly with this issue as a social problem. orsteinn Ingimundar-
son tries to prevent his sons unwelcome visits to his girlfriend, and when he talks to his son, his words indicate
that his sons behaviour was becoming a social problem among young men of higher rank: Annan htt hafi r
en vr hfum unga aldri, geri yr at ginnungum, er hfingja efni eru; lt af tali vi dttur ttars bnda.
(F VIII, p. 143). His words also reveal that this problem was related to the younger generation and not his own.
5
Einar l. Sveinsson (1939) believed that it was Kormkrs skldlund that prevented him from marrying Ste-
ingerr: N dgum munu menn yfirleitt heldur leita annarra skringa: til skldlundar Kormks. Hallfreur
vildi ekki kvnast Kolfinnu, egar kostur var, a v er saga hans segir, og fleiri vlk dmi kynni a mega
finna r vi annarra sklda. stur eirra eru margvslegar. Hj Kormki hefur a n efa mtt sn mest a
hann var ekki til ess binn a skipta skldskap snum og hversdagslfi. (lxxxi). See also Hans E. Kinck
(1921: 6177) and Gurn Lange (1992: 85106).

232
 But despite of the fact that orkell does not accept the relationship, Kormkr continues to
visit his daughter. His behaviour demonstrates that he has no respect for Steingerr or her
family and this attitude further underlines their class differences; in terms of his social author-
ity it does not seem have bothered Kormkr what kind of attitude orkell had in this issue.
Because of the different social status of the lovers, Steingerr is not an appropriate wife for
Kormkr. In spite of the fact that Kormkr was deeply in love with Steingerr, her lower so-
cial status seems to have prevented a marriage agreement. The pre-history section of the saga
includes an introduction of Kormkrs noble family in Norway. His grandfather, Kormkr
was rkr ok kynstrr (203) who had participated with king Haraldr hrfagri in many battles.
Kormkrs father, gmundr, is then described as a promising young man who also partici-
pated in Viking expeditions with the king. But after gmundr settles down in Iceland the saga
has not much to say about him; he is married to Dalla and he dies when their two sons, Kor-
mkr and orgils, reach maturity. Dalla then takes over the farm along with her sons under the
protection of Mifjarar-Skeggi who was the most powerful chieftain in Mifjrr at that
time. The pre-history emphasises Kormkrs noble origin, and later in his life, especially on
his Viking expeditions abroad, Kormkr demonstrates that he has the same ambituous nature
as his father and grandfather had.
Steingerr is a farmers daughter but despite her lower status it could nevertheless be main-
tained that the author did not seek to specifically highlight the social differences of the two
lovers. This is at least true of Steingerrs lower status as the authors explanation of the
seir-incident later in the saga suggests. The introduction of Steingerrs family shows that
the author had limited knowledge about her family background. For instance, he does not
mention the social status of her father: orkell ht mar, er bj Tungu; hann var kvngar,
ok ttu au dttur, er Steingerr ht; hn var Gnpsdal at fstri. (206). But from the context
of the saga, we understand that he was a farmer and probably a rather poor one, as indeed his
conflict with Kormkr illustrates.
orkell does not seem to have any powerful supporters to assist him in his conflict with
Kormkr. The only available help he can muster comes from a certain Narfi, who seem to
have been orkells workman, and the two sons of the witch rveig, Oddr and Gumundr.
Despite their efforts to keep Kormkr away from Steingerr, they do not succeed.6 Finally, the
sons of rveig try to ambush Kormkr but he kills both of them. After their killing, the dif-
ferent social status of Kormkr and his opponents is underlined. When Kormkr later visits
rveig and because of his power he is able to chase her away from Mifjrr and then refuse
to pay fines for the killing of her sons: skaltu flytja ik brott at kveinni stundu, en ek vil
allra bta varna um sonu na. (221). As noted the same pattern is found in Hallfrear saga,
in the episode which describes the lawsuit between orsteinn and ttarr. In this respect one
could also mention the last conflict between Snorri goi and Bjrn Breivkingakappi in Eyr-
byggja (ch. 47), when Snorri forces Bjrn to leave the country. This episode also underlines
Snorris powerful position in the district. An underlying theme in these narratives thus ap-
pears to be conflict about social power between chieftains and farmers which results in dem-
onstration of power from the former.
rveig had no other choice than to move from the district, but as a witch she was able to
use her own power by putting a curse on Kormkr: at er lkast, at v komir lei, at ek

6
A similiar episode is to be found in Ljsvetninga saga (ch. 1) where a farmers vulnerability is underlined in his
conflict against the visitor of his daughter. This episode (ch. 5) in Kormks saga also has another parallell with
Fstbrra saga. Both Steingerr and rds alert their lovers about a planned ambush by their opponents but
neither Kormkr nor ormr take the warnings seriously (Fstbrra saga, ch. 9). As Bjarni Einarsson (1961:
6465) has pointed out, there is also a parallell between these sagas in relation to the account of the first meeting
of the lovers, i.e. Kormkr and Steingerr (ch. 3) and ormr and orbjrg kolbrn in Fstbrra saga (ch.
11).

233
vera herafltta, en synir mnir bttir, en v skal ek r launa, at skalt Steingerar aldri
njta. (222). Despite of Kormkrs social standing in the district, he is not able to have the
last word in this conflict.

rveigs curse (seir)

Despite their class differences, Kormkr seems in the end to have persuaded to propose to
Steingerr. A scene (ch. 6) suggests that it was Steingerr who made him do it: N bir Ste-
ingerr Kormk stunda til fur hennar ok f hennar, ok fyrir sakar Steingerar gaf Kormkr
orkatli gjafar. (223). But when Kormkr formally proposes to Steingerr and the wedding
day has been planned, he gets cold feet. The reason for his change of hear was, according to
the saga, the following: N fara or milli eira, ok vera nkkurar greinir um fjrfar, ok
sv veik vi breytiliga, at san essum rum var rit, fannsk Kormki ftt um, en at var
fyrir sk, at rveig seiddi til, at au skyldi eigi njtask mega. (223).
The only thing mentioned here which could explain why Kormkr changed his mind, re-
lates to money. This in turn could support the notion that it was actually their different social
status which made Kormkr change his mind; it was not economically advantageous for him
to marry Steingerr.7 Although class differences seems to have been the reason for the fact no
marriage took place, the author, provides another explanation when he blames it on rveigs
curse. It could therefore be argued that the authors explanation contradicts the source he had
on this matter. But since the curse is directly linked to Kormkrs conflict with rveig, the
question arises if the curse did also belong to the narrative pattern that this episode seems to
be based on?
Bjarni Einarsson (1961) believed that the seir-incident in Kormks saga derived from the
Tristran story. In both stories curse plays an important role in deciding the fate of the lovers
but, as Bjarni points out, it also works in the opposite way:

Hr verur a g a hvers konar sgu hfundurinn er a setja saman og hver er hin lklega
hfufyrirmynd hans. lgin sem elskendurnir vera fyrir og eiga san vi a stra til viloka
minna beint sgur um Tristran. ar er a tfradrykkur sem veldur eim lgum a Tristran
og kngsdttir f me engu mti ri vi stru starinnar og eru hennar valdi til hinztu
stundar. En Kormks sgu meina lg elskendunum vilangt a njtast. Hin beinu hrif
laganna eru lk essum tveim sgum, en bum eru afleiingarnar vilng gfa sguhet-
junnar og konunnar sem hann elskar. (8283).

Since supernatural power is used in some of the sagas, which include the venja kvmur-
pattern, the question arises whether the curse is part and parcel of this pattern? Besides Kor-
mks saga, supernatural power, including witchcraft (fjlkynngi), plays an important role in
Fstbrra saga and Hallfrear saga, and although a different motif is used in Kormks
saga, the function of the supernatural power is used as a punishment of people that have done
other people harm. The use of supernatural power in this way could in fact suggest that the
venja kvmur-pattern is originally based on folk narrative, since it is used by people in order
to punish people that have done them harm. That fact could suggest that these texts are origi-
nally told from the peoples point of view.

7
Scholars have believed that it was because of Kormkrs status as a skald that he did not want to marry at all
(Hans E. Kinck 1921: 6177, Einar l. Sveinsson 1939: lxxxi). But it is worth noting that later in the saga (ch.
89), Kormkr considers a marriage to another woman. Hlmgngu-Bersi, Steingerrs first husband, offers
Kormkr to marry his sister, Helga, in his attempt to reconcile with him. Surprisingly, Kormkr considers ac-
cepting Bersis offer, presumably because Bersi was a wealthy man which made his sister an appropriate mar-
riage for Kormkr.

234
 Although Kormkr is the main character in Kormks saga, he is usually not described in a
positive way in that sense that he does not have the sympathy of the sagas author. The depic-
tion of him is actually rather negative. For instance this presentation of Kormkr can be ob-
served in his conflicts with Steingerrs husbands, Hlmgngu-Bersi and orvaldr tinteinn.
The negative attitude towards Kormkr is, on the other hand, particularly noticeable in sec-
tions of the saga which takes place at home rather than abroad. In his travels abroad, Kormkr
is represented as a successful Viking who participates in expeditions in many countries. The
attitude towards him in these episodes is accordingly much more positive than in other parts
of the saga. A distinction must therefore be made between what we could term the private and
a more public traditions about Kormkrs life (Meulengracht-Srensen 1995: 6263; Baetke
1956: 1526). The private tradition seem to be originally based on folk narrative, since the
point of view of the people is a dominant element in these narratives.
Witchcraft plays an important role in the venja kvmur-pattern in Fstbrra saga. When
ormr Kolbrnarskld refuses to stop visiting rds, her mother, Grma in gur, has her
slave, Kolbakr, to fight with ormr. But without Grmas witchcraft, Kolbakr would never
have been successful against ormr. In their fight, Kolbakr causes an injury to ormr and
for that he is later sentenced to outlawry. But Grma rewards Kolbakr for his help, by giving
him his freedom and ensuring that he is able to leave the country. In this episode, it is Grma
and her people who have the sagas sympathy. Her witchcraft is shown to be a positive ele-
ment in her dispute with ormr and it is her witchcraft which ensures her success against
ormr. The sagas sympathy is with the people. At the end of the episode, even the slave
Kolbakr is described in a positive way compared to ormr: Kolbakr rzt li me vkin-
gum ok reynisk harfengr mar llum mannraunum. [] Eigi hfum vr heyrt getit at or-
mr hafi fengit meiri sm sns verka en sekir Kolbaks. (F VI, p. 169).
In an another episode (ch. 11), supernatural power is used as a punishment for ormr.
He composed the so-called Kolbrnarvsur for his second girlfriend, orbjrg kolbrn. His
first girlfriend, rds, becomes very jealous and ormr then amends the verses, kva n
Kolbrnarvsur ok snr eim rendum til lofs vi rdsi, er mest vru kvein or, at hann
hafi um orbjrgu ort. Gefr hann n rdsi kvit til heilla stta ok heils hugar hennar ok
sta vi sik. (F VI, p. 173). orbjrg kolbrn then appears to ormr in a dream and pun-
ishes him for what he has done to her:

N mun ek launa r v lausung na ok lygi, at skalt n taka augnverk mikinn ok strangan,

sv at bi augu skulu springa r hfi r, nema lsir fyrir alu klkisskap num, eim
er tkt fr mr mitt lofkvi ok gefit annarri konu. Muntu aldregi heill vera, nema fellir
nir r vsur, er hefir snit til lofs vi rdsi, en takir r upp, er hefir um mik kveit,
ok kenna eigi etta kvi rum en eim, sem ort var ndveru. (F VI, p. 17475).

During the night, ormr wakes up with so much pain in his eyes that he is unable to sleep
for the rest of the night. He is then forced to change the poem to its original form and explain
the whole situation to the people: N lsir hann fyrir alu, hversu farit hafi um kvit,
ok gefr af nju vi mrg vitni orbjrgu kvit. ormi batnai skjtt augnaverk-
jarins, ok verr hann alheill ess meins. (F VI, p. 17677). In Hallfrear saga we find a
similar incident, where Hallfrer vandraskld is punished by King lafr Tryggvason (ch.
10). The missionary king appears to Hallfrer in a dream and punished him for his bad behav-
iour in his conflict with Grss, Kolfinnas husband. Hallfrer then follows the kings advice
and stops bothering Kolfinna and her husband.
Fstbrra saga and Hallfrear saga share the same motif of the main-protagonists of
these sagas being punished for their bad behaviour. In Kormks saga, we find a different mo-

235
tif, but the function of the use of supernatural power is still the same; people that have done
harm to others are punished and the punishment is directed at men that have dishonoured
women and their families. The curse in Kormks saga could therefore originate in the venja
kvmur-pattern that seems to be the source for this episode in the saga.
As a result, the episode (ch. 46) in Kormks saga which tells about the love-affair be-
tween Kormkr and Steingerr has many parallels with a narrative pattern in the sagas of Ice-
landers, or the venja kvmur-pattern as I chosen to label it. Since this is a common pattern in
the sagas, we have to assume that it derives from oral tradition. In some extent, this pattern
could reflect social reality at the oral stage, or a social problem relating to sexual behaviour.
The pattern reveals a class divided society that, for instance, refuses men of high rank to
marry women of a lower rank. The analysis of the particular episode in Kormks saga indi-
cates that it was class differences that prevented Kormkr from marrying Steingerr. On the
other hand, the author of the saga did not seek to emphasise the different social status of the
lovers, especially Steingerrs low status. He moreover explains Kormkrs refusal of Ste-
ingerr by blaming it on rveigs curse. Since supernatural power does occur in some of the
narratives, containing the venja kvmur-pattern, and the function of it is the same, the follow-
ing question arises: Does the curse also derive from the same narrative pattern that appears to
be the source for the whole episode in the saga?

Many thanks to Haki Antonsson for valuable comments on my English in this article.

Bibliography
Andersson, Theodore M. 1969. Skalds and Troubadours. Mediaeval Scandinavia 2. Odense Univer-
sity Press.
Austfiringa sgur. 1950. Jn Jhannesson gaf t. slenzk fornrit XI. Hi slenzka fornritaflag, Reyk-
javk.
Baetke, Walter. 1956. ber die Entstehung der Islndersagas. Akademie-Verlag, Berlin.
Bjarni Einarsson. 1961. Skldasgur. Um uppruna og eli staskldasagnanna fornu. Menningars-
jur, Reykjavk.
Bjarni Einarsson. 1971. The Lovesick Skald: A Reply to Theodore M. Andersson. Mediaeval Scan-
dinavia 4. Odense University Press.
Einar l. Sveinsson. 1939. Formli. Kormks saga. slenzk fornrit VIII. Hi slenzka fornritaflag,
Reykjavk.
Finlay, Allison. 2001. Skald Sagas in their Literary Context 2: Possible European Contexts. Skald-
sagas. Text, Vocation, and Desire in the Icelandic Sagas of Poets. Walter de Gruyter, Berlin, New
York.
Gurn Lange. 1992. Andleg st. Arabsk-platnsk hrif og integumentum slenskum fornbk-
menntum? Skrnir. 166 r. Reykjavk.
Kinck, Hans E. 1921. Kjrligheten i Kormaks Saga. Mange Slags Kunst. H. Aschehoug, Kristiania.
Meulengracht Srensen, Preben. 1995. Fortlling og re. Studier i islndingesagaerne.
Universitetsforlaget, Oslo.
ODonoghue, Heather. 1991. The Genesis of a Saga Narrative. Verse and Prose in Kormaks
saga.Clarendon Press, Oxford.
Torfi H. Tulinius. 2001. The Prosimetrum Form 2: Verses as an Influence in Saga Composition and
Interpretation. Skaldsagas. Text, Vocation, and Desire in the Icelandic Sagas of Poets. Walter de
Gruyter, Berlin, New York.
Vatnsdla saga, Hallfrear saga vandrasklds, Kormks saga. 1939. Einar l. Sveinsson gaf t.
slenzk fornrit VIII. Hi slenzka fornritaflag, Reykjavk.
Vestfiringa sgur. 1943. Bjrn K. rlfsson og Guni Jnsson gfu t. slenzk fornrit VI. Hi slen-
zka fornritaflag, Reykjavk. www.lexis.hi.is/corpus/leit.pl?lemma

236
 The ethical map of the Hrlfs saga Gautrekssonar
Alexey Eremenko, Institute of Slavic Studies, Russian Academy of Sciences
A less-researched aspect of the saga ethics is its connection with ethno-geography that is,
inhabitants of different countries have different ethical features attributed to them. I attempt to
outline some points of an ethical saga map, which encompasses Gotaland, Sweden, Russia
and Ireland, using Hrlfs saga Gautrekssonar, a late 13th-century fornaldarsaga, as a source.
Sagas of this kind offer a highly stylized picture of the world, but they still reflect the ideas
the medieval Icelanders had about other lands, even if in a simplified form. Hrlfs saga
Gautrekssonar, in particular, focuses on the three bridal quests (Kalinke 1985; Kalinke 1990),
and the regularity of heroes travels offers perfect comparative material. I also include a short
inheritance episode predating the quests and providing an important moral counterpoint to the
three quests. The saga references two more geographic locations England, and a nameless
island, which is the dwelling place of the sorcerer giant Grimnir, but Im omitting them here,
since the four stories I work with cover the bulk of the saga text and plot.
The inheritance story takes place in Gotaland (Gautland), which is where the saga pro-
tagonist comes from. Hrlf Gautrekssons qualities make him the perfect king, as explicitly
stated by his father (Hrlfs saga 57), and his fitness to govern overrides even the nominal
rules of inheritance in the country, which prioritize the eldest son (Hrlfs saga 56) in this
case its Hrlfs brother Ketil, described in the saga as an adorable, but not very competent
sidekick of Hrlf (Hrlfs saga 150). The ascension of Ketil to the throne could prove trouble-
some, but his father manages to solve the problem while avoiding conflict: upon his deathbed
he simply asks his sons, the queen and the important people of the country to let Hrlf be-
come his successor instead of Ketil. The suggestion is supported by all parties involved, in-
cluding the elder son himself, and proves to be a wise decision indeed, because Hrlf eventu-
ally achieves dominance over one more kingdom (Hrlfs saga 150) a feat Ketil is hardly
expected to accomplish and settles there, returning Gotaland to his brother as a reward for
the moderation Ketil had shown.
Settling the matters of the crown, the brothers embark on bridal quests. The first one finds
the titular hero wooing ornbjrg, the daughter of Eirek, king of Sweden (Svj). The prob-
lem with ornbjrg is that she became a maiden king, or meykngr (Wahlgren 1938; Matyu-
shina 2006), identifying as male and excelling as a warrior. She received a third of Sweden
from her father, and began to reign there as king (not a queen) in her own right, ridiculing and
maiming all male suitors. Hrlf is pushed into courting ornbjrg by his brother Ketil, despite
objecting that the time isnt right and the enterprise is doomed to fail. It really fails, and Hrlf
spends several years accumulating forces and battle experience in viking raids, before making
another move, at a time he deems appropriate. Freed from outside interference, he succeeds in
defeating ornbjrg and taking her captive, all only to ask her to consider his marriage pro-
posal, to which she agrees, abandoning the crossdressing. Of note is the fact that king Eirek,
father of ornbjrg, disapproves of her behavior from the beginning and thinks good of Hrlf,
although the suitor gets no help from the Swedish king while battling his daughter.
The second bridal quest takes place in Russia (Gararki). The suitor in this case is Ketil,
the prospective bride lof, daughter of Hlfdan, the ruler of Russia, and the root of all prob-
lems lies with the berserks living at the latters court. The leader of berserks is secretly plot-
ting to marry lof himself, and therefore scorns and bans all other suitors as unworthy of her.
The princess and the king, even though he is described in the saga as a wise man and well-
liked (Hrlfs saga 91; translation after: Hrlf Gautreksson 77), become influenced by this
behavior and begin to refuse the suitors, making no exception for Ketil, even though this
raises the objection of rir, king Hlfdans mightiest follower, who points out that the rejec-

237
tion will harm the king of Russia, because he has no chance victory in the fight with Hrlf-
backed army of Ketil. When ignored, rir withdraws his support for Hlfdan, who is
promptly defeated. rir is still pushed into fighting Hrlf by lof, who invokes his earlier
moral obligation to her, but when the two heroes (reluctantly) go into combat, rir loses.
Hrlf, however, spares the king of Russia, and hires rir into his own army after the battle.
The third bridal quest takes place in Ireland (Irland). Asmund, the blood-brother of earlier
suitors Hrlf and Ketil, woos Ingibjorg, the daughter of Hrlf, king of Ireland, who is de-
scribed as a great man for sacrifices, able to see into the future with the help of his evil and
depraved belief (Hrlf Gautreksson 102). The saga also mentions his ruthlessness
(harfengi) (Hrlfs saga 113). Hrlf of Ireland, just as the ruler of Russia, rejects the suitors
of his own daughter. Hrlf Gautreksson advises against the expedition to Ireland, but eventu-
ally takes part in it, fulfilling his obligations to the blood-brother. His pessimism proves right
again: his forces are outnumbered and annihilated by the Irish, and he and Asmund are cap-
tured by the king of Ireland, who dooms them to a dishonorable death by starvation. Their
survival and victory are ensured by two things: first, Hrlf Gautrekssons far-sightedness,
which led him to keep a military reserve in Gotaland, and second, the sympathy of Ingibjorg,
described as an intelligent good-looking girl, (Hrlf Gautreksson 102) who saves Hrlf and
Asmund. In the end, the quest is accomplished, Asmund and Ingibjorg marry, and the king of
Ireland is defeated, but spared.

Table 1. Summary of the four conflicts of Hrlfs saga Gautrekssonar:

Country Nature of conflict Opponents Power of Peacemakers Peacemakers
opponents influence
GOTALAND Inheritance debate two princes no conflicting both princes, the general consensus
parties queen, all vassals
SWEDEN Bridal quest princess control over 1/3 king of the coun- nominal sovereign
(maiden king) of the country try
RUSSIA Bridal quest berserk vassals control over the top vassal strong standing at
king & princess whole country the court
IRELAND Bridal quest king of the country control over the princess no open influence
whole country

The key ethical categories of Hrlfs saga Gautrekssonar are wisdom and moderation, on one
hand, and arrogance on the other. All of these concepts were well-represented in medieval
Icelandic culture: for instance, wisdom, usually denoted with the words speki / rspeki and
vsdmr, is given a lot of attention in Eddaic poem Hvaml, described by Thomas Anderson
as an archaic moral treatise (Andersson 1970:588). The principal role played by moderation
(hf) in saga culture was conclusively proven by a host of researchers during the debate of the
19601970s, although saga scholars failed to reach a consensus concerning its roots, traced
either to Christian influence (Hermann Plsson 1966; Lnnroth 1969; Fulk 1986; Ciklamini
1988) or authentic features of the Icelandic society (Andersson 1970; Karlsson 1985;
Vilhjlmur rnason 1991; Gurn Noral 1998). Wisdom and moderation were also inter-
connected semantically, since moderate behavior, which required the individual to be capable
of adequate estimation of ones surroundings and rationalization of the psychological im-
pulses (anger, desire etc.), was impossible without a certain level of intellectual capacity.
Arrogance in the saga tradition served as the opposite of moderation, representing over-
blown and limitless pride. The main Old Norse words for this concept were ofsi and ofmet-
nar, their semantic field also incorporating tyranny and pride; a number of researchers
also pointed to its connection with the Christian concept of superbia (Ciklamini 1988;
Vsteinn lason 1998:177).

238
 Hrlfs saga Gautrekssonar contains a number of lexemes relevant to these ethical catego-
ries, namely vsdmr wisdom (Hrlfs saga 132, 140), rligr cleverly (Hrlfs saga 51,
61), hgvr equanimity (Hrlfs saga 62), dramb arrogance (Hrlfs saga 61), and leitni
pettish disposition (Hrlfs saga 70). This linguistic evidence proves that the author(s) of the
saga gave thought to ethical matters; however, since Hrlfs saga Gautrekssonar conforms to
the typical saga pattern of describing the moral principles without naming them (Vsteinn
lason 1998:166), linguistics alone are not enough to reconstruct the ethical map of Hrlfs
saga Gautrekssonar in its entirety, and must be supplemented by the analysis of the narrative,
which is why I included plot descriptions of three bridal quests and the inheritance episode in
Gotaland, which sets the moral blueprint for the rest of the saga.
Wisdom in Hrlfs saga Gautrekssonar is presented as an individuals ability to judge the
circumstances adequately, i.e. perceive events, persons and their interrelations in an undis-
torted way, thus being able to make correct forecasts concerning the situation ones dealing
with. This allows the character to avoid excesses in his or her behavior (theres an impres-
sive number of wise female characters in Hrlfs saga Gautrekssonar, albeit supporting ones),
which brings wisdom in close connection with moderation a trait that has its importance for
the saga further emphasized by the mercy Hrlf and his brothers-in-arms show to their ene-
mies.
Arrogance in Hrlfs saga Gautrekssonar, on the other hand, is the result of an erroneous
judgment of circumstances. More precisely, an arrogant individual is the one that overesti-
mates his or her social status (again, no gender discrimination here) and considers other
people to be inferior, even if ones own personal achievements and qualities arent sufficient
to justify the claim of superiority. Arrogance is usually associated with people from outside of
the titular heros native Gotaland; this may be interpreted as a reflection of the ethnic xeno-
phobia, which the Gotalanders (with whom the audience is expected to associate with) faced
during their travels. However, even assuming the dislike was initially ethnic in nature, in
Hrlfs saga Gautrekssonar it was transferred into the realm of morals.
The perception of ethics in Hrlfs saga Gautrekssonar has a distinct geosocial dimension:
the moral qualities are projected on the society of each described country, reflecting the level
of its otherness in the eyes of the author(s) of the saga. The titular hero is shown as the perfect
wise man, to the point where he becomes compatible with the noble heathen type of charac-
ters, singled out by Lars Lnnroth in the slendingasgur (Lnnroth 1969), where he serves as
a moral example for the other characters and the audience of the saga. Hrlf and his allies can
perhaps be labeled true Scandinavians, as opposed to the backward Swedes, satirized in
Gautreks saga, a text interconnected with Hrlfs saga Gautrekssonar (it tells of Hrlfs fa-
ther) (Gautreks saga 311).
Hrlf and, to a lesser extent, his brothers-in-arms serve as Kulturtregers, or, rather,
Moraltregers, following the right ethical code and bringing it to the three regions they visit.
With each bridal quest, the virtues of the visitors wisdom and moderation are matched
against the arrogance of local inhabitants. The aboriginal vice, however, differs in form and
scope.
In Sweden, the visitors have to deal with womanly arrogance based on gender reversal: the
kings daughter, a maiden king, considers herself a man (this is actually a stock motif for
fornaldarsgur and other younger sagas). The princess has the support of only one third of the
land, and the supreme ruler of the country disapproves of her behavior, although has no means
to stop her.
In Russia, the king and the princess both adopt an arrogant model of behavior, but its
source is found in the kings berserk vassals, not the king himself. Still, since the sovereign is
involved, arrogance becomes the law of the land and of its entire people. The kings domi-
nance and moral influence, however, are not absolute: he is opposed by the countrys second

239
man his own top vassal, who also shows virtues of wisdom and moderation, advising his
king against looking down upon suitors which dont deserve to be looked down upon. The
vassal fails to change the kings mind, but the attempt is still notable. Also, while the king
becomes arrogant, hes not aggressive (unlike the berserks): he refuses Ketils suit, but avoids
insults and shows no desire to fight, asking politely to be left alone.
Ireland comes across as the most hostile place of all. The source of its hostility is the arro-
gance of its sorcerer king, who is devoid of any positive traits found in the rulers of Sweden
and Russia. Not only he uses witchcraft, which is emasculating and disgraceful for a warrior
(Clunies Ross 1998:3233), and is an explicit and devout pagan, which could elicit no sympa-
thy from the Scandinavian audience of the 13th century, hes also unnecessarily cruel, as
shown by his dishonorable and cruel treatment of Hrlf, his war prisoner, who could have at
least hoped to be granted a noble death for his valiant slaying of the attacking Irishmen, but
gets no mercy from his Irish namesake. Theres no-one to stop the Irish king: his court has no
open opposition, and the princess, which helps the imprisoned suitors, is powerless and has to
act in secret relying, it should be noted, on her own wit and cunning, i.e. showing a form of
wisdom.
The settings and plots of the three bridal quests in Hrlfs saga Gautrekssonar obviously
dont mirror the social realities of the 13th century (or the times before the settlement of Ice-
land, which the saga was supposed to describe) literally. The maiden king, the berserks, the
powerful evil sorcerer are all stock characters that abound in the fornaldarsgur. However,
they still may yield some information about the ways in which the author(s) of the saga per-
ceived ethics, ethnogeography and their relationship.
All European countries shown in the bridal quests of the saga share the same social struc-
ture consisting of a ruler, his offspring, a number of mighty vassals, possibly split into several
factions, the simple subjects of the king, who appear in the story only as nameless soldiers,
and the aliens, represented by the people of Gotaland, which has the same structure as well,
except for the aliens. The latter should be considered a separate category, since they have a
significant influence on the social context of the countries they travel to. The relationship be-
tween Gotalanders and the inhabitants of other lands is established on the ethical plane: the
moral foreignness, which determines the level of the countrys enmity, increased from Swe-
den to Russia and then to Ireland, with the most significant watershed found between Sweden,
which is still largely cordial to outsiders, and the other two regions, where the dislike of
aliens, caused by the wrong ethical position, determines the whole countrys policy. Russia
and Ireland have their own differences as well, with the Russian ruler shown in the saga as a
man subjected to bad influence, but not devoid of good qualities, and the Irish one coming
across as a model villain.
The moral gap between the main heroes of Hrlfs saga Gautrekssonar and the inhabitants
of the lands they visit is never absolute: the characters that personify the countries are ar-
ranged and shift on a wisdom arrogance scale, influenced by external factors, the prime
one being the actions of the protagonists, who are guided by wisdom and moderation and
transmit the right ideals to others (this doesnt exclude military conflicts, but Hrlf and his
friends try to avoid those and usually strive to minimize their consequences, refraining from
unnecessary killings). Besides, constructive cooperation between the people from Gotaland
and other countries is also possible: they intermarry and befriend each other, with the Gota-
landers even sometimes becoming rulers of these lands, all of this proving that the moral
and therefore ethnic otherness can be overcome.

Bibliography

240
Andersson, Theodore M., 1970: The Displacement of the Heroic Ideal in the Family Sagas. In: Specu-
lum. Vol. 45, No. 4. Cambridge, Mass. Pp. 575593.
Ciklamini, Marlene, 1988: Sturla Sighvatssons Chieftaincy. A moral probe. In: Sturlustefna:
Rstefna haldin sj alda rt Sturlu rarsonar sagnaritara 1984. Ritstjrar Gurn sa
Grmsdttir, Jnas Kristjnsson. Reykjavk. Pp. 222241.
Clunies Ross, Margaret, 1998: Prolonged Echoes. Old Norse myths in medieval Northern society.
Volume 2: The Reception of Norse myths in medieval Iceland. Odense. Pp. 3233.
Fulk, Robert D., 1986: The moral system of Hrafnkels saga Freysgoa. In: Saga-Book of Viking Soci-
ety for Northern Research. Vol. 22, part. 1. London. Pp. 132.
Gautreks saga. In: Fornaldarsgur Norurlanda. Ed. Guni Jnsson, Bjarni Vilhjlmsson. B. 3. 1944.
Reykjavk. Bls. 341.
Gurn Noral, 1998: Ethics and Action in Thirteenth-Century Iceland. Odense.
Hermann Plsson, 1966: Sifri Hrafnkels sgu. Reykjavk.
Hrlf Gautreksson: a Viking Romance. Trnsl. Hermann Plsson, P. Edwards. 1972. Toronto.
Hrlfs saga Gautrekssonar. In: Fornaldarsgur Norurlanda. Ed. Guni Jnsson, Bjarni Vilhjlmsson.
B. 3. 1944. Reykjavk. Bls. 45151.
Kalinke, Marianne E., 1985: Riddarasgur, Fornaldarsgur And the Problem of Genre. In: Les sagas
de chevaliers (Riddarasgur). Actes de la Ve Confrenece Internationale sur les Saga. Prsents par
Rgis Boyer. (Toulon, Juillet, 1982). Paris. Pp. 7791.
Kalinke, Marianne E., 1990: Bridal-Quest Romance in Medieval Iceland. Ithaca.
Karlsson, Gunnar, 1985: The Ethics of the Icelandic Saga Authors and Their Contemporaries: A
comment on Hermann Plssons Theories on the Subject. In: The Sixth International Saga Confer-
ence 28.7 2.8 1985. Workshop Papers 12. Kbenhavn. Pp. 381399.
Lnnroth, Lars, 1969: The Noble Heathen: A Theme in the Sagas. In: Scandinavian Studies. Law-
rence, Kansas. Vol. 41. Pp. 129.
Matyushina, Inna, 2006: Magic Mirrors, Monsters, Maiden-kings (the Fantastic in Riddarasgur). In:
The 13th International Saga Conference. Durham and York, 6th12th August, 2006. Conference
Papers. Durham. Pp. 660670.
Vsteinn lason, 1998: Dialogues with the Viking Age. Narration and Representation in the Sagas of
Icelanders. Reykjavk.
Vilhjlmur rnason, 1991: Morality and Social Structure in the Icelandic Sagas. In: Journal of English
& Germanic Philology. Bd. 90. Urbana, Ill. Pp. 157174.
Wahlgren, Erik, 1938: The Maiden King in Iceland. Diss. Chicago.

241
 Reception and function of stories about the East
Stefka G. Eriksen, University of Oslo, Norway
In this paper I will discuss the intended reception mode and function of texts based on the
materiality of the manuscripts they appear in. The texts I will concentrate on are two ver-
sions of the Old Norse Els saga. The first version is the Old French chanson de geste Elye de
Saint-Gille as it appears in the manuscript BNF 25516, from North East France from ca. 1280.
The second version is the Old Norse Els saga, as it appears in the Norwegian manuscript De
la Gardie 4-7 folio, dated to ca. 1270. In the following, I will first present the theoretical start-
ing point for my argument, then the relevant evidence from the two manuscripts, and thereaf-
ter comment on the implications such evidence may have.

Introduction
The first issue to address is of methodological nature and concerns the relationship between
the materiality of a manuscript and the intended reception mode of the same manuscript. What
aspects of a text and a manuscript may be interpreted to elucidate the issue of reception mode
and on what grounds?
One method is to study the prologues and epilogues, and other relevant parts of a text, for
information on reception, like for example: As you will hear/ As you have heard/ []
those who read and hear, etc.1 Another method is to search for various rhetorical devices,
like address to the audience, narrators comments, use of direct speech, use of temporal vs.
spatial adverbs, etc. Such rhetorical devices may be regarded as tools used by a consciously
writing author, who intended a specific reception mode for his text.2 Yet another method is to
study whether various grammatical aspects of the language in a text may seem to have been
adapted to a certain mode of reception. For example, using various colloquial grammatical
constructions in a written text may have been a way of adapting it to an intended listening
audience.3 Further, the style and the rhythm of a text, like for example extensive use of allit-
erations, end rhyme, assonance and other metrical patterns, may also indicate that the text was
intended to be listened to. A fifth possible method for revealing intended reception mode is
studying the narrative structure of a text and discussing how the dramaturgical line of the plot
may have functioned if the text was listened to or read privately. A final method, which dis-
tinguishes itself slightly from the others, is investigating the way the text appears in the manu-
script the way it is written (e.g. abbreviated and punctuated4), and the way it is arranged and
structured on the manuscript page (by means of illustrations, rubrics, initials, etc.5).
Since the methods for discussing reception are rather manifold, one problem that appears is
that the different types of evidence may occasionally point in different directions. This has
been commented by for example D.H. Green (1994, 2002) and rendered unproblematic. He

1
Such studies of Old French sources have been done by Crosby (1936:105), Coleman (1996), Vitz (1999);
Clanchy (1993) discusses the significance of such terms in Latin, Middle English and Old French; D.H. Green
(1994, 2002:3653) focus on Middle High German language and texts; Terje Spurkland (1994, 2000) has studied
the respective Old Norse terms in various sources.
2
Various rhetorical devices have been discussed from orality-literacy perspective, both when it comes to compo-
sition, transmission and reception. Some scholars that have discussed rhetoric aspects of various texts in relation
to reception are D.H. Green (1994), Suzanne Fleischmann (1989, 1990), Keith Busby (2002), just to mention a
few.
3
See Fleischman (1990a, b).
4
See M. Parkes (1992), Derolez (2003).
5
See Paul Saenger (1982, 1997), Keith Busby (2002).

242
claims that one and the same text may have been intended to have been both publically and
privately read, both listened to and looked upon, in various contexts and settings.
Still, a frequent drawback in the existent research is the failure to combine and consider all
the different types of evidence a text in a manuscript presents. How does the materiality of a
manuscript relate and correspond to the rhetorical, lexical, grammatical, or structural aspects
of the text? In this paper I will therefore adopt such an eclectic method of reading a text. I will
study one aspect of the materiality of the manuscript initials and discuss their correspon-
dence to various textual aspects, in order to comment on the intended reception mode of the
text. Further, it will be interesting to see whether the correspondence between the graphical
and the textual is the same or different in the two versions that will be studied. What may the
similarity/ difference in this relationship indicate of the reception and function of the two ver-
sions in their respective historical and geographical contexts?

The text versions and their content

The Old French chanson de geste Elye de Saint-Gille is extant in only one medieval manu-
script, which is dated to the second half of the 13th century and seems to have originated in
North-East France, possibly by commission of a member of the House of Flanders (Raynaud
1879:13; de Winter 1985:234; Busby 2002:5345). In the beginning of the poem, it is retold
how Elye, who is a son of a feudal lord, proves his worthiness of being a knight, is dubbed
and leaves home to find adventures, after being challenged by his father. He encounters a se-
ries of troublesome episodes on his journey he fights against Saracens to help a group of
Christian warriors and gets captured on a Saracens ship, from which he escapes. This leads to
his meeting his fellow comrade, Galopin, a former thief, with whom Elye fights yet more
Saracens. Unfortunately, Elye is wounded and Galopin helps him hide in the gardens of a cas-
tle, which turns out to be the home of his enemies, but also of Rosamunde, a famously beauti-
ful Saracen princess. She not only heals his wounds and saves his life but eventually is also
baptised because of her love for Elye. Elye fights several more battles against various Sara-
cens, but at the end, he is compelled to wait for military assistance. A great Christian army,
consisting of the king and his well-known vassals, comes to his help. They take over the Sara-
cen castle, baptise all the people and turn it into a Christian stronghold. Since Elye has over-
seen Rosamundes christening, the two can not get married. Galopin and Rosamunde are, in-
stead, married and become the rulers of the new Christian stronghold, while Elye, the king
and the army return to France, where Elye is married to the kings sister.
The Old Norse version of the same poem is in prose and it appears in one Norwegian and
several Icelandic manuscripts.6 In this context, I will study the Norwegian manuscript as a
base of comparison with the Old French manuscript. The manuscript DG 47 is dated to ca.
1270 and it was written in south-west Norway, possibly around the town of Bergen, either in
the royal chancellery or in the nearby Lyse abbey, both of which were scribal milieus of suffi-
cient stature to produce a manuscript like DG 47 (Tveitane 1972:13; Holm Olsen 1940:83).
Note that the two manuscripts are rather contemporary and they are both related to great po-
litical and cultural milieus.
The content of Els saga in DG 47 is, however, somewhat different. First, the text is not
entirely extant. There is a lacuna, which must have contained the section where Els is cap-
tured on the Saracens ship and his escape, in addition to some other information.7 We are
therefore unfamiliar with the exact reading and structuring of this section. The other major
difference concerns the ending of the Old Norse saga. The saga ends prior to the grand battle

6
See Introduction in Kblings (1881) edition.
7
The lacuna in DG 47 is filled by the version from AM533 4to in Kblings edition.

243
between Christians and Saracens, and thus appears as a different story than the Old French
poem. The battle is very weakly suggested as a possible future plan, besides the potential for
future marriage between Els and Rosamunda.8

Initials size and styles

The Old French poem is structured in terms of initials in various sizes. The first initial spans
over eleven lines and is as wide as one of the two columns the text is written in. The majority
of the initials in Elye are two-line capitals. These initials are simple with no extra decoration,
appear alternatively in red and blue, and visualise the textual partition of the poem. Besides,
there are a few nicer three-line pen-flourished initials which appear in connection to illustra-
tions. There are in total six illuminations in the poem which contribute further to a specific
structuring of the text.
The initials in the Old Norse version are all two-line initials, with the exception of the ini-
tial in the beginning of the saga, which spans over five lines. The latter is however missing, as
are some of the two-line initials.9 The initials that are inserted are relatively simple, either
green with red details or red with green details.
In the following the placement of the initials in the two text versions will be related to four
different textual aspects: the rhythm in the text, various rhetorical devices, lexical evidence on
reception mode and the narrative structure of the text-versions.

Initials and rhythm

The initials in the Old French version correspond to the rhythmical pattern of the poem, since
there is a change in assonance with all initials, but one.10 Besides, while the majority of the
poem is written in dodecasyllable form, with the caesura dividing the verse into 6+6 syllables,
some of the verses appearing in relation to initials are decasyllable (6+4 or 4+6).11 Thus the
graphical emphasis of the initials would have been audible because of the assonance change
or/ and the shortness of the verse.
The Old Norse version of the text is in prose and may in general be characterised as less
rhythmical than the poem-version. But still there are a couple of rhythmical patterns that are
used in relation to the initials. First, alliterations are often used in the beginning of chapters in
order to render the text rhythmical and audibly appealing. Besides, at the end of some of the
chapters, one may unveil a certain prose rhythm/ cursus, possibly originating from Latin.
Kirsten Berg points out that determining conscious use of cursus depends on the punctuation
principles in the manuscript, the merging of words and the style of the text (Berg 1999:171
72). She claims to find a conscious use of cursus in several translated sagas from the second
half of the 13th century Barlaams ok Josaphats saga, Konungs Skuggsj and Els saga and
Els saga in DG 47 (Berg 1999:175, 177).

Initials and rhetorical devices

The initials in the Old French text correspond to various rhetorical devices as well. Some of
these are narrators comments, which appear either at the end or in the beginning of chapters.
The comments vary in character, some include emotional exclamation and appeals to God,

8
Note that the ending of the Old Norse in the Icelandic manuscripts differs from the DG version, since the final
grand battle is accounted for. However, the end differs from the Old French version as well.
9
See Tveitane (1972) and Holm-Olsen (1940) for possible explanations.
10
For a presentation of the various assonances, see Raynaud 1879:1011.
11
Note that such shorter verses appear also independently from the initials, and may rather be explained by their
special content and narrative function in the poem.

244
others conatin a qualitative evaluation of an event or a protagonists, others refer backwards
and forward in the story, yet others have proverbial qualities.
Narrators comments occur in relation to initials in the Old Norse version of the text as
well. These are in general less emotional than in the Old French. Some of the comments have
intertextual character and refer to the Christian faith of the narrator. Sometimes the narrator
expresses his personal evaluation of the situation. The whole of chapter 14, for example, is an
address to the audience by the narrator, which has proverbial character (50a1):

Nu lyit gogfliga. betra er fogr fre en kuiar fylli. o scal vi saugu spa. en ofmikit
drecka sm. er saugu at segia ef hyrenr til lya. en tapat starfi at hafna at hyra ------------

Now listen carefully! It is better to be knowledgeable than full in the stomach; one may drink
when stories are told, but not too much; it is honourable to tell stories, if people listen, but it is a
wasted effort, if nobody listens.

Note also that this chapter is an addition in the Old Norse version, which in style and content
may remind a traditional so-called Hvaml-wisdom.
A second rhetorical device that appears in relation to initials is repetitions. Various types of
repetitions appear in the Old French text. The first type is when important information from
one chapter is given in the beginning of the next. Sometimes the repetition involves also lexi-
cal repetition, with variation. Some phrases are repeated often and may be considered formu-
las, but these appear not only in correspondence to initials. Another type of repetition that
occurs in the Old French text is of episodic nature, when a similar event occurs two or three
times. Some new details or a tiny twist takes place, and thus distinguishes the new episodes
from the previous and drives the narration forward. It is the frame or the structure of the epi-
sodes that is repeated. It has been observed that the length in number of verses of such re-
peated episodes may vary, and amplifying the episode may be a means of emphasizing it
(Heinemann 1987:24).
Repetitions occur in the beginning of chapters in the Old Norse version as well, but the na-
ture of the repetition is different from those in the Old French version. Many of the repetitions
are much shorter, and constitute just a clause of the first sentence of a chapter, in structures
like When he heard that []/ When he saw[]. On several occasions there are also repeti-
tions on lexical level, with variation or oppositions. There are also some examples of the epi-
sodic repetition, when a similar event occurs two times. But not all episodic repetitions from
the Old French are preserved in the Old Norse version.
Another rhetorical tool which appears very often in correspondence to initials is temporal
adverbs and connectives. In the Old French version, the two temporal adverbs are quant and
or. In the Old Norse version, however, the temporal adverbs are more varied, and include n,
sem, n sem, a, en n sian, etc.
Thus, it seems that the same types of rhetorical devices are used in correspondence to the
initials in both versions, but these are used in different modes the narrators comments are
fewer and less emotional, the repetitions are also fewer, while more types of temporal adverbs
are used. Note also that the extensive use of temporal adverbs, narrators comments and repe-
titions, have been interpreted, independently from the materiality of the manuscripts, as tools
which render a text suitable for listening audience

Initials and lexical evidence

Further, in the beginning and end of the Old French poem, and other places (possibly places
for breaks, but not always in relation to initials), there is explicit lexical evidence that the text
was to be listened to. Here are some examples:

245
 Or faites pais, Signor, que Dieus vous benie / Now, be silent lords, may God bless you (ch. 1,
1)

Or escouts, Signor, que Dieus grant bien vous don/ Now Listen, my Lords, and may God pro-
vide well for you (ch. 19, 572)

Or vous dirons noveles du bachelor Elies/ Now we will tell you the news of the young knight
Elye (ch. 60, 2083)

The very beginning and end of the poem are also rich in lexical evidence relevant for the in-
tended reception mode. Regarding all the examples at once, the intended reception by hearing
of the text is plausible. Note also that such comments appear also in the middle of various
chapters and are not always corresponding to initials
As in the case of the Old French poem, the beginning and ends of the chapters in the Old
Norse version do often contain relevant lexical information. The first, last and several other
chapters contain constructions of the type Listen []/ Hear []. The saga starts (ch. 1,
43a15):

HYRIT horskir menn. ina fagra saugu. dyrlegs drengscaps. um raustan riddera scap & lof-
sla atger. ins uirulegs hertoga. er stiornn & riki.

LISTEN, wise men, to a beautiful story of glorious manhood, of valiant chivalry, of the glorious
and praiseworthy deeds of a virtuous count, who is powerful and mighty []

The short chapter 14, which was cited above (see p. 3), also contains interesting lexical evi-
dence.
The information regarding the reception of the two texts is, thus, not corresponding fully
some pieces are missing and other new ones are present in the Old Norse vs. Old French ver-
sion. In general, however, such information is less abundant in the Old Norse than in the Old
French version.

Initials and narrative structure

The initials divide the Old French poem in laisses/ chapters between five and 376 verses.
With regard to chanson de geste, it has been suggested that sometimes these may be struc-
tured by means of long or short laisses in order to provide a rhythm and variation in the
poems flow, which possibly corresponds to the narrative structure of the poem. In order to
comment on such a hypothesis with regard to Elye, I have studied the dramaturgical plot of
the poem, defined by means of the length of the chapters. The poem is, thus, given a narrative
structure with clear summits, that are reached sometimes gradually by means of secondary
peaks preceding and following them (section 2240) and are other times set up more inde-
pendently (section 4448). If the content of the long/ short chapters are studied a certain ten-
dency may be revealed. The sections consisting of laisses which are of similar and relatively
low length contain similar type of information a series of events in a single scene, in which
the narration is driven forward at a constant speed, episodic repetitions which have similar
narrative significance, or alternating voices in one dialogue. Thus, these sections seem to set
up a rhythm to the narrative structure of the whole poem, due to their content.
The next question concerns then the content of the absolutely longest laisses in the poem
and the secondary summits, which may be regarded as long in comparison to their surround-
ings. The plot of the poem, if described in terms of these peaks, is a story about Elye who is

246
off on his way to find adventures; he is taken prisoner on a ship, from which he escapes; he
meets his comrade Galopin; they end up at Sobrieborg, where they meet Rosamunde; the
Saracen princess saves Elyes life and is later about to be married to another man. The poem
ends with a great fight and victory for the Christian army, but a somewhat unhappy ending for
the couple. Besides, the illustration program of the poem seems tightly related to the narrative
structure of the poem, as defined by the length of the laisses, since the theme illustrations cor-
respond very closely to the themes of the peak-laisses. Thus, I would claim that there exists
certain correspondence between the placement of the initials and the narrative structure of the
story, since the series of laisses of similar length contain episodes of similar function, provid-
ing a regular speed to the evolvement of the plot, while the longer laisses correspond to the
most significant episodes, where there is a break in the regularity of plot evolvement, and thus
provide a skeleton of the narrative.
The Old Norse text is divided in chapters of various lengths as well these vary from four
to 105 lines. Note that while the Old French poem had 69 chapters, the Old Norse text has
only 54. The difference may be explained, to some degree, by the lacuna in the manuscript.
Besides, as already mentioned, the Old Norse version is shorter than the Old French.
Besides, the narrative structure of the saga, if defined by means of the length of the chap-
ters, is not marked by distinguished summits as clearly as the Old French poem. Thus, it may
be said that the correspondence between the placement of the initials and the narrative struc-
ture of the text is not as strong as in the Old French poem. One explanation for that may be
that the scribe did not consciously employ a strategy to structure his text by means of the
lengths of the chapters. This was possibly a less common strategy when a prose text was writ-
ten than when a chanson de geste was written. If the theme of the narrative peaks in the Old
French and Old Norse texts are compared, it turns out that some of the summits are the same,
but not all. There is absolute correspondence in three of these summits when Els meets the
messenger on the road (chapter 5 in both versions), when he meets the thieves and Galopin
(ch. 29 in Old French, ch.24 in Old Norse), and when Rosamundes father decides to marry
her off to another man (ch. 47 in the Old French poem and ch. 37 in Old Norse saga). Other-
wise, the description of the starting point for Els seek for adventure is important in the Old
Norse text, but is not emphasised equally in the Old French. Still, the same scene was de-
picted in the first illustration of the poem. The episode when Els and Galopin meet
Rosamunda and she heals Els appears as a summit in the Old Norse version, but only as a
secondary summit in the Old French. The episode when Galopin steels the horse is a secon-
dary summit in the Old French poem, and does not appear as a major episode in the Old
Norse version. Further, the episode when Els goes out to fight a duel with a Saracen in order
to protect Rosamunde from being married is central in the Old Norse version and is not im-
portant at all in the Old French version. The Old French version ends with a description of a
grand battle, a newly Christianised land and a couple of marriages; none of these episodes
occur in the Old Norse version, even though the battle is suggested as a possible continuation
of the plot. Thus, despite the similarities, the narrative structures of the two versions defined
by means of initials are different. While the Old French text appears as a description of a se-
ries of adventures of one protagonist, that culminates with a grand battle between Christians
and Saracens, the Old Norse version appears more as a story about a hero and his relations to
his father, his mother, his friend, his lady and his enemies.

Implications
In the previous, I have studied and compared the correspondence between one graphical as-
pect (the initials) and four textual aspects (rhythm, rhetorical devices, lexical evidence, and
narrative structure) of two versions of one text the Old French chanson de geste Elye de

247
Saint-Gille and the Old Norse Els saga. The main result from this investigation is that in both
versions a certain correspondence was at stake, but the type and degree of correspondence
differed. These results have several possible implications
When it comes to intended reception mode of the two versions, the Old French version
seems to have been intended for both ocular and aural reception, based on the presence of
illustrations and the correspondence between the graphic and rhetoric emphasis. One possibil-
ity is then that it was to be publicly read to audience that had visual access to the manuscript
prior to or during the reading sance. The Old Norse manuscript seems to have been intended
for being listened to rather than looked at.
When it comes to the possible function of the two versions, the difference in their structure
renders their core message rather different. The Old French poem is a story about crusades,
about taking and Christianisation of new land, about strong Christian norms and a powerful
king. The Old Norse saga, on the other hand, is a story of personal adventure, inner conflicts,
friendship and love relationship. These themes also appear in the Old French poem, but re-
main secondary having in mind the final scene in the poem, which is entirely missing from
the Old Norse saga.
Since the Old Norse text is different than the Old French, it seems that it was not trans-
lated/ rewritten word for word, but changes were made, as seemed appropriate. One obvious
reason for that is the difference in form (verse vs. prose) which would have necessitated some
changes. But still, this may not be an explanation for omissions, additions, difference in narra-
tive structure and in used rhetorical techniques.
Such changes may further indicate that translating/ copying (since we do not know who
made the changes the translator or the copyist) were activities similar to composing, which
required competence in available rhetorical tools and their appropriate use. The translator/
scribe may thus be characterised as competent and erudite in classical composition tech-
niques, such as change of narrative structure, abbreviation and addition, just to mention a few.
Finally, the materiality and textual style of the Old Norse saga has, in the previous, been
claimed as different from the Old French poem. Some aspects of the Old Norse text, such as
the prose-rhythm, the variation of temporal adverbs, however, may remind of other Old Norse
texts from the same period. Therefore, it may be said that the scribe of the Old Norse version
complied with local standards for retelling which may testify that Old Norse language and
literary tradition had greater status than the original Old French language and literary tradition
for the scribe. It should be said that the conclusion may be quite different if another aspect of
the text, or a different text, is investigated. Such a conclusion has been made by other scholars
working on translations, but none of them have studied the same aspects of the texts as I have.
To summarise, based on the correspondence between the materiality of the two manu-
scripts BNF 25516 and DG 47 and the textuality of the two texts Elye de Saint-Gille and Els
saga, it appears that more-or-less the same story about travelling to the East was intended to
be received in different ways and possibly to fulfil different functions. In the transmission
process, the story has changed character from being a political narrative of national impor-
tance to being a didactic story of personal conflicts and ideals.

Bibliography
Berg, Kirsten M., 1999: Cursus i norrnt? en metodediskusjon med eksempler. Maal og Minne
1999:165186.
Busby, Keith, 2002: Codex and Context. Reading Old French Verse Narratives in Manuscripts.
Clanchy, Michael, 1997: From Memory to Written Record. England 10661307.
Coleman, Joyce, 1996: Public Reading and the Reading Public in Late Medieval England and France.
Copeland, Rita, 1991: Rhetoric, Hermeneutics and Translation in the Middle Ages.
Crosby, Ruth, 1936: Oral delivery in the Middle Ages. Speculum 11 (1):88110.

248
de Winter, Patrick, 1985: La bibliothque de Philippe le Hardi, duc de Bourgogne (13641404): tude
sur les manuscrits peintures dune collection princire lpoque du style gothique international.
Derolez, Albert, 2003: The Paleography of Gothic Manuscript Books. From the Twelfth to the Early
Sixteenth Century.
Fleischman, Suzanne, 1989: A Linguistic Perspective on the Laisses Similaires: Orality and the Prag-
matics of Narrative Discourse. Romance Philology 42:7089.
Fleischman, Suzanne, 1990a: Philology, Linguistics, and the Discourse of the Medieval Text. Specu-
lum 65(1): 1937.
Fleischman, Suzanne, 1990b: Tense and Narrativity: From Medieval Performance to Modern Fiction.
Green, D.H, 1994: Medieval Listening and Reading. The Primary Reception of German Literature
8001300.
Green, D.H, 2002: The Beginnings of Medieval Romance. Fact and Fiction, 11501220.
Heinemann, Edward A., 1987: Measuring Units of Poetic Discourse: Analogies between Laisse and
Verse in the Chanson de Gests. In: Romance Epic. Essays on a Medieval Literary Genre. Edited by
Hans-Erick Keller. Pp. 2134.
Holm Olsen, Ludvig, 1940: Den Gammelnorske Oversettelsen av Pamphilus, med en Underskelse av
Paleographi og Lydverk.
Kbling, Eugen, 1881: Elis saga ok Rosamundu, mit Einleitung, Deutscher bersetzung und
Anmerkungen zum Ersten Mal Herausgegeben.
Malicote, Sandra., Forthcoming: Elye de Saint-Gille. Critical Edition with an English Translation.
Malicote, S.C.O., 2006: Visual and Verbal Allusion. Disputatio and the Poetics of Elie de Saint Gille
and Aiol. Romania 124:77111.
Parkes, M.B., 1992: Pause and Effect. An Introduction to the History of Punctuation in the West.
Raynaud, Gaston, 1879: Elie de Saint Gille: chanson de geste, publ. avec introduction, glossaire et
index par Gaston Raynaud ; accompagnee de la redaction norvegienne, trad. par Eugene Koelbing.
Saenger, Paul, 1982: Silent Reading: Its Impact on Late Medieval Script and Society. Viator 13:367
414.
Saenger, Paul, 1997: Space Between Words: The Origin of Silent Reading.
Spurkland, Terje, 1994: Mtte Herren hjelpe den mann som ristet disse runer og liks han som leser
dem! Norsk Lingvistisk Tidskrift 12:315.
Spurkland, Terje, 2000: eir bru fram brf ok segja rendi au sem fylgu. Om brevveksling i
middelalderen. In: Den Nordiske Renessansen i hymiddelalderen. Edited by Jn Viar Sigursson
and Preben Meulengracht Srensen. Pp. 4763.
Tveitane, Mattias, 1972: Introduction. In: Elis saga, Strengleikar and Other Texts. Uppsala University
Library Delagardieska Samling Nos. 47 Folio and AM 666 Quarto. Pp. 934.
Vitz, Evelyn Birge, 1999: Orality and Performance in Early French Romances.

249
 From saga to Chronicle: Motif Migration inside Medieval
Scandinavia
Fulvio Ferrari, Universit di Trento, Italy
Three monumental historiographical works enlighten us on the history of the Norwegian and
the Danish royal dynasties during the Viking age and the Middle Ages: Saxos Gesta Da-
norum, dated to the beginning of the 13th century; Heimskringla, written by Snorri Sturluson
around 1230; and Kntlinga saga, perhaps composed by Snorris nephew, lfr rarson,
some decades after the composition of Heimskringla.
The first Swedish comparable work, Erikskrnikan, was written more than half a century
after Kntlinga saga in the period 1320 1335, and most probably soon after King Birger
Magnussons expulsion from Sweden and his son Magnus Birgerssons execution, during the
years 1320 1322 (Ferrari 2008). Whilst the strong similarities between Heimskringla and
Kntlinga saga have already been recognized and pointed out (Bjarni Gunason 1982: 115
117), as well as the circulation of literary, mythological, and folkloric motifs between the
Latin chronicle and the vernacular sagas, the relationship between the Swedish chronicle, and
the preceding Danish and Norwegian historiographical works is less investigated.
The differences between the Swedish chronicle and the older Scandinavian histories are, in
fact, striking: whereas Saxo alternates prose and verse, following the literary tradition of the
prosimetrum, and the two king sagas are mainly works of prose with poetic insertions,
Erikskrnikan is a verse composition which fits in the genre of the rhymed chronicle. Both the
identity of the patron, and of the poet of Erikskrnikan, are still a matter for discussion among
scholars, but no serious doubts exist with regard to the social and cultural context in which the
text was composed and performed: not only the hints contained in the text itself, but also the
whole ideology of the chronicle, point to the aristocratic milieu which held power in the
Swedish kingdom after King Birgers flight to Denmark and the acclamation of Magnus
Eriksson then still a child as the new king of Sweden (Vilhelmsdotter 1999: 4147;
Pneau 2005: 3440; Ferrari 2008: 5556).
Since the reign of Magnus Laduls Magnus Erikssons grandfather the Swedish court
and aristocracy had rapidly adopted chivalric norms, language and habits, thus conforming to
the continental standards of political organization and to the ideology of power which already
had spread into the neighbouring kingdoms of Denmark and Norway (Harrison 2002: 150
160). Such political and cultural reshaping was strongly facilitated by the dynastic intertwin-
ing of the reigning families of Northern Europe (Mitchell 1997; Wrth 2000). Literature
played a relevant role in this process of modernization and Europeanization of Scandinavia,
and the relationships between the courts of Norway, Sweden and Northern Germany were
surely responsible for the use of the stylistic devices of the genre rhymed chronicle in order
to present Swedish history from the beginning of the 13th century up until the more recent and
dramatic events of the first two decades of the 14th century. In the period immediately preced-
ing the composition of Erikskrnikan, the German-born queen Eufemia of Norway had given
a strong impulse to the formation of a Swedish literary system by commissioning the transla-
tion into Swedish of three courtly romances the so called Eufemiavisor which she in-
tended to give as a gift to her future son-in-law, the very same Duke Erik Magnusson who is
the central figure of Erikskrnikan and the father of King Magnus Eriksson. With these three
romances, which adapted to the Swedish language the metrical rules of the contemporary
German narrative poetry, Eufemia contributed in a decisive way to differentiating the Swedish
literary system from the West Norse tradition. The influence of the Eufemiavisor upon
Erikskrnikan is recognized by all the scholars in the field, but it is difficult to think that the

250
author of the chronicle had no acquaintance with some of the German rhymed chronicles.
Eufemia was not the only link between the Swedish and the North German courts: king Mag-
nus Laduls married Helvig, the daughter of Count Gerhard of Holstein, and his sister Rikiza
was married, the first time, to the Norwegian King Hkon hinn ungi, and the second time to
Heinrich I of Werle. From this second marriage was born Rixa von Werle, which married
Albrecht II, Duke of Braunschweig-Lneburg. Albrecht was in Stockholm in 1289 and, as the
Erikskrnikan relates (ll. 11461161), on that occasion he was made a knight by King Mag-
nus Laduls son, the future King Birger Magnusson. Considering the tight relationships be-
tween the courts of Sweden and Braunschweig, it seems most likely that the author of
Erikskrnikan had the opportunity to get acquainted at least with the Braunschweiger Reim-
chronik, a rhymed chronicle which was written in the circle of the Dukes of Braunschweig
during the second half of the 13th century (Pipping 1926: 774776; Lnnroth 1987: 107;
Mitchell 1996: 27; Pneau 2005: 1314).
The political and personal connections between the aristocratic lites of Sweden, Norway
and Northern Germany can thus explain the formal and stylistic choices made by the author of
Erikskrnikan and the decision to import into the young and weak Swedish literary system
a new literary genre, that of the rhymed chronicle, which had already established itself in the
neighbouring German cultural milieu (Even-Zohar 1990a and 1990b). Just as important as the
relationships between the Swedish and the German dynasties, however, were the complicated
and quite often stormy relationships among the Scandinavian courts. Valdemar Birgersson,
the elder brother of King Magnus Laduls, married princess Sofia, the daughter of the Danish
King Erik Plogpenning. Sofias cousin, the Danish King Erik Klipping, actively took part in
the conflict between Valdemar and his younger brother, supporting first Valdemar and then
Magnus. After conquering the Swedish throne, Magnus married his daughter Ingeborg to Erik
Klippings son and successor, Erik VI Menved, and arranged the marriage between Erik
Menveds sister, Margrete, and his own heir, Birger (Bagge 2007: 8). Consequently, Erik
Menved supported his brother-in-law to the very end in his fight against the brother, Duke
Erik of Sdermanland, who, on his part, for most of the war was supported by the Norwegian
King Hkon V and by his influential wife, the already mentioned Eufemia of Rgen. More-
over, the tempestuous relationships inside the dynasties as well as between the royal families
and the aristocratic lites were the cause for the presence in the Scandinavian courts of nu-
merous exiles, who looked for shelter in the neighbouring lands.
Such a vast, interconnected social and political network makes it unlikely that the great
historiographical works which were composed in connection with the Danish and Norwegian
courts, or directly under the patronage of Danish and Norwegian kings, were unknown in the
Swedish aristocratic milieu and, more specifically, to the author of Erikskrnikan. Further-
more, the genre of konungasgur was productive throughout the 13th century, as Sturla
rarsons Hkonar saga Hkonarsonar and Magns saga lagabtir testify. It is therefore
reasonable to wonder whether, in the case of Erikskrnikan, traces of an intertextual practice
which involved not only other Swedish texts the Eufemiavisor and texts which belong to
the same literary genre the German rhymed chronicles , but also the previous Scandinavian
historiographical works are perceivable.
With regard to its structure, Erikskrnikan consists of a series of episodes, organized ac-
cording to the guiding principle of the succession of the kings. This structural principle is
shared by the rhymed chronicles as well as by Heimskringla, Kntlinga saga and Saxo, and
therefore it cannot by itself say much about any intertextual relationships between the Swed-
ish chronicle and some particular text. In one respect, however, Erikskrnikan shows an inter-
esting similarity with the Kntlinga saga. In order to relate the vicissitudes and the deeds of
the Danish kings, this saga follows in the steps of Snorris Heimskringla: Snorri had put the
figure of the holy king Olaf Haraldsson in the middle of his history, thus making of this kings

251
life the very core of his entire work. Likewise, the author of Kntlinga saga puts the figure of
Knut the Saint in the middle of his saga, assigning to his life the same ideological and struc-
tural function that Olafs life played in the Heimskringla. For the author of the Kntlinga,
however, it was of fundamental importance to give a prominent position also to another figure
of the Danish past, the figure of Knut Lavard. We cannot with certainty know to which pur-
pose the saga was written, and to which audience it was addressed. Possibly it was commis-
sioned by King Valdemar the Victorious in order to have a vernacular counterpart to the his-
tory of the Norwegian kings written by Snorri Sturluson, and possibly it played a role in the
Icelandic political public discourse in the middle of the 13th century (the last hypothesis does
not exclude the first one). In any case it was necessary for the author of the saga to single out
and to exalt also the figure of Knut Lavard, who was the father of Valdemar the Great and
thus the direct ancestor of Valdemar the Victorious (Bjarni Gunason 1982: 144). To this end
the author deviates from the adopted narrative scheme: as he comes to the reigning period of
King Niels (11041134) he does not focalize on the king, but on Knut, making of Niels and of
his son Magnus the antagonists of the Duke and the villains of the tale. A comparable narra-
tive strategy, on the other hand, is carried out also by Saxo in book 13 of Gesta Danorum. The
same shift in focalization occurs in Erikskrnikan as Duke Erik enters the scene, with the sub-
stantial difference that the story of his conflicts with his brother, King Birger, and of the civil
war following his murder in prison, occupies the greatest part of the chronicle, almost reduc-
ing the preceding sections, devoted to the stories of the previous kings of Sweden, to the rank
of a long prologue.
It must be admitted that the story of Erik Magnusson presents striking similarities with that
of Knut Lavard, which had occurred almost two centuries before: both noblemen were the
ambitious and resolute sons of a king, they had been excluded from the succession, but had
succeeded in creating a strong power base and a vast network of alliances, thus representing a
danger for the ruling kings. Both Erik and Knut Lavard were deceitfully imprisoned and killed
by close relatives a brother, in the case of Erik; an uncle and a cousin in the case of Knut
and in both cases civil war had been the result of the murder. Furthermore, in both instances
the war concluded with the defeat of the murderers, and the sons of the murdered dukes fi-
nally ascended to the throne. This apparent parallelism in the two stories is rooted in the his-
torical events of which Erik and Knut were the protagonists, and it cannot therefore demon-
strate by itself the existence of intertextual relationships between the Swedish chronicle,
Kntlinga saga and Saxos Gesta Danorum. However, if we take into consideration the liter-
ary motifs and the explicit intertextual references present in Erikskrnikan, the hypothesis of a
relationship between the Swedish chronicle and the older Danish histories appears to be more
convincing.
First of all, it is worth noticing that even if Duke Erik Magnusson is never explicitly pre-
sented in Erikskrnikan as a saint, the narrative of his death and that of his brother Valdemar
shows them as innocent victims which suffer martyrdom by command of their own wicked
and treacherous brother. The parallel established by the narrating voice between King Birger
and his accomplices on the one side and the biblical figure of Judas on the other side suggests
that a religious interpretation of the events is implied. Judas is mentioned the first time as the
narrator, anticipating the imprisonment of the two unaware brothers, curses the author of such
a perverse deception:
han wari fordmpder nw ok
med iudasse ok slla thee
Ther nidre sithia i heluitis poth
ok haffua ilt ok aldregh goth

252
 (May he be damned forever / together with Judas and the wicked / which sit down in hells
abyss, / they always suffer and never have any joy, ll. 36323634)

The comparison is proposed a second time soon after, as the narrator describes the banquet
offered by King Birger to his brothers. In commenting on the scene, the narrating voice first
states that the king, his wife and his men treated Erik and Valdemar as Judas treated Jesus
(the haffdo som iudas wider crist, l. 3767) and then adds:
[J]ak hrde for pascha at man laas
i scriptenne aff iudas
at han swek van herra i tro
thy skal han i heluite boo
ok haffua ewynnelika wee
swa skulo ok alle the
ther orlika myrda ok forradha
j then pina ther aldreg komber nade
The mogha sarlika jwa om sik
som forraddo hertogh Erik

(At Easter I have heard read / about Judas from the Holy Scriptures: / that he betrayed Our
Lord / and therefore he will dwell in hell / and suffer forever. / The same will happen to every-
body / who shamefully kills and betrays: / they will all suffer pain without mercy. / Especially
they have to fear for themselves / who betrayed Duke Erik, ll. 37683777)

The motif of Judas is present in Saxo as well as in lnoths Gesta Swenomagni Regis et
Filiorum eius et Passio gloriosissimi Canuti Regis et Martyris and in Kntlinga saga (Ferrari
1998: 96100). In Saxo, in fact, Knut the Saint is deceived and betrayed by Blacco, a charac-
ter which actually has two counterparts in Kntlinga saga sbjrn Eydanajarl and Eyvindr
Bifra whereas Knut Lavard is deceitfully killed by his cousin Magnus.
Of particular interest for our discussion is the description, in Erikskrnikan, of Erik and
Valdemars reaction to the unexpected and treacherous aggression of their brothers men. The
dukes are sleeping together, naked and defenceless, as a group of warriors break into their
room. Valdemar reacts immediately, trying to strike back, but Erik dissuades him from fight-
ing:
Hertogh Erik sagde lat wara som er
war stridh dugher ekke nw hr

(Duke Erik said: let things go their own way / there is no use to fight now, ll. 38583859)

An analogous distribution of roles between the brothers the younger more impulsive and
willing to fight, the elder wiser and ready to accept the impending death is present also in
the narratives about the last hours of Knut the Saint. Kntlinga saga, in particular, emphasizes
the difference in attitude between King Knut and his brother Benedikt in a series of dialogues
in which the younger brother exhibits his courage and his will not to surrender to the assail-
ants, whereas Knut calmly prepares himself to die.
It is unlikely that the authors of Kntlinga saga and of Erikskrnikan were unaware of the
structural similarity between their narratives and the episode of the arrest of Jesus as related in
all four gospels (Matt 26, 5155; Mark 14, 47; Luke 22, 4951; John 18, 1011). This similarity ac-
tually suggests a confrontation between the deceitfully killed noblemen and Jesus Christ as a
model of sanctity, thus contributing to exalting the protagonists of the narratives (Jansson
1971: 172173). The sanctity of Knut IV and Erik Magnusson, moreover, is confirmed by

253
further allusions contained in the description of their death. Like the first Christian martyr,
Stephan, Knut is hit by a stone, which is thrown through a window of the church in which he
has found shelter, while like the holy King David he is singing psalms, and finally he is
treacherously pierced by the sword of the Judas-like character Eyvind Bifra. Also Saxo makes
use of symbolic elements in order to draw a parallel between the king and Jesus: in his de-
scription of his death, Knut is pierced by a spear (immiss per fenestram lance mortifico
iactu confossus, Liber XI, 15) while he is lying on the floor with his arms spread out
(propassis utrimque brachiis, ante aram securus occubuit, Liber XI, 15). As to Erik, accord-
ing to Erikskrnikan he is shut in a tower together with his brother Valdemar, and their hands
are nailed to a stock in front of them (thera hender waro ngelda a stokkin fram, l. 3948), so
that their suffering is alike to that of Jesus nailed on the wood of the cross (Jansson 1971:
174).
The parallel strategies carried out by Erikskrnikan on the one hand, and by Saxo and the
author of Kntlinga saga on the other one, make it at least thinkable that the Swedish chroni-
cler had some knowledge of the Danish tradition. This supposition is enhanced by an interest-
ing similarity that we can find between the narrative about King Birgers deceitful invitation
to his brothers Erik and Valdemar in Erikskrnikan and Prince Magnus invitation to his
cousin Knut Lavard in Saxo Grammaticus Gesta Danorum. In both cases the hero of the nar-
rative falls in the trap set by a close relative, and in both cases he is uselessly warned about
the danger. Erik and Valdemar are riding on the road to Nykping as a young knight (en
vnger riddare, l. 3708) rides towards them and tries to dissuade them from entering the
kings castle, saying: ridhin i bade i husit snder / thz skal ider angra ok alla idra frender
(if you ride to the castle / both you and your relatives will regret it, ll. 37123713). Never-
theless, the dukes do not listen to the unknown knight, and Erik even reproaches him for sow-
ing discord between them and their brother.
Sven-Bertil Jansson, in his interesting and stimulating study of the genre medieval
chronicle pointed out some structural similarities between the story about a treacherous invi-
tation as narrated in Erikskrnikan and the same motif in the Nibelungen matter, and came to
the conclusion that the Swedish chronicler probably had the opportunity to be acquainted with
the Norwegian ireks saga (Jansson 1971: 176184). Already Saxo Grammaticus, however,
had recognized a similarity between the story of Knut Lavards murder and the one of Krim-
hilds vengeance against her brothers, and he not only explicitly mentioned Krimhilds story,
but also made use of it in the construction of his narrative. In Saxos account of Knut
Lavards death, Magnus sends to the Duke a Saxon singer, who has to bring him the message
that he will wait for him in the woods. In vain one of Knuts servants tries to convince his lord
not to go unarmed to the meeting, the Duke answers that he does not need to entrust his life to
the sword (nequaquam se in salutis tutelam ferro opus habere respondit, XIII, 6). The singer
himself attempts then to warn Knut by singing a song about Krimhilds notorious intrigue
against her brother, in the hope that he would understand the allusion and save himself (Igi-
tur speciosissimi carminis contextu notissimam Grimild erga fratres perfidiam de industria
memorare adorsus, famos fraudis exemplo similium ei metum ingenerare tentabat, XIII, 7).
Saxo, actually, mentions the story of the Nibelungen without retelling it, but we can as-
sume that, given the strong structural similarity between the narratives about Knut Lavards
and Eriks imprisonments, the Swedish chronicler took Saxos reference to the Nibelungen
matter as a hint to shape his narrative according to the structure of the story about Krimhilds
vengeance.
Furthermore, the possibility that the author of Erikskrnikan knew the story of the Nibe-
lungen is made plausible by an allusion contained in the passage where the narrating voice
refers to Erik and Valdemars army:

254
 skulle man leta wt til riin
han kunne ey btre hlade faa
n hertogane haffdo mz sik tha

(even if one had searched as far as the Rhine / he could not find better heroes / than the ones
the dukes had in their army, ll. 22672269)

His knowledge of Gesta Danorum, on the other hand, seems to be confirmed by another allu-
sion contained in line 3916. In describing Birgers joy after his brothers imprisonment, in
fact, the narrating voice points out that the king behaves exactly as if he were a Hamlet (rt
som han ware en amblodhe). The word amblodhe is a hapax legomenon in the Old Swedish
text corpus, and one cannot be completely sure that it did not generically mean madman, but,
taking into account the other intertextual clues present in the chronicle, a reference to Saxos
work looks much more credible (Pipping 1926: 680681; Jansson 1971: 176; Jansson 1993:
236).
Even if we are not in possession of unquestionable proof, we find several clues in the text
of Erikskrnikan which point to an intense cultural interchange between the Northern Euro-
pean courts and the intellectual circles connected to them. In his choice of the literary genre,
its author was inspired by German rhymed chronicles which, at least, were known to the
German noblemen present at the Swedish court in the second half of the 13th and at the begin-
ning of the 14th century; in his use of metrical and stylistic devices he was surely indebted to
the Eufemiavisor (Vilhelmsdotter 1999: 3647), whose composition was commissioned by
the German-born queen of Norway for her future Swedish son-in-law. Finally, the strategy of
presenting the hero of the narrative as a saint-like figure, the structural organization of some
episodes in the text, and the exploitation of motifs present in Saxos Gesta Danorum and in
Kntlinga saga make it very likely that the Swedish chronicler had the possibility of acquiring
knowledge also of the historiographical tradition connected with the Danish court.

Bibliography
Bagge, Sverre, 2007: Aims and Means in the Inter-Nordic Conflicts 13021319. In: Scandinavian
Journal of History, 32 (2007). Pp. 537.
Bjarni Gunason, 1982: Formli. In: Danakonunga sgur (slenzk fornrit 35). Reykjavk. Pp. 5194.
Erikskrnikan enligt Cod. Holm. D2 jmte avvikande lsarter ur andra handskrifter. Nytryck (med ett
tillgg). Ed. by R. Pipping. 1963. Stockholm.
Erikskrnikan. Ed. by S. B. Jansson. 1993. Stockholm.
Even-Zohar, Itamar, 1990a: Laws of Literary Interference. In: Poetics Today, 11 (1990). Pp. 5372.
Even-Zohar, Itamar, 1990b: Interference in Dependent Literary Polysystems. In: Poetics Today, 11
(1990). Pp. 7983.
Ferrari, Fulvio, 1998: Modelli di santit e di eroismo nella Kntlinga saga. In: Tra edificazione e
piacere della lettura: le vite dei santi in et medievale. Ed. by A. DeglInnocenti & F. Ferrari.
Trento. Pp. 71101.
Ferrari, Fulvio, 2008: Literature as a Performative Act. Erikskrnikan and the Making of a Nation. In:
Lrdomber oc skmptan. Medieval Swedish Literature Reconsidered. Ed. by M. Bampi & F.
Ferrari. Uppsala. Pp. 5580.
Harrison, Dick, 2002: Sveriges historia. Medeltiden. Stockholm.
Jansson, Sven-Bertil, 1971: Medeltidens rimkrnikor. Studier i funktion, stoff, form. Stockholm-
Gteborg-Lund.
Lnnroth, Lars, 1987: Det hviska tilltalet. In: Den svenska litteraturen. Frn forntid till frihetstid,
8001718. Stockholm. Ed. by L. Lnnroth & S. Delblanc. Pp. 93123.
Mitchell, Stephen A., 1996: Literature in Medieval Sweden. In: A History of Swedish Literature. Ed.
by Lars G. Warme. Lincoln-London. Pp. 157.

255
Mitchell, Stephen A., 1997: Courts, Consorts and the Transformation of Medieval Scandinavian
Literature. In: Germanic Studies in Honor of Anatoly Liberman. Ed. by K. G. Goblirsch, M.
Berryman Mayou & M. Taylor. Odense. (NOWELE 31/32) Pp. 229241.
Pneau, Corinne, 2005: Introduction. In: Erikskrnika. Chronique dErik, premire chronique rime
sudoise (premire moiti du XIVe sicle). Paris. Pp. 592.
Pipping, Rolf, 1926: Kommentar till Erikskrnikan. Helsingfors.
Saxonis Gesta Danorum. Tomus I textum continens. Ed. by J. Olrik & H. Rder. 1931. Hauni.
Vilhelmsdotter, Gisela, 1999: Riddare, bonde och biskop. Studier kring tre fornsvenska dikter jmte
tv nyeditioner. Stockholm.
Wrth, Stefanie, 2000: Eufemia: Deutsche Auftraggeberin schwedischer Literatur am norwegischen
Hof. In: Arbeiten zur Skandinavistik (13. Arbeitstagung der deutschsprachigen Skandinavistik 29.
7. 3. 8. 1997 in Lysebu/Oslo). Texte und Untersuchungen zur Germanistik und Skandinavistik.
Frankfurt am Main. Pp. 269281.

256
 Narrative Trajectories between Nodal Points in the Cultural
Landscape
The Eriksgata of King Ingjald
Svante Fischer, Muse dArchologie nationale, Saint-Germain-en-Laye, France
This paper will discuss the fictional narrative of king Ingjald in Heimskringla and Ynglinga-
tal. The story of Ingjald may best be described as a mercurial career in arson. In my interpre-
tation of this text as an archaeologist, I would like to emphasize the narrative structure in rela-
tion to archaeological finds and place names rather than get entangled in the spurious historic-
ity of the protagonists. The episode of Ingjald is there as a narrative concept in the shape of a
lifes journey. When analyzing this, it becomes apparent that it is not the traditional cyclical
itinerary where a kleptocratic ruler travels between his farms that is under the loupe. By con-
trast, the narrative is linear with three consecutive arsons serving as nodal foci. This trajectory
can be applied to the cultural landscape of the Mlar Valley. In this paper, it will be argued
that the trajectory of the story essentially follows the Eriksgata from Uppsala to Rn in time
and space (Fischer 2005:18890). Although the Eriksgata is first mentioned only in the law
code of Uppland, Upplandslagen of 1296, new research on the Migration Period and Vendel
Period burials and gold hoards, as well as Viking Period runestones, will highlight the impor-
tance of certain relevant central places and place names along this route that are also men-
tioned in the written sources (Arrhenius 2004, Fischer and Victor 2008). The first journey of
the Eriksgata has traditionally been described as departing from Uppsala across Lake Mlaren
into Sdermanland. In its second leg in the inland hundred of Daga, it was said to pass be-
tween Norrtuna and Sdertuna in Frustuna parish before turning southwest by Bjrnlunda via
the S 2 Axala inscription on a rock wall along the route. Then the Eriksgata arrived in Aspa,
with four runestones along the route, past the Aspa hg, a possible legal mound. The exact
itinerary of the next step is more uncertain. The story unfolds as follows:
The other children at play pick upon Prince Ingjald. At the insistence of his foster father
Svipdag the blind, young Ingjald eats a wolfs heart as a child (Ynglingasaga 34).
Ingjalds father Anund dies (Ynglingasaga 35).
Ingjald vows to enlarge his inherited kingdom in front a group of invited neighboring po-
tentates. These petty kings are then subject to arson in Uppsal (Ynglingasaga 36).
A rival king, Hiorvard Ylving arrives in Myrkvafiorr (Ynglingasaga 37). This ought to be
Mrkfjrden.
War in Sviio (Ynglingasaga 38). This name first appears on S Fv1948:289, the third
node discussed below.
Granmar is arsoned in Sili (Ynglingasaga 39). This ought to be in the area of Kolsundet on
Selan, but as Wessn notes (1952:43): Frsta vokalens kvantitet ses ej av hs. This is the
site of Tuna in Yttersel parish on Selan in Sdermanland. Here, I shall focus on the crema-
tion burial ground of Tunaby Mellangrd (SHM 9435), excavated at the turn of the 20th cen-
tury by the Swedish archaeologist Bernhard Salin and his Finnish collegue Alfred Hackman.
Grave 24, with an the unusual find of a barbaric imitation of a solidus struck for emperor
Theodosius II sometime after 441 AD has remained somewhat of a riddle (e.g. berg 1953). I
have since been able to show that the imitation is die-identical with a 1991 find from the
solidus hoard of Botes, Etelhem parish, Gotland (Fischer 2008). With 84 solidi from the pe-
riod 402533, the Botes hoard is the largest solidus hoard in Scandinavia (Fagerlie 1967,
hoard 137b).

257
 The solidus imitation from Tunaby Mellangrd. After Janse 1922.

Ingjald and his daughter sa are arsoned in Rningi (Ynglingsaga 40). The third important
central place is to be found in Aspa and Ludgo parishes in Rn, Sdermanland. Here, the
Daga hundred legal assemblies and execution grounds and runic inscriptions line the Eriks-
gata before the journey stretches across Rn south via Jnker and Kolmrden to Svintuna in
stergtland. Wessn (1952:44) shows that this place name is sometimes spelled differently:
rningi, reiningi. Cf. Codex Frisianus (attributed by Wessn to an Icelandic scribe, 1325):
rauningi. Cf. S Fv1948;289 ostrith : lit : -ira : ku(m) usi at : anunt auk : raknu-
alt : sun : sin : urthu : taR : (t)an-(k)u : ua-u : rikiR : o rauniki : ak : snia-
lastiR : i : suithiuthu. Jansson (1987:105) argues for Codex Frisianus as the correct source
and argues that its location is identical to Aspa, Ludgo parish.

Bibliography
berg, N., 1953. Den historiska relationen mellan folkvandringstid och vendeltid. KVHAA. Stock-
holm.
Arrhenius, B., 2004. Det forntida Sveariket: en myt eller en arkeologisk realitet? KVHAA rsbok,
pp. 203227. Stockholm.
Fagerlie, J.M., 1967. Late Roman and Byzantine Solidi Found in Sweden and Denmark.
American Numismatic Society 157. New York.
Fischer, S., 2004. Kulturell mngfald i Sdermanland under folkvandrings- och vendeltid?
Agneta kerlund, ed. Kulturell mngfald i Sdermanland, del 2. Lnsstyrelsen i Sdermanland, rap-
port. Nykping.
Fischer, S., 2005. Roman Imperialism and Runic Literacy The Westernization of Northern Europe
(150800 AD). Aun 33. Uppsala.
Fischer, S., 2008. Brief Events and Long-Term Change The Slow Impact of Foreign Influences in
Valsgrde. S. Norr, ed. Valsgrde Studies. the Place and its People, Past and Present. OPIA 43.
Uppsala.
Fischer, S. and Victor, H., 2008: De gnistrande kammargravarna i Lilla Sylta. M. Olausson, (ed.).
Hem till Jarlabanke. Jord, makt och evigt liv i stra Mlardalen under jrnlder och medeltid.
Stockholm.
Jansson, S. B. F. 1987. The Runes of Sweden. Uppsala.
Janse, O., 1922. Le travail de lor en Sude lpoque mrovingienne. Etudes prcdes dun
mmoire sur les solidi romains et byzantins trouvs en Sude. Orlans.
Wessn, E., ed. 1952. Ynglingsaga. Stockholm.

258
 Odin eller Kristus p Vindgameir?
Rune Flaten, Department for Linguistic and Scandinavian studies,
University of Oslo, Norway

1. Sophus Bugge og Hvamls kristne opprinnelse

Sophus Bugge hevdet i 1889 at myten om Odin p treet, kjent fra Hvaml 138141, hadde
en kristen bakgrunn og var inspirert av korsfestelsen. Bugges tolkning mtte levende, for
ikke at sige harmfuld, Modsigelse (Ohrt 19291930:273), og har p bakgrunn av den sterke
kritikken nrmest forsvunnet helt fra forskningen. I leksikonet Medieval Scandinavia fra
1993 nevnes for eksempel ikke teorien med ett ord, ei heller Bugge. (Mitchell 1993:444445)
Bugge forutsatte en betydelig bibelsk og kirkelig lrdom hos dikteren. Om vi vender p
dette, og isteden sker finne hvordan diktstrofene, slik de fremstr i manuskriptet, ville blitt
mottatt av et publikum med noe bibelsk og kirkelig lrdom p 1200-tallet, vil vi kanskje
kunne bruke Bugges argumentasjon p en annen, og mer gyldig mte. Det er i denne
sammenhengen jeg vil ske finne ut hvordan datidens mennesker forsto versene. Kunne
man p 1200-tallet unng tenke p Kristus nr diktet forteller om Odin hengende p treet,
sret med spyd, og nektet mat og drikke? Mytens opprinnelse er i denne sammenhengen ikke
viktig for forstelsen av hvordan strofene om Odin p treet ble oppfattet i kristen middelalder.
Diktet finnes i Codex Regius (R) av Eddadiktene. Diktets frste strofe er ogs sitert i be-
gynnelsen p Snorres Edda, og den siste delen av strofe 84 finner vi Fstrbrra saga kap.
21. Diktet bestr av flere forskjellige deler, kanskje s mange som seks, og er antagelig satt
sammen av en redaktr p 1200-tallet (Evans 1993:272). Jeg forutsetter at versene 138141
utgjr en helhet, og nyer meg med g dypere inn i disse. I tillegg velger jeg forutsette at
strofene handler om Odin, uten g inn i andre mulige forklaringsmodeller.1
Manuskriptets publikum var antagelig geistlige, prestelrde, eller i det minste etter
forholdene godt utdannede mennesker, godt trent i tekstlige tolkningsteknikker. Bare
kostnadene ved produksjonen av et slikt manuskript forutsetter et hyere sosialt sjikt, og
lesekunsten var fremdeles forbeholdt de f.2 Det var i Norden p 1200-tallet stor interesse for
fortidens kultur og historie, noe som ga seg utslag i blant annet Snorres Edda og i den store
interessen for sagalitteratur som vi nyter godt av den dag i dag. R er antagelig samlet som et
ledd i denne interessen for den fortidige hjemlige kultur, og for kunne forst de kompliserte
poetiske omskrivningen i skaldediktningen, kenningene.

2. Typologi og euhemerisme.
En av mtene man i middelalderen forsto historie p, var gjennom typologi, med elementene
typus og antitypus, samt det sammenbindende begrepet praefiguratio. Gerd Weber har i en
artikkel fra 1987 brukt denne modellen for forklare middelalderens nordiske
historieskrivning. Jeg vil her ske bruke samme modell p de fire strofene fra Hvaml.
Typen prefigurerer antitypen, vanligvis slik at passasjer fra det Gamle Testamentet sies
prefigurere hendelser i det Nye Testamentet (Weber 1987:97). En eventuell bruk av denne
typologien som en historisk typologi ogs utenfor en bibelsk kontekst er kontroversielt, da
man kan hevde at prefigurasjon kun kan forsts som en sammenheng mellom det Gamle
Testamentet og frelsen, som gjengitt i det Nye Testamentet. Etter Weber vil dette synet ta fra
oss et viktig instrument for forst middelalderens historiografi og historiske tankegang.

1
For eksempel Grnvik 1999
2
down to the thirteenth century, written traditions were largely islands of higher culture in an environment that
was not so much illiterate as nonliterate. Stock 1983:7

259
Augustin, som ble sett p som en ugjendrivelig autoritet i perioden, hadde vist at det ikke
fantes noe slikt som profan historie. All historie foregikk etter Guds Store Plan. Prefigurasjon
var ikke begrenset til bibelsk historie, men gjaldt hele frelsningshistorien, fra Adam og Eva
ble drevet ut av Paradiset og til Dommedag. Frelsningshistorien, og dermed grunnlaget for en
forstelse av prefigurasjon, omfattet hele menneskehetens historie (Weber 1987:98).
Etter hvert som Europa ble kristnet, opplevde de forskjellige nasjonene Kristi komme
gjennom evangeliet. Verdens frelsningshistorie kunne dermed deles inn i flere mer eller
mindre parallelle nasjonale frelsningshistorier, hvor det nasjonale kristne vendepunkt foregikk
p et senere stadium av verdenshistorien sammenlignet med det historiske yeblikk hvor
Kristus ble inkarnert i denne verden. Som Weber fremhever, er det da bare naturlig at kristne
historieskrivere i middelalderen ville brukt typologi som et bindeledd mellom deres frkristne
nasjonale historie og tiden etter omvendelsen (Weber 1987:99100).
Denne generelle historiske typologien var mindre penbar enn den bibelske, i dette
tilfelle da de germanske folkene var lenger vekk fra Gud. Allikevel kunne Guds plan for
verdens frelse manifestere seg, dog i en mer obskur form, ogs i den frkristne historien til en
nasjon. For en kristen middelalderhistoriker, opplrt i typologisk tankegang, var det mulig
skjelne sammenhengen, ogs der den ikke var umiddelbart tydelig. Prinsippet om bibelsk
typologi brukt ogs utenfor den rent bibelske sfre var alts helt i trd med det generelle
verdenshistoriske perspektivet hevdet av Augustin, og ogs av senere autoriteter som
Gregorius av Tours og Otto av Freising (Weber 1987:100). Som eksempel p hvordan denne
typologiske tankegangen ble brukt ogs i Norden p 1200-tallet, viser Weber til Snorres
fremstilling av Kong Frode i Skldskaparml. Frodefreden kobles her til Pax Romana,
fredsperioden under Augustus, som ble forsttt som et resultat av Kristi fdsel. Snorre knytter
her den norrne historien til verdenshistorien, og kobler indirekte Frode til Guds strre plan
(Weber 1987:102103).
Ogs i kirkekunsten kan vi spore den samme tankegangen. Nes kirke i Telemark fra ca.
1150 har p portalene bde Davids kamp mot Goliat og Gunnar i Ormegrden, som gjennom
harpen kan kobles typologisk til David. Gunnar gr igjen p flere stavkirkeportaler, gjerne
sammen med Sigurd Fvnesbane (Weber 1987:104105). Kan hende har vi ogs den samme
tankegangen bak fremstillingene av Tor og Midgardsormen p steinene fra Hrdum i
Danmark, Altuna i Sverige og Gosforth i England (Janson 2005, Flaten 2007). Den norrne
helten eller euhemeriserte guden fr gjennom prefigurasjon et kristent tolkningsinnhold. lfr
hvtaskld gjorde det samme i den tredje grammatiske avhandling, hvor han knyttet den
norrne poesien til den klassiske, gjennom senes utvandring fra Asia (Fidjestl 1999:11).
Slik oppndde man knytte den norrne forhistorien til den kristne verdenshistorien.
Ved hjelp av euhemerisme kan alts ogs ikke-kristne myter vre sanne, i det de egentlig
handler om menneskers verk i frkristne tider, og da ndvendigvis m vre en del av Guds
strre plan. Middelalderens historieskrivning blir da et sprsml om tolkning, slik Snorre ogs
sier det i Skldskaparml 66 (at ra skldskapinn).3
Jeg vil bruke den gamle norske homilieboken (GNH)4 som et eksempel p hva slags tekster
som var tilgjengelige p 1200-tallet. Manuskriptet med denne samlingen er antagelig blitt til i
et monastisk milj, men har flere tekster felles ogs med den gamle islandske homilieboken. I
tillegg er flere av homiliene oversettelser av tekster fra kjente kristne tenkere som Gregor den
store, Alkuin og Wulfstan, og br kunne sees som et forholdsvis representativt eksempel p
vestnordiske, norrne religise tekster.

3
Weber 1987:108 Se hele Webers 1987 for flere eksempler.
4
Henvisningene er til side:linje

260
 3. Forskningshistorie
Sophus Bugge hevdet alts i 1880-rene at myten om Odin p treet var forst som inspirert
av beretningen om Kristus p korset (Bugge 1889). Med sin store kildekunnskap trakk han
frem mange likhetstrekk som virker tilforlatelige, men ogs en del som ved nrmere yesyn
kanskje ikke vil holde ml. Slik tidens skrivemte var, tok han en del sjanser, og fremmet
flere antagelser. Som et eksempel kan vi lese Vi tr ogsaa tro, at Nordboerne videre har
udviklet det rent germanske Grundlag uden Indflydelse udenfra, at Sagn om Guder og Helte
er spirede frem af udelukkende nordisk Rod. (Bugge 1889:3).
Bugges hovedtese var at mange av de nordiske gude- og heltesagn gjengir eller har oppsttt
under pvirkning fra de britiske yer, hvor hel- eller halvhedenske nordboere har vrt i
kontakt med kristne menn, munker og prester. Fra disse har nordboerne hrt religise eller
verdslige sagn, dikt, legender og overtroiske forestillinger som har blitt grunnlag for enkelte
av de norrne gude- og heltesagn vi kjenner i dag (Bugge 1889:89). Bugges fokus var i
hovedsak p mytenes opprinnelse, ikke p hvordan de var blitt forsttt i en gitt periode
(Bugge 1889:16).
I det store lente Bugge seg mye p utenomnorrne kilder. Dette er en alvorlig svakhet.
Flere av parallellene han brukte for underbygge sine ideer er ogs fjerne fra den norrne
verden rent tidsmessige. Tidlige kristne tekster og senmiddelalderske dikt er vanskelige
bruke som kilder til frkristen tro. Bugge var allikevel ikke i tvil om at myten har fungert i et
frkristent samfunn, uavhengig av mytens opprinnelse.
Bugges teori fikk bde tilhengere og motstandere, og en debatt raste til dels kraftig
gjennom hele starten p 1900-tallet.5 I mer moderne tid er det mange som ikke nevner teorien
om et kristent opphav i det hele tatt,6 mens andre kun nevner den kort.7 Gabriel Turville-Petre
gikk i 1964 ganske grundig gjennom myten, og tok ogs Bugges teori med i betraktningen.
Han s klart likheten, og skrev som flger: If the myth of the hanging inn did not derive
from the legend of the dying Christ, the two scenes resembled each other so closely that they
came to be confused in popular tradition. (Turville-Petre 1975:43). Som en slags konklusjon
kom han til at det ikke var umulig at myten om Odin var pvirket av fortellingen om Kristus
p korset (Turville-Petre 1975:50).
I hovedsak har imidlertid de fleste forskere etter 1930 sett p myten som rent frkristen,
eller trukket inn indoeuropeiske paralleller. Ferdinand Ohrts ord summerer ganske treffende
opp forskningstendensen: Til Slut, og vel til Overflod, skal ett fremhves. Kvadet ikke alene
vil vre hedensk men det er hedensk. (Ohrt 19291930: 285).

4. De enkelte strofer
Det er en jeg-person som taler i strofene, og en som tydeligvis fremdeles er i live, ogs etter
ha hengt, sret med spyd, uten mat og drikke i ni netter. Dette minner om GNH, hvor for
eksempel deler av De natiuitate domini sermo er gjengitt i veldig direkte tale: Ec var grafen.
ok ris ec upp af daua (GNH 34:25).

138. Veit ec, at ec hecc Jeg vet at jeg hang

vindgameii i det vindkalde tre
ntr allar no, ni hele netter,
geiri undar med odd sret,
oc gefinn ni, til Odin gitt,

5
Se Reichardt 1957 for en gjennomgang av denne debatten frem til 1957.
6
For eksempel de Vries 1934, Dumezil 1969, Strm 1999, Clunies Ross 1994, Schjdt 1995, Steinsland 2005
7
For eksempel Reichardt 1957, Davidson 1990:144, Holtsmark 1970:125. Kure 2006:45 sammenstiller Odin
og Kristus, uten g inn p Bugges teori.

261
 silfr silfom mr, sjl gitt til meg sjl,
eim meii, i det treet
er mangi veit, som ingen veit
hvers hann af rtom renn.8 av hvilke rtter det rant.9

Jeg vet at jeg hang p det vindblste tre. Dette forblste treet er tradisjonelt forsttt som
galgen, og Odin er vel belagt som for eksempel glga farmr galgens byrde,10 hangago og
hangatr de hengtes gud.11
Men ogs Kristus kan henge i galgen. Bugge viste til flere eksempler fra angelsaksisk,
oldsaksisk, oldhytysk og gotisk hvor korset omtales som en galge (Bugge 1889: 293294).
Ogs i den mer hjemlige litteraturen finner vi uttrykket, som i et senmiddelaldersk islandsk
dikt referert av Bugge, hvor vi kan lese at Kristus bar sjlfs sns glga bar selv sin galge. Av
Bugge ble det argumentert filologisk og meget teknisk for likheten p de to
retterinstrumentene (Bugge 1889:292296). I mine yne er den enkleste forklaring at korset
som torturinstrument var ukjent i germanske omrder. Som en forklarende oversettelse har
man da satt galge i stedet for kors, som ikke har gitt de germanske kristne noen
assosiasjoner utenfor kristen bruk. I s fall vil det si at galgen her i Norden var alminnelig
kjent og hadde et semantisk innhold som kunne brukes ogs i fortellingen om Kristi endelikt.
Uavhengig av kulturforskjeller var det n konkret gjenstand misjonrer hadde med seg
som symbol p sin tro: krusifikset. At frkristne nordboere med et forhold til en hyeste gud
som henger torturert p et mystisk tre for oppn visdom (og dermed makt) sammenblander
Odin og Kristus i denne sammenhengen er ikke uforstelig (Reichardt 1957:26). Om dette er
opprinnelsen til versene som behandles her, fr vre usagt i denne sammenheng. Men det er
da ikke underlig at ogs kristne mennesker som omtrent to hundre r senere hrer eller leser
Hvamls strofe 138141 ubevisst gjr akkurat den samme velsen, bare med motsatt
fortegn. I diktet om Odin vil de naturlig se Kristus.
Korset kalles flere steder et tre, som for eksempel i den gamle norske homiliebokens
Dominica palmarum sermo, hvor det heter tre pinslar (GNH 80:27), og i Jn inuentione sancte
crucis sermo, hvor vi ogs finner []bte hann at pnslartre er hinn fyrsti mar misgere
girnar-tre (GNH 103:1819). Her ser vi ikke bare at korset likestilles med et tre, men
kunnskapens tre fra Skapelsesberetningen prefigurerer korset; Adam som type og Kristus som
antitype. P bakgrunn av dette kan vi anta at 1200-tallets mennesker, og da i sr prestelrde,
har kjent prefigurasjon som en i det minste bibelsk historisk forklaringsmodell.
At Odin hang ni netter p treet, legger Bugge ikke srlig vekt p (Bugge 1889:304).
Imidlertid er det i evangeliene fremhevet at Jesus hang ni timer p korset. Nr man i
middelalderen hrer om en som henger ni netter, vil dette, satt i en sammenheng som denne,
vre med p forsterke inntrykket av at vi har med en typologisk Kristusfortelling gjre.
Odins ni netter, ntr allar no, kan ha funnet gjenklang i bde kristne og eldre, hjemlige
forestillinger. Nitallet gr igjen i mange norrne frkristne forestillinger, og har sannsynligvis
en annen, frkristen bakgrunn ogs i Hvaml. I alle evangeliene, bortsett fra
Johannesevangeliet, finner vi at Jesus roper ut i den niende timen p korset, og deretter dr.12
Noen norrn oversettelse av dette er s vidt meg bekjent ikke bevart, men da det har en spass
sentral plass i fortellingen rundt Jesus p korset, ville det ikke vre unaturlig om nitallet ogs
var med i en ikke bevart oversettelse13, eller vi kan regne med at prestelrde med

8
Alle sitater fra Hvaml er hentet fra Jn Helgasons utgave.
9
Oversettelsen ved den norrne teksten er fra Holm-Olsen 1993. Oversettelsene ellers i teksten er mine.
10
Eyvindr Finsson skldaspillir: Hleygjatal 1. Alle henvisninger til skaldedikt er til Skaldic Poetry of the
Scandinavian Middle Ages
11
Hvaml 14, Vga-Glmr Eyjlfsson 10
12
Matt. XXVII:4650, Mark. XV:3437, Luk. XXIII:4446
13
Det har for eksempel antageligvis eksistert en oversettelse ogs fr Stjrn. Se Seip 1980:520

262
latinkunnskap kjente dette fra Vulgata. Dette blir dog bare spekulasjoner, og vi kan ikke uten
videre g ut fra at 1200-tallets mennesker ville gjort den mentale koblingen mellom Odin og
Kristus p bakgrunn av nitallet i evangeliene.
Et merkelig sammenfall er det mellom Odin og Kristus i at begge fremstilles som sret
med spyd. For Odin er spydet i flere sammenhenger fremhevet som guddommens spesielle
vpen (Strm 1999:108). Blant annet kalles han geirs drttin (spydets herre), i Sonatorrek 22,
og noe mer usikkert Gungnis vfar (Gungners rister eller vind) i et fragment tilskrevet Bragi
inn gamli.14 Det ser ut til at det har vrt mulig ofre mennesker, da gjerne motstanderne i
kamp, til Odin ved hjelp av et spyd.15 For Kristus del er det Longinus som med sin lanse gir
ham et sr i siden etter at dden har inntruffet.16 Longinus ser ut til ha vrt viktigere i andre
deler av den vestlige kristenheten enn i det norrne omrdet. Ogs at Odin henger sret p
treet, geiri undar, kan vi finne igjen i homilieboken. eir logu spiote siu mer sva at blo
ran ior finner vi i De natiuitate domini sermo (GNH 34:2425).
I verset heter det ogs gefinn ni, silfr silfom mr gitt til Odin, selv til meg selv. Det
kan vel sies at Jesus ofret seg selv p korset til Gud, for menneskenes skyld. Med treenigheten
i bakhodet, kan man absolutt ogs si at Kristus ofret seg til seg selv.17 Men hvor utviklet var
denne forestillingen p 1200-tallet? At Odin er gitt til seg selv, krever en forstelse av
treenigheten for at koblingen mellom Odin og Kristus skal fungere. Dette kan vi vel regne
med at menn med noen grad av prestelrdom hadde. Vi finner som et eksempel et ekko av
denne forestillingen i Snorres Gylfaginning, hvor Hrr, Jafnhrr og rii kan sees som et
litterrt grep for vise en ufullkommen norrn variant av treenigheten.
At ingen vet hvor treets rtter finnes, kan igjen gi assosiasjoner til homilieboken, hvor det
forklares hvordan den delen av korset som ikke kunne sees, allikevel holdt det hellige korset
oppe, slik den usynlige guds makt styrer alt synlig (GNH 104:1823).
Sett under ett gir hele strofen store muligheter for forsts som en typologisk
Kristusfremstilling.

139. Vi hleifi mic sldo Brd fikk jeg ikke,

n vi hornigi, de brakte ikke horn;
nsta ec nir; speidet ned fra treet,
nam ec upp rnar, tok s opp runer,
pandi nam, tok dem med skrik,
fell ec aptr aan. og fra treet falt jeg.

Denne strofen er mer tvetydig. At Odin hverken fikk brd eller drikke, kan selvsagt ha blitt
sett i sammenheng med Kristus som fikk eddik p en svamp drikke,18 men det er ikke
vektlagt i homiliebokens fremstilling av korsfestelsen. Vi kan kanskje regne med at
prestelrde med kunnskap fra Vulgata kan ha koblet de to fremstillingene sammen, srlig p
bakgrunn av det foregende versets mer tydelige mulige referanser til Kristus.
Etter Bugge fikk Kristus galle spise og eddik drikke,19 men det blir i begge
sammenhenger en negativ opplevelse. Odin fr ingenting, Kristus fr uspiselig mat og
udrikkelig vske.

14
Bragi inn gamli, fragments 4
15
Se for eksempel Turville-Petre 1975:47 for en oversikt
16
Bugge vil her ogs vise til engelsk og irsk middelaldersk tradisjon, hvori Kristus fikk banesr av dette spydet.
Dette velger jeg ikke ta hensyn til her, da det ikke kan godtgjres noen sammenheng med norrne kilder.
Bugge 1889:3941
17
Se for eksempel Joh. XX:30
18
Matt. XXVII:48, Mark. XV:36, Luk. XXIII:36, Joh. XIX:29
19
Bugge 1889:346, jeg har ikke funnet kilden til forestillingen om at Kristus fikk galle spise.

263
 Nsta ec nir, jeg speidet ned, kan vanskelig passes inn i fortellingen om Kristus p korset.
Bugge ville se en sammenheng med Joh. XIX:30, hvor det str at den dende Jesus byde
hodet ned. I neste linje finner vi nam ec upp rnar, tok jeg opp runer. Runer m i denne
sammenhengen forsts ikke bare som skrifttegn, men som hemmelig kunnskap. Denne kan
godt ha hatt en sammenheng med runestavene, men det gr ikke klart frem av diktet hvordan
det er tenkt at runene blir tatt opp. Vi m vel allikevel forst det slik at han hentet ny
kunnskap nedenfra, fra dypet eller underjorden. Odin tilegner seg kunnskap, slik som i s
mange av de andre historiene vi kjenner om ham. Men kunnskapen er bare til ham, den
kommer ikke umiddelbart noen andre til gode. N kan det sies at det er ndvendig at Odin
oppnr ny viten for bevare verdensordenen, men det gr ikke frem av strofene. I de fire
versene som her behandles, er det Odin selv som er i fokus. Det virker ikke som om noe av
det han foretar seg er til gagn for andre enn ham selv. Bugge henviser til
Nikodemusevangeliet, hvori det gr frem at Jesus etter sin dd skal f herredmme over hele
verden. De underverk han gjr kalles mysteria per mortem crucis, og skal alts ha inspirert
historien om Odin som mottar mystisk hemmelig kunnskap gjennom en slags ddssituasjon
p det vindblste treet. Deler av dette evangeliet ble tidlig oversatt til norrnt, som
Nirstigningarsaga, og har antagelig vrt godt kjent for 1200-tallets norrne elite. Her
fortelles det hvordan Kristus farer ned til helvete etter sin dd, slipper fri alle rettferdige
sjeler, og siden inntar sin maktposisjon i himmelen. Men sammenhengen med Hvaml 139
er ikke umiddelbart penbar.
Nr Odin skrikende, pandi, tar opp runene har vi en tydeligere sammenheng. At Odin
tok i mot kunnskapen s hylydt kan selvsagt settes i forbindelse med at Jesus i flge
evangeliene ropte ut med hy rst fr sin dd. Det fortelles i evangeliene at Kristus ropte p
korset, en eller to ganger. Imidlertid er ordet som brukes i Hvaml nrt knyttet til pi, som
vi finner senere i R, i Skirnisml 29. Her er det i en ganske annen sammenheng, i en regle
eller besvergelse med sterke seksuelle undertoner, brukt av Skirne for tvinge Gerd til
godta Frys amorse tilnrmelser. pi oversettes som en trolldomsrune som forrsaker skrik
og yl (Heggstad 1958:508) eller vanvidsskrig (LP 446). Sammenhengen mellom skrik og
runer virker reell, srlig nr vi kan legge til at en av de mest produktive svenske runeristerne
signerte steinene som pir. At det er en sammenheng i samme manuskript m veie tyngre
enn et mer usikkert bibelsk samsvar. Imidlertid er det ingenting i veien for at en skriftkyndig
ogs her hadde sett en sammenheng.20
I siste del av verset sies det fell ec aptr aan falt jeg derifra igjen. Kristus falt p ingen
mte ned fra korset, men ble hentet livls ned. At Odin s faller ned fra treet igjen, ser jeg
liten penbar kristen symbolikk i.
Om vi ser p hele strofen, s er det her mulig tolke Odin som Kristus, men det er slett
ikke ndvendig, og det blir bare aktuelt om vi ser strofen i sammenheng med den foregende.

140. Fimbulli nio Ni storgaldrer

nam ec af inom frgia syni fikk jeg av den navngjetne snnen
Blors, Bestlo fur, til Bltorn, Bestlas far;
oc ec drycc of gat og en drikk fikk jeg
ins dra miaar, av dyrebar mjd,
ausinn reri. st av drre.

Sophus Bugge hoppet behendig over dette verset i sitt resonnement, og behandlet det heller
senere i boken. Det er fullt forstelig, da det ikke passer godt inn i noen kristen kontekst. I
stedet gikk Bugge rett p vers 141. Her m imidlertid ogs vers 140 tas med i betraktningen i
den rekkeflgen den str p i R.

20
Jeg forstr her skriftkyndig som en som er kyndig i Skriften, alts bibelen.

264
 Fimbulli er store eller mektige sanger, forbundet med trolldom. Nr vi blir fortalt at det
er ni av dem, s kan dette ikke lenger p noen mte sammenstilles med Kristi ni timer p
korset. At Odin fr ni kraftige trolldomssanger fra den vidgjetne snnen til Boltor, Bestlas far,
passer ikke inn i noen kjent forestilling om Kristus p korset. Her er det tilsynelatende Odins
onkel p morssiden som er opphavet til den hemmelige kunnskapen, mens Kristus
maktposisjon flger av hans rolle som Guds snn.
Bugges eneste kristne tolkning i verset var hvordan den dyrebare mjden, inn dri mjr,
kan sammenlignes med episoden nr disiplene mottok den hellige nd (Apostl. II:33), i et
latinsk dikt beskrevet som den guddommelige nektardrikk (Bugge 1889:353354). Han var
imidlertid uvanlig ydmyk p dette punkt, og ga ikke inntrykk av ha funnet den eneste rette
tolkning av strofen (Bugge 1889:354). Det er da heller ikke srlig sannsynlig at dette latinske
diktet skal ha pvirket forestillingen om den hellige nd i norrne omrder p 1200-tallet.
Nr strofen forteller at Odin fikk drikke av det dyrebare let, vektlegges det at det er st fra
kjelen rerir hvor en del av skaldemjdet befinner seg. Vi er her midt i en av de viktigste
Odinsmytene, om hvordan han ved svik fikk skaldemjdet fra Suttungs datter Gunnl.21
Denne myten var ikke bare viktig for forstelsen av skaldediktningens opprinnelse, men blir
ogs presentert bare noen vers tidligere i diktet slik det fremstr i R. Vi m regne med at
manuskriptets mlgruppe har kjent denne myten, og derfor har koblet strofen til
skaldediktningen. Koblingen mellom skaldskap og kunnskap var sterk, da historie og annen
kunnskap i stor grad ble formidlet gjennom dikt. nevne Suttungs mjd her ville virket
forsterkende p viktigheten av den kunnskapen Odin tilegner seg p treet.
Det er med andre ord ingenting som knytter denne strofen til Kristus. Her spilles det p
andre, kjente, Odinsmyter, av hvilke myten om skaldemjdet ogs er belagt ellers i diktet.

141. nam ec frvaz Da ble jeg frodig,

oc frr vera og fikk meg kunnskap,
oc vaxa oc vel hafaz; vokste og trivdes vel;
or mr af ori av ord kte ord
orz leitai, meg ord p ny,
verc mr af verki av verk skte verk
vercs leitai. meg verk p ny.

Denne siste av vre strofer virker mer nytral. Fruktbarhet og kunnskapsrikdom, vekst og
trivsel er religionsnytrale, positive begrep. At det ene ordet ledet til det andre, den ene
handlingen til den andre likes. Bugge vil ogs her se en likhet med Nikodemusevangeliet,
hvor Kristus str opp fra de dde som den guddommelige, altbeseirende Herre. (Bugge
1889:352). En 1200-talls leser eller lytter kan godt ha lest et kristent innhold inn i strofen,
men det var ikke opplagt at vedkommende gjorde det.

Konklusjon
Noe av kritikken mot Bugge gikk ut p at han forutsatte en veldig hy grad av kristen lrdom
hos dikteren bak strofene. Om vi alts i stedet forutsetter en ganske god kjennskap til kristen
lrdom hos strofenes publikum p 1200-tallet, blir sammenstillingen av Odin og Kristus mer
fruktbar. Imidlertid viser det seg da at Bugges argumentasjon i liten grad kan brukes, etter
som hans grunnlagsmateriale bde tids- og stedsmessig er vanskelig knytte til 1200-tallets
norrne omrder.
Da det frankiske imperium og andre europeiske nasjoner etter hvert ble kristnet, ble det
ndvendig integrere deres nasjonale historier i den kristne verdenshistorien. Typologi ser ut

21
Hvaml 104110, og i gjenfortalt med flere detaljer i Snorres Skldskaparml

265
til ha utviklet seg fra en rent bibelsk tolkningsprosess til en overordnet forstelsesmetode for
generell historie i lpet av denne prosessen (Weber 1987:131, note 19). Om vi gr gjennom
strofene med et typologisk 1200-talls utgangspunkt, basert p tidens tilgjengelige religise
skrifter og teorier, ser vi at strofene hver for seg tilsynelatende gir et forvirrende uttrykk.
Bare strofe 138 gir alene et klart kristusbilde. Strofen har s mange likhetstrekk med
kristusfremstillinger som var kjent i det aktuelle omrdet under den aktuelle perioden, at det
nok var vanskelig for en leser ikke se dette som en typologisk kristusfremstilling. Strofe 140
gir umiddelbart et lite kristent inntrykk. Vers 139 og 141 kan godt forsts som omhandlende
Kristus, sett i sammenheng med strofe 138. Ogs strofe 140 kan, med ideen om Odin som en
ufullkommen historisk type til antitypen Kristus passes inn i sammenhengen. Sett under ett,
gir de fire versene et betinget kristent uttrykk. En leser med noe kirkelig kunnskap p 1200-
tallet har alts ikke kunnet unng tolke Hvaml 138141 som en typologisk
kristusfremstilling. Og dette er den mest sannsynlige lesergruppen for manuskriptet hvor
diktet er bevart.
Vi har her gjre med et transformasjonsprodukt, hvor en frkristen myte fikk et nytt
meningsinnhold da samfunnet hvor myten fungerte fikk en ny religis ideologi. Ferdinand
Ohrt var i slutten p 1920-rene nr tanken, nr han p sprsmlet om hvorvidt rnar i strofe
139 gjelder skriftsystemets (futharkens) opphav eller bruken av visse Magtruner som Odin
fandt svarer: Digteren kan have ment det sidste, selvom Hvamls Redaktor har tnkt sig
det frste. (Ohrt 19291930:277). Dikter, redaktr og leser vil nesten uvegerlig ha hatt
forskjellige oppfattelser av innholdet i disse strofene, avhengig av tid, sted og vedkommendes
bakgrunn. Isolert fra resten av Hvaml ville disse fire strofene antagelig blitt forsttt som en
kristusfremstilling av de som arbeidet med manuskriptet som nytt, selv om myten om Odin p
treet hadde hatt et helt annet meningsinnhold i frkristen tid.

Litteratur
Kilder:
Eddadikte, oversatt av Holm-Olsen, Ludvig. Oslo 1993 (1975)
GNH = Gamal Norsk Homiliebok. Ved Gustav Indreb, Oslo 1966
Gylfaginning: http://www.heimskringla.no/original/snorre/gylfaginning.php
Hvaml, i (utg.) Jn Helgason: Nordisk Filologi, Eddadigte I. Vlusp Hvaml. Oslo 1971
Holy Bible. King James version. London 1872
Skaldic Poetry of the Scandinavian Middle Ages: http://skaldic.arts.usyd.edu.au
Vulgata: http://www.intratext.com/IXT/LAT0001/_INDEX.HTM

Oppslagsverk:
Heggstad, Leif: Gamalnorsk ordbok. Samlaget 1958
LP = Lexicon Poeticum, ved Finnur Jnsson. Kbenhavn 1966

Litteratur:
Andrn, Anders: I skuggan av Yggdrasil, i (red.) Andrn, Anders, Jennbert, Kristina og Raudvere,
Catharina: Ordning mot kaos. Vgar till Midgrd 4. Lund 2004. ss 389430
Bugge, Sophus: Studier over de nordiske Gude- og Heltesagns Oprindelse. Frste rekke, Christiania
18811889
Clunies Ross, Margaret: Prolonged Echoes. Old Norse Myths in Medieval Northern Society. The Vi-
king collection; vol. 710. Odense 1994
Davidson, H.R. Ellis: Gods and Myths of Northern Europe. London 1990 (1964)
de Vries, Jan: Odin am Baume, i Studia Germanica, tillgnade Ernst Albin Kock. Lund 1934
ss.393395
Dumezil, Georges: De nordiske Guder. Odense 1969 (1959)
Evans, D.A.H.: Hvaml, I (red.) Pulsiano, Phillip: Medieval Scandinavia, an Encyclopedia. New
York 1993, ss. 272

266
Fidjestl, Bjarne: The Dating of Eddic Poetry. A Historical Survey and Methodological Investigations.
Bibliotheca Arnamagnana 41 1999
Flaten, Rune 2007: rr dr Migarsorm: hedensk myte og kristen allegori. Masteravhandling, ILN,
UiO, Oslo
Grnvik, Ottar: Hvaml. Oslo 1999
Holtsmark, Anne: Norrn mytologi. Oslo 1970
Hultgrd, Anders: Old Scandinavian and Christian Eschatology, i (red.) Ahlbck, Tore: Old Norse
and Finnish Religions and Cultic Place-Names, Scripta Instituti Donneriani Aboensis 13,
bo/Stockholm 1990 ss. 344357
Janson, Henrik: Snorre, Tors fiskafnge och frgan om den religionshistoriska kontexten, i (red.)
Raudvere, Catharina, Andrn, Anders og Jennbert, Kristina: Hedendomen i historiens spegel. Bil-
der av det frkristna Norden. Vgar til Midgrd 6. Lund 2005. ss. 3355
Jolly, Karen Louise: Popular Religion in Late Saxon England. Chapel Hill 1996
Kure, Henning: Hanging on the World Tree. Upublisert artikkel 2006
Mitchell, Stephen A.: inn, i (red.) Pulsiano, Phillip: Medieval Scandinavia, an Encyclopedia.
New York 1993, ss. 444445
Ohrt, Ferdinand: Odin paa tret, i (red.) Brndum-Nielsen, Johs.: Acta Philologica Scandinavia,
fjerde bind. Kbenhavn 19291930 ss. 273286
Reichardt, Konstantin: Odin am Galgen, i (red.) Faber du Faur, Curt et al: Wchter und Hter, Fest-
schrift fr Hermann J. Veigand. New Haven 1957 ss. 1528
Schjdt, Jens Petter: Odin p tret, i Religionsvidenskabeligt tidsskrift nr. 27, 1995 ss. 520
Seip, D.A. Bibelversttning, Norge, i Kulturhistorisk leksikon for nordisk middelalder, bind 1,
Rosenkilde og Bagger 1980 ss. 520
Steinsland, Gro: Pagan Myth in Confrontation with Christianity, i (red.) Ahlbck, Tore: Old Norse
and Finnish Religions and Cultic Place-Names, Scripta Instituti Donneriani Aboensis 13,
bo/Stockholm 1990 ss. 316328
Steinsland, Gro: Norrn religion. Oslo 2005
Stock, Brian: The Implications of Literacy. Written Language and Models of Interpretations in the
Eleventh and Twelfth Centuries. Princeton 1983
Strm, Folke: Nordisk hedendom. Gteborg 1999 (1961)
Weber, Gerd Wolfgang: Intellegere historiam. Typological perspectives of Nordic prehistory, i
(red.) Hastrup, Kirsten og Meulengracht Srensen, Preben: Tradition og historieskrivning. Kilderne
til Nordens ldste historie. Aarhus 1987 ss. 95141

267
 The Story of Hrlf Saga Kraka as the manuscripts tell it
Charlotte Frantzdatter, Department of Scandinavian Research, The Arnamagnan Institute,
Copenhagen University, Denmark
I will by looking at the manuscripts wherein Hrlf saga Kraka appears, try and tell the story of
the saga within the North. Instead of focusing on the text, I will look at the material: the
parchment, the paper, the ink, the scribe, the collation of the books and more, and try to an-
swer:
When was it popular?
Why was it popular?
With which other sagas was it associated?
How did it migrate around the North?
All of the answers to these questions and other questions will, I hope, add up to a picture of
the life of the saga.

268
A New Etymology for Hamlet? The Names Amlethus, Amli
and Admlithi
Lisa Fraser, University of Aberdeen, Scotland
The name of the Gesta Danorum character Amlethus (predecessor of Hamlet) is often held to
be related to the Icelandic Amli, which is found in a single skaldic verse. Two of the most
influential explanations for the apparent connexion between Amlethus and Amli hinge on
the Irish name Amlaidhe, which has been viewed as a complex corruption of Scandinavian
*Anlaifr or *Anl(e) e. However, in more recent research Amlaidhe is interpreted as a
straightforward Irish rendering of Old Norse Haflii; and I wish to propose that a different
Irish name may provide simpler etymologies for Amli, Amlethus, and ultimately Hamlet.
My suggestion is that Amlethus and Amli both derive from Irish Admlithi (Great Grind-
ing), and are therefore equivalent to the mill name Grotti which appears with Amli in
skaldic verse.

269
 Snorri Sturluson and oral traditions
Frog, Dept. Scandinavian Studies, University College London, England; Dept. of Folklore,
University of Helsinki, Finland
Snorri Sturluson was a poet of exceptional skill. This is generally overlooked when consider-
ing Snorris knowledge of poetry and its content. Edda will be taken to reflect his knowledge
of and competence in the poetic system, and specific poetry and poems known or otherwise
available in his immediate environment: it is approached as a concrete example of applied
knowledge. Snorris knowledge of oral traditions and their cultural activity is viewed through
patterns of quotation as well as Snorris own composition in order to develop an overview.
The poetic system is addressed as a coherent and dynamic whole which is best approached as
systems of registers, meters and prosodies with conventional constellations according to genre
and application (Frog 2009:280288). However, the conventional binary division of the po-
etic system into eddic and skaldic categories retains a practical (though not analytical)
value for discussion.

1. The Corpus
The Codex Regius (R) manuscript of Snorra Edda will be treated as the model text for discus-
sion, excluding Grottasngr and the ulur at the end of Edda. Extended quotations from
ecphrasis poems (49 stanzas) are distinguished in statistics.

Table 1. Extended quotations from ecphrasis poems.

Subject poem quoted total stanzas stanzas1
rr and Geirrr rsdrpa 17 7391
The Rape of Iunn Haustlng 13 92104
rr and Hrungnir Haustlng 7 6571
Srli and Hamir Ragnarsdrpa 5 154158
Hjaningavg Ragnarsdrpa 5 250254

All statistics are considered approximations presented in the concrete terms of a defined cor-
pus. Counting stanzas is inherently problematic. Stanza is used inclusive of stanzas not
quoted in full. Eddic stanza divisions and numbering follow Neckel & Kuhn 1962; quotation
follows Faulkes 1988 and Faulkes 1998, with the addition of the long line of verse attributed
to Hrlfr Kraki (Faulkes 1998:59.15). Skaldic stanzas are counted and numbered according to
Faulkes 1998, adding the repeat quotation of 111.1 in prose (Faulkes 1998:36.8) and counting
the Bragi-trllkona exchange (300ab) as two stanzas. This gives Bjarne Fidjestls
(1985:323) total of 373 skaldic stanza quotations. 19 quotations repeat a helming,2 2 quota-
tions of different helmings from the same stanza can be identified (though there are probably
more),3 totalling 21 repetitions, 3 from Haustlng (93.14, 94.14, 65.58).

1
All stanzas in Skpm are numbered according to Faulkes 1998.
2
In Gylf.: Vsp 52; Faulkes 1988:9, 5152. In Skpm: 3=226; 5.58=278; 12=308; 33.58=40; 65.58=108; 93.1
4=341; 94.14=305; 111.1=p.36.8; 118=291; 126=347; 146=232; 186.45=389; 196=287; 210=316; 217=314;
279.12=394; 282=344; 386.12=411. The prose introducing 357 (Faulkes 1998:95.9) indicates a repeat quota-
tion (cf. Faulkes 1998:36.8; 92.26; 93.13) but the lines are not found elsewhere. As there may be as many as 9
additional stanzas from the same poem in Edda, this may be an honest mistake.
3
Skpm 3(=226)//337; 117//143.

270
Table 2. Stanza quotations in Edda: / indicates totals include/exclude quotations in Table 1;
indicates totals include exclude repeat quotations.
Work total stanzas skaldic stanzas eddic stanzas
Edda 467/418 446/400 373/324 353/307 94 93
Skldskaparml 390/341 370/324 371/322 351/305 19 19
Gylfaginning 77 76 22 75 74

2. Edda and the orality of poetry

As Bengt R. Jonsson has stressed, the natural state of oral poetry is to remain oral and[]
such poetry is rarely written down without a real incitement (Jonsson 1991:145, his empha-
sis). Snorri exhibits no interest in documenting whole poems except Httatal. Codicological
evidence seems to imply that the documentation of complete oral poems was a development
which followed a gradual rise in poetic quotation and pedagogy within the manuscript tradi-
tion, a process which only seems to have come into full bloom with Snorris works (cf. Quinn
1997; Harris 1985; Nordal 2009). The documentation of complete oral poems (distinct from
quoting them) requires them to have value and relevance as metrical texts in a written mode
of expression to fulfil a function related specifically to reading almost certainly different
from their significance and functions as oral phenomena.
Poetry was quoted on the basis of relevance. There is very little overlap in verses quoted in
different genres of medieval Icelandic literature (Fidjestl 1985:322333). Snorris work is
the first vernacular ars poetica and unique for its extensive presentation of mythological ma-
terial. Bjarne Fidjestl (1985:323) found that Snorri is our only source for 317 of Eddas 373
skaldic stanzas, and that his work had a tremendous impact on uses of mythology in oral po-
etry and manuscript contexts (cf. Fidjestl 1999:293). The vast majority of poetry quoted in
Snorris exegesis of poetics appears to derive from oral traditions which the unique work gave
cause to write down.

3. Skaldic quotations
Without repetitions, 353/307 skaldic stanzas are quoted in Edda. Attributions are made to 67
poets. Only 20 stanzas remain anonymous, including Eirksml (20), the trllkonas verse
(300b), and three stanzas (200, 236, 375) attributed to poets in other manuscripts or sources.
Authorship was clearly essential to the identity of verses. Snorri seems to have anticipated
that his audience would recognize poets by name, and perhaps could recognize a poet through
his verse: cf. 13 attributions to Einarr when Einarr Sklaglamm (15 stanzas) and Einarr
Sklason (35 stanzas) are both often quoted. Poem titles only appear introducing extended
quotations (Table 1) with one exception (137); the patron/subject of the poem may be indi-
cated when his identity is relevant to a circumlocution under discussion (282, 293, 296, 297):
the poet is named in all these cases. Snorri appears to have had a significant arsenal of poetry
at his disposal, quoting from more than 85 different poems,4 yet the poem title or other con-
text of the quotation is only stated for 8 quotations and those in Table 1.
Skaldic stanzas are primarily quoted for examples of nominal circumlocutions. This pro-
vides the basis for their selection and organization. Half to two-thirds of the stanzas quoted
can be described as recalled according to the content:5 e.g. poems, stanzas and lines on sea
faring for ship- and the sea-circumlocutions. The majority of stanzas treating mythological
subjects preserved only in Snorra Edda are quoted in this capacity. Within these sections,

4
This minimal estimate is inclined to reduce quotations to a minimum of possible poems; 16 stanzas are treated
as lausavsur; 50 are removed from consideration as problematicly ambiguous.
5
A practical and effective system for efficiently, distinguishing, classifying and describing these patterns of
quotation is still being developed, hence they will only be addressed here in general terms.

271
Snorri moves between content-based quotation and elements of the surface texture such as
heiti (cf. 5) and kennings (cf. 13) within kennings. The series of 22 quotations (16 poets) for
two specific indices of lord-heiti associated with dynasties (390411) are indicative of an abil-
ity to recall stanzas on the basis of specific elements in the surface texture of a poem. The
relationship between surface-texture and referent combine, for example, in the series of man-
(195200) and woman-kennings (201) with gold as a modifier (6 poets, one unattributed;
most include the lexeme gull-) followed by 17 man-/woman-kennings with a tree-heiti as a
base-word (202218: 10 poets; 3 unattributed). The latter appear to actively avoid repeating
heiti and are therefore unlikely to represent more than a portion of examples Snorri was pre-
pared to produce. It implies that his memory was trained in recalling kenning-referent or heiti-
referent combinations. Quotations repeatedly seem selected on a much narrower basis than the
circumlocution addressed: e.g. 9 of 12 examples of sky-kennings appear in variations of ruler
of/under the sky (105116: 8 poets). These are clearly recalled on the basis of the circumlo-
cution in which the sky-kenning occurs. The examples are summoned on such specific terms
that Snorri must have quite literally known a tremendous amount of poetry inside and out.

4. Medieval Icelandic poetry: A coherent system

Quotations in Skldskaparml (Skpm hereafter) focus on providing examples of poetic cir-
cumlocutions. This presentation occasionally blurs into presenting or authenticating mytho-
logical information6 or interpreting specific circumlocutions (e.g. 153). This is symptomatic
of the intimate relationship between the system of poetic circumlocutions and the pre-
Christian cultural milieu in which their referents and patterns of association developed
(Clunies Ross 2005:114115, 134138). Snorri reinforces and promotes this relationship
though Edda, and promotes the use of mythological and heroic material associated (by Snorri)
with eddic verse in skaldic compositions. This is emphasized in the introductions to quota-
tions in Table 1 by the formula eptir essi/eiri sgu orti/hefir ort, rather than sv kva etc.
This would seem to be a natural development from the apparent in the rise of interest in ma-
nipulating eddic poetry and its subjects by skaldic poets during the 12th century (Fidjestl
1980 passim.; Faulkes 1998:223), although Snorri and his nephews appear largely responsible
for the significant revival of interest in mythology and mythological reference in skaldic verse
(Fidjestl 1999:293).
Only 19 eddic stanzas are quoted in Skpm: ula-like stanzas, situational verse, more rarely
authenticating verse for mythological information, but only a gnomic verse (240) is quoted for
a single heiti. With the exception of 240, the pattern of eddic quotation is consistent through-
out Edda although skaldic poets composed, knew and referred to poetry which we class as
eddic (Clunies Ross 2005:628). Neither Snorri nor other medieval treatments of metrics
differentiate between eddic and skaldic meters (Nordal 2009:33n). lfr rarsons ars po-
etica quotes eddic examples alongside skaldic without distinction (Sigursson 2004:100
114), and in addition to two skaldic stanzas, the First Grammarian quotes two probable long
lines of eddic poetry to demonstrate a minimal pair.7 Snorri is inclined to divide the corpus as
part of his compositional strategy, but his division does not entirely coincide with a modern
eddic-skaldic dichotomy. He treats Eirksml (20) as an eddic composition: it is referred to
6
E.g. 11 of the first 15 stanzas for rr (4256: 7 poets) are associated through the battle with Migarsormr;
one stanza (50) lacks a circumlocution for rr, implying emphasis momentarily shifted to content. Con-
versely, the impoverished section on Freyr (6063) presents three stanzas authenticating mythological informa-
tion (one of which is eddic) and only one kenning-example, which may reflect difficulty (or bias) in producing
examples of Freyr-circumlocutions.
7
Sigursson 1966:42: hdu er Hlgatrll d (fornyrislag?); en heyri til hoddu er orr bar hverinn
(ljahttr?). It is reasonable to assume that this comparison of two alliterating expressions of mythological
content is not random. On the The First Grammatical Treatise and vernacular poetics see Nordal 2009.

272
by title without author and lacks a circumlocution. The Bragi-trllkona exchange (300ab) is
in an anecdote and included for short ulur. The narrative context and mythic being identify it
with eddic situational verse as opposed to skaldic quotation which did not distinguish
lausavsur. Snorri recognizes and promotes the poetry as a coherent and interrelated system,
but his pattern of quotation appears to have been influenced by his general inclination to sys-
tematize and categorize (cf. Clunies Ross 1987).

5. Variation in eddic and skaldic verse

The eddic-skaldic dichotomy is bound up with romantic ideas of skaldic poetry as the poetry
of the aristocracy in opposition to eddic poetry as the poetry of das Volk. Oral-Formulaic
Theory revised this dichotomy into improvisational and memorized traditions. Oral-
Formulaic Theory as outlined by Albert Lord (1960) was developed form South-Slavic epic
and is not appropriate for the description of eddic poetry (cf. Mellor 1999; Haymes 2004;
Thorvaldsen 2006) as is the case for other oral poetic traditions more similar to eddic poetry
such as byliny (cf. Vesterholt 1973) or kalevalaic epic (cf. Harvilahti 1992). Skaldic verse was
subject to a highly crystallized process of transmission, particularly on the level of coherent
syntactic expressions (14 lines) and the 4-line helming as a sequential series (cf. Rubin
1995). Metrical constraints can be attributed a significant role in the stability of the line and
couplet in drttkvtt. By association, stability is projected onto skaldic fornyrislag and
skaldic ljahttr where it is attributable to convention and raises the question of equivalent
stability in eddic poetry.
Unless all Icelandic examples of eddic poetry are traced back to a single manuscript exem-
plar for each poem, eddic poetry exhibits stability on the level of the line comparable to
skaldic verse. This includes the Lokasenna stanza quoted by Snorri, which verbally corre-
sponds almost verbatim to Ls 21.12, 47.3 (legskau:lezkau), followed by Ls 29.46 as a
second helming. The strong the verbal, acoustic and functional similarities of Ls 21.12 to Ls
47.12 and Snorris use of legskau rather than letskau in Ls 47.3 (the only lexical variation)
make a manuscript dependence unlikely. Eddic stability on the level of the line seems rea-
sonably attributable to that tradition. Comparisons across sources reveal that lexical variation
within the eddic line only rarely exceeds that found in skaldic verse. Consequently, ver-
bally equivalent or near-equivalent lines cannot be assumed to indicate dependence on a
manuscript exemplar (cf. kv 14.16 and Bdr 1.16). Comparisons of skaldic verse across
sources reveals inconsistencies in attribution and apparent variation in the organization of
stable syntactic units (particularly helmings) into stanzas and larger compositional structures.
We are therefore most likely to encounter evidence of oral variation or dependence on a
manuscript exemplar in larger compositional structures. The Lokasenna quotation is unusual
for its mix-matched helmings, but this is also found in other eddic quotations where its oc-
currence may be statistically related to the number of stanzas quoted from a poem (Table 3)
and the nature that poems of stanzas.8
Snorri documents 5 stanzas of the poem Alsvinnsml across three quotations (328330,
332, 380). Two (332, 380) correspond to Alv 20 and 30. Both stanzas exhibit variation, which
might be considered extreme in 380/Alv 30 (l.2 me goom/ helju; l.3 kllu er grma me
goum/kalla grmo ginregin) although it remains within the repeating patterns of the poem
(e.g. Alv 30.3=Alv 20.3; kalla helju in Alv 20.6). Stanzas 328330 are not found in Alv nor
have any place in that poem. In AM 748 I (A), Alsvinnsml is changed to Klfsvsa for
328330 and two lines are added to 330, implying familiarity with a relevant poem; Als-
vinnsml is changed to Alvssml for stanzas 332 and 380. Lines 2 and 3 of 380 are
8
Cf. the organization of helmings in his quotation of Vsp 26; Vsp 46.58+47.14 (apparently) as a single stanza;
the order of long lines in Vsp 5.510; see also variation in ula quotations in Gylf.

273
changed to agree with Am 30.23 (cf. l.4 sorg/ljs). This is only one of many revisions un-
dertaken by this enthusiastic scribe and may be evidence that a copy of Alv was in the AM
748 I collection of eddic poems. Codex Upsaliensis (U) gives lvssml for Alsvinnsml
without revising the text (380; 332 is omitted) and leaves 328330 unattributed. The attribu-
tion of an additional situational ljahttr stanza to rr indicates revision by someone with
knowledge of eddic poetry, while the curious suspensions in Vsp quotations are clearly mod-
elled on manuscript traditions and seem applied by analogy in Gm quotations, indicating good
familiarity with manuscript copies of eddic poems.9 The lvssml attribution may be asso-
ciated with manuscript influences applied from memory. Snorris Alsvinnsml appears to have
been a different poem from the Codex Regius Alv. Manuscript redactor(s?) attempted to cor-
rect this according to their knowledge of Alv. This example is important because Snorri may
not have associated other stanzas with the same poems we do: a range of variation on the level
of compositional units or segments may have been a natural part of this oral tradition, result-
ing in Hvaml, Hyndlulj and Ffnisml. The same phenomenon is common in kalevalaic
poetry and should not be dismissed as corruption, particularly when we do not know how
extant variants may or may not reflect the broader cultural activity of a poem.

Table 3. Eddic stanza quotations. S indicates quotations in Skpm; all others found in Gylf. * indi-
cates the poem in which a specific stanza is preserved may not be the poem from which Snorri
quoted it. # indicates situational verses attributed to speakers in the narrative frame of Gylf.
Poem evidence total ula authent. situational attributed unattributed speechact
Vlusp E 31/30 7 24/23 28/27 3
Grmnisml E *24 6 *17+*S1 11 *12+*S1
Vafrnisml E 9 8 #1 1 7 #1
Vsp in skamma / 1 1 1
Lokasenna E 1 1 1
Heimdalargaldr U 1 1 1
Njrr/Skai A 2 2 2
Gns ride U 2 2 2
Skrnisml E 1 1 1
kks verse U 1 1 1
Ffnisml E *3 *1 S2 *1 S2
rrs verse U10 1 S1 S1
Hrlfrs verse E 1 S1 S1
Topographical (142) U 1 S1 S1
Alsvinnsml / 5 S3 S2 S5
orgrmsula U 4 S3 S1 S4
Bjarkaml in fornu A 3 S3 S3
*Hvaml E or / 1 #1 #1
*Wisdom competition U 1 #1 #1
*Gnomic verse (240) U 1 S1 S1

6. Eddic quotations
Snorri quotes 94/93 stanzas of eddic verse from ca. 20 poems (as known to him). Table 3
shows that complete variants are preserved of 8 (E), 2 are attested in other sources (A), corre-
sponding stanzas of 2 are found in extant poems other than Snorri knew them (/), and 8 are
otherwise unknown. The total quotations are divided according to application (authenticating

9
Lasse Mrtensson and Heimir Plsson (2008) examine the use of suspensions in stanzas in U quoted from Vsp,
Gm and Sexstefja. They argue that the suspensions were copied directly from manuscript versions of Vsp and
Sexstefja into the original Edda. I would like to thank Heimir Plsson for supplying me with a copy of this article
before it had become publicly available.
10
A second stanza is attributed to rr later in the same narrative in U.

274
or situational) and attribution (attributed, unattributed or speech-act/situational verse). ula
indicates lists of heiti longer than one stanza and may be considered a subclass of authenticat-
ing verse.
Snorri opens Gylf with two skaldic stanzas and quotes eddic verse in the dialogue between
Gylfi and Trinity-inn. This is clearly part of his organizational strategy. 63 of the 74 eddic
stanzas quoted in Gylf derive from three Odinnic wisdom poems which provide overviews of
cosmogony, cosmography and eschatology. These poems are quoted exclusively as authenti-
cating verse. The pattern of quotation appears directly related to Trinity-inn as the speaker
and the function of Gylf. Only four additional poems are quoted as situational verse. Vlusp
in skamma is preserved as Hdl 33, where a vlva-poem has been assimilated into the narrative
framework. It may have been an Odinnic wisdom poem or the vlva may derive authority
through Vlusp. The Lokasenna stanza is attributed to inn. Heimdalargaldr is a statement
by Heimdalr about himself. Snorri emphasizes these speakers with the demonstrative pronoun
sjlfr. In Edda, the only other use of this pronoun in stanza attribution is *Gm 44 spoken by
the sir themselves. This use of sjlfr is found in all manuscripts, with slight variation in
U.11 *Gm 44 is spoken by inn in Gm, and Snorri appears to have employed sjlfr exclu-
sively to qualify an exceptional attribution for authenticating verse, hence he appears to have
known this stanza from a poem other than Grmnisml. Thus 62 stanzas quoted derive from
three Odinnic wisdom poems, one from the associated Vlusp in skamma, and three which
qualify an exceptional attribution with a verbal cue. Fm 13 is the single exception. Like stan-
zas from Grmnisml and Vafrnisml (rarely Vlusp), it is unattributed, and it is the only
stanza quoted in Gylf not found in a mythological poem. Fm 13 is an atomic answer associ-
ated with the question in Fm 12.46: the formulaic helming Fm 12.13 contextualizes it in the
poem and could be easily exchanged. Vafrnisml only receives attribution in an excep-
tional context. If Snorri knew Fm 12.413.6 from Vafrnisml, it would explain the
stanzas appearance in a context where quotation is otherwise highly systematic.
Snorris deployment of situational verse is both sparing and strategic. Three situational
verses are attributed variously to Gylfi, Hr and rii in the beginning of Gylf. This appears to
be associated with his narrative strategy and art of intertextual relations. *Hv 1 is attributed
to Gylfi as the first direct speech. Potential verses from an unknown wisdom competition are
the first direct speech attributed to Trinity-inn. The first quotation from Vafrnisml is
Vm 30.4631: Vm 30.46, spoken by inn, is attributed to rii introducing his quotation
of the giants response. The giants response is the only attribution to Vafrnisml in the
text. Strategies are more evident in mythological narratives: attributions are restricted to a
single figure or two figures in a single exchange; no figure is attributed with more than one
stanza; attribution always occurs at a narrative climax; situational verse and authenticating
verse are never attached to the same narrative irrespective of its length i.e. incorporating
kks verse was a choice which according to Snorris narrative strategy excluded the quota-
tion of any additional poetry in relation to the Baldr-Cycle. The same pattern of situational
verse appears in Skpm (cf. also 300ab), although narratives with situational verse are accom-
panied by related skaldic verses (apparently as exemplars of application or to be explicated by
the prose rather than authenticating it per se).
Snorri subjected his quotation of eddic poetry to self-imposed restrictions which signifi-
cantly limited the number of verses quoted. He avoided their use as examples of poetic lan-
guage; limited the poems appropriate for authenticating verse in Gylf and authenticating verse
11
Following R: Sv sem hr er sagt at inn mlir sjlfr vi ann s er Loki heitir (Faulkes 1988:26); ok enn
segir hann sjlfr Heimdalargaldri (Faulkes 1988:26; compare U: ok enn segir sjlfum Heimdallargaldri); sv
er hr sagt orum sjlfra sanna (Faulkes 1988:34; this and the rest of the prose passage between the Vm 41
and *Gm 44 quotations has been replaced by ok enn segir in U). It is interesting but circumstantial that omissions
in U resolve conflicts between Edda and the Codex Regius poems.

275
(for which eddic material was appropriate) was not a priority in Skpm; avoided quoting more
than 12 stanzas from any narrative poem; and his abstinence from combinations of situ-
ational and authenticating verse may reflect an avoidance of over-quotation. Snorri clearly felt
no need to affirm his mythological narratives with eddic verse: several of his longest and
most prominent narratives are unaccompanied by verse although it is probable that he knew
some.12 His self-imposed restrictions imply that, like his skaldic quotations, Snorri had the
option of selectivity. It is fairly certain that he knew far more eddic verse than is quoted, par-
ticularly considering the intimate relationship between his knowledge of eddic and skaldic
poetry. This should not, however, be confused with his applications of that knowledge.

7. Eddic poetry and mythological narrative

A poem is not identical to a myth. Mythological narratives are best approached as extra-
textual entities which are adapted, applied and communicated in range of contexts, applica-
tions and modes of expression factors which can effect form, organization, emphasis and
interpretation. Snorris emphases and priorities are not necessarily identical to poems and
other cultural activity associated with the extra-textual entities which he adapted and applied
in Edda (and e.g. Ynglinga saga). For example, Snorris knowledge of Skrnisml has been
criticized because he quotes Skm 42 but scarcely mentions Skrnirs adventure, focusing on
Freyr and the loss of his sword (e.g. Gunnell 1995:222). However, Snorri has applied his
knowledge of the (quoted) poem and the extra-textual entity according to the narrative priori-
ties of Gylf: social relations and Freyrs weapon at Ragnark are relevant while Skrnir and
Gerr are secondary. Similarly, the Njrr-Skai stanzas are quoted in Gylf while the prose on
their marriage appears independently in Skpm. Snorri may have known a tremendous amount
of eddic poetry, but his applications are not necessarily intended to represent the poems them-
selves. His prose is generated in relation to his knowledge and understanding of specific po-
ems and the extra-textual entities (developed through a range of genres and modes of expres-
sion) which they reflect. He also adapted and manipulated these as culturally loaded referents
in relation to other models, referents and the strategies and priorities of Edda as a work.

8. Snorri and oral traditions

Snorris citations exhibit tremendous facility with existing skaldic poetry. His description of
the kenning system is both intuitive and guided by an academic agenda (Clunies Ross
1987), but he has clearly internalized this formulaic system, its conventions and strategies of
application (Fidjestl 1997). Some observations, such as differences in conventions for con-
sonant and vowel alliteration (Faulkes 1999:4), could come from active instruction or the
conventional discourse surrounding poetics, while his presentation of the metrical extremes of
the drttkvtt line appear derive from his own sensitivity to relationships between meter and
language (Kjartansson 2009). Httatal shows that he had internalized idiomatic conventions
of word-placement which only emerge through statistical analysis (Frog 2009:288289), indi-
cating a high (or even unconscious) sensitivity to the relationship between lexical item and
metrical position. He utilizes traditional compositional strategies in drttkvtt such as the
multiform, a constellation of associated verbal systems and formul on the level of tex-
ture.13 This degree of internalization is indicative of an extensive knowledge of poetry far in
12
The banishing of Migarsormr, Hel and the binding of Fenrir (cf. Ls 38, 39, 41); rrs visit to tgarar (cf.
Ls 60, 62); rrs fishing expedition (cf. Sigursson 2004:1317).
13
Httatal 39.4, mlmskrar dyn hjlmar, is one of 10 examples of a full-line drttkvtt multiform which
involves the even-line dyn X formula, the lmr-mlmr-hjlmr aalhending frame (cf. the lmr-mlmr-hjlmr-
hilmir skothending frame) and a conventional lexeme which completes the 2nd, 2nd3rd or 3rd4th metrical posi-
tions (skr- fills this function in 4 of the 10 examples; vi- and ing- each in two). This variation on the metri-

276
excess of the of stanzas quoted and raises questions about genres and contexts through which
Snorri and other poets must have exercised their art in order to develop their competence (cf.
Frog 2009): it seems probable that during his life, Snorri was responsible for slightly more
than the nine stanzas attributed to him (averaging 1 every 7 years) in addition to Httatal. His
internalized understanding implies that he grew up in an environment where the poetry was a
living tradition in which he participated observing that the core of Snorris knowledge of
poetry and poetics was almost certainly established in his youth, before the turn of the 13th
century, and Edda may reflect back on ideal conceptions of that time.
Edda is clearly concerned with the relationship between the poetry of hfuskld and
anonymous wisdom and narrative poetry, attesting to an intimate relationship between them.
The prose presentation emphasizes the content of this poetry and the extra-textual entities of
its referents. This content is treated as essential for understanding the system of poetic circum-
locutions, both for the interpretation of earlier poets and the generation of new compositions.
It is also promoted as material suitable for treatment in skaldic composition (in addition to its
independent entertainment value). This appears to reflect Snorris knowledge and understand-
ing of the poetic system and its circumlocutions, with the implication that eddic poetry played
a significant role in the development of his own internalized understanding. There is a certain
incongruity between the practical quotation of eddic ulur and ula-like stanzas as examples
of poetic language and the general restriction of eddic quotation to narrative or as examples of
literal information in a stanza for use as a heiti or in a kenning. Snorri appears to be con-
sciously de-emphasizing the independent value of eddic poetry in the promotion of skaldic
art. This may be related to his emphasis on tradition and traditional models found explicitly in
the works of hfuskld (Faulkes 1998:5, 85), leading to an assertion of hierarchical relation-
ships among genres. Consequently, Edda does not present displays of knowledge of eddic
poetry, but almost the reverse. Knowledge of and familiarity with eddic poetry and eddic sub-
jects appears essential to Snorris competence in the poetic system, and his knowledge of ed-
dic verse probably far exceeds that quoted. Conversely, Snorris prose is generated according
to the various priorities and intentions of Edda, which inclines toward practical (and entertain-
ing) referents for the poetic system (cf. Fidjestl 1997:4850) rather than the specific contents
of individual poems. Snorri was not attempting to reconstruct the pre-Christian mythology
or give an accurate representation of it for its own sake, and we should not underestimate
that part of the frleikr ok skemtun of Edda (Faulkes 1998:5) was the skemtun of throwing
things at Baldr (Faulkes 1988:45).

Bibliography
Clunies Ross, Margaret, 1987: Skldskaparml. Snorri Sturlusons ars poetica and medieval theories
of language. Odense.
Clunies Ross, Margaret, 2005: A history of Old Norse poetry and poetics. Cambridge.
Dewey, T. K., & Frog (eds.), 2009 in press: Versatility in versification. Multidisciplinary approaches
to metrics. New York.
Faulkes, A., (ed.), 1988: Edda. Prologue and Gylfaginning. London.
Faulkes, A., (ed.), 1998: Edda. Skldskaparml. Vol. 1. London.
Faulkes, A., (ed.) 1999: Edda. Httatal. London.
Fidjestl, Bjarne, 1980: Det norrne fyrstediktet. PhD. diss. University of Bergen.
Fidjestl, Bjarne, 1985: On a new edition of skaldic poetry. In: The 6th International Saga Conference
28.7.2.8. 1985. Copenhagen. Pp. 319335.

cally bound dyn X formula (in which X = a 2-syllable genitive forming a battle kenning) is one of four varia-
tions on the formula that replace the preposition (of 21 examples from 24 occurrences of dynr as an uncom-
pounded noun in drttkvtt). This example is of particular interest because the unique syntactic break between
dynr and hjlmr appears to be an artful display of skill which requires the recognition of the multiform to be
fully appreciated. (Frog 2009:288299.)

277
Fidjestl, Bjarne, 1997: The kenning system. In: Selected papers. Ed. by O. E. Haugen & E. Mundal.
Odense.
Fidjestl, Bjarne, 1999: The dating of eddic poetry. Ed. by O. E. Haugen. Copenhagen.
Frog, 2009 in press: Speech-acts in skaldic verse. In: Dewey & Frog 2009, 275303.
Grape, A., G. Kallstenius & O. Thorell (eds.), 1977: Edda. Uppsala-handskriften DG 11. Stockholm.
Gunnell, Terry, 1995: The origins of Scandinavian drama. Woodbridge.
Harris, Joseph, 1985: Eddic Poetry. In: Old Norse-Icelandic literature. Ed. by C. J. Clover & J. Lin-
dow. London. Pp. 68156.
Harvilahti, Lauri 1992: Kertovan runon keinot. Helsinki.
Haymes, Edward R., 2004: The Germanic heldenlied and the poetic Edda. Oral tradition 19:4362.
Jonsson, Bengt R., 1991: Oral literature, written literature. In: The ballad and oral literature. Ed. by J.
Harris. Cambridge. Pp. 139170.
Kjartansson, Helgi Skli, 2009 in press: No royal road. The extremes of dtttkvtt lines in Snorris
Httatal. In: Dewey & Frog 2009, 305318.
Lord, Albert B., 1960: The singer of tales. Cambridge.
Mellor, Scott, 1999: Function and formula. PhD diss. University of Wisconsin.
Mrtensson, Lasse, & Heimir Plsson 2008: Anmrkningsvrda suspensioner i DG 11 4to. Scripta
Islandica 59:135155.
Neckel, Gustav, & Hans Kuhn (eds.) 1963: Edda. Heidelberg.
Nordal, Gurn, 2009 in press: Metrical learning and the First Grammatical Treatise. In Dewey &
Frog 2009, 2948.
Rubin, David C., 1995: Memory in Oral Traditions. Oxford.
Sigursson, Gsli, 2004: The medieval Icelandic saga and oral tradition. Cambridge.
Sigursson, Jn, et al. (eds.), 1966 [184887]: The First Grammatical Treatise. In: Edda Snorra Stur-
lusonar 13. Osnabrck. Vol. 2, pp. 1042.
Thorvaldsen, Bernt yvind, 2006: Sv er sagt fornum vsindum Tekstualiseringen av de mytolo-
giske eddadikt. PhD. diss. University of Bergen.
Vesterholt, Ole, 1973: Tradition and individuality. Copenhagen.

278
 Hundingr und Saurr: zum Mythologem Hund im Altnordi-
schen
Natalija Ganina, Dept. of Germanic & Celtic Philology, Lomonosov Moscow State Univer-
sity, Russia
Die Episode in Hkonar Saga Ga, wo der Knig Eysteinn seinen Hund Saurr den Ein-
wohnern von Trondheim zum Knig aufzwingt, wird von den Forschern mit den Berichten
von Plutarch und Claudius Aelianus ber dem Hundeknig der thiopiern verglichen (Krappe
1942, Alexeev 1969). Wie soll aber diese Saurr-Episode im Vergleich zu der altnordischen
berlieferung und Mythologie bewertet werden?
Snorris Erzhlung hat direkte Parallelen in einer Reihe altnordischer Quellen: Af Upplen-
dinga konungum I, Hversu Noregr byggist, das Skldatal, Gesta Danorum VII, Chro-
nicon Lethrense III (Annales Lundenses), Rydrbogen (Annales Ryenses). In einigen
Chroniken (Chronicon Lethrense V, Rydrbogen) heit der Hund Raki/Rakke, eigentlich
Hund, kleiner Hund. Im letzteren Fall kann man von einem Apellativum oder von einer
bergangsstufe vom Apellativum zum Eigennamen sprechen: Vgl.

Rakk koning sprang i bland andr hund (Karker 2001: 114),

was in der modernen Textausgabe als hundekongen und nicht als Rakke kongen bersetzt
wird, also der Hundeknig sprang unter andere Hunde. Bei Snorri Sturluson, im Skldatal
und in Af Upplendinga konungum trgt aber der Hund einen konzeptuell wichtigen Eigen-
namen: Saurr/Sorr Mist, Dreck. In lafs Saga ins Helga 137 wird dasselbe Motiv knapp
erwhnt, und der Name des Hundes dient bei Snorri als Hinweis auf die ganze Geschichte.
Aus den Texten darf gefolgert werden, dass die Legende vom Hundeknig vor allem mit
schwedischen Knigen verbunden ist: Gunnarr (Gesta Danorum, der Sieg ber den Norwe-
gern), Eysteinn (Af Upplendinga Konungum), Adils (Rydrbogen), Hkon (Hakun, Er-
niedrigung der Dnen in Chronicon Lethrensis; in Annales Lundenses, 14. Jh. Athisl).
Es ist bemerkenswert, dass in allen Fllen, wo der Hund Saurr genannt wird, der Knigsname
Eysteinn lautet.
Snorri Sturluson schrieb die Figur des Knigs Eysteinn, der seinen Hund den Trndern
aufgezwungen hatte, der Epoche des Hakons des Guten zu (10. Jh.). Denkbar wre, dass die
Aufmerksamkeit des Autors zu Eysteinn fast aussschlelich durch den Zusammenhang dieses
Knigsnamens mit dem Motiv vom Hundeknig Saurr verursacht worden war.
Das Namenpaar Eysteinn Saurr fhrte Snorri zu einem Missverstndis. Laut anderer
Texte ist Eysteinn ein schwedischer Knig einer wesentlich frheren Epoche. So wird in Af
Upplendinga Konungum Eysteinn der Bse aus der Dynastie der Ynglinger genannt (die
einmalige Erwhnung Eysteins konungs illra). Eysteinn der Sohn von Adils (vgl. Erwh-
nung von Adils in Rydrbogen und in Annales Lundenses) war der Schwiegervater von
Hlfdanr Hvtbein (Weiknochen). So stellt sich heraus, dass dieser Eysteinn nicht ein unbe-
kannter Gegner Hakons des Guten (Mitte des 10. Jhs), sondern ein legendrer Knig aus dem
ausgehenden 6. Jahrhundert ist. Dabei ist fr Snorri der schwedische Knig Eysteinn nur mit
der Thematik des Inglingatals verbunden.
In knappen Zeilen des Skldatals, wo eine Menge Skalden nur bei Namen genannt wird,
kann man aber die Geschichte des Skalden namens Erpr Ltandi finden, dessen Werke nicht
erhalten sind. Dem Skldatal gem schuf Erpr eine Drpa nach dem Befehl des Knigs
Eysteinn beli (Eysteinn Vielfra, auch Eysteinn hinn illri, Eysteinn der Bse genannt), des
Zeitgenossen von legendren Ragnarr Lodbrok (Ende des 8. Jhs.).

279
 Im Ganzen zeichnet sich in diesem Sagenkreis eine invariante legendre Gestalt des bsen
Knigs namens Eysteinn, der eine ganze Reihe Beinamen hat: harri der Gestrenge, (die
Ynglingasaga 44 ber Eysteinn dem Sohn von Adils), illri der Bse (Herrscher) (die drei
obengenannten Quellen mit dem Namenpaar Eysteinn Saurr), hinn illi der Bse (die
Saga von Hakon dem Guten), beli Vielfra (das Skldatal). Die Situation wird durch die
Verwechslung dieses Knigs mit dem Eysteinn dem Mchtigen (hinn rki), einem legendren
Knig aus der Dynastie der Ynglinger (9. Jh.), verwickelt.
Diese Namenverhltnisse bezeugen, dass die von Snorri mitgeteilte Geschichte in der alt-
nordischen belieferung tief eingewurzelt ist.
Die altnordische Legende vom Hundeknig wird gewhnlich als Widerspiegelung des in-
terkulturellen Mythologems Hund als Knig interpretiert. In der gelehrten Tradition des
klassischen Altertums ist dieses Motiv dank Plutarch und Aelianus bezeugt, die eine thiopi-
sche Stammsage mitgeteilt haben (Literatur s. oben). Da handelt es sich aber nur um eine lo-
kale Sage und um keine universelle Mythe. Bei genauer Betrachtung der altnordischen Ge-
schichte vom Hundeknig knnen wir einen wesentlichen Unterschied zur thiopischen Sage
sehen: Die Wahl (eine gewaltsame Ernennung) des Hundes zum Knig hat zum Zweck, laut
allen altnordischen Quellen, die Erniedrigung des besiegten Volkes (vgl. die Wahl zwischen
dem Sklaven und dem Hund). Analogischerweise dient das Knigsbefehl dem Skalden Erpr,
eine Drpa zum Ehren des Knigshundes schaffen, zur Erniedrigung des Skaldes, der ein
schweres Verbrechen (einen Mord im Heiligtum) begangen hat. Die Erniedrigung wird durch
den pejorativen Rufnamen des Hundes besonders betont.
Darin besteht die Eigenart der Saurr-Geschichte. Ein Hund und ein Rufname des Hundes
an sich tragen sowohl in der altnordischen, als auch in der altgermanischen Tradition keine
negative Bedeutung. Wenden wir uns an Beispiele.
In Edda-Liedern gibt es keine Vergleichung Sklave Hund und keine Vergleiche derart.
In allen Kontexten wird der Hund als mchtiges chthonisches Wesen dargestellt: Der chthoni-
sche Hund Garmr tritt zusammen mit dem Wolf Fenrir hervor, Hyndla Hndchen ist der
Name einer Riesin, die im Hyndlalied von Freyja als mr meyja Maid der Maiden, mn vina
Freundin und Hyndla systir Hyndla, Schwester bezeichnet wird. Hundingr heit der Held,
dessen Geschlecht mit den Nachkommen des Wolfs, Ylfingar, kmpft. Diese Sippenfehde
kann man als Widerhall eines altgermanisches Ritus betrachten (Hfler 1992), aber der Name
Hundingr hat in der Tradition nichts Entehrendes. Im Ganzen wird der Hund dem Wolf
gleichgestellt. Vgl. Snorris Beschreibung der Berserkern in der Ynglingasaga 6, die sich sem
hundar ea vargar, wie Hunde oder Wlfe benehmen.
Wie Otto Hfler erwies, war der Hund ein Machtssymbol der Langobarden: Vgl. Beschrei-
bung der Langobarden als Cynocephali bei Paulus Diakonus (Historia Langobardorum I,
11), ihren alten Nebennamen Winnili wtende Hunde (Etymologie von Rudolf Much) und
Hundsymbolik in den Namen und Beinamen der Scaligerfrsten und ihrer Verwandten
(Cangrande I., Cangrande II., Mastino I., Mastino II., Cansignorio, Canfrancesco, Cagnola)
neben dem Hundsymbol als Bildzeichen der Familie (der Hundehelm von Cangrande und
gekrnte Hunde, Wappenschilde haltend). Der Forscher betonte, dass in der selbst in der sp-
ten dnischen berlieferung der Wilde Jger (= Odin) Kong Hundhoved Knig Hundekopf
heit (Hfler 1973: 249). Hfler machte auch darauf aufmerksam, dass der berhmte Fhrer
der nord-norwegischen Bauern in ihrem Kampf gegen Olaf den Heiligen rir Hundr Thorir
Hund genannt war:

Personennamen wie Wolf, Wolfgang oder das im Norden so gelufige Bjrn bezeugen noch in
spten Zeiten alte Ideen von irgendeiner Wesensverwandtschaft. Hund aber, in so vielen
Sprachen als Schimpfwort verwendet, scheint uns wenig geeignet als Mannsname. Und doch

280
 begegnet er, und in einem sehr handhaften Sinn, auch in hohen Schichten der alten Kul-
tur(Hfler 1992: 52).

Die Erbdrpa des Skalden Sighvat berichtet, Thorir Hund sei durch zauberkrftiges Fell gegen
Schwerthiebe geschtzt gewesen (Hfler 1992: 53).
In den islndischen Sagas ist eine Vergleichung Sklave Hund auch nicht bezeugt. Aber
man kann feststellen, dass die Gestalt des Hundes ihre Verbindung mit der archaischen kon-
zeptuellen Sphre nach und nach verliert und ambivalent wird. Einerseits, erhlt die Tochter
einer Knigin ihren Namen nach dem Rufnamen des Hundes (Yrsa, Hrlfs Saga Kraka ok
kappa hans Helga ttr 9), andererseits, ist das adlige Mdchen von ihrer Mutter ver-
schmht und darum ist ihr Hundename ursprnglich pejorativ eingeschtzt (obwohl die ei-
gentliche Bedeutung des Namens Yrsa Brin die Mglichkeit einer Umwertung gibt). Diese
durch die synchrone Stellung eines Haustieres bedingte Ambivalenz des Hundes liegt dem
Gestalten- und Wortspiel in den Sagas zugrunde. Vgl. den Kontext der Vlsungasaga, wo
Gunnarr die Mannen von Atli mit gewisser Verachtung, aber im Kontext des Kampfes als
smir hundar, kleine Hunde bezeichnet, und die Verheimlichung der Kinder unter den Hun-
derufnamen Hoppr und H (Hrlfs saga kraka ok kappa hans Fra ttr 2). Es sei aber
zu bemerken, dass als Welfe, kleine Hunde die Knigsshne, Nachkommen eines adligen
Geschlechts bezeichnet werden: Vgl. gnadda niflfarna die ins Dunkel weggegangene kleine
Tiere, ber die Shne von Atli in Atlakvia 33. Die von Jakob Grimm und Otto Hfler
behandelten Geschichten vom Langobardenknig Lamissio und vom Ursprung der Welfen,
wo die adligen Jungen fr blinde Welfe um Rettung willen ausgegeben oder gehalten worden
seien (Hfler 1992: 58), bezeugen den altgermanischen Charakter dieses Motivs.
In einigen islndischen Sagas werden solche Eigenschaften des Hundes wie sein Verstand
und seine Fhigkeit, fast menschliche Gefhle zu empfinden, besonders hervorgehoben. So
sind Gunnars Hund Smr in Njals saga, den Gunnarr geschenkt bekommen hat (Njals sa-
ga 70), und der getreue Hund in Hlfdanar saga Eysteinssonar 17, den Hlfdanr vom Tode
rettet, wobei der Hund vor Freude weint und spter sein Leben fr Hlfdanr opfert. In diesen
Fllen wird aber der Hund wie ein reelles Haustier mit hervorragenden oder sogar magischen
Eigenschaften beschrieben (aber kein chthonisches Wesen oder hamrammr, Werwolf). Der
Hund ist hier eine Art Eigentum: Vgl. Selsnautr Sels Gabe/Eigentum vom Hund in
Hlfdanar saga Eysteinssonar. Die Hunderufnamen in den Sagas haben keine negative Fr-
bung: Hoppr und H (s. oben) eine auf die ursprngliche Alliteration der Namen Helgi und
Hrarr anspielende Onomatopie (weiter, in der Episode mit den Pferden, werden Hlfdans
Shne Hamr und Hrani genannt), Vgi in lafs saga Tryggvasonar 32 vom an. vgja trei-
ben, Smr in Njls saga 70 eigentlich der Schwarze.
Wie Otto Hfler fr den Namen Thorir Hund nachgewiesen hat, ist der Vergleich eines
Menschen mit dem Hund nicht schmhlich (s. oben). In Fra ttr finden wir weitere Be-
lege: Helgi und Hrarr, die Shne von Hlfdanr, werden nicht blo fr Hunde ausgegeben,
sondern vielleicht sogar in die Hunde verwandelt, weil ihr Beschtzer Vfill als Zauberer be-
schrieben wird, der Feinde magisch ablenken kann. Die Hunderufnamen Hoppr und H gelten
als Signal zum Fliehen der Jungen vom Knig Fri. Die Lebensrettung unter dem Hunderuf-
namen ist also ein besonderes Glck, das mit Hilfe von Magie erreicht wird und keinen Schat-
ten auf die Knigssohne wirft.
Das Wort Hund (an. hundr) ist in den Sagas kein Schimpfwort. Vielmehr, zeigen die
Hundezge im alltglichen Benehmen eines Menschen auf etwas Auerordentliches: So
handelt z.B. rndr in Freyinga saga 38. Er riecht seine Hand, mit der er die Erde betastet
hat, und entdeckt die Menschenspuren, dann schnffelt er die Erde wie ein Hund und findet
die Verfolgten. rndr wird in der Saga wie ein kluger und wrdiger Mann beschrieben, er ist

281
schn und kann hflich reden, und die Erzhlung von seinen Hundeeigenschaften trgt kei-
ne negative Bewertung.
Wie kann man denn auf diesem Hintergrund die Geschichte des Hundeknigs Saurr inter-
pretieren?
Das Motiv vom Hundeknig passt in den allgemeinen Rahmen hinein, wo der Hund als
Eigentum des Knigs begriffen wird (vgl. z. B. rymskvia 6). Beachtenswert ist, dass in al-
len Fassungen der Legende vom Hundeknig der Hund als Knigseigentum beschrieben wird:
Vgl. Saur konungshund (im Akkusativ, als Objekt) Saur den Knigshund im Skldatal. Of-
fensichtlich aus diesen Grnden wird er dem Sklaven gleichgestellt. Eine Mglichkeit solcher
Gleichsetzung finden wir in Helgakvia Hundingsbana II, wo der besiegte Hundingr die
Sklavendienste erfllt (wobei der Text doch keine direkten Vergleiche mit dem Hund bietet).
Der Hund Saurr wird bei Snorri absolut negativ begriffen. Darauf gibt es eine ganze Reihe
deutlicher Hinweise.
Der Rufname Saurr Mist, Dreck korrespondiert mit an. saurr m. Schmutz, feuchte Er-
de, Mist, Dreck, saurigr, saurugr schmutzig, saurga verunreinigen, saurnir Schild
eigentlich der Befleckte (de Vries 1962: 464); vgl. auch saur-reiir Dunggrube, sauru-ligr
unrein, unkeusch, sauru-liga in unreiner, unkeuscher Weise, saur-yri unreine, unkeusche
Worte, feste Wendung aurr ok saurr Schmutz und Mist (Cleasby, Vigfusson 1957: 515,
525). Das altnordische Wort wurde ins Alt- oder Mittelenglische entlehnt: Vgl. me. sowre,
soore, ne. dial. saur Kuhmist, soor Dreck (de Vries 1962: 464). An. saurr gehrt zu der
breiten Sphre der gut erhaltenen und motivierten Kontinuanten von idg. *seur/-l- Feuchte,
Schmutz mit den Ablautvarianten *sour/-l- und *sur/-l-: Vgl. got. *bisauljan verunreinigen,
beschmutzen, bisaulnan sich beschmutzen, sich besudeln, an. sol- Schlamm, Pftze im
Ortsnamen Soleyar (de Vries 1962: 529), norw. dial. syla beflecken, ae. sol n., solu f.
Sumpf, as. sulwian beschmutzen, ahd. sol m./n. schmutzige Pftze (Lehmann 1986: 72),
im Ganzen zu idg. *seu-/*sou- flieen, rinnen (Pokorny 1959: 91213, de Vries 1962: 464,
Lehmann 1986: 72).
Deutliche negative Konnotationen hat der Name Saurr in Eirks Saga Raua, wo das
Wort als mnnlicher Beiname gebraucht wird: Eyjlfr Saurr ist ein Verwandter Valjfs, des-
sen Hof von Eyrks Sklaven beschdigt wird. Eyjlfr ttet diese Sklaven und wird von Eyrkr
aus Rache umgebracht.
Im Kontext von Hkonar Saga Ga bekommt der Name Saurr eine entwickelte Moti-
vierung und bildet das semantische Zentrum der Episode. Auer der bestimmt negativen Fr-
bung des Namens selbst tritt das Verhltnis von Eigennamen Saurr zum Apellativum saurr
hervor: Laut Snorri, mssten die Gefolgsmnner von Saurr den Hundeknig auf ihren Schul-
tern da tragen, wo es schmutzig (saurugt) wre. Der Ortsname Saurshaugr spielt dabei eine
bedeutende Rolle.
Der Ortsname kommt in Norwegen oft vor, weil er eigentlich feuchter/schmutziger Hgel
bedeutet. Aus genauer Betrachtung der Landkarte Norwegens geht hervor, dass der Ortsname
Saurshaugr, heute Srhaugen, sehr hufig bezeugt ist und allgemeine Verbreitung hat, weil er
mit solchen Naturobjekten, wie Flsse, Moore, Smpfe und Inseln fest verbunden ist. Es gibt
also eine ganze Reihe Srhaugen (nicht weniger als 15), die nicht auf die Unterdrckung der
Norweger, sondern auf die Moorgegend hinweisen. Der Ortsname wurde von Skandinaven
nach England bertragen: Vgl. Sowerby (West Yorkshire, England, eine Siedlung auf der
Sandbank des Flusses Calder) < *Saurr-by Siedlung in Moorgegend.
Als Magnus Olsen nachwies, waren die Ortsnamen mit dem Bestandteil -haugr bei Trond-
heim mit dem Ahnenkult verbunden und wurden als Sippenheiligtmer verehrt; in der Regel
waren es Grabhgel der frheren Epoche (Olsen 1926:263271, de Vries 1937:102103). Ein
solcher Hgel diente als sakrales Zentrum:

282
 Es ist [] bezeichnend, dass die heidnischen Germanen fr offentliche Handlungen, die einen
mehr oder weniger religisen Charakter gehabt haben, gerne hochgelegene Stellen gewhlt ha-
ben. In erster Linie gilt das von Dingversammlung; sie fand auf einem natrlichen oder sogar
von Menschenhnden aufgeworfenen Hgel statt[] Ein Hgel in der Nhe des Kniggehftes
ist[] wohl als Grab eines Vorfahrs zu betrachten[] Wir wissen sogar, dass Knigsgrber
mehrfach als Dingshgel benutzt worden sind[] Der nordische Knig, der auf dem Ahnengrab
thront[], konnte[] in unmittelbare Berhrung mit den Ahnengeistern seiner Sippe treten[]
(de Vries 1937:101102).

Hier sehen wir die Gedankenweise, auf die der landschaftliche Ortsname Saurshaugr von
Snorri in diesen Koordinaten reinterpretiert werden knnte. So ist Saurshaugr in seiner Erzh-
lung zu einem Ahnenhgel geworden, wohin der fremde Knig einen Hund setzte, mit ande-
ren Worten zu einer entweihten sakralen Stelle der Trnder.
Also kommt der Hundeknig Saurr bei Snorri nicht als ein Widerhall der altgermanischen
Vorstellungen vom Hund als Machtssymbol, sondern als eine originelle Darstellung einer
wandernder altnordischen Stammsage. Die Geschichte von Saurr ist keine Mythe und keine
gelehrte berlieferung, sondern eine gesellschaft-kritische Empfindung, in knstlerischen
Form verkrpert.
Snorris Erzhlung vom Hundeknig ist die ausfhrlichste in der ganzen altnordischen
berlieferung, und hier wie in anderen Fllen wirkt er als wahrer Knstler. Es scheint nicht
zufllig zu sein, dass sich markante typologische Parallelen zu diesem Kontext in der Erzh-
lung von Michail Bulgakov Das Hundeherz zur Erscheinung treten. Der Hund trgt einen
Namen des niedrigen Status an. Saurr Mist, Dreck: russ. < grau, bunt,
ein gewhnlicher Rufname fr den Kter (vgl. Vasmer 1973: 406, 411). In beiden Fllen wird
in den Hund mit Hilfe der Magie der Verstand hineingelegt. Die menschliche Rede fllt aber
dem Hund schwert, es bleiben Rudimente des Bellens. Der Hund gert auf den Hhepunkt der
macht hin, wobei der neue Herrscher mit groben und unvernnftigen Sklaven assoziiert wird
rir Faxi bei Snorri: Proletarier mit Schwonder bei Bulgakov. Dann kehrt aber der Hund
in ursprnglichen tierischen Zustand zurck, und die neue erniedrigende Regierung hat ein
rasches und ruhmloses Ende. Besonders charakteristich ist der Schlu der Bulgalkovs Erzh-
lung, wo der Professor mit dem konzeptuellen Familiennamen Preobrazhenskij (ein russi-
scher Priesterfamilienname vom aksl., russ. Verklrung [Christi], Ver-
wandlung), der den Hund zu einer Persnlichkeit der gesellschaftlichen Lebens verwandelt
hat, eine umgekehrte Verwandlung ausfhrt und dabei vom Autor Zauberer genannt wird.
Es sei aber betont, dass diese hnlichkeit rein typologisch ist, weil Michail Bulgakov nie
Skandinavistik studierte und die Heimskringla kaum lesen knnte (die russische bersetzung
erschien 1980; wenn auch Bulgakov den Bericht vom Hundeknig bei Plutarch und Aelianus
bercksichtigt hat, sehen wir beim russischen Verfasser eine selbstndige Begrndung des
Themas).
Die Parallelen auf dem Gebiet der sozialen Satire und Antiutopie werden damit nicht er-
schpft. So wird z. B. im Roman vom modernen sterreichischen Schriftsteller Christoph
Ransmayr Morbus Kitahara, wo es sich um eine Alternativweltgeschichte handelt, eine
Welt beschrieben, in der ein im Krieg besiegtes Land (Parallelen zu Deutschland und ster-
reich nach dem Zweiten Weltkrieg) nach der Niederlage deindustrialisiert und in eine Agrar-
gesellschaft zurckverwandelt wird (Parallelen zum Morgenthau-Plan). Eine der drei Haupt-
figuren, der ehemalige Hftling Ambras, der von den Siegermchten befreit worden ist, und
dem die Verwaltung des Steinbruchs versklavten Kriegsverlierer obliegt, wird der Hundek-
nig genannt.
Die Autorschaft von Snorri gehrt aber, wie Michail Steblin-Kamenskij fr die Sagas for-
muliert hat, zum Typ der unbewuten Autorschaft des Mittelalters (vgl. Steblin-Kamenskij

283
2003:125129). Mit knstlerischen Mitteln schafft Snorri die Illusion einer tiefen Sachkennt-
nis und sogar eines absoluten Wissens, der Weisheit der weisen Leuten des Altertums (vgl.
Prolog zu der Heimskringla). Dazu dienen sowohl auffallende, auf die Einbildungskraft wir-
kende Details der Erzhlung (eine bergangsstufe zu der Autorschaft der Neuzeit), als auch
Eigennamen: Hkon der Gute, Eysteinn der Bse, rir Faxi, Saurr, Ortsnamen: Upplendin-
gar, rndheimr, rndir, Eynafylki, Sparbyggjafylki, eyjunn Iri, Saurshaugr. Die Hinweise
auf die berlieferung (Variante des Beinamen von Knig Eysteinn) sind auch vorherbe-
stimmt, das Vertrauen zur Erzhlung und zum Erzhler zu steigern.
Im Ganzen lt die Betrachtung der skandinavischen berlieferung vom Hundeknig zum
Schlu kommen, dass die Funktionierung des Mythologems Hund im Altnordischen fol-
genderweise formuliert werden kann: Von Kong Hundhoved Hundingr zum Hundeknig
Saurr.

Literatur
Alexeev, Michail P., 1969: Jurij Krizhani i folklor moskovkoj inozemnoj slobody. In: Trudy Otdela
drevnerusskoj literatury Instituta russkoj literatury Akademii Nauk SSSR. T. 24. Leningrad. S.
299304.
Bulgakov, Michail A., 1992: Sobae serdce [Hundeherz]. In: Bulgakov, Michail Afanasevi. Werke
in 5 Bnden. Bd 2. Moskau. S.119208.
Cleasby, Richard, and Vigfusson, Gudbrand 1957: An Icelandic-English Dictionary. Initiated by R.
Cleasby; revised, enlarged and completed by Gudbrand Vigfusson. Oxford.
Hfler, Otto, 1973: Verwandlungskulte, Volkssagen und Mythen. Wien.
Hfler, Otto, 1992: Cangrande von Verona und das Hundsymbol der Langobarden. In: Hfler, Otto.
Kleine Schriften. Hamburg. 4282.
Karker, Allan, 2001: Dansk i tusind r. Et omrids af sprogets historie. Modersml-Selskabets rbog
1993. Ny revideret udgave. C. A. Reitzels Forlag A/S.
Krappe, Alexander H., 1942: The Dog King. In: Scandinavian Studies. 17. Provo, Utah. P. 148153.
Much, Rudolf, 1920: Der germanische Osten in Heldensage. In: Zeitschrift fr deutsches Altertum. Bd
57. S. 145176.
Lehmann, Winfred P., 1986: A Gothic Etymological Dictionary. Based on the 3d ed. of Vergleichen-
des Wrterbuch der gotischen Sprache by Sigmund Feist. Leiden.
Olsen, Magnus, 1926: ttegrd og Helligdom. Oslo.
Pokorny, Julius, 1959: Indogermanisches etymologisches Wrterbuch. Bd 12. Bern Mnchen.
Ransmayr, Christoph, 1995: Morbus Kitahara. Frankfurt am Main.
Steblin-Kamenskij, Michail I., 2003: Trudy po filologii [Werke in Filologie]. Sankt-Petersburg.
Vasmer, Max, 1973: Etimologieskij slovar russkogo jazyka. Perevod s nem. i dopolnenija O. N.
Trubaova [Etymologisches Wrterbuch der russischen Sprache. Aus dem Deutschen bersetzt und
erweitert von Oleg N. Trubaov]. Bd 4. Moskau.
Vries, Jan de, 1962: Altislndisches etymologisches Wrterbuch. Leiden.
Vries, Jan de, 1937: Altnordische Religionsgeschichte. Bd 2. Berlin Leipzig.

Summary
The article is concerned with the investigation of the episode about the dog king in Hkonar
Saga Ga against the background of the mythical functions of dog in Old Germanic tradi-
tion. Here is the pejorative dog name Saurr mud, dirt, excrements of primary importance, for
it obtains a crucial lexical support in the context and becomes both the conceptual focus of the
narrative and the basis of the poetic etymology given by Snorri Sturluson to the local Norwe-
gian place name Saurshaugr damp/dirty hill.

284
 The Good, the Bad and the Undead
New Thoughts on the Ambivalence of Old Norse Sorcery
Leszek Gardea, Department of Archaeology, University of Aberdeen, Scotland

Introduction
This short paper presents, in a condensed form, a review of my most recent studies on the
aspects of Old Norse sorcery and the initial results my PhD project which is currently under-
taken at the Department of Archaeology, University of Aberdeen1.

The ambivalence of Old Norse sorcery

Numerous Old Norse accounts such as sagas, skaldic poems and Eddic poetry but also medie-
val Norwegian chronicles (for example Historia Norwegie, Historia de Antiquitate Regum
Norwagiensium and grip af Noregskonungasgum) and rune-stones contain information on
the enigmatic performers of a very special magical craft often referred to as seir. When taken
collectively those sources imply that seir was a kind of operative magic which among
other things enabled its practitioners to foresee the future, heal the sick, change weather
conditions, reveal the hidden, shift into animal form or travel to other worlds in a state of
trance. Seir, however, also had a darker side and could be employed to inflict physical or
mental harm. At present, the darker aspect (or as Dag Strmbck 1935; 2000 would see it:
black seir) of this practice lies at the core of my studies.
The undoubted existence of the two distinct facets of seir, which are so evident in the
written accounts, has recently led me to reinterpreting a number of very atypical Scandinavian
burials (Gardea 2008b: 60; 2009a: 208209; 2009b; 2009c). After having conducted a pre-
liminary analysis of the available archaeological material I am inclined to believe that when
given a closer look and viewed from an interdisciplinary perspective those graves may pro-
vide actual, material evidence for what some scholars understand as social ambivalence of
Old Norse sorcery (Dillmann 2006: 457586). Furthermore, they imply that there existed
multiple forms of treating the deceased sorcerers and that the manner of burying the dead was
dependant not only on the role which they played during their lives but also on a social per-
ception of their actions and the very nature of their craft.

The archaeology of sorcerers

In 2002 Neil Price published his influential book entitled The Viking Way. Religion and War
in Late Iron Age Scandinavia where he convincingly argued that it is possible to identify a
number of Viking Age graves as belonging to ritual specialists involved in the practices of
seir.
The graves discussed within his book (Price 2002: 127161, 191203 ) can be divided into
a number of categories. Some of the alleged seeresses and sorcerers were buried in wagons,
others in wooden chambers and a few of them were even interred on boats. Alongside several
extremely rich graves, there are also less elaborate inhumation and cremation burials. In a
number of cases the deceased were accompanied by animals such as horses or dogs. Although

1
Due to the review-form of the present paper I will only reference the most vital literature and avoid debates of
more general nature. History of research on Old Norse sorcery as well as the latest advancements in the studies
of seir can be found in the works of Price (2002); Solli (2002); Heide (2006a); Dillmann (2006) and Gardea
(2008a; 2008b; 2009a; 2009b; 2009c).

285
all those graves are in many respects different from one another, there exist a number of inter-
esting confluences. It is impossible to elaborate on them here, but it is significant to note that
the main argument that makes it possible to view them as a special, coherent group of burials
is the presence of iron rods in each one of them.
Those rods, which in several cases were decorated with bronze knobs, are currently be-
lieved to be attributes of the ritual performers and labeled as staffs of sorcery (a term first in-
troduced by Neil Price in 2002). As Price (2002: 175180) argued, the staff was one of the
main attributes of the Late Iron Age performers and there exist many sources which confirm
that they were strongly associated with the practices of seir. Furthermore, an account from
Laxdla saga (ch. 76) suggests that the deceased seeresses were actually interred with their
staffs. This piece of literary evidence strongly supports the archaeological interpretations of
the graves with iron rods as belonging to seir-workers.

New perspectives on the staffs of sorcery

In my masters thesis (Gardea 2008a) and a number of academic papers (2008b; 2009a) I
aimed at expanding the earlier interpretations of the staffs and argued that the iron rods
from the graves of potential ritual performers possessed an extremely rich symbolism. I sug-
gested that they corresponded with the symbolic concepts that were related to both domestic
and military tools and activities. The staffs, in my view, worked as multi-layered metaphors
and could have been perceived by the contemporary societies as objects of truly otherworldly
qualities.
In one of my most recent articles (Gardea 2009a) I argued that it is possible to discuss the
staffs of sorcery in the light of the archaeology of personhood a theory recently developed
in the works of Chris Fowler (2006) and perceive them as persons in their own right. This
approach has led me to a reconstruction of the complex processes of creating, using and aban-
doning/depositing/destroying/killing the staffs. The most remarkable result which this re-
search has revealed is that some of the staffs known from the archaeological contexts actually
died in the same way as the human sorcerers. This is particularly apparent in the case of the
staff found in the grave Ka. 294296 in Kaupang-Skiringssal (Vestfold, Norway), which was
found lying under a large rock (Stylegar 2007: 96; Gardea 2009b: 193195). As we shall
later see, there exist a number atypical burials in Scandinavia, where individuals are also liter-
ally crushed with large stones. In my opinion, such graves could have belonged to malevolent
seir performers and form a very distinct new category of sorcerers burials.
Furthermore, it can be added that there exist very interesting parallels to the Scandinavian
staffs of sorcery in the Slavic and Baltic archaeological and ethnographical material (Gardea
2008b: 5152; 2009a: 201). I believe that by viewing the seir-staffs in a cross-cultural con-
text, we might also arrive at a better understanding of the nature of their owners.

The Kriwe priest and the concept of divine crookedness

The names Krvis in Lithuania and Kriwe in Prussia were sometimes attributed to a pagan
high-priest. However, it has been argued that Kriwe was not really a name of a particular per-
son, but rather a term used for defining a certain category of ritual specialists (Tomicki 2000:
472; Banaszkiewicz 2002: 3943; Kowalik 2006: 395397; for an earlier hard to accept
interpretation rejecting the existence of Kriwe, see Rowell 1994: 128). The prefix kriv- seems
to be related to Indo-European concepts of twisting, turning or crooking (Tomicki 2000: 471
472).
One of the most famous sources which discusses the role of Kriwe is the Chronicon terrae
Prussiae written by Petrus de Dusburg in the 14th century (Rowell 1994: 3839, 125128;
Kowalik 2006: 395397). In his description the author mentions a pagan temple in

286
Romuva/Romowe where a priest named Kriwe resided. He was the guardian of the sacred
fire, possessed divinatory skills and was greatly respected by the society. The most important
tool of his trade was a crooked staff. Another account can be found in the work of Simon
Grunau from the 16th century, where we read about Kriwe-kriwaito, referring to a pagan priest
in the temple of Perkunas in Wilno/Vilnus (Tomicki 2000: 472).
In his paper, Tomicki (2000: 472) interestingly argued that the name Kriwe could be re-
lated to the particular features of his staff. In his view the original Kriwe staffs might have
been very similar to the staffs which were used as symbols of power over village communities
in Poland and Lithuania until the 20th century. In Poland such items were known as krzywula
(krzywua), kluka, kula and in Lithuania krivule, krievas or krive (Tomicki 2000: 427). Staffs
of this kind were often made from a very unusually shaped and twisted branch or a root
(Tomicki 2000: 427).
Tomicki (2000: 428) also mentions that references to sticks or clubs used as symbols of
power could also be seen in the names of mythical or semi-historical characters such as Kij,
Krak, Krok, Klukas and others (see also Banaszkiewicz 2002: 3943).
He further argues that the pagan practice among the Baltic peoples could have also been
referred to as krzywanie (Tomicki 2000). This term is closely related to everything that is un-
usual or supernatural, but also to looking inside a web or reaching into the other world.
All this implies that both the ritual practitioner Kriwe, his practice krzywanie, and the
crooked staff krzywula were connected to the concepts of physical and metaphorical crook-
edness. This crookedness was however not seen as a fault at all, but rather as a complex
metaphor of supernatural qualities of the ritual performer as well as his actions and tools.
I strongly believe that the Viking Age seir performers and their staffs of sorcery recently
identified in the archaeological material could also relate to concepts of divine crookedness.
As we have seen, this idea of crookedness seemed to be vital in the representations of tools
for sorcery or authority among the Baltic peoples and also in the later Balto-Slavic folklore. It
could also explain why most of the Viking Age iron staffs have a rather strange looking ex-
panded handle construction. Apparently it is very similar to some of the 19th and 20th cen-
tury staffs of the krzywula-type (Mierzyski 1885; Moszyski 1968: 897). Thus the physical
and metaphorical crookedness of the seir-staff might actually prove to be an another way
of expressing the otherness of the ritual specialist to whom it belonged.

The malevolent sorcerers and stones in graves

The Old Norse sources which contain information on the lives of seir-workers strongly sug-
gest that there was a certain ambivalence in the perception of their actions. On one hand there
existed greatly respected specialists whose main domain was conducting divinatory rituals
and helping the contemporary societies in overcoming various problems related to their eve-
ryday lives. On the other hand, however, there were also a number of sorcerers who got in-
volved in malevolent actions and committed acts of theft or murder. The saga accounts imply
that there were specific methods of punishing the evil sorcerers. In most cases the punishment
for practicing evil sorcery was stoning to death (Strm 1942: 102115)2. From an archaeo-
logical perspective it is striking that the sagas provide rather precise details about the ways in
which the evil sorcerers were interred. As the sources suggest, after the stoning procedure, the
bodies of the deceased were also covered up with stones. It is also significant to note that the
burial often occurred in a secluded place were people were least likely to pass by. Already
during their lifetimes, the sorcerers were seen as rather ambiquous and marginal figures and
this aspect of marginality seems to have also been apparent after their death.

2
Other forms of punishment involved outlawry, drowning or burning.

287
 In my latest research I have aimed at listing all the available Old Norse accounts which
contain the motif of punishment by stoning and later comparing the evidence with a number
of atypical burials from Late Iron Age Scandinavia (Gardea 2009b; 2009c). So far, I have
been able to identify seven Scandinavian graves in which the individuals (both women and
men) were buried under large stones. Two such graves are known from Iceland3, four from
Gotland4 and one from Gerdrup in Denmark. It is remarkable that the stones were placed di-
rectly on the bodies of the deceased, as if they were intended to pin them to their graves.
What this meant is hard to interpret, but it is possible that the crushing with stones was done
to avoid the dead from returning to the world of the living. However, such a strange ritual
could have also had a number of other meanings (Gardea 2009c). For example, according to
several Old Norse accounts (Hamisml st. 26, Ragnarsdrpa, Skaldskaparml ch. 7, Vlsun-
gasaga ch. 44) stones seem to possess the capacity of breaking magic spells and piercing
magically enhanced armor. If this was the case, perhaps the stones were placed in the graves
to once and for all neutralise the magical skills of the deceased? Or maybe the stones pre-
vented other people (other sorcerers?) from raising them from the dead, through the practice
known as tiseta? Because the Viking mentalities were so diverse, we must always remain
open to many interpretational possibilities.

The sorcerers from Gerdrup a special case study

One of the most remarkable graves, which I consider as belonging to potentially malevolent
sorcerers, was found in 1981 (Christensen 1982) in the village of Gerdrup (Zealand, Den-
mark) to the north from Roskilde. The burial mound was originally located on an beach ridge
near an old tributary of the fjord.
The grave was more than a metre deep, filled with blocks of grass peat (Christensen 1997:
34) and it contained well preserved skeletal remains of two individuals a man and a woman.
They were both aligned NW. In the northern part of the grave against the mans head there
was also a large stone (Christensen 1981: 21).
According to Christensen (1981: 21) the interred man was hanged, as suggested by twisted
cervical vertebrae. Due to the peculiar position of his legs it is likely that his feet were bound5
with a rope or some organic material (which unfortunately did not survive until the present
day). He was buried lying on his back, aged around 3540 years old and equipped only with a
knife (16cm long) placed on the bottom part of his chest (Christensen 1981: 2122). In Chris-
tensens view the man was a rall (a slave), but this might not necessarily be the only plausi-
ble interpretation (Gardea 2009b; 2009c).

3
Grave KT-25: 1 from Haugava near Traarholt (rnesysla, South-West Iceland) and Grave KT-145: 2 from
Va (South-Mlassla, East Iceland) (ra Ptursdttir 2007: 39, 54).
4
All graves of this kind were found in Frjel, Gotland: Grave 32/88, Grave 9/89, Grave 19/89 (Carlsson 1999)
and an unnumbered grave discovered in 1998 (Carlsson 1998: 1011).
5
As Gade (1985: 161) observed, hanging in medieval Scandinavia was a penalty for treason, insolence, murder
and offences of sexual nature like: adultery, seduction or abduction. It was also inflicted upon those who com-
mitted acts of theft, plundering or marauding (Gade 1985: 161). It is interesting to note, that in the Late Iron Age
some individuals could have been hanged by the feet (Gade 1985: 173). Hanging by the feet would not cause the
twisting of cervical vertebrae, but it is not impossible that in case of the man from Gerdrup the braking of the
neck occurred after the hanging procedure, when the rope was cut. Perhaps the man fell from the tree and hit his
head against the surface? Some scholars have (in my opinion very convincingly) suggested that in the memora-
ble episode from Hvaml (st. 137) inn was also hanging with his head down from the tree Yggdrasil (Fleck
1971: 142; Supecki 2003: 120) and this motif of ritual inversion has many parallels around the world (Fleck
1971). It seems to me that the hanging of the man from Gerdrup had some very strong ritualistic overtones. On
the possible relation between hanging and the initiations of seimenn see Solli 2002; 2008.

288
 45cm to the east from the skeleton of the man there laid a skeleton of a woman. She had
her head placed to the north and feet to the west. What is most remarkable, however, is that
the woman was lying on her back and her body was crushed by two large boulders. One of the
stones (30x45cm) was placed directly on her chest and the other one (20x30cm) was lying on
her right leg (Christensen 1981: 21). Another boulder was placed to the east from the woman,
several centimetres from her waistline. She was roughly 40 years old and equipped with a
knife (14cm long). On her waistline the woman had a bone case containing small iron pins.
Additionally she was given a roughly 40cm spear which was lying around 510cm from her
right leg (Christensen 1981: 22).
Another ambiguous feature of the grave is that both the man and the woman seem to be
covering their genitals. Furthermore, the poses in which they were interred appear to mirror
each other: the man has his right hand placed on his right lap and the woman has her left hand
placed on the pelvic girdle. The left hand of the man is under his pelvic girdle and the
womans right hand is under her pelvic girdle as well.
In this context we might recall the account of Ibn Fadln who had a unique opportunity to
observe a funeral of a Rus chieftain by the river Volga in the year 922. In his elaborate de-
scription of the burial ceremony he mentioned how the closest relative of the deceased while
being completely naked approached the funeral pyre walking backwards and covering his
anus6. It is quite possible that the covering of the anus was done to avoid the penetration by
spirits7.
Eldar Heide (2006b: 355356) recently suggested that perhaps the reason why seir was
seen as a perverse practice resulted from the fact that the practitioner was believed to become
possessed and penetrated by spirits during the ceremony. Due to this act of metaphorical
penetration a male seir-worker was immediately ascribed a feminine role. Furthermore, as
Heide argues (2006: 356), while some spirits entered the body through the respiratory pas-
sages others the more hostile ones perhaps did so through the backside. We cannot be sure
why the two individuals from Gerdrup had their hands placed under the pelvic girdle but per-
haps this had something to do with the notions sketched above. In this context we may also
recall the finds of Migration Period golden bracteates, such as the one from Alles in Den-
mark, where a man is covering his genitals with his right hand (Duczko 2002: 176).
Christensen argued that it is likely that the deceased woman was considered as a sorceress
and that the peculiar deposition of her body within the grave might reflect the acts of stoning
that we know from the written accounts (1981: 2728). I agree with Christensen but I am
convinced that the spear placed by her right leg was in fact a special kind of staff of sorcery
(Gardea 2008b: 5960; 2009b: 209). We know from the Old Norse written accounts that
staffs often transformed into spears (Gardea 2008b: 59) and that the spear Gungnir was an
important attribute of the god inn an undoubted master of seir.
To conclude, I find it incredibly remarkable how closely this burial reminds the passage
from Eyrbyggja saga (ch. 20) where there is a mention of a man named Oddr who was ac-
cused of cutting off a womans arm and for that act sentenced to death by hanging. Immedi-
ately afterwards his mother, a sorceress named Katla, was stoned to death for helping Oddr to
hide from his pursuers. The hanged man, the stoned woman and the location of the grave it-
self at a beach near to the fjord reflect the grim story of the death of Katla and her son Oddr
very clearly. No other potential sorcerers burial seems to parallel the written accounts as

6
As we read in the account of Ibn Fadln (translation after Montgomery 2000: 20) : Then the deceaseds next
of kin approached and took hold of a piece of wood and set fire to it. He walked backwards, with the back of his
neck to the ship, his face to the people, with the lighted piece of wood in one hand and the other hand on his
anus, being completely naked.
7
On this notion see for example: Price 2002: 360361; Duczko 2004: 151152; Heide 2006c: 168.

289
closely as that from Gerdrup. Of course we must be mindful that the events in Eyrbyggja saga
take place in Iceland and not Denmark, but it is apparent that the same custom of treating the
dead sorcerers was known in many parts of Late Iron Age Scandinavia.
My hypothesis, which shall be expanded in the nearest future, is that from among all the
burials of the alleged Viking Age ritual performers, there could be distinguished two basic
categories: graves with staffs and graves in which the individuals were crushed with stones.
The first category could perhaps be ascribed to highly regarded seers and seeresses whereas
the other category belonged to malevolent sorcerers who committed some violent acts. The
existence of such distinct types of burials in the archaeological material reflects the ambiva-
lent nature of Old Norse sorcery which is observable in the written accounts8.

Conclusions and future research

The archaeological evidence for the ambivalence of Old Norse sorcery has until now never
been discussed in the academic literature, but the problem of ambiquity of Viking Age magi-
cal practices was certainly observed in the earlier works of philologists and historians of relig-
ions. In my opinion by trying to build bridges between those disciplines it is possible to create
a new and fascinating picture of the Viking Age realities.
In the further stages of my research the evidence from the Scandinavian world will be
viewed in comparison with the worldviews and burial practices of the Slavs9 as well as the
Baltic and Finno-Ugric peoples.

8
Of course this view is open to expansions and alterations. The past was not simply black and white and in
many respects the Viking mentalities were as complex as those of our own. It is quite probable that there may
exist more types of sorcerers burials. In my view, female graves with spears or graves in which the spears were
thrusted into the ground could perhaps also belong to potential ritual specialists. This problem, however, requires
further research and cannot be discussed here.
9
In the Slavic archaeology there exists a concept of the so-called vampire burials (Wrzesiski 2008). In those
graves, the individuals are often found with their heads chopped off or they are buried facing the ground. In
some cases the bodies of the interred were also covered with stones. My preliminary hypothesis is, that perhaps
some of the alleged vampire burials (especially the ones with stones), actually belonged to Slavic sorcerers or
pagan priests.

290
Figure 1. An iron staff of sorcery from grave Bj. 834 at Birka (Uppland, Sweden). Photographs
Leszek Gardea, drawings after Price 2002: 182183.

291
Figure 2. A selection of nineteenth-century staffs of the krzywula-type from Lithuania and Poland.
After Mierzyski 1885: figure 1.

292
Figure 3. The remarkable Gerdrup grave (Zealand, Denmark). After Christensen 1997: 35.

Bibliography
Primary sources
Driscoll, M.J. (ed. and trans.), 1995: grip af Nregskonungasgum. A Twelfth-Century Synoptic
History of the Kings of Norway. Viking Society for Northern Research Text Series. Volume 10.
London.
Dronke, Ursula, (ed. and trans.), 1969: The Poetic Edda, I: Heroic Poems: Oxford.
Einar lafur Sveinsson (ed.) 1934: Laxdla saga. slenzk Fornrit 5. Reykjavk.
Einar lafur Sveinsson, Matthias rarson, (ed.) 1935: Eyrbyggja saga. slenzk Fornrit 4. Reykjavk.
Ekrem, Inger, Mortensen Lars, Boje, Fisher, Peter, (ed. and trans.), 2003: Historia Norwegie. Copen-
hagen.
Faulkes, Anthony, (ed.), 1998: Edda. Skldskaparml. 1. Introduction, Text and Notes. London.
McDougall, David, McDougall Ian, (trans.), 1998: Theodoricus Monachus. Historia de Antiquitate
Regum Norwagiensium. An Account of the Ancient History of the Norwegian Kings. London.
Neckel, Gustav, Kuhn Hans, (eds.), 1983: Edda. Die Lieder des Codex Regius. Nebst Verwandten
Denkmlern. Band I. Text. Heidelberg.

Secondary sources
Banaszkiewicz, Jacek, 2002: Polskie dzieje bajeczne Mistrza Wincentego Kadubka. Wrocaw.
Carlsson, Dan, 1998: Frjel Discovery Programme. Forskning og historieformedling, (no publisher,
available online: http://www.hgo.se/frojel/excavation.html).
Carlsson, Dan, 1999: Ridans. Vikingahamnen i Frjel. ArcheoDok Skrifter nr 2. CCC Papers 2.
Visby.
Christensen, Tom, 1981: Gerdrup-graven. Romu: rsskrift fra Roskilde Museum 2(1981). Pp. 1928.
Christensen, Tom 1997: The Armed Woman and the Hanged Thrall. In: The Ages Collected from the
Roskilde Museum. Ed. by F. Birkebk. Roskilde. Pp. 3435.
Dillmann, Franois-Xavier, 2006: Les magiciens dans lIslande ancienne: tudes sur la reprsentation
de la magie islandaise et de ses agents dans les sources littraires norroises. Uppsala.
Duczko, Wadysaw, 2002: Taczcy wojownicy: ikonografia rytuaw kultowo-militarnych w
skandynawskiej sztuce wczesnego redniowiecza. In: Imago Narrat. Obraz jako komunikat w
spoeczestwach europejskich. Ed. by S. Rosik, P. Wiszewski. Wrocaw. Pp. 165187.
Duczko, Wadysaw, 2004: Viking Rus. Studies on the Presence of Scandinavians in Eastern Europe.
Leiden Boston.
Fleck, Jere, 1971: Oinns selfsacrifice a new interpretation. Scandinavian Studies 43. Pp. 119142.
Fowler, Chris, 2006: The Archaeology of Personhood. An anthropological approach. London-New
York.
Gade, Kari, Ellen, 1985: Hanging in Northern Law and Literature. Maal og Minne 1985. Pp. 159183.

293
Gardea, Leszek, 2008a: Brzemi seir. Oblicza magii w epoce wikingw. Pozna (Masters thesis
defended at the Institute of Prehistory, Adam Mickiewicz University in Pozna).
Gardea, Leszek, 2008b: Into Viking Minds: Reinterpreting the Staffs of Sorcery and Unravelling
seir, Viking and Medieval Scandinavia 4. Pp. 4584.
Gardea, Leszek, 2009a: A biography of the seir-staffs. Towards an archaeology of emotions. In:
Between Paganism and Christianity in the North. Ed. by L.P. Supecki, J. Morawiec. Rzeszw. Pp.
188217.
Gardea, Leszek, 2009b: Ukamienowani. mier czarownikw w islandzkich sagach a archeologia
epoki wikingw. In: IV Oglnopolska Studencka Konferencja Antropologiczna, Od narodzin do
mierci. Czowiek w naturze i kulturze. Ed. by H. Gb, M. Dbrowska, M. Kpa & al. Krakw (in
press).
Gardea, Leszek, 2009c: Buried with Honor and Stoned to Death. The Ambivalence of Viking Age
Sorcery in the Light of Archaeology. In: Analecta Archaeologica Ressoviensia 4. Ed. by S.
Kadrow, W. Rczkowski, Rzeszw (in press).
Heide, Eldar, 2006a: Gand, seid og ndevind, Bergen.
Heide, Eldar, 2006b: Spirits Through Respiratory Passages. In: The fantastic in Old Norse/Icelandic
Literature. Sagas and the British Isles. Preprint Papers of the 13th International Saga Conference,
Durham and York, 6th12th August 2006. Ed. by D. Ashurst, D. Kick, J. McKinnel. Durham. Pp.
350358.
Heide, Eldar, 2006c: Spinning Seir. In: Old Norse religion in long-term perspectives : origins,
changes and interactions : an international conference in Lund, Sweden, June 37, 2004, Vgar till
Midgrd 8. Ed. by A. Andrn, K. Jennbert, C. Raudvere. Lund. Pp. 164170.
Kowalik, Artur, 2006: Kosmologia dawnych Sowian. Prolegomena do teologii politycznej dawnych
Sowian. Krakw.
Mierzyski, Antoni, 1885: Nuncius cum baculo. Studjum archeologiczne o krywuli. Wisa 9. Pp. 361
397.
Montgomery, James, 2000: Ibn Fadln and the Rssyyah. Journal of Arabic and Islamic Studies 3. Pp. 1
25.
Moszyski, Kazimierz, 1968: Kultura Ludowa Sowian 2(2). Warszawa.
Price, Neil, S., 2002: The Viking Way: Religion and War in Late Iron Age Scandinavia. Uppsala.
Rowell, S.C., 1994: Lithuania Ascending. A pagan empire within east-central Europe 12951345.
Cambridge.
Supecki, Leszek, Pawe, 2003: Mitologia skandynawska w epoce wikingw. Krakw.
Solli, Brit, 2002: Seid. Myter, sjamanisme og kjnn i vikingenes tid. Oslo.
Solli, Brit, 2008: Queering the Cosmology of the Vikings. A Queer Analysis of the Cult of Odin and
Holy White Stones. Journal of Homosexuality 54(1/2). Pp. 192208.
Strm, Folke, 1942: On the Sacral Origin of the Germanic Death Penalties. Lund.
Strmbck, Dag, 1935: Sejd: Textstudier i nordisk religionhistoria. Stockholm.
Strmbck, Dag and others, 2000: Sejd och andra studier i nordisk sjlsuppfattning. Hedemora.
Stylegar, Frans, Arne, 2007: The Kaupang Cemeteries Revisited, In: Kaupang in Skiringssal, Kaupang
Excavation Project Publication Series vol. I. Norske Oldfunn vol. XXII. Ed. by D. Skre. Aarhus.
65101.
Tomicki, Ryszard, 2000: Czy Krive krzywa? Kilka uwag o poytkach z przeytkw. In: Archeologia
w teorii i praktyce. Ed. by A. Buko, P. Urbaczyk. Warszawa. Pp. 459481.
ra Petursdttir 2007: Deyr f, deyja frndr. Re-animating mortuary remains from Viking Age
Iceland, MA-Thesis in Archaeology. Faculty of Social Sciences. University of Troms. Troms.
Wrzesiski, Jacek, (ed.) 2008: Czarownice Prba podsumowania warsztatw. In: Funeralia
Lednickie spotkanie 2. Ed. by J. Wrzesiski. Pozna. Pp. 143156.

294
 The Mental Map of Greenland in the Icelandic Sagas
Gsli Sigursson, Stofnun rna Magnssonar slenskum frum, Hskla slands, Iceland
and Universitetet i Stavanger, Norway

In my paper I shall discuss the mental map of Greenland as it is presented in the Icelandic
Sagas. I have previously studied the sagas mental map of the lands south and west of
Greenland as well as of the British Isles. This I have discussed in view of historical memory
and contemporary knowledge with emphasis on one of the social roles of oral storytelling
about remote places and voyages to faraway lands: Namely that stories inform the audience of
the worlds geography, that is in which direction people can sail and which features can
be expected to be outstanding to the seafarers eye when he comes up to previously unknown
coasts. Still another way to experiment with this line of thought is to analyse stories in Iceland
about characters who are in or visit Greenland: Do these stories draw up or reflect
a comprehensive mental map of the area and if so, could that map serve as a realistic back-
ground for the travels and movements described and thus be of informative value for those
who have not visited this part of the world themselves? For this purpose it is irrelevant
whether or not the stories reflect a profound knowledge of the historical reality in Greenland,
which narrative function the landscape may have or if supernatural phenomena play a consid-
erable role in the stories. The important question is if the stories can be regarded as
an encyclopedic medium of traditional geographical knowledge about Greenland in the minds
of the Icelandic audience.

295
 Sensory deceptions. Concepts of mediality in the Prose Edda
Jrg Glauser, University of Basel and University of Zrich, Switzerland

1. The rubric
Gylfaginning is preceded in the U manuscript of the Prose Edda by a well-known and much-
cited rubric, which gives the work the title it still bears today, and ascribes it to an author: Bk
essi heitir Edda. Hana hefir saman setta Snorri Sturluson eptir eim htti sem hr er skipat.
Er fyrst fr sum ok Ymi, ar nst Skldskaparml ok heiti margra hluta, sast Httatal er
Snorri hefir ort um Hkon konung ok Skla hertoga. (Faulkes 1988:xiii; cf. Snorri Sturluson
1977:1), This book is called Edda. It was compiled by Snorri Sturluson in the manner in
which it is arranged here. There is told first about the sir and Ymir, then Skaldskaparmal
[the language of poetry] and terms for many things, finally Hattatal [list of verse-forms]
which Snorri has composed about King Hakon and Duke Skuli (Faulkes 1987:[vi]).

Illustrattion 1. DG 11 (Snorri Sturluson 1962:[unpag.])

If we turn our attention for once away from the content of these statements if we refrain,
that is, from enquiring into the likelihood of Snorris authorship of the Edda (cf. on this
Faulkes 1988:xiiif.) it is apparent that the rubric in its first sentence refers deictically to the
works status as object (bk essi), as a concretely existing manuscript or book, and in its
second, draws attention to the composition of this book as a whole: it is a text made up of
several parts. The writer of this note, using highly stereotyped formulations, emphasises at the
outset that a (medieval) text always exhibits a material dimension as well as a thematic-
aesthetic one; then he reviews the origins of the work, ascribes it to an author/compiler (one
who he assumes will be known to the audience), provides the text as a whole and its parts
with titles (cf. Faulkes 1988:xviif.; Faulkes 1998:vii) and gives an overview of its structure.
With setja saman compile, in the sense of add, bring together, arrange, and yrkja to
compose, he avails himself of a specifically poetological terminology.1

1
Anthony Faulkes translates setja saman with compile, yrkja with compose (Faulkes 1988:xiiif.), Arnulf
Krause uses zusammenstellen and dichten (Snorri Sturluson 1997:7), Karl G. Johansson and Mats Malm
render the two terms with stta samman and dikta in their Swedish translation of the Prose Edda (Snorri
Sturluson 1997a:24).

296
Illustration 2. DG 11 (Snorri Sturluson 1977:1)

The first sentences of DG 11 show, then, a scribe with a decided awareness of a process
which one can call medial. A move to the performative can already be observed in this short
passage at the beginning of U, a performativity which, with its stress on the elements of
staging, framing, and reflection, is characteristic for the specifically pre-modern mediality of
the Prose Edda as a whole.2 The fragmentary Edda manuscript AM 748 I b 4to, from c.
130025 (cf. Faulkes 1988:vi), for example, contains the following passage, which bears out
the medial significance of the opening rubric in U: Hr r lykt im lvt bokar r Olafr
orarson hfir samansett ok vpphefr skalldskaparmal ok knningar ptir vi sem fyri
fvndi var kvvm hfvtskallda ok Snori hfir sian samanfra latit (Snorri Sturluson
1852:427f.; cf. Faulkes 1998:xii) Here is ended that part of the book which lfr rarson
has put together, and begins the language of poetry and the kennings, according to that which
was found in the poems of the chief skalds and which Snorri later caused to be brought
together.
Numerous similar passages in other texts3 show that the rubric in U is no random scribal
note. The following section at the beginning of Skldskaparml especially deserves attention.
It not only functions as a kind of prologue to this part of the work, but also refers to the
conception of the Prose Edda as a whole (this book, at the beginning of this book): En
etta er n at segja ungum skldum eim er girnask at nema ml skldskapar []: skili
hann essa bk til frleiks og skemtunar. [] En eigi skulu kristnir menn tra heiin go
ok eigi sannyndi essarar sagnar annan veg en sv sem hr finnsk upphafi bkar er sagt er
fr atburum eim er mannflkit viltisk fr rttri tr [] hvernig Asiamenn eir er sir eru
kallair flsuu frsagnir r fr eim tindum er gerusk Troju [] (Snorri Sturluson
1998:5), But these things have now to be told to young poets who desire to learn the
language of poetry [] Then let such a one take this book as scholarly enquiry and
entertainment. [] Yet Christian people must not believe in heathen gods, nor in the truth of
this account in any other way than that in which it is presented at the beginning of this book,
where it is told what happened after mankind went astray from the true faith [] how the
people of Asia, known as sir, distorted the accounts of the events that took place in Troy
[] (Faulkes 1987:645).
An awareness of the significance of the medially reflective and therefore literary-
theoretical and poetological aspects of medieval texts achieves explicit formulation in such
passages of the Prose Edda (it probably also finds expression in its title, as edda is already
used in the fourteenth century in collocations such as eddu list (art of edda) or eddu reglur
(rules of edda) to mean poetics). They draw on an already well-developed terminology and
deliver precise insights into the formation and function of narratives in their capacity as

2
On mediality in premodern literary texts, cf. especially Kiening 2007.
3
For a discussion of the historical terminology for the various aspects of writing in the prose texts of the Norwe-
gian and Icelandic Middle Ages, see Glauser in press.

297
written texts. Whoever composed the rubric of the U manuscript of the Prose Edda was well
aware that his text would be received in the context of a differentiated written culture.4

2. Gangleri asks
In contrast to this discussion of mediality, which assumes an established written culture as its
reference point, another well-known and oft-cited instance in the Prose Edda presents a
medial situation which appears to be part of a quite different tradition. The drawing which
occupies page 50 of the U manuscript of the Prose Edda (cf. Snorri Sturluson 1962:[unpag.];
Snorri Sturluson 1977:XVIII), the oldest
(fourteenth-century) and also by a
significant margin the best-known
illustration of the Codex Upsaliensis, shows
on the left a standing figure, supporting
himself on a staff, who is named as Gangleri
the legend over this figure reads Gangleri
spyrr Gangleri asks, and on the right
three crowned figures, named as Hr,
Jafnhr and rii, sitting on a sort of triple
throne.
From a media-historical perspective, we
are confronted here, for one, with a highly
consequential shift in medium from text to
image (translation of content from one
medial form to another), and/or with mixed
media, text and image (simultaneous
application of several medial forms on the
same occasion). Furthermore, the non-
linguistic medium of the drawing brings a
medial situation into focus in which verbal,
close-range communication is represented in
the context of a dialogue between physically
present speakers and listeners. In this simple
drawing the human voice, intimately bound
to a physically present body the clear
gestures of the figures, for instance, signpost
this is the centre of attention. This is a
completely different kind of medial situation

4
Faulkes (1998:xix) has already pointed out that Skldskaparml in particular makes numerous references to the
text as written object. A few of the many examples: Hr skal heyra hv skldin hafa kennt skldskapinn eptir
essum heitum er r eru ritu (Snorri Sturluson 1998:11), Now examples will be given of how the poets have
referred to poetry using such terms as were noted above (Faulkes 1987:70); [] sv sem kva Eyvindr ok fyrr
var ritat (Snorri Sturluson 1998:14), as in the poem of Eyvind quoted above (Faulkes 1987:72); Hr er ess
getit [] sem fyrr er ritat (Snorri Sturluson 1998:18), Here reference is made [] which was written about
above (Faulkes 1987:75); N skal enn segja dmi af hverju r kenningar eru er n vru ritaar, er r vru
eigi dmi til sg (Snorri Sturluson 1998:20), Now there shall be told more of the underlying stories from
which those kennings just listed have originated, and of which the origins have not already been told (Faulkes
1987:77); essi nfn himins eru ritu, en eigi hfum vr fundit kvum ll essi heiti (Snorri Sturluson
1998:85), The following names for the heavens are written down, but we have not found all these terms in po-
ems (Faulkes 1987:133).

298
from that presented in the introductory rubric of U, in which an act of communication
between an author and belated, perhaps physically distant in any case, not necessarily
present readers is sketched out.5
In the Prose Edda, learned written culture is supplemented by the textual imagining of
traditional oral communication. The various frame and interior narratives, in Gylfaginning as
well as in Skldskaparml, map this quasi pre-written form of communication on to the the-
matic level, when Gylfi/Gangleri talks to Hr, Jafnhr and rii, or gir/Hlr to Bragi.6 The
fundamentally dialogical principle of the sections couched as wisdom conversations (to which
Httatal also belongs) receives explicit emphasis in innumerable scenes of narrating in the
elaborate frame-stories of Gylfaginning and Skldskaparml, and contributes significantly to
the poetological and medial self-reflexivity of the text. The dialogical structure is directly
thematised in the second chapter of Gylfaginning by one of the figures involved, namely, Hr:
Hann [Gangleri] segir at fyrst vil hann spyrja ef nokkvorr er frr mar inni. Hr segir at
hann komi eigi heill t nema hann s frari, ok Stattu fram mean fregn, / sitja skal s er
segir. (Snorri Sturluson 1988:8), He [Gangleri] said that he wished first to find out if there
was any learned person in there. High said he would not get out unscathed unless he was more
learned, and Stand out in front while you ask:/ he who tells shall sit. (Faulkes 1987:8).
At the beginning of Skldskaparml the text also explicitly signals that the following nar-
rative is a rendering of a conversational situation, when it says that [t]he person sitting next
to gir was Bragi, and they drank and conversed together. Bragi related to gir many events
in which the sir had been involved. He began his account where[] (Faulkes 1987:59).
(Nsti mar gi sat Bragi, ok ttusk eir vi drykkju ok oraskipti. Sagi Bragi gi fr
mrgum tindum eim er sir hfu tt. Hann hf ar frsgn at [], Snorri Sturluson
1998:1). And when Hr closes his portrait of the god Thor in chapter 21 of Gylfaginning as
follows: En engi er sv frr at telja kunni ll strvirki hans, en segja kann ek r sv mrg
tindi fr honum at dveljask munu stundirnar r en sagt er allt at er ek veit. (Snorri Stur-
luson 1988:23), But there is no one so wise that he can recount all his exploits, though I can
tell you so many stories about him that much time will be taken up before all I know is told.
(Faulkes 1987:223), the text refers to an oral, improvised story-telling which rests on a res-
ervoir, in principle inexhaustible, of orally-delivered narratives which can be recalled contex-
tually.
A sophisticated narrative about the genesis of mythic narratives, that is, nothing less than a
metapoetic, medially-reflexive, self-displaying history of textual origins (cf. Kiening
2007:345), is to be found in the conclusion of Gylfaginning which, after staging the final dis-
illusionment of the questioner, suggests transmission from mouth to mouth as the kernel of
tradition: Gengr hann [Gangleri] lei sna braut ok kemr heim rki sitt ok segir au
tindi er hann hefir st ok heyrt. Ok eptir honum sagi hverr mar rum essar sgur. En
sir setjask tal ok ra rum snum ok minnask essar frsagnir allar er honum vru
sagar [] (Snorri Sturluson 1988:54), Then he [Gangleri] went off on his way and came
back to his kingdom and told of the events he had seen and heard about. And from his account
these stories passed from one person to another. But the sir sat down to discuss and hold a
conference and went over all these stories that had been told him[] (Faulkes 1987:57).

3. Sensory deceptions

5
On the significance of bodily presence and non-linguistic communication in the medial context of saga litera-
ture, cf. Glauser 2007.
6
Faulkes distinguishes three extended conversational sequences in Skldskaparml (cf. Faulkes 1998:xixxxi).

299
The distancing ironisation implicit in the dissolution of the illusion-creating frame at the end
of Gylfaginning refers to a further fundamental condition of narrative, insofar as it shows how
literature (oral and written) rests at heart on deception; he who believes in it succumbs to a
fraud. The text, that is to say, repeatedly makes clear that the stories about Hr, Jafnhr and
rii or about Bragi are myths, which must be considered under the sign of sensory deception,
sjnhverfingar (where the English term, in contrast to the Icelandic one, includes other senses
as well as seeing, for example, hearing). At the beginning of Gylfaginning, the magically-
skilled Swedish king sets out on his way in order to discover why the sir are so
knowledgeable: Hann byrjai fer sna til sgars ok fr me laun ok br sik gamals
manns lki ok dulisk sv. En sir vru v vsari at eir hfu spdm, ok s eir fer hans
fyrr en hann kom, ok geru mti honum sjnhverfingar. En er hann kom inn borgina s
hann ar hva hll [] (Snorri Sturluson 1988:7), He set out to Asgard and travelled in se-
cret and assumed the form of an old man and so disguised himself. But the sir were the
wiser in that they had the gift of prophecy, and they saw his movements before he arrived, and
prepared deceptive appearances for him. When he got into the city he saw there a high
hall[] (Faulkes 1987:7). This episode is highly interesting from a media-theoretical view-
point, because it demonstrates by means of a small scene how unreliable the human senses,
here the sense of sight (he saw) are in actual fact.
This pattern of deception and countering deception is repeated at the beginning of
Skldskaparml, as gir like Gylfi, a master of magic approaches Asgard: Hann var mjk
fjlkunnigr. Hann geri fer sna til sgars, en er sir vissu fer hans var honum fagnat vel
ok margir hlutir me sjnhverfingum. (Snorri Sturluson 1998:1), He was very skilled in
magic. He set out to visit Asgard, and when the sir became aware of his movements, he was
given a great welcome, though many things had deceptive appearances (Faulkes 1987:59).
As Hr, Jafnhr and rii break off the conversation at the end of Gylfaginning and the
illusion which has sustained the dialogue is lifted, the text, significantly, uses verbs of
perceiving (hear, look, see), which refer to the medial situation, []v nst heyri Gan-
gleri dyni mikla hvern veg fr sr, ok leit t hli sr. Ok er hann ssk meir um stendr
hann ti slttum velli, sr nga hll ok nga borg. (Snorri Sturluson 1988:54), [n]ext Gan-
gleri heard great noises in every direction from him, and he looked out to one side. And when
he looked around further he found he was standing out on open ground, could see no hall and
no castle (Faulkes 1987:57).
Literary fictions, as these stories from the Prose Edda demonstrate in an admirably preg-
nant fashion, rest on the basic principle of trickery of the senses; they are, in fact, phantasma-
goric. The multiply framed and interrupted narratives about the old stories of the gods stage
rhetorical and medial situations and show how myths arise out of such illusions. Perhaps the
best example of reflection on medial conditions in the Prose Edda is given by the story of the
origin of the mead of poetry in Skldskaparml. The story of the precious mead itself a
powerful medium, mediating between the earthly and supernatural is dominated by fraud,
deceit and aggression. One of the central myths of the Prose Edda, it is at the same time a
story about how stories come to be, how they develop and are passed on, a multi-dimensional
meta-narrative about literary art and its medial forms, which carries out a complex poetologi-
cal and medial discussion. In mythic narratives this is one of the lessons Gangleri may draw
from his conversation with the sir (njttu n sem namt (Snorri Sturluson 1988:54), may
the knowledge you have gained do you good (Faulkes 1987:57)) but, of course, not only
there, one should not believe too readily in what one thinks oneself to see and hear.7 The nar-

7
Faulkes (1988:xx) briefly discusses the fictional nature of the frame story in Skldskaparml and asks among
other things whether the absurdities were deliberately intended as a joke or included for ironical purposes.

300
ratives of Gylfaginning and Skldskaparml construct phantasms, then promptly deconstruct
them again.
As Christian Kiening has shown in a groundbreaking article on mediality in the pre-
modern era, medieval texts tend to display, stage and reflect medial conditions (cf. Kiening
2007:esp. 33943). The tutorial in deconstructive reading offered by the dialogically con-
ceived sensory deceptions of the Prose Edda is at the same time self-displaying, as regards
basic medial situations and processes of literature in the Middle Ages. This reflexivity con-
cerns not least the numerous performative aspects of medieval texts, also treated in the Prose
Edda.
This implicit poetics of the Prose Edda suggests a further circumstance which is of impor-
tance in the current connection. The work contains no myths in themselves and does not
consist of direct, immediate myth-tellings. Rather it belongs to the genre of mythography, and
presents a prototypical medial situation: the Prose Edda de- (or in-)scribes and discusses the
narrating of mythic narratives. This literally mythographic aspect refers in itself to the medial-
ity of the text. In this way the Prose Edda shows very clearly that, as the German philosopher
Hans Blumenberg pointed out in his now-classic work Arbeit am Mythos (Work on Myth,
Blumenberg 1981), mythological content is never available as such, but rather is always re-
ceived, that is to say, mediated through a middle term. From the viewpoint of religious his-
tory, this creative process of reworking appears as a rule in a negative light, when a one-sided
view is taken of the process of dismantling old forms in favour of different accentuations of
the same content in new media. This is the case, for example, in Jn Hnefill Aalsteinssons
investigation of the history of the mead of poetry, whose committing to writing in the medial
context of the Icelandic High Middle Ages he interprets as the desecration of an ancient
northern myth (cf. Jn Hnefill Aalsteinsson 2005). But from the perspective of literary and
media history, the processes of revision and transmission, treated on various levels in the text
itself, yield extremely interesting indications of how the Middle Ages imagined the coming
into being of mythological narratives. When we are told at the end of Gylfaginning, as cited
above, that the sir sat down to discuss, took counsel, and recalled the stories they had previ-
ously told Gangleri, it is the multilayered character and complexity of narration in different
media which is the theme. Terms like minnask (recall), frsagnir (stories), segja (say,
narrate), setjask tal (sit down to discuss), ra rum snum (take counsel) make evi-
dent the significance of memory, the human voice, and the collective, performative act of
story-telling for the genesis and transmission of literature. Hr, Jafnhr and rii and Bragi, as
mediators in the narrative frame of the Prose Edda, mark the (often precarious) medial
situations at the intersections between individual medial forms. From the point of view of the
history of mediality, the Prose Edda is a text which both has a rich array of contextualisations
in a written culture, and simultaneously bears in itself many traces of other medial forms;
furthermore, in its dense network of representations and reflections of different medial
situations, it displays a sophisticated textual self-consciousness.

Bibliography
Blumenberg, Hans, 1981: Arbeit am Mythos. Frankfurt am Main: Suhrkamp.
Faulkes, Anthony, tr. 1987: Snorri Sturluson: Edda. London: Everyman.
Faulkes, Anthony, 1988: Introduction. In: Snorri Sturluson, xixxxiv.
Faulkes, Anthony, 1998: Introduction. In: Snorri Sturluson, viilxxii.
Glauser, Jrg, 2007: The Speaking Bodies of Saga Texts. In: Learning and Understanding in the Old
Norse World. Essays in Honour of Margaret Clunies Ross. Ed. by J. Quinn, K. Heslop & T. Wills.
Turnhout: Brepols, Pp.1326.

301
Glauser, Jrg, in press: Staging the Text. On the development of a consciousness of writing in the
Norwegian and Icelandic literature of the Middle Ages. In: Along the Oral-Written Continuum.
Types of texts, relations and their implications. Ed. by E. Mundal & S. Rankovic. Turnhout: Bre-
pols.
Jn Hnefill Aalsteinsson, 2005: Vanhelgun norrnnar gosgu. In: Skrnir 179: Pp. 5976.
Kiening, Christian, 2007: Medialitt in medivistischer Perspektive. In: Poetica. Zeitschrift fr
Sprach- und Literaturwissenschaft 39: Pp. 285352.
Snorri Sturluson, 1852: Edda Snorra Sturlusonar. Edda Snorronis Sturli. II Hafni: Sumptibus Lega-
ti Arnamagnani.
Snorri Sturluson. 1962: Snorre Sturlasons Edda. Uppsala-handskriften DG 11, I. Facsimileedition i
ljustryck. [] utg. med historisk inledning av Anders Grape, Stockholm: Generalstabens Litogra-
fiska Anstalt, Almqvist & Wicksells Boktryckeri AB.
Snorri Sturluson, 1977: Snorre Sturlasons Edda. Uppsala-handskriften DG 11, II. Transkriberad text
och Paleografisk kommentar av Anders Grape, Gottfrid Kallstenius och Olof Thorell. [] Uppsala:
Almqvist & Wicksell International.
Snorri Sturluson, 1988: Edda. Prologue and Gylfaginning. Ed. by Anthony Faulkes. London: Viking
Society for Northern Research, University College London.
Snorri Sturluson, 1997: Die Edda des Snorri Sturluson. Ausgewhlt, bersetzt und kommentiert von
Arnulf Krause. Stuttgart: Philipp Reclam jun.
Snorri Sturluson, 1997a: Snorres Edda. versttning frn islndskan och inledning av Karl G. Johans-
son och Mats Malm. Stockholm: Fabel.
Snorri Sturluson, 1998: Edda. Skldskaparml. 1. Introduction, Text and Notes. Ed. by A. Faulkes.
London: Viking Society for Northern Research, University College London.

302
 On the Reception of Eastern Europe
in Pre-Literate Iceland1
Galina Glazyrina, Russian Academy of Sciences
The presence of East-European matter in Old Norse-Icelandic writings reflects a sphere of
specific interest to Icelanders. In spite of its distance from other countries, Iceland never be-
came a culturally isolated country and was in constant communication with the external world
(Vsteinn lason 1998:31). The abundance of travel stories which are a dominating theme in
Icelandic sagas of different genres (Andersson 2000; Jesch 2005:119; Zilmer 2003, 2006), is
clear evidence of this. The sagas testify that news about Eastern Europe was brought to Ice-
land by informants (both Icelanders and foreigners) who either traveled austr themselves or
heard stories told by or about someone who did. The reasons for traveling to this area much
depended on various forms of activity, and almost all of them can be exemplified by referring
to the sagas where they are preserved both as episodes of different length and as isolated de-
tails. Icelanders traded regularly in Rus this is attested by Grgs (Ch. 259) and reflected in
many sagas; in Iceland they met merchants who brought there not only various goods but also
new stories (a merchant brought from Sweden to Iceland the story of Yngvar which is told in
Yngvars saga vfrla). Icelanders took part in viking raids (Gunnarr of Brennu-Njls saga
raided in the Eastern Baltic) and in military expeditions (Ketill the Icelander was a member of
Yngvars detachment in Yngvars saga vfrla); they served as body-guards of foreign kings
(Icelanders were in the service of Haraldr Sigurarson; the main hero of Bjarnar saga Hit-
dlakappa and Bari from Heiarvgasaga give service to the Russian Kings Valdamarr and
Jarizleifr). With the advent of Christianity some pilgrims who traveled to Holy Places reached
Eastern Europe and Rus (orvaldr, a hero of Kristni saga and orvalds ttr vfrla, trav-
eled as far as Polotsk and died there; another traveler by name of Brandr vfrli visited the
place where orvaldr was buried and composed a verse about him). The role of skalds in
transmitting East-European subjects must have been very high in Iceland, and people with this
professional activity should be mentioned in the list of informants who contributed to enrich-
ing the knowledge of Icelanders about distant worlds (Jesch 2005:119); thanks to their artistic
talent and trained memory many East-European themes may have been introduced in Iceland
and other Nordic countries.
For those who listened to tales about a persons travels there must have existed a conven-
tional set of key features distinguishing a particular region from all others. There is no doubt
that often-repeated place-names assisted in creating a mental picture of the region. Even if a
person had a very vague idea about an exact geographical location of a town, a river, a lake or
a mountain, these places were recognised by him as belonging to a certain area. The first to be
remembered were general terms, such as the names of the countries (Gararki) and of the
region (Austrlnd, Austrvegr), the most often visited towns (altogether 10 Russian towns are
mentioned in the sources, among which Hlmgarr dominates), etc. (see Table 7 in:
Drevnjaja Rus 1999:466474) An argument for an easy comprehension of East European
place-names by the audience, even if they were not part of an immediate saga context, can be
found in such textual evidence as, for example, the use of the nickname Hlmgarsfari (Hrafn
from Vk in Freyinga saga, ch. 8; Skinna-Bjrn in Landnmabk, ch. 55) attached to the
names of people who traveled to Rus. In spite of the fact that the written sagas do not include

1
I would like to aknowledge the financial support by OIFN RAN. Project: Istoricheskaja tradizija v
dopismennyh i pismennyh obschestvah: reprezentazija, vzaimodejstvije, transformazija. Komparativnoje issle-
dovanije.

303
any expanded narratives about the travels undertaken by these saga heroes, one can be sure
that they did exist, and that Hlmgarr as the destination of Hrafn and Bjrn could not have
been misunderstood by the audience of their lost tales. Unusual objects, originated in the
East-European region (such as a hat of special form or cloth, gerzkr httr: Brennu-Njls
saga, ch. 46; Gsla saga Surssonar, ch. 55; Laxdla saga, ch. 12, etc.), also bore a connection
to a certain geographical place in Rus.
Similar to place-names, the names of several historical persons, Russian rulers of the tenth
to eleventh centuries, were often repeated in the written sagas. It is not at all clear whether
these people became as easily recognizable as Russian place-names and whether their person-
alities were as easily identified in Iceland at the time preceding the creation of the corpus of
written kings sagas. In the sagas of Icelanders the name of Jarizleifr (Yaroslav the Wise,
10191054), the most popular of Russian rulers with the saga authors, is not often met. Val-
damarr (Vladimir Svyatoslavitch, ?1015), Jarizleifrs father, the second most popular Rus-
sian personage, is mentioned only in Bjarnar saga Htdlakappa (c. 1230). Of these two, lets
look more precisely at the narrative tradition connected with Yaroslav.
The historical foundation for literary plots connected with Yaroslav were intensive con-
tacts between the European North and Rus during the period of Yaroslavs reign. The name
of Jarizleifr (Jaritlfr, Jarizlfr, etc.) is mentioned in a wide variety of Old Scandinavian
writings of different genres, such as early historiography (the synoptics), hagiography, skaldic
poetry, sagas of different types. Special attention given to the personality of Yaroslav in the
works of modern scholars naturally has resulted in the fact that practically every scrap of writ-
ten evidence concerning him has been known and commented on. In Russian historical schol-
arship priority can be given to Elena Rydzevskaja who in her article Yaroslav the Wise in
Old Icelandic Writings (1945) summarized the material known about Jarizleifr. Rydzevskaya
showed that kings sagas are the richest in the plots in which Yaroslav is involved. The his-
torical evidence given by them constitutes the major part of everything known to us about the
contacts of Rus and Scandinavian countries in the first half of the eleventh century. At the
same time all the diverse connections that took various forms were presented as descriptions
of the personal relations of Yaroslav with Norwegian and sometimes Swedish kings. Ry-
dezevskaya grouped the data and revealed that in the written texts there are four principal
situations around which Yaroslavs plots appear:
1. The marriage (c. 1019/1020) of Yaroslav to Ingigerr, the daughter of Olaf the Swede.
Ingigerr had been betrothed to lfr the Saint of Norway, but her father broke the engage-
ment and arranged her marriage to Yaroslav who, intending to marry a Swedish princess, sent
messengers to Sweden with his proposal.
2. The arrival of lfr of Norway (c. 1029) in Rus after a series of failures in his struggle
with Kntr. On his return to Norway in 1030 lfr left his little son Magns with Jarizleifr
and Ingigerr who took very good care of him.
3. After lfr died in 1030, his brother Haraldr fled to Jarizleifr who made him the leader
of his Varangian troops. Twelve years later on his return from Byzantium to Rus Haraldr
married Elizabeth, Yaroslavs daughter.
4. The arrival of Norwegian chieftains (1034) who decided to make Magns lfsson the
king of Norway. Yaroslav and Ingigerr let Magns go with them on their promise to be
faithful to the son of lfr (Rydzevskaya 1945:6768).
Each of these four historical situations is described in the sagas with a different degree of
detail: the later the text, the more developed are its East-European plots. Thus, the synoptics
primarily enumerate the facts and dont contain evaluative statements, individual characteris-
tics or dialogues. On the other hand, in Snorri Sturlusons Heimskringla and in Flateyjarbk
the episodes with Yarizleifr are rich in details and descriptions. The continued use of the plots
in later texts seems to testify to a better knowledge and understanding of the East-European

304
material, compared to that in the earlier writings, thanks to its repetition in a fixed literary
form. To make a survey of Yaroslavs appearance in Old Norse-Icelandic writings, lets sup-
ply the four groups singled out by Rydzevskaya with such cases as the use of Yaroslavs
name in, for instance, informative formulas which explain the relationship between Russian
princes or between members of Russian and Scandinavian ruling dynasties. Sometimes refer-
ences to Yaroslav are aimed at determining place (Old Rus) and time (the epoch of Yaroslav)
and cannot be classified in any of the four groups.
One finds Yaroslavs episodes in the earlier writings, created at the end of the twelfth or at
the beginning of the thirteenth century, i.e. in those writings which were composed at the be-
ginning of the Icelandic written tradition. Medieval authors do not disclose their sources of
information about Yaroslav and his time. Memories of this Russian ruler might have been an
element of oral stories about Icelanders or their Norwegian or Swedish relatives and friends
who were in service in Rus, who traded there or visited it for other reasons. Its not easy to
decide how widely such stories were known in different parts of the European North, on the
one hand, and in different parts of Iceland on the other. With the development of literacy his-
torical topics received their fixed form and more people were able to learn stories told in a
uniform manner, but shortly before that, in the second half of the twelfth century, Russian
history could hardly have had the form of a common body of learning shared by people in
different localities in Iceland, due to the facts that, firstly, communication between Rus and
Iceland was sporadic, and, secondly, because people in various parts of Iceland must have
enjoyed listening to different stories told by different story-tellers who, in their turn, had dif-
ferent backgrounds, and sometimes actual knowledge of the events they described.
For an illustration lets read a passage in lfs saga Tryggvasonar by Oddr monk (c. 1180)
which seems to elucidate the reception of Jarizleifr by the audience, different from that of the
learned clerics. The episode refers to the time of St Vladimir, the father of Yaroslav the Wise,
but it still has a bearing on Yaroslav.
Text in AM 310, 4to (12501275, ONP:450):

I enna tima re firir Gara riki Valldamarr konungr me miclum ueg. Sua er sagt (my italics.
G.G.) at moir hans var spakona oc er at callat ibokum phitons andi er heinir menn spau. at
geck miok eptir er hon mlti. oc var hon a a orvasa alldri. at var sir eira at iola aptan hinn
fyrsta scylldi bera hana astoli firir hasti konungs. Oc ar menn tki til dryckio a spyrr
konungr mour sina ef hon si ea vissi nocquorn hasca ea scaa yfir gnapa sinu riki. ea
nolgaz me nocquorum ufrii ea tta. ea arir agirntiz hans eigu. hon suarar. Eigi se ec at
son minn er ec vita er horva meinsamliga ea inu riki oc eigi oc at er skelfi ina hamingio.
En o se ec micla syn oc agetlega Nu er borinn aessum tium eiN konungs s. iNoregi oc a
essu ari sa er her man up faz iessu landi. oc sia man vera agetligr mar. oc dyrlegr
hofingi. oc eigi man hann scaa gera inu riki. helldr man hann at margfalliga auca yr til
handa. oc sian man hann aptr huerva til sins lanz a er hann er a ungum alldri. oc mun hann a
olaz riki sitt at er han er til boriN. ok mun hann konungr vera oc skina me mikilli birti. oc
morgum mun hann hialpare vera inorralfu himsins. En scamma stund mon hans riki standa
yfir Noregs ueldi. beri mek nu abrot ui at ec man nu eigi framarr segia. oc rit sagt er nu. essi
Valldamarr var fair Iarizleifs konungs.

(lfs saga Tryggvasonar by Oddr monk: 20. Ch. 6).

Text S in Holm. perg 18, 4to (c. 1300, ONP:355):

Nv aessi ti re austr firir Gara rike Valldamarr konungr oc var hann agtr mar. moir hans
var spakona ok sa hon marga luti firir oc geck at eptir er hn melti. hon var rvasa. ok var at
sir hennar. at hn var borin i hllina hvern iola aptan oc skylldi segia hvat a veri tienda vm

305
 heiminn. ok sat hon a stoli firir haseti konungs oc er men voro komnir isti sin oc bunir til
drykio. a melti konungr. Hvat ser u moir. ea er nockot haskasamlikt mino rike. Hon melti
at se ekki at eigi stande itt rike me soma ok veg. Hit se ek at aessi stvndv ti er borinn
iNorege konungs s. me biortvm fylgiom ok hamingiom ok er mikit lios yfir honum. hann mvn
her upp fzaz i esso lande oc styia itt rike a marga lvnd ok mvn sian aptr hverfa til sinna
ttlanda. ok vera ar konungr agtligr oc dyrligr. ok mvn skiot o missa ok er hann er kvaddr
af heimi. a ligr fyrir honum miklo meiri dyr en ek kvnna vm at tala oc berit mik nv a brt. oc
mvn ek nv eigi segia framaR. esse ValldamaR var faer Iarizleifs f. Holta f. Valldamars f.
Harallz f. Ingibiargar mour Valldamars Dana konungs.

(lfs saga Tryggvasonar by Oddr monk: 2021. Ch. 5).

A popular folklore motif (Boberg 1966:84; D1812: magic power of prophecy) structures the
episode predicting the future for lfr Tryggvason. What differs the story told by Oddr munk
from other stories of the same sort is that the future for the perspective king of Norway is be-
ing predicted by a spkona a pagan sorcerer, the mother of King Valdamarr r Gararki,
who doesnt belong to the same cultural tradition as lfr himself (E. Rydzevskaya found an
analogue to this episode in smundar saga kappabana. Rydzevskaya 1935:14). The texts in
A and in S stem from a common original (Bjarni Aalbjarnarson 1936:5768), of which S is a
shortened version. They reveal different stages in the presentation of the whole episode. An
indication of this is the way in which the author of the first written saga had to explain the
name of Valdamarr to his readers. In A he uses a simple fatherson relationship. In S he
substitutes it with the evidence of the genealogy of the Russian ruling dynasty, in that part of
it which has connections with the dynasties of Scandinavian countries (Pashuto 1968:passim,
419420; Jackson 1991:159163; Nazarenko 2001:582):
esse ValldamaR (Russian king Vladimir Svjatolavitch) var faer Iarizleifs (Russian king
Jaroslav the Wise) f. Holta (Russian prince Vsevolod Yaroslavitch) f. Valldamars (Russian
king Vladimir Monomakh) f. Harallz (Russian king Mstislav Vladimirovitch) f. Ingibiargar
(Russian princess Ingibjrg, the daughter of Mstislav, the wife of Danish king Knut Lavarde)
mour Valldamars Dana konungs (Danish king Valdemar I).
The dates of the rule of the last mentioned Danish king Valdemar the Great (11571182)
provide evidence that this genealogy was compiled earlier than the ascendancy of Valdemar
II, the son of Valdemar the Great and queen Sofia, Russian by birth, that is before 1202. Had
it been written at the time of Valdemar II, the author would have mentioned eight stages in the
pedigree and marked four, not three, Russian and Danish rulers named Valdemar/Valdamarr.
The accuracy in S cannot be explained by its dependence on a contemporary written
source, as no other source contains this combination of names (Jackson 1993:187188). The
genealogy is definitely not of Icelandic origin: in this case it would have been easier for an
author at the end of the thirteenth century to refer to the marriage of Ingigerr, a Swedish
princess, and Yaroslav, who was Valdamarrs son. An attempt to draw the genealogy as far as
to king of Denmark Valdemar I suggests the Danish line of borrowing.
In A the prophecy episode ends by stating: essi Valldamarr var fair Iarizleifs konungs.
Here the reference to Jarizleifr has not been supplied with additional commentary, as if the
author was sure that the name he mentions is one of those which do not need comment. He
himself knows the person he is referring to, and he has no doubt that his audience shares his
knowledge. The name of Jarizleifr, he is sure, can be faultlessly associated with only one
person, the famous king of Rus. The introductory sentence in A I enna tima re firir Gara
riki Valldamarr konungr me miclum veg was evidently not enough to explain which of the
several personalities who bore this name. By mentioning the name of king Valdamarrs son
Jarizleifr the author identifies essi Valdamarr. A special interest of the saga authors and their

306
audience in king Jarizleifrs time naturally resulted from the popularity of kings sagas
devoted to the lives of St lfr, Magns lfsson and Haraldr Sigurarson, whose life and
adventures often brought them to Rus during Yaroslavs reign. Some of the stories might
have been supported by the evidence of the ancestors of the informants who themselves took
part in these travels. Stories about lfr Tryggvason, on the other hand, referred to earlier
times and belonged to a different tradition which did not coincide with entertaining stories of
other Norwegian kings exploits in Eastern Europe.
The prophecy tale in manuscript A comprises a sva er sagt formula which points to an oral
tradition as the source of the story. In the sagas of Icelanders, however, as Th. Andersson
showed (Andersson 1966; see also: Manhire 19741977), such oral references might become
a literary convention; in the earlier writings other than the sagas (Nikuls Bergsson. Leiar-
vsir: 12, 17; Ari orgilsson. slendingabk. ch. I:48, ch. VII:53) their basic function of
refering to generally known facts or traditions is clearly revealed (similar formulas are known
in the early Slavic traditions; see: Schaveljov 2007:8996). For the authors of the first written
sagas, who worked in the period of a recording of available tradition (Andersson 2006:23),
the original function of the sva er sagt reference seems to be more than natural. And if this is
so, the prophecy tale would have been borrowed by Oddr Snorrason from an oral source; this
gives additional support to the idea that East-European themes were indeed part of the earlier
oral tradition.
From the fact that the author had to specify the identity of king Valdamarr, whose mother
predicted a glorious future for lfr Tryggvason, by using the name of Yaroslav (in A), it
follows that this tale did not belong to the most popular stories about lfr Tryggvason. The
changes introduced by the author of S point to the version in A as the earlier version of the
prophecy tale, which the audience in Iceland was not yet familiar with. It might have been
borrowed by Oddr Snorrason from someone who knew the Norwegian oral traditional tales.
The results of J. McKinnells study of the use of the vlva as a free-standing motif in the sa-
gas (McKinnell 2003) give additional support to this conclusion. E. Rydzevskaya classified
both this tale and the legend of lfrs activity in baptizing Rus as belonging to the circle of
Christian legends about the king which had a limited dissemination (Rydzevskaya 1935:12).
The identification of Valdamarr by means of his son Yaroslav breaks a standard rule of
son father genealogical reference. It gives grounds for believing that at the end of the
twelfth century, connecting the names of these Russian kings in narrative and written practice
was not yet habitual. We can also assume that by this time oral stories about East-European
and Scandinavian contacts of the first part of the eleventh century, in which Yaroslav appears
as an active figure, became widely known in Iceland. Consequently, there is not enough
evidence to speak about the popularity of the Russian king Vladimir.
And finally, the tale illustrates that at the time when Oddr Snorrason composed his saga the
narrative tradition of Iceland did not rely exclusively upon local samples. It was attracting and
adapting oral legends and stories from other Nordic regional traditions.

Bibliography
Andersson, Theodore M., 1966: The Textual Evidence for an Oral Family Saga. In: Arkiv fr nordisk
filologi. 1966. B. LXXXI. Pp. 123.
Andersson, Theodore M., 2000: Exoticism in Early Iceland. In: International Scandinavian and Me-
dieval Studies in Memory of Gerd Wolfgang Weber. Ed. by M. Dallapiazza & al. Trieste. Pp. 19
28.
Andersson, Theodore M., 2006: The Growth of the Medieval Icelandic Sagas. 11801280. Ithaca and
London.
Ari orgilsson. slendingabk. Edited and translated with an introductory essay and notes by Halldr
Hermannsson. 1979. Millwood, N.Y.

307
Bjarni Aalbjarnarson, 1937: Om de norske kongers sagaer. In: Skrifter utgitt av Det Norske Videns-
kaps-Akademi i Oslo, II. Historisk-filosofisk Klasse. 1936. No. 2. Oslo. Pp. 5768.
Boberg, Inger M., 1966: Motif-Index of Early Icelandic Literature. Copenhagen.
Drevnjaja Rus v svete zarubezhnyh istochnikov. Red. E.A. Melnikova. 1999. Moskva.
Jackson, Tatjana N., 1991: Islandskije korolevskije sagi kak istochnik po istorii Drevnej Rusi i jeje
sosedej. XXIII vv. In: Drevnejshije gosudarstva Vostochnoj Evropy. 19881989. Moskva.
Jackson, Tatjana N., 1993: Islandskije korolevskije sagi o Vostochnoj Evrope (s drevnejshih vremen
do 1000 g.). Teksty, perevod, kommentarij. Moskva.
Jesch, Judith, 2005: Geography and Travel. In: Old Norse-Icelandic Literature and Culture. Ed. by R.
McTurk. Oxford. Pp. 119135.
Manhire, W., 19741977: The Narrative Functions of Source-References in the Sagas of Icelanders.
In: Saga-Book. Vol. XIX. Pp. 170190.
McKinnell, John, 2003: Encounters with Vlur. In: Old Norse Myths, Literature and Society. Ed. M.
Clunies Ross. Pp. 110131.
Nazarenko, Aleksandr V., 2001: Drevnjaja Rus na mezhdunarodnyh putjah. Mezhdisziplinarnyje
ocherki kuljturnyh, torgovyh, politicheskih svjazej IXXII vekov. Moskva.
Nikuls Bergsson. Leiar-vsir. In.: Alfri slenzk: Islandsk encyclopdisk literatur. I. Cod. Mbr.
AM 194, 8vo. Udg. Kr. Klund. 1908. Kbenhavn. Bls. 1223.
ONP Ordbog over det norrne prosasprog. Registre. Kbenhavn, 1989.
Pashuto, Vladimir T., 1968: Vneshnjaja politika Drevnej Rusi. Moskva.
Rydzevskaja, Elena A., 1935: Legenda o knjaze Vladimire v sage ob Olafe Tryggvasone. In: Trudy
otdela drevnerusskoj literatury. T. II. Leningrad. Pp. 520.
Rydzevskaja, Elena A., 1945: Jaroslav Mudryj v drevnesevernoj literature. In: Kratkije soobschenija
instituta materialjnoj kuljtury. 11. Leningrad. Pp. 5165.
Saga lfs Tryggvasonar af Oddr Snorrason munk. Udg. af Finnur Jnsson. 1932. Kbenhavn.
Schaveljov, Aleksej S., 2007: Slavjanskije legendy o pervyh russkih knjazjah. Sravniteljno-
istoricheskoje issledovanije modelej vlasti u slavjan. Moskva.
Vsteinn lason, 1999: Dialogues with the Viking Age. Narration and Representation in the Sagas of
the Icelanders. Reykjavk.
Zilmer, Kristel, 2003: Representations of Intercultural Communications in the sagas of Icelanders. In:
Scandinavia and Christian Europe in the Middle Ages. Papers of The 12th International Saga
Conference. Bonn. Pp. 549556.
Zilmer, Kristel, 2006: Icelandic Sagas and the Narrative Tradition of Travelogue. In: The Fantastic in
Old Norse/Icelandic Literature. Sagas and the British Isles. Preprint Papers of the 13th International
saga Conference. Durham and York. Pp. 11051113.

308
 Saintly Exile: the commemoration of King lfr inn helgi in
the poetry of Heimskringla
Erin Goeres, Lincoln College, University of Oxford, England
Skaldic verse dominates Snorri Sturlusons saga about lfr Haraldsson in Heimskringla:
while recounting the kingship, exile, and sainthood of his protagonist, Snorri cites fifteen
named skalds and weaves a total of 178 skaldic stanzas into the prose text. This is one of the
reasons Carl Phelpstead has called the saga a mixed text in which verse and prose, history
and hagiography are brought into dialogue so as to present a realistically paradoxical portrayal
of lfr as holy Viking, beatissimus tirannus (Phelpstead 2007:118). Snorri emphasizes and
even enhances the conflicting aspects of lfrs life through a heteroglossia of different
voices throughout the narrative (PhelPstead 2007:127). Such conflicting voices can be heard
even within the small corpus of one skalds poetry in the saga: the lausavsur of Sigvatr
rarson jockey with stanzas from his named drpur, while political statements collide with
personal elegies as the prose text binding them in place creates only the illusion of chrono-
logical order in his work. Constantly reinvented as missionary king, saintly exile, and royal
martyr throughout the course of the saga, lfrs identity is unstable; his poet is likewise
forced to adapt his poetry and the role his poetry plays in the celebration and commemoration
of his patron.
The preservation of skaldic verse within later prose narratives has in large part been re-
sponsible for the way in which the skaldic stanzas were traditionally read, with far more criti-
cal attention paid to the so-called authenticating verses long, formal poems such as the
Erfidrpa lfs helga and the Kntsdrpa than to the situational verses, namely the
lausavsur.1 Diana Whaleys division of the verses into authenticating and situational is
based less on the content of the individual stanzas and more on their prose introductions
(Whaley 1993:245266). Whaley theorises that authenticating verses validate or elaborate
on the events of the prose narrative; situational verses tend to represent dialogue integrated
more fully into the text (Whaley 1993:251). However, such divisions suggest that there is
something inherently authenticating or situational about a verse, an interpretation which
ignores the stanzas composition before it was incorporated into the later saga. There is a gap
between the composition of a skaldic stanza and its preservation in a written saga, and these
gaps are too often ignored or glossed over. I would like to open up the gaps between prose
and poetry by examining the relationship between Sigvatrs verses and Snorri Sturlusons
Heimskringla, particularly as Sigvatrs expressions of grief and loss chronicle the exile and
death of lfr, and the subsequent return of his son Magns from his exile in the east to re-
claim the Norwegian throne.2
Chapters 181229 of lfs saga helga tell the story of lfrs first defeat in 1028, his exile
in Russia, and his final battle at Stiklarstair in 1030. The remaining twenty-one chapters re-

1
The Erfidrpa is Sigvatrs most popular poem in terms of critical attention (Phelpstead 2007, Rainford 1996,
and Naumann 1986). Sigvatrs Kntsdrpa has also prompted a significant amount of investigation (Townend
2001 and Jesch 2001). A recent article by Judith Jesch is particularly revealing of the scholarly bias in favour of
Sigvatrs political drpur: examining three lausavsur by Sigvatr which praise lfrs queen strir, Jesch at-
tempts to raise the profile of the verses by objecting to their status as lausavsur. Noting their similarities with
conventional praise-poems, she suggests that they may have been drawn from a longer, formal poem and are
thus, she implies, more worthy of attention (1994:118).
2
The question of Snorris authorship is too large to treat fully here. Sverre Bagge has discussed the exact extent
to which Snorri was the author or compiler of Heimskringla, as well as the many different voices in that debate.
He concludes that we can safely consider Snorri as the main authorial influence behind the text (Bagge 1991:23
25). I will refer to Snorri as the author of the prose text in this paper.

309
cord lfrs posthumous miracles and the Danish occupation of Norway. Snorri first cites
three stanzas from Sigvatrs Erfidrpa in his lfs saga helga after an account of the kings
defeat in 1028. The stanzas praise lfrs adherence to traditional forms of kingship and criti-
cise those who forced him into exile.3 The first of these is typical:

Goll buu opt eirs ollu

thlaupum gram kaupask
rautt, en rsir ntti,
rklunduum undan;
skr ba hann me hjrvi
(her land skal sv verja)
rns biu rekkar sna
refsing, firum efsa (4.18).

Often those who went on raids offered red gold to the high-spirited ruler to buy themselves off,
but the king refused. He bade the hair to be cut off from men with the sword. So should the land
be guarded here. The men got an obvious punishment for robbery.

The gnomic approval of so should the land be guarded casts the kings actions in the pre-
Christian tradition of the land-guardian, as depicted most famously by Einarr sklaglamm in
Vellekla:

Engi var jru

ttum gr, nema Fri,
gti-Njrr, ss geri,
geirbrkar, fri slkan (17.14).

Except Fri, there was never in the world a guarding-Njrr of the spear-board of a good fam-
ily who protected such peace.

Sigvatr echoes the earlier poem even more closely in the following stanza when he says, frir
bttisk sv [] fylkis lands [the peace of the kings land was thus restored] (5.34). Folke
Strm has demonstrated that the skalds eulogising Jarl Hkon stressed the earls adherence to
the pagan religion and its beneficial effect on the land (Strm 1981:44058). However, by the
eleventh century the motif of the land-guardian had been incorporated into praise-poems for
the early Christian kings, as in Hallfrer vandrasklds Erfidrpa lfs Tryggvasonar.4
Sigvatrs lausavsa 26 is not quoted in Heimskringla but it takes the idea even further, explic-
itly equating Danish rule with famine, in marked contrast to the prosperity of lfrs Chris-
tian reign. By recycling such an image in his praise of lfr, Sigvatr locates his patron within
the model of traditional Scandinavian kingship and thus shows the loss the country has sus-
tained by his departure.
Such skaldic parallels between the saintly lfr and the pagan Jarl Hkon are not, how-
ever, obviously conducive to the promotion of lfrs cult. In the surrounding prose narrative
Snorri betrays a great anxiety about these three stanzas and the image of kingship they pro-
mote: he identifies the kings actions three times as rtt [just] (Heimskringla 1945:32830).
This frames the land-guardian role promoted by the verses within a Christian interpretation
of just kingship and prevents the reader from interpreting lfrs rather violent actions as un-

3
These stanzas are also grouped together as stanzas 4, 5 and 6 of Sigvatrs Erfidrpa by the editors of the
Skaldic Project Homepage. All Old Norse citations in this paper are taken from the Skaldic Project. Translations
are my own.
4
See stanzas 12, 19, and 28.

310
kingly. The verses are neither authenticating (they do not support the description of an im-
portant historical event) nor situational (they are not integral to the plot). Rather, they pro-
vide Snorri with an opportunity to excuse his heros sudden reversal of fortune: he asserts that
the king vildi heldr lta af tgninni en af rttdminu [wished rather to give up his kingdom
than (his) just rule] (Heimskringla 1945:330). In this way Snorri reinterprets lfrs exile not
as a military defeat, but as a choice on the part of the king, a voluntary laying down of his
power in deference to Christian justice. Separated from the prose text, Sigvatrs verses evoke
a connection between lfr and traditional Scandinavian kingship models; framed by the text,
they are reinterpreted to explain his exile and defeat through quite a different lens.
After the loss of their king, lfrs followers are transformed from warriors into worship-
pers as lfr himself transforms from defeated ruler into royal saint. Sigvatr, the kings most
devoted follower, takes the lead in this process. His most famous possession is his gilded
sword, a gift from King lfr. The first verse Sigvatr speaks in the saga is a lausavsa that
encapsulates the traditional poet-patron relationship as the young poet thanks the king for his
gift:

Ek tk lystr n lastak
(leyf es) at san,
sknar Njrr, vi sveri
ss mnn vili nu;
ollr, fekt hskarl hollan
(hfum rit vel bir)
ltrs, en ek lnardrttin,
linns bla mr gan (3.18).

Eagerly I took your sword, Njrr of battle; I do not speak ill of that afterwards. Praise is my
occupation and my wish. Fir-tree of the blood of the serpents litter, you win a faithful follower
and I a good lord. We have both benefited.

It is not surprising therefore that swords and other such weapons figure prominently in the
kennings of the Erfidrpa and are key to the transformation lfrs followers undergo after
his death. The first stanza to describe the battle of Stiklarstair in Heimskringla shows lfrs
standard-bearer rr Flason leading the charge into battle (7.48). Like Sigvatrs sword, the
standard is fagrla gylta [beautifully gilded] (7.7); also like the sword, which cements the
kings relationship with the poet, the battle-standard draws the drtt together as saman
hjrtu [hearts together] they enter into the conflict (7.4). The king heroically advances first
behind this standard:

Mest frk merkjum nstan

mnn drttin framm snum,
stng fyr gram, gingu,
(gngr styrr vas ar) fyrri (12.14).

I heard my lord went the very nearest in front of his standard; the pole went forth before the
lord; there was enough tumult there.

In this stanza, the possessive mnn drttin further emphasises the collective nature of the
warrior band and in particular the poets close relationship with the king.
In contrast to the kings golden swords and standards, Sigvatr uses metaphors of unadorned
poles to describe the enemys forces:

311
 r vitu eigi meiar
gnar skers n hersa
(j r engils daua)
ann styrk bandmanna (20.14).

Beams of the skerry of terror did not know before then the strength of the farmers or of warri-
ors. The people planned the kings death.

The use of the word meir to describe the enemy echoes again the pre-Christian kennings of
Vellekla: in that poem, the skald praises Jarl Hkon as a rngvimeir gunnar lunda [op-
pressing-pole of the trees of battle] (24.14). Suggesting also the verb meia [to injure or
maim], Sigvatrs use of the word characterises lfrs enemies as pagan and deadly. Simi-
larly, the second helmingr of that stanza calls the enemy swords srelds viir [trees of
wound-fire] (20.6). The enemys poles of terror and wound-branches are very different from
the golden swords and standards of King lfrs army. Their swords are not expensive works
of art, but deadly instruments of destruction.
There is, however, a reversal in the pole and sword imagery in Snorris Magnss saga ins
ga when King lfr has become the venerated royal saint. The word meir, previously used
to describe the rebellious warriors, appears in a subsequent stanza to describe the pilgrims
coming to lfrs tomb:

Grts, eims gtt bar hjarta,

gollit skrn of mnum
(hrsak helgi rsis;
hann stti go) drtni;
r gengr margr fr mru
meir ess konungs leii
hreins me heilar sjnir
hrings, es blindr kom ingat (24.18).

A gold shrine is made for my lord, who carried a good heart. I praise the sanctity of the king; he
sought God. Many a pole of the ring who came there blind walks early from the bright kings il-
lustrious way with healed eyes.

Apart from stanza 12, cited above, this is the only other instance of the phrase mnn drttin
[my lord] in Sigvatrs Erfidrpa. In the previous instance, the king was seen advancing at the
head of his troops under a golden standard; here, the king lies in a golden shrine. Whereas
meir formerly described the hostile forces driving lfr out of his kingdom, here the same
word is part of a kenning for the pilgrims coming to visit the saints relics; it suggests the pil-
grims staff rather than the soldiers sword. Such a comparison implies that the same men
who fought against the king now pay their respects at his tomb, in keeping with Sigvatrs as-
sertion that irask n [] ess verks bendr [the farmers now repent of this deed] (11.24).
Such a transformation is likely a nod towards reconciliation under the reign of Magns and re-
focusses attention from the kings ignominious flight into exile to the pilgrims holy journey
towards his shrine.
Snorri, however, primarily quotes stanzas from the Erfidrpa not in the saga of King
Magns, during whose reign it was composed, but in the saga of King lfr to support his
own account of the kings final battle. At first Snorri seems to emphasise the authority of
skaldic stanzas in general and to justify his use of them in a historical text: King lfr deploys
his skalds in the field, saying,

312
 Skulu r [] hr vera ok sj au tendi, er hr gerask. Er yr eigi segjandi saga til, v at
r skulu fr segja ok yrkja um san (Heimskringla 1945:358).

You will be here and see the events of what happens. Then the news will not need to be told to
you, because you will tell and recite it afterwards.

Each skald immediately recites a verse and the warriors learn by them by heart (Heimskringla
1945:35860). However, Sigvatr is clearly a special exception to the rule: he has gone on pil-
grimage to Rome. When the skalds mock his absence, the king says,

Ekki arf Sigvatr at sneia, tt hann s eigi hr. Opt hefir hann mr vel fylgt. Hann mun n
bija fyrir oss, ok mun ess enn all mjk urfa (Heimskringla 1945:358).

It is not necessary to scorn Sigvatr, although he is not here. He has often followed me well. Now
he will pray for us, and it will be a very great need (for us).

This episode in the saga indicates a re-imagining of the role of the court poet and of how Sig-
vatr is most useful to a king who will soon become a saint. The three skalds who fulfil the
traditional role of remaining on the battlefield with the king also die with him; their poems are
not remembered. Sigvatrs new identity as the Christian pilgrim and his voluntary exile from
the battle ensure not only that his prayers may help the king, but also that he will survive.
Rather ironically, in light of the kings words to his three skalds, Snorri primarily uses Sig-
vatrs Erfidrpa to corroborate his account of the battle.5 The Erfidrpa then goes on to de-
scribe the first miracles that take place after lfrs death. It is clear therefore that as the Vi-
king warrior lfr transforms into a saint, his skalds role also changes: Sigvatr has become
not only a eulogist but also a hagiographer.
The Erfidrpa is not the only sequence of verses to commemorate the kings death. Sig-
vatrs absence from the battle also prompted him to compose a series of lausavsur mourning
the death of the king, which Snorri incorporated into the beginning of his Magnss saga ins
goa, the next saga in Heimskringla.6 Each of the verses is identical in structure, contrasting
past happiness with present sorrow. In the first helmingr of each stanza, Sigvatr describes a
circumstance that reminds him of his grief: he hears a man mourning the death of his wife
(20); ravens fly over the harbour (21); the kings warriors play war games (22). Verbs are
primarily in the present tense: mar missir [a man loses], sk [I see], geng ek fr [I turn
from] (20.1; 21.1; 22.1). The second helmingr of each stanza then shifts into the past tense
when Sigvatr meditates on the loss that has caused this grief. Words such as forum and
endr [formerly] emphasize the past (21.4; 21.7); the repetition of the verb minna [to remem-
ber] further highlights the poets separation from earlier and happier days (21.2; 22.5). This
contrast between past and present is a distinctive characteristic of Old English elegy, which,
as Joseph Harris has observed, consists predominantly of dramatic monologues in which a
human speaks in the first person about the past[]the contrast with the speakers present, a
contrast invested with sadness, is constant (Harris 1983:47). Harris observations on the simi-
larity of elegy in Old Norse is based largely on Eddic poems and on the longer erfidrpur, but
this opposition between past and present is certainly in evidence in Sigvatrs lausavsur as
well.

5
Heather ODonoghue observes that Snorris description of this battle is the most factual among the three prose
narratives lfs saga helga, the Legendary Saga of St. lfr, and Fstbrra saga that describe it, and that
Sigvatrs verses play a particularly important corroborating role in his account (ODonoghue 2005:7273).
6
These are lausavsur stanzas 18 and 2024 in the Skaldic Project. Chapters 910 continue with lausavsur 25,
26, and 2830.

313
 These stanzas also contain moving descriptions of Sigvatrs physical reaction to his grief.
Discussing the paucity of descriptions of emotion in the sagas, William Ian Miller notes that
physical indicators rather than emotion words are often used to signal feeling in a sort of so-
matic semiotics (Miller 1992:100). However, Sigvatrs poetic descriptions are more complex
than Millers saga examples: not only does Sigvatr describe his physical reaction to grief, he
metaphorically compares grieving with being wounded in battle. In lausavsa 20 the poet en-
counters a man weeping over his wifes death; he then compares the husbands grief to a war-
rior who has lost his lord and who cries vgtr [battle-tears] (20.8). This word could be both
a kenning for the blood a warrior would shed as he continued to fight a losing battle, and also
the tears that would be wept over the lords death as a result of that battle. In lausavsa 22, the
poet says, emk sem bazt, brjsti, / bleikr [Pale, I feel as though wounded in the heart]
(22.34). This metaphorical use of the verb binda in conjunction with a paleness that could be
the result either of grief or of the loss of blood further compares injuries suffered in war to the
bodys expression of grief. The poetic language thus blurs the distinction between emotional
injury and physical injury in Sigvatrs verses. Countering the accusations of desertion Sigvatr
faced for his absence from the battle, such metaphors re-imagine the poet as one of the warri-
ors who fought with the king. Poetic grief thus becomes a substitute for heroic death.
Snorri structures the prose text around these lausavsur in a way similar to the battle se-
quence surrounding Sigvatrs Erfidrpa. In this instance, the prose narrative describes not a
battle but a journey, locating the recitation of each lausavsa progressively closer to the poets
farm in Norway as he returns from Rome. Sigvatr speaks the first verse in Rome itself before
he moves on to an unnamed village; he then travels to Hillarsund, Kaupang and Trondheim
before arriving at his own farm, reciting a verse at each location.7 It is surely no coincidence
that when the prose text finally locates Sigvatr at home in Norway, the stanza Sigvatr recites
is one which extends his grief to his entire native land:

H tti mr hlja,
hll, of Nreg allan,
fyrr vask kenndr knrrum,
klif, mean leifr lifi.
N ykki mr miklu,
mitt str es sv, hlir,
jfurs hylli vark alla,
blari san (24.18).

When lfr lived, the high, sloping cliffs seemed to me to laugh through all of Norway. Earlier,
I was known on the ships. Now the mountainsides seem greatly unhappy to me, such is my ad-
versity since I lost all the lords favour.

These stanzas form a bridge in the narrative between the death of King lfr and the corona-
tion of his son Magns five years later. By including skaldic verse that mourns the death of
lfr rather than celebrating the Danish occupiers who briefly follow his rule, Snorri keeps
his focus firmly on the succession of the Norwegian royal house. In this case, sorrow func-
tions as a refusal to eulogize: the narrative emphasis shifts to the poet and then via the poets
grief to the gap created by the kings absence, but never to the king who killed him.8
7
Lausavsur 18 and 2023 respectively.
8
Snorri also uses poetic lament in lfs saga helga when describing lfrs victory over Jarl Svein. Corroborat-
ing his account of the battle with Sigvatrs Nesjavsur, the only stanzas Snorri cites to represent the enemys
experience come from Bersi Skldtorfusons flokkr about lfr helgi in which the skald laments the fall of his
patron (Heimskringla 1945:6567). In this episode, the focus of the verse is similarly on the poets grief rather
than on the enemy kings prowess.

314
 Despite the similarity of his lausavsur to the Old English elegies, Sigvatr differs from the
Anglo-Saxon poets in one very important respect: he does not present the kings death as the
passing of an age that can never come again. Sigvatrs Bersglisvsur stands in marked con-
trast to The Seafarers celebrated assertion that:

nearon nu cynigas ne caseras

ne goldgiefan swylce iu wron
onne hi mst mid him mra gefremedon
ond on dryhtlicestum dome lifdon.
Gedroren is eos dugu eal, dreamas sind gewitene,
Wunia a wacran ond as woruld healda (8387).

There are not now kings nor emperors nor gold-givers as there once were, those who performed
among themselves the greatest of glorious deeds and lived in the most noble renown. All that
company has perished; joys are departed; weaker people remain and inhabit the world.

Rather, in Magnss saga ins ga, Snorri uses Sigvatrs Bersglisvsur to show how the
kings son gradually learns to step into his fathers shoes, re-establishing not only the heredi-
tary line of kingship, but also the poet-patron relationship that was destroyed by lfrs death.
The nine stanzas of Bersglisvsur incorporated in Heimskringla are almost more interesting
for what they leave out than for what they include, suggesting deliberate choice on the part of
the saga-author. The reconstructed poem in the Skaldic Project database contains a number of
stanzas that, like the lausavsur discussed above, express the theme of past happiness and pre-
sent sorrow.9 However, Snorri incorporated none of those stanzas into Heimskringla. Instead,
the stanzas he presents from Bersglisvsur are those which echo the models of kingship pro-
moted earlier by the Erfidrpa.
The first stanza of Bersglisvsur reiterates the connection between the kingship and the
well-being of the land when Sigvatr asks, tyst / hv lengi skal / hans grund? [How long shall
his land be sad?] (1.68). In the Erfidrpa lfr maimed thieves and robbers (5.18); in
Bersglisvsur his son Magns is similarly a veltir jfs [toppler of the thief] (10.13). Sig-
vatr locates this model of kingship within a long line of Norwegian rulers when he reminds
Magns of the successful reigns of his predecessors: underlining the importance of father-son
relationships, Sigvatr calls Hkon gi fstr Aalsteins [elstans foster-son] (5.58),
while the two lfrs are Haralds arfi and sonr Tryggva [Haralds heir and Tryggvis son]
(5.58; 6.56). In the final stanza quoted in Heimskringla, Magns has stepped fully into his
fathers shoes when Sigvatr calls him, as he once called his father, mnn drttin [my lord]
(14.2). In this phrase, the poet-patron relationship that formerly applied to lfr and Sigvatr
has been re-applied to the skald and Magns. This is Sigvatrs last verse in Heimskringla and
it marks the beginning of a new reign as the prose comments,

Eptir essa minning skipaisk konungr vel[]Magns konungr gerisk vinsll ok stsll llu
landsflki. Var hann fyrir sk kallar Magns inn goi (Heimskringla 1951:31).

After this admonition, the king changed for the good. King Magns became popular and be-
loved by all the people of the land. Because of that reason he was then called Magns the good.

9
See for example stanza 16: goll bark jafnt of allan / aldr ok herverk sjaldan / hrygg hvrri / tveggja hendi
flotna sendis [Always through the course of his life I, seldom sad, wore gold on both of my two hands, the
plunder of the sailors messenger] (16.58). This stanza, along with many of the others, is preserved in Flatey-
jarbk.

315
Having established Magns as a law-abiding king, Snorri no longer uses Sigvatrs verses and
the poet disappears from the pages of Heimskringla.
Sigvatrs verses have thus been used to trace the demise of one ruler and the re-
establishment of another, law-abiding king in Snorris narrative. His Erfidrpa chronicles the
exile and death lfr; his expressions of grief in the subsequent lausavsur during his own
temporary exile allow the saga-author to concentrate on the gap left by the downfall of lfr
rather than on the rule of the Danish occupier. This gap can then be filled by lfrs son when
he has learned how to emulate his fathers method of kingship, allowing Sigvatr, in Bersglis-
vsur, to transfer the poet-patron relationship successfully to a new king of Norway. In
marked contrast to the mournful lausavsur, there are no expressions of the poets grief in this
sequence of verses as they are recorded in Heimskringla. In Snorris narrative, the grief ex-
pressed by Sigvatr merely acts as an interim state between the downfall of one good king and
the installation of another; such expressions are not used when a model king sits on the
throne. Skaldic grief has thus become a narrative device through which to discuss a change in
rulers and to manoeuvre between the old state of affairs and the new. Sigvatrs function in the
saga-narrative is clearly as mutable as his own role as a poet; chronicling the lives of kings,
saints and exiles, Sigvatrs poetry is as mixed as Snorris later text.

Bibliography
Bagge, Sverre. 1991. Society and Politics in Snorri Sturlusons Heimskringla. Berkley: University of
California Press.
Harris, Joseph. 1983. Elegy in Old English and Old Norse: A Problem in Literary History. The Old
English Elegies. New Essays in Criticism and Research. Ed. M. Green. Mississauga: Associated
University Press. Pp. 4656.
Heimskringla. 1945. Vol. 2. Ed. Bjarni Aalbjarnarson. slenzk fornrit 27. Reykjavk: Hi slenzka
fornritaflag.
Heimskringla. 1951. Vol. 3. Ed. Bjarni Aalbjarnarson. slenzk fornrit 28. Reykjavk: Hi slenzka
fornritaflag.
Jesch, Judith. 2001. Skaldic Verse in Scandinavian England. Vikings and the Danelaw: Select Papers
from the Proceedings of the Thirteenth Viking Congress, Nottingham and York, 2130 August
1997. Ed. J. Graham-Campbell et al. Oxford: Oxbow. Pp. 313325.
. 1994. In praise of strr lfsdttir. Saga-Book of the Viking Society 24:1. Pp. 118.
. 1992. Skaldic Verse and Viking Semantics. Viking Revaluations. Society Centenary Sympo-
sium 1415 May 1992. Eds. A. Faulkes and R. Perkins. London: Viking Society for Northern Re-
search. Pp. 16071.
Miller, William Ian. 1992. Emotion and the Sagas. From Sagas to Society: Comparative Approaches
to Early Iceland. Ed. G. Plsson. Enfield Lock: Hisarlik Press. Pp. 89110.
Naumann, Hans-Peter. 1986. Nordische Kreuzzugsdichtung. Festschrift fr Oskar Bandle. Ed. H.
Naumann et al. Beitrge zur nordischen Philologie 15. Basel: Helbing und Lichtenhahn. Pp. 175
189.
ODonoghue, Heather. 2005. Skaldic Verse and the Poetics of Saga Narrative. Oxford: Oxford Uni-
versity Press.
Phelpstead, Carl. 2007. Holy Vikings: Saints Lives in the Old Icelandic Kings Sagas. Medieval and
Renaissance Texts and Studies 340. Tempe, Arizona: Arizona Center for Medieval and Renais-
sance Studies.
Rainford, Jessica. 1996. lfr Haraldsson, king and saint of Norway, and the development of skaldic
style (ca. 1015ca. 1153). Diss. University of Oxford.
Skaldic Project Homepage. 2007. Margaret Clunies Ross et al. Skaldic Project Academic Body. 26
November 2007. <http://www.skaldic.arts.usyd.edu.au/db.php>.
Strm, Folke. Poetry as an instrument of propaganda. Jarl Hkon and his poets. Speculum Norroe-
num. Norse studies in Memory of Gabriel Turville-Petre. Ed. Ursula Dronke. Odense: Odense Uni-
versity Press. Pp. 440458.
The Seafarer. 2004. The Cambridge Old English Reader. Ed. R. Marsden. Cambridge: Cambridge
University Press. Pp. 221231.

316
Townend, Matthew. 2001. Contextualizing the Kntsdrpur: Skaldic Praise-poetry at the Court of
Cnut. Anglo-Saxon England 30. Pp. 145179.
Whaley, Diana. 1993. Skalds and Situational Verses in Heimskringla. Snorri Sturluson: Kolloquium
anllich der 750. Wiederkehr seines Todestages. Ed. Alois Wolf. ScriptOralia 51. Tbingen: Gun-
ter Narr. Pp. 245266.

317
 Ekki ntr slar
Nr himlen frgades rd av gudarnas blod
Bo Grslund, Dept. of Archaeology and Ancient History, Uppsala University, Sweden

Bakgrund
Jag har tidigare argumenterat fr att de norrna myterna om Fimbulvintern och Ragnark yt-
terst tergr p erfarenheter av en lngvarig solskymning ren 536537 e Kr orsakad av ett
vulkanutbrott eller mjligen en asteroid eller meteorit. Kraftiga utfllningar av sulfat i isarna
p Antarktis och Grnland bekrftar att en mycket kraftig eruption av ngot slag omkring r
536 (Grslund 2008).
Uppgifterna i Gylfaginning att solen var utan verkan tv r i fljd r s precisa och mlan-
de att de mer ger intryck av faktisk beskrivning n ngot sgenartat. Just detta tog jag som en
maning lyfta fram de uppgifter hos senantika frfattare som Prokopios, Cassiodorus, Zacha-
riah frn Mytilene, Johannes Lydos, Johannes av Efesos/ Mikael Syriern som talar om att
solen i Medelhavsomrdet var permanent skymd i nra tv r, kanske frn mars 536 till sen-
ret 537. Jag noterade ocks att en mycket likartad tradition terberttas i Kalevala.
Den indonesiska vulkanen Tamboras utbrott 1815 brukar anfras som det vrsta under de
senaste rtusendena. Sulfatiska gaser som d slungades upp i stratosfren hindrade solvrmen
att n jorden med en mycket kall sommar som fljd ret drp, 1816, ret utan sommar,
med missvxt och svlt i Europa och Nordamerika. nd var solen inte synbart skymd en
enda dag. Kontrasten r enorm till 536 rs utbrott, d solen knappt syntes p nra tv r och
som enligt studier av fossila trdringar fljdes av iskalla somrar p norra halvklotet under hela
tio r, 536545. I norra Sverige sjnk medeltemperaturen under sommaren 34 grader, en
vldsam klimatfrsmring.
De klimatiska fljderna av 536 rs utbrott uppfyller alla krav p en nukler vinter, det
slag av lngvarig global nedkylning som antas flja av ett allmnt krnvapenkrig, med kallare
och fuktigare vder under alla rstider. Men helt avgrande r de kallare och fuktigare som-
rarna med begrnsad vxtssong. Frstrd betesvxt och skrd tv r i fljd ledde i ldre tid
ofrnkomligen till svr nd. Om myten om Fimbulvintern ytterst tergr p konsekvenserna
av vulkanutbrottet 536 med full effekt i tio hela r, var det en katastrof av rent gammaltesta-
mentliga dimensioner.
Vidare framhll jag att stora, tidigare ppna arealer i Skandinavien och Mellaneuropa un-
der mitten och andra hlften av 500-talet vxte igen med skog, ett tydligt uttryck fr en be-
folkningskatastrof med effekter ver flera generationer. Hrtill kommer den av klimatkata-
strofen troligen utlsta Justinianska pestepandemin, som kraftigt minskade befolkningen i
tminstone sdra och mellersta Europa.
Till bilden hr ocks att majoriteten av jrnlderns mnga byar i stra Mellansverige un-
der 500-talet vergavs fr att i reducerad storlek flytta till litet hgre och torrare mark i nrhe-
ten, en bebyggelseomlggning av enorma dimensioner. Likas tycks s gott som all fast bo-
sttning norr om mellersta Hlsingland ha slagits ut i ett par generationer. Andra klimatiska
marginalomrden, som delar av sydsvenska hglandet, kom sedan att lnge ligga fr ffot, i
vissa fall till vr tid. Till detta kommer fr Sveriges del frre bosttningsspr, gravar och forn-
saksfynd. (Fr referenser till denna bakgrundsteckning, se Grslund 2008.)
I det fljande terkommer jag till frgan om myterna om Fimbulvintern och Ragnark yt-
terst tergr p denna klimatkatastrof och kan d ocks visa att likartade mytiska teman upp-
trder ver stora delar av vrlden, lngt bortom indoeuropeisk tradition. Enligt min mening

318
strker det i hg grad hypotesen. Jag frsker hr ocks p ett litet tydligare stt frklara
vergngen frn reell upplevelse till mytisk tradition.

Gylfaginning
I Gylfaginning i Snorres Edda omtalas fimbulvetr, med ungefrlig betydelse den stora, den
mktiga vintern. I tre av huvudhandskrifterna sgs Fimbulvintern komma som ett frebud
.
om Ragnark och i alla fyra att den frebdas av krig och mnsklig ondska Beskrivningen i
Gylfaginning (kap. 51) r densamma i alla huvudhandskrifterna, medan de inledande fraserna
r mer kortfattade i Codex Upsaliensis (Dillmann 1991: 187; Hultgrd 2004: 51 ff.).
Under namnet Gangleri trder kung Gylfe in i en hall hos asarna och stller frgor till tre
mn som sitter dr. Texten terges efter Codex Regius (Holtsmark & Helgason 1950: 68 f.):

mlti Gangleri: Hver tindi eru at segia fr um ragnarkr? ess hefi ek eigi fyrr heyrt ge-
tit.

Hr segir: Mikil tindi eru aan at segia ok mrg. au en fyrstu at vetr s kemr er kallar
Fimbulvetr; drfr snr or llum ttum, frost eru mikil ok vindar hvassir. Ekki ntr slar.
eir vetr fara .iii. saman ok ekki sumar milli.

D sade Gangleri: Vad finns att bertta om Ragnark? Jag har inte hrt ngot berttas frut.

Hr sger: Mnga och stora hndelser r att bertta om det. Det frsta r att den vinter som
kallas Fimbulvintern kommer. D yr sn frn alla hll, d blir det mycket kallt och vassa vindar.
Solen har ingen verkan. Tre vintrar kommer i fljd och ingen sommar emellan. (Johansson
1997 med korrigering av frf., Grslund 2008, not 10).

Beskrivningen av Fimbulvintern inskrnks till de tre sista meningarna, drfr snr or llum
ttum, frost eru mikil ok vindar hvassir. Ekki ntr slar. eir vetr fara .iii. saman ok ekki
sumar milli. Av dessa framstr Ekki ntr sla helt klart som en frklaring till upplysningen
om sn, kyla och vassa vindar. S har ocks Finnur Jnsson (1902: 71 f.) och Bjrn Collinder
(1978: 81) uppfattat det nr de i sina versttningar lter denna mening bilda bisats till den
fregende. Men att solen inte r srskilt framtrdande om vintern och inte skiner nr snn yr
frn alla hll r ju inte frvnande, det vet varje nordbo sedan barnsben. Och d frasen Ekki
ntr slar samtidigt frklarar varfr det under tv r inte blir ngon sommar, mste slutsatsen
bli att alla tre meningarna talar om samma sak, nmligen att det rder vintervder under som-
maren tv r i fljd p grund av att solen inte har ngon verkan.
Den ganska vanliga versttningen av Ekki ntr slar med solen skiner inte r en ngot
fr fri tolkning. Hr sgs endast att solen inte gr ngon nytta, inte har ngon verkan. Det
lmnar ett visst utrymme fr att solens sken kanske kunde sknjas svagt, tminstone sommar-
tid.

Vlusp
I Vlusp 4041 finns en notis om Fenrisulven och dess avkomma som rimligen ocks har
med Fimbulvintern att gra. Partiet terges hr efter Hultgrd 2004 som fljer Codex Regius i
enlighet med Helgason 1971; svensk tolkning Williams 2007):

Austr sat in aldna

Irnvii
ok fddi ar
Fenris kindir.

319
 Verr af eim llom
einna nkkorr
tungls tigari
trollz hami.

Fylliz firvi
feigra manna,
rr ragna sit
rauom dreyra;
svrt vera slskin
of sumor eptir,
ver ll vlynd
vito r enn, ea hvat?

Den gamla satt ster i Jrnskogen och

fdde dr Fenres slktingar. En enda av
dem alla blir i trollgestalt mnens rvare;

fyller sig med dda mnniskors lik,

rdfrgar makternas boning med rtt blod.
Solskenet var d svart somrarna drefter,
alla vindar onda. Fattar ni, va?

Hr redovisas allts att mnen frsvinner och att solens sken frmrkas efterfljande somrar
och allt vder blir oplitligt. Draget av saklighet blir inte mindre av uppgiften att himlen sam-
tidigt frgas dramatiskt rd, ngot som brukar ske just efter stora vulkanutbrott (se vidare
nedan).
Tidsordningen, att mne frsvinner innan solen svartnar, kan naturligtvis vara en tillfllig-
het. Men d det tar en viss tid innan stratosfren efter en kraftig vulkanisk eruption fylls upp
helt av sulfatiska aerosoler, r det naturligt att mnen med sitt svaga sken frsvinner ur mn-
niskors syn strax innan solen gr det.
Senantika kllor talar om att solen och dess sken, liksom mnen, blev mrka, svarta eller
blaktiga. Det stmmer vl med nrvaron av torra vulkaniska sulfatpartiklar i atmosfren. Det
ger ocks drag av verklighet t uppgiften i Vlusp att solskenet svartnar i samband med Rag-
nark liksom t frasen Ekki ntr slar i Gylfaginning, att solen inte har ngon verkan. Sam-
mantagna framstr dessa uppgifter i det annars s gtfulla Vlusp mer som en saklig rapport
av nyktra gonvittnen n som frvirrad mytisk verdrift.
Att Vlusp endast nmner rstiden sommar i samband med att solen svartnar ter sig ock-
s fullt rimligt eftersom fenomenet leder till katastrof bara om det drabbar sommaren.
Det faktum att Vlusp talar om somrar i pluralis framstr som ett viktigt komplement till
Gylfaginning, dr tiden fr solfrmrkelsen preciseras till tv somrar i fljd, precis som i
senantika kllor.

Hyndlulj
Uppgiften i Hyndlulj om snfall och skarpa vintrar som ett frebud om Ragnark framstr
ocks som en del av traditionen om Fimbulvetr (hr efter Hultgrd 2004 som fljer Helgason
1971):

Haf gengr hrum

vi himin silfan
lr lnd yfir,
en lopt bilar;
aan koma snvar

320
 ok snarir vindar;
er ri
at rgn um riti.

Havet slr med stormvgor mot sjlva himlen,

stiger ver landen och luften ger vika
drav kommer snfall och skarpa vindar
d r gudarnas undergng bestmd

Fr att en klimatsituation verkligen skall uppfattas som ett frebud om vrldens och gudarnas
undergng, mste den vara abnorm i den meningen att den strider kraftigt mot naturens vanli-
ga ordning. Denna vers kan drfr knappast syfta p kalla vintrar som sdana utan rimligen p
Fimbulvinterns tv vintriga somrar.

Vafrnisml
I Vafrnisml frgar Oden jtten Vafrnir (efter Hultgrd 2004 som fljer Helgason
1971):

hvat lifir manna

er inn mra lr
fimbulvetr me firom?

Vilka mnniskor lever kvar

nr den beryktade Fimbulvintern
lider mot sitt slut?

Fr vi tro Vafrnisml har det allts funnits en tradition som frknippat Fimbulvintern med
en befolkningskatastrof. Ja, tanken ligger implicit redan i frgan. Oden ber inte att f veta hur
mnga mnniskor som verlevde Fimbulvintern, bara vilka som gjorde det.
Vafrdnir svarar:

Lf ok Lfrasir
en au leynaz muno
holti Hoddmmis;
morgindggvar
au sr at mat hafa;
aan af alder alaz

Liv och Livtrasir, och de skall gmma sig i Hoddmimes lund; morgondagg har de att ta; frn
dem fds nya slkten

De nordiska namnen Liv och Livtrasir antyder att det rr sig om en hednisk tradition, en slags
terskapelseberttelse som knyter an till myten om Ask och Embla. Att bara tv individer sgs
verleva fr ses som en naturlig frenklingsmetafor fr en svr befolkningskatastrof.

Frn verklighet till myt

Ragnarksmytens slutscener dr sol, jord, gudar och mnniskor teruppstr i en paradislik-
nande milj ser jag som en metaforisk minnesbild av den lyckliga dag d Fimbulvintern ntli-
gen r slut, solen har kommit tillbaka och bttre tider stundar fr dem som verlevt.
Enligt Vafrnisml (1112) och Snorres Gylfaginning (9) dras solen under dagen ver
himlen av hsten Skinfaxe fr att sedan om natten fras tillbaka t ster av hsten Rimfaxe
som vid ankomsten lter sin fradga falla som morgondagg. Fradgan r nrande drfr att hs-
ten om natten betar p de underjordiska fltens grs som r bestrtt med honungsdagg, just

321
dr Mimer hller till vid Yggdrasils nedersta grenar. S nr man poetiskt skulle beskriva den
minnesvrda dag d Fimbulvinterns mrker gav vika fr den tervndande solen, kan den
livgivande morgondaggen i [Hodd]Mimers lund ha varit en fr alla begriplig metafor fr natu-
rens och mnniskoslktets pnyttfdelse. Scenen fregrips i tv efterfljande verser i
Vafrnisml (4647) dr solen, just innan hon slukas av ulven, hinner fda en dotter som
efterhand ska verta rollen som sol (Johansson 1997: 91):

hvaan kmr sl
inn sltta himin,
er essa hefir Fenrir farit?
Eina dttur
berr lfrull,
r hana Fenrir fari;
s skal ra,
er regin deyia,
mur brautir, mr.

hur kommer sol

p den slta himlen,
nr ulven denna sol hunnit upp?

En dotter fr Alfrdul
Innan Fenre frgr henne.
Den mn skall rida
moderns vg
nr makterna dr.

Talet i Volusp om att himlen frgas rd av gudarnas blod r helt i verensstmmelse med
hypotesen om ett vldsamt vulkanutbrott. De stora mngder svaveldioxid som vid kraftiga
utbrott slungas upp i de hgre luftlagren brukar p nordliga breddgrader under lng tid tfljas
av vldsamt rdflammande spektakulra kvllshimlar. Sdana syner blev omtyckta motiv fr
mnga konstnrer efter utbrotten av Tambora 1815 och Krakatoa 1883. Med tanke p att tra-
ditionen om Ragnark i frsta hand fr antas ha rtter i Norge, tar jag som exempel Edvard
Munch. I sin dagbok har denne vittnat om hur han hsten 1883 vid solnedgngen promenera-
de med tv vnner utanfr Kristiania. Det var strax efter Krakatoas utbrott:

s gik solen ned himmelen ble pludselig blodrd jeg stanset, lnet mig til gjrdet trt til
dden over den blsvarte fjord og by l blod och ildtunger mine vnner gik videre og jeg
stod igjen skjlvende av angst og jeg flte at det gik et stort uendleig skrig gjennem naturen.
(Eggum 2007)

Det drjde tio r innan Munch frskte befria sig frn sin kusliga upplevelse genom att mla
den. Frst kom Skriet 1893 och de fljande ren ytterligare mlningar med vldsamt blodrda
himlar, ven de med knslomssigt laddade namn som Frtvivlan, ngslan och Melankoli.
Hr terger jag n en gng de aktuella stroferna ur Vlusp i svensk tolkning:

Fyller sig med dda mannars lik,

rdfrgar makternas boningar med rtt blod.
Solskenet var d svart somrarna drefter,
alla vindar onda

D makternas boningar bara kan syfta p gudarnas himmel och d Fenre hr hemma i gudar-
nas vrld, kan det bara vara gudarnas blod och gudarnas lik saken hr gller. Sprkligt sett

322
tycks det ocks fullt mjligt att manna hr syftar p gudar och inte p mnniskor (Grslund
2008: 117, not 80). I versttningen ovan, hmtad frn Williams 2007, har jag drfr ersatt
ordet mnniskor fr det islndska manna med det mer neutrala mannar, som kan funge-
ra ven fr gudar.
ven i Snorres version av Volusp i Gylfaginning (vers 56) kopplas solens och stjrnornas
frdunkling till eld som flammar mot himlen. Hr i tolkning av Williams (2007):

Solen svartnar
Jorden sjunker i havet,
Klara stjrnor frsvinner frn himlen.
Eld rasar mot livgivaren [eld]
Lgan flammar hgt mot sjlva himlen

Munchs chockartade reaktion ger en levande bild av hur jrnlderns mnniskor i Norden kan
ha upplevt den blodrtt flammande himlen efter eruptionen r 536, fr vilken just vulkanen
Krakatoa brukar gras ansvarig. Munch och Vlusp talar samma knslosprk.
Skert fylldes jrnlderns mnniskor av fasa nr de, redan plgade av svlt, kvll efter
kvll, mnad efter mnad liksom p storbildsskrm sg himlen frgas rd av gudarnas blod i
deras kamp mot onda makter, som i en vldsam upptakt till vrldsalltets undergng. Men nr
nu trots allt vrlden inte gick under, kom minnet av fimbulvetr i stllet att lnge svva ver
nordborna som ett skrmmande apokalyptisk hot om en kommande undergng, Ragnark, ett
hot som efterhand blev allt blekare fr att till slut finnas kvar bara som rester av en myt.
Att man verhuvudtaget kom p tanken att de oddliga gudarna kunde d ser jag som ett
uttryck fr den bittra erfarenheten att gudarna tagit emot mnniskornas bner och dyrbara
offer fr att de skulle avvrja katastrofen men sedan visat sig ofrmgna att gra det. Gudarna
var lika maktlsa som mnniskorna. Jag hller det fr troligt, att makternas frtroendekapital
p jorden efter detta kom att till stor del vara frbrukat under flera generationer framver.
Lsta p traditionellt vis, r dikternas uppgifter om fimbulvetr och Ragnark precis s
kryptiska som man kan frvnta sig av myter. Men insatta i ett bredare skrifthistoriskt och
naturhistoriskt sammanhang ter de sig klara, logiska och utan sakliga motsgelser, precis som
r att vnta fr en tradition som ytterst tergr p en fr alla mnniskor gemensam faktisk
erfarenhet.
I Vlusp och Vafrnisml frklaras solens och mnens frsvinnande med att de slukas
av odjur. Mtt med dtidens naturvetenskapliga kunskapsmtt r det en hgst rimlig frklar-
ing.
Avstndet frn r 536 till nedtecknandet av myterna om Fimbulvintern och Ragnark kan
frefalla lngt. Men en stor del av den historiska ramen till den poetiska Eddans hjltedikt-
ning, som hnfr sig till 300400-talen, r ju nnu ldre trots att hndelserna bakom mste ha
haft en lngt mindre inneboende minneskraft n ngonsin den fruktansvrda Fimbulvintern.
Frleden *fimbul-, stor, mktig, skulle d vara ett frstrkande epitet fr en samman-
hngande vinterperiod p omkring tv och ett halvt r, en lngvinter som vergick all mnsk-
lig hgkomst. Nr man i efterhand kom att tala om fimbulvetr var det drfr skert inte i be-
mrkelsen den ovanligt svra vintern utan som den ondliga lngvintern, den som aldrig
tycktes vilja ta slut. P samma stt br d inte heller uttrycket inn mri fimbulvetr i
Vafrnisml tolkas som den bermda eller ryktbara storvintern utan snarare som den
beryktade eller omtalade lngvintern.
Ragnark har denna form i den ldre Eddan medan Snorres Edda har ragnarkkr. Som
Bernharsson noterar har det ldre rk traditionellt ansetts hra samman med FE racu f., l-
pande, flodbdd, berttelse och med FS raka f. och FHT rahha f., berttelse, tal
varfr Ragnark tolkats som gudarnas den. Den yngre skrivningen rkkr har d frklarats
som en folketymologisk frvxling med ordet rkr, mrker. Nyligen har dock Bernhars-

323
son visat att detta inte har std i det sprkliga materialet. Rk r inte en neutral a-stam utan en
neutral wa-stam och grundordet r substantivet FI rk(k), rek(k), mrker med drtill hran-
de starkt verb rk(k)va, rek(k)va, mrkna.
Ordet rk r semantiskt komplext och kan ocks betyda orsak, ursprung, eller de,
och Bernharsson antar att det r denna spnnvidd som kommer till uttryck i att substantivet
rk(k)r frutom skymning ocks kan betyda gryning. I valet mellan dessa alternativ tolkar
han Ragnark som gudarnas gryning gudarnas pnyttfdelse med hnvisning till att Rag-
nark avslutas med att gudarna tillsammans med solen, naturen och mnniskorna d terupp-
str. Men som Bernharsson noterar har verbet rk(k)va, rek(k)va ingen annan knd betydelse
n mrkna samtidigt som betydelsen gryning fr substantivet rk(k)r r belagd endast i
ett fall och d frn 1600-talet. (Bernharsson 2007: 34 f.) Med hnvisning till den lngvariga
solskymning som enligt Gylfaginning inleder Ragnark, till traditionen om att solskenet dr-
vid svartnar, till den permanenta solskymning som senantika kllor klart belagt fr ren 536
537 samt till det dunklare dagsljus som lnge brukar rda efter vldsamma eruptioner, frefal-
ler det drfr rimligare att ta fasta p den fr bde rk och rk(k)r sprkligt vl belagda
grundbetydelsen skymning.
Som Ragnark(k)r och fimbulvetr framtrder i Gylfaginning, stmmer allts den sprkliga
innebrden i det frra begreppet vl verens med den reella innebrden i det senare. Ragna-
rk br d i frsta hand tydas som gudarnas skymning, den frmrkelse som inte ens gu-
darna kan avvrja och som frebdar deras undergng. Jag hller det drfr fr troligt att be-
nmningen Ragnark terspeglar den stora solskymningen ren 536 och 537 e. Kr.

Liknande myter i vrlden

Om traditionen om Fimbulvintern och Ragnark ytterst tergr p erfarenheter av en lngva-
rig solfrmrkelse som drabbat hela jorden, borde liknande myter terfinnas utanfr det indo-
europeiska traditionsomrdet. Redan ett hastigt skummande av ytan av det ofantliga mytmate-
rialet visar ocks att s r fallet. Motivet att solen frsvinner under lng tid s att jorden fr-
snks i mrker och kyla och livet blir outhrdligt men sedan efter mycket besvr terfrs till
sin plats p himlen har nstan global spridning, bortsett frn Afrika sder om Sahara och Au-
stralien. Hr endast ngra exempel.
I Kalevalas 47:e sng enligt Lnnrots redigering, beger sig solen och mnen ner till jorden
fr att lyssna p Vinmoinens kantelespel. Men bda tas till fnga av Pohjagumman, som
stnger in solen i ett berg och mnen i en sten, 47:2129. versttning av Lars och Mats Hul-
dn:

Lnge var det natt i vrlden,

nedmrkt utan ljus fr jmnan ;
det var natt i Kalevala,
natt i Vinslktens stugor,
liksom ock i hga himlen,
dr i himlagubbens hemvist.

verguden Ukko frstr inte vad som hnt (47: 5158):

Ukko grubblar nu och grunnar
vad slags moln det r fr mnen,
vad slags dunst som dljer solen,
d man ej frmrker mnen,
solen inte lngre lyser.

Solen och mnen r frsvunna lnge, s nnu i 49:e sngen (110):

324
 Solen lyser alltjmt icke,
milda mnen skiner icke
ver Vinfolkets grdar,
Kalevalas kulna moar;
sden frs p alla krar,
krken kunde inte trivas,
det var hemskt fr himlens fglar,
sorgesamt fr alla mnniskor,
d man ej sg solen lysa,
mnens sken ej nnsin sknjdes.

Efter mnga om och men befrias till slut bde sol och mne och terfrs till himlen.
I Kalevala sgs allts uttryckligen att solen r skymd av dis eller tjocka av ngot slag och
att sol och mne r borta lnge. Grdan fryser, boskapen far illa och mnniskorna r djupt
oroade. Talande r notisen det var hemskt fr himlens fglar, en reaktion typisk fr
eklipstillstnd. Hr mter vi inte heller ngot tal om vinter, allt utspelar sig sommartid.
D det i andra delar av Kalevala beskrivs detaljer i drkt och vapen som pekar tillbaka
mot den yngre jrnldern, kan det inte vara omjligt att berttelser om s enastende och
skrmmande hndelser som ren utan sommar kunnat traderas i sngform i fyrtio genera-
tioner. Min hypotes r drfr att ven dessa partier i Kalevala r fragment av goda gonvitt-
nesskildringar av hndelserna r 536537.
Likartade frklaringsmyter som de i Eddan och i Kalevala r knda frn mnga andra hll
p norra halvklotet. Intressant r att det bara r i myter hos nordliga folk, som ju drabbas extra
hrt av klimatkatastrofer, som gudarna gr under tillsammans med mnniskorna som i Ragna-
rk. Sdana teman mter vi t ex hos Ainu p Hokaido och hos Kamtchuk, Koryak och andra
nordostsibiriska folk (Witzel 2005: 47).
S har vi den forniranska traditionen om vad som brukar uppfattas som speciellt kalla
vintrar (Hultgrd 2004; 2007). Men i Pahlavi-traditionen talas inte bara om kyla och sn utan
ocks om ett evigt regnande och haglande, ngot som mer hr samman med uselt sommarv-
der n smllkalla vintrar. Dr talas ocks om tre p varandra fljande katastrofala vintrar,
frdande fr mnniskor och djur. Men strnga vintrar leder sllan till katastrof, dremot is-
kalla somrar med frstrd vxtssong. Till samma mnster ansluter Avesta-mytens tal i
Vidvdd om en vinter som varar i tio mnader, vilket skulle innebra vintervder under strre
delen av den lnga iranska sommaren.
Vlknd r myten hos nordvstkustindianerna Haida om solen som lnge hlls gmd av
en gammal man medan jorden ligger i mrker. Till slut lyckas Korpen rva tillbaka solen och
terfra den till himlen. Varianter p samma tema mter vi hos bde Inuiter och nordliga
skogsindianer som Cree, Crow och Cherokee. (Reid & Bringhurst 1996; Witzel 2005: 49 f.)
Liknande myter frekommer ver stora delar av Eurasien, Japan, Sydostasien liksom i
Mellanamerika. Ofta omtalas de p ett missvisande stt som skapelsemyter, men vrlden med
gudar, mytiska eller ibland riktiga mnniskor finns redan. Myterna handlar i stllet om hur
solen rvas bort av onda makter och gms, eller gmmer sig sjlv, i en grotta, ask eller lik-
nande. Genomgende r att vrlden tidigare varit ljus men nu frsnks i mrker med katastro-
fala fljder fr allt levande. Det hela slutar alltid med att gudar eller mytiska mnniskor mste
befria solen och terfra den till himlen s att livet kan terg till det normala (Witzel 2005).
Nr det en enda gng sgs hur lnge solen r borta, Miao (Hmong) i sydstra Kina, fr vi
uppgiften tv r, precis i enlighet med senantika kllor, Gylfaginning och Volusp. Hos Yaba-
rana i Amazonas sgs att solen r borta i flera r (Witzel 2005, 47, 54).
I en myt frn Shao, centrala Taiwan, berttas att sol och mne stulits av tv drakar och att
vrlden d faller i mrker s att skrdarna frstrs och folket lider. Efter stora mdor lyckas

325
ett ungt mnniskopar dda drakarna och placerar ut sol och mne p himlen igen. Intressant r
uppgiften att man i samband med att sol och mne frsvann hrde tv fruktansvrda buller
och knde att marken skakade (http://www.taiwandc.org/folk-sun.htm). Eftersom bullret av
det ganska mttliga utbrottet p Krakatoa vid Java 1883 bde hrdes och kndes s lngt
norrut som Saigon och Bangkok och p Mauritius utanfr stra Afrika (Winchester 2003,
260, 262) mste den mycket kraftigare eruptionen r 536 ha hrts och knts tydligt p Taiwan
om den, som man ofta antar, gde rum just i Indonesien.
ven om dessa myter och mytritualer delvis kan ha spridits och pverkat varandra frn
omrde till omrde och ibland kan ha vissa drag av skapelsemyter, r det grundlggande temat
i allt vsentligt sdant att det kunde terg p faktiska upplevelser av ren 536537 d hela
vrlden under lng tid var bervad bde sol, mne och stjrnor.
Lt oss hr ocks nmna den grekiska vulkann Santorinis vldsamma utbrott ca r 1628
fre vr tiderkning. Trdringsserier visar att det nstan hade samma effekt som 536 rs ut-
brott, med en efterfljande kraftig global nedkylning under 46 r. Det r fullt mjligt att so-
len ocks d var skymd helt eller delvis en viss tid. Hur som helst mste man rkna med svra
fljder det nordiska samhllet vid brjan av bronsldern, det samhlle som under det fljande
rtusendet visade sdant ptagligt intresse fr solkult. verhuvudtaget finns det skl att beakta
detta utbrott nr det gller tidiga myter inom den indoeuropeiska sprkkretsen om solen som
frsvinner och sedan terfrs till sin plats. Eventuella rester av myter frn en sdan hndelse
kan mycket vl ha funnits kvar nnu vid mitten av frsta rtusendet e. Kr. fr att revitaliseras i
samband den nya solkatastrofen ett och ett halvt rtusende senare.

Referenser
Bernharsson, Haraldur 2007: Old Icelandic ragnark and ragnarkkr. In: Verba docenti. Studies in
historical and Indo-European linguistics presented to Jay H. Jasanoff by students, colleagues, and
friends. Beech Stave Press. Ed. A. J. Nussbaum. S. 2538.
Collinder, Bjrn, 1978: Snorres Edda. versttning och inledning. Stockholm
Dillmann, Franois-Xavier, 1991: L Edda. Rcits de mythologie nordique. Traduit du vieil islandais,
introduit et annot. Paris.
Eggum, Arne 2007: Edvard Munch som maler. Munchmuseet. http://www.
munch.museum.no/content.aspx?id=15
Faulkes, Anthony, 1982: Snorri Sturluson, Edda: Prologue and Gylfaginning. Oxford.
Grslund, Bo, 2008: Fimbulvintern, Ragnark och klimatkrisen 536537 e. Kr. In: Saga och Sed, Vol.
2007: 93123.
Helgason, Jn, 1971: Eddadigte 2. Gudedigte. 3. udg. Kbenhavn/Oslo/Stockholm.
Holtsmark, Anne & Helgason, Jn, 1950: Snorri Sturluson. Edda, Gylfaginning og prosafortellingene
av Skldskaparml. Kbenhavn/Oslo/Stockholm.
Hultgrd, Anders, 2004: Fimbulvintern ett mytmotiv och dess tolkning. In: Saga och Sed, 2003: 51
69.
Hultgrd, Anders, 2007: The mythic theme of the great winter in ancient Iranian traditions. In: Reli-
gious texts in Iranian languages. Symposium held in Copenhagen May 2002. Ed. By F. Vahman &
C. V. Pedersen. Det Kgl. Danske Videnskabernes Selskab. Kbenhavn. Pp. 101120.
http://www.taiwandc.org/folk-sun.htm,1996. The legend of the sun-moon lake. In: Folk stories of
Taiwan
Johansson, Karl, G., 1997: Snorres Edda. versttning frn islndskan och inledning. Stockholm
Jnsson, Finnur, 1902: Snorre Sturluson. Gylfaginning. Den gamle nordiske gudelre (frste del av
Snorres Edda). Oversat af Finnur Jnsson. Kbenhavn.
Kalevala. Finlands nationalepos. versttning av Lars och Mats Huldn. Stockholm 1999.
Reid, Bill & Bringhurst, Robert, 1996: The Raven steals the light. Vancouver.
Williams, Henrik, 2007: Vlusp i Uppsalaeddan. En nyfilologisk underskning. Manuskript. Uppsala
Winchester, Simon, 2003: Krakatoa. The day the world exploded: August 27, 1983. New York.
Witzel, Michael, 2005: Vala and Iwato. The myth of the hidden sun in India, Japan, and beyond. Elec-
tronic Journal of Vedic Studies (EJVS) 121, March 2005: 169.

326
 Recreating Tradition: Sigvatr rarsons Vkingarvsur and
ttarr svartis Hfulausn
Jonathan Grove, Dept. of Anglo-Saxon, Norse, & Celtic, University of Cambridge, England

Introduction
Unlike other oral poets, court skalds in the Viking age expected their work to be disseminated,
re-performed and perpetuated in a more or less stable form. Verse-composition was tightly
bound by conventional metrical, syntactic and stylistic constraints, but at the same time the
finished poem was conceived as a unique artefact that proclaimed the virtuosity of its maker.
The need to satisfy conflicting demands of conformity and individuation of expression consti-
tuted a fundamental creative tension for the poets. Traditional formulas and verse-conventions
were not employed mechanically. The craft of verse-making lay in varying established pat-
terns of kennings, epithets and rhyme-combinations, producing originality within narrow for-
mal limits. As Roberta Frank has put it, the skaldic poets competed with each other in recre-
ating tradition (2006:190), fixing the poetic art under their own names. This was not simply a
matter of aesthetics; the mastery of the skaldic rtt bore with it unusual potential for the
enrichment and social advancement of its practitioners.
Court poetry from as early as the 10th century evinces patterns of echoes and parallel us-
ages between colleagues and rivals and between successive generations of poets. It is not al-
ways clear, however, whether such echoes are deliberate, or what purpose they served if they
were. Despite the confidence of much scholarship that the texts of Viking-age poems are re-
coverable from their anachronistic scribal contexts, fragmentary preservation makes it virtu-
ally impossible to ascertain the original distinctiveness and intertextual resonances of any
piece, rendering the complexity of interactions between poets largely indiscernible. Further-
more, although twentieth-century scholars often inferred direct connections between poems
on the basis of the slightest lexical parallels (e.g. de Vries 1999 I:99281, passim), many of
these are probably the inevitable symptom of shared formal constraints. Nevertheless, the case
for direct indebtedness does seem reasonably secure where marked similarities occur between
poems in material that is thematically significant or otherwise particular to the works in ques-
tion, especially where it is possible to identify a context in which borrowing may readily be
conceived to have taken place (Fidjestl 1982:2402). Resemblances of this sort often seem
pointed, indicating that court poets engaged directly with the works of their contemporaries
and forebears, advertising their capacity to match or surpass them by referencing their verse
and developing new formulations from old (Frank 1978:278). The clearest examples occur in
poems composed for rival dynastic groups in Norway. In Eyvindr skldaspillirs textual ap-
propriations in his poems for Hkon gi and Hkon jarl (Fidjestl 1997), or Hallfrer van-
drasklds polemical citations in his erfidrpa for lfr Tryggvason (Von See 1977), the
poetics of outdoing comply with the agonistic function of verse-making designed to counter
the panegyrics of political adversaries. But similar correspondences are also commonplace in
works composed by poets who shared the same patrons, and it is worth considering to what
extent this reflects a more general recursive mode of composition in skaldic verse-making.1
I wish to focus here on a single example connecting poems composed in the first half of
the 10th century by Sigvatr rarson and his nephew ttarr svarti, the two foremost skalds
of lfr Haraldsson. The case is particularly intriguing because an account of their poetic
fraternization is actually registered in the medieval narrative sources. The sagas relate that

1
Note for instance the repetitive features in poetry connected with the court of Kntr inn rki (Frank 1994a).

327
Sigvatr helped his nephew compose and win a hearing for his Hfulausn (Hfl), and it seems
clear from the internal evidence of the poem that ttarr took as his principal model Sigvatrs
Vkingarvsur (Vkv), the earliest surviving poem composed in praise of lfr (Finnur Jnsson
192024 I:5767). Nevertheless, there are considerable differences between the two works.
These arise partly from their divergent performance contexts, since they were composed at
quite different stages in the reign of lfr; but the stylistic contrasts arguably also stem from
ttarrs individuation of a derivative work that appears to flaunt his extension of the poetic
effect of his model.2

The head-ransom narrative

The versions of the tale of Sigvatr and ttarr, principally represented in article 10 of the notes
on the life of lfr Haraldsson attributed to Styrmir Krason (H:6889, cf. 7023), and ch.
57 of the Legendary Saga (HLeg:130), refer to ttarrs arrival at the court of lfr in the
early 1020s, soon after the death of his patron lfr Eirksson of Sweden (1021/22).3 The
story goes that ttarr found himself clapped in irons for having once composed a mansngr
concerning strr, lfrs queen and the daughter of the late Swedish king. The two versions
of the tale agree that Sigvatr came to the help of his nephew, and literally saved his neck. t-
tarr recites the poem on strr to his uncle, and Sigvatr offers one of the only pieces of con-
structive professional advice and assistance from one poet to another recorded in saga litera-
ture: miok er kudit ort ok eigi er vndarligt ott konunginum misliki kudit. nu skulum vit
snua eim visum sem mest eru akuedin ord i kudinu (H:688; cf. 702).4 ttarr edits the of-
fending mansngr on his uncles instructions, and following his advice he also composes a
praise-poem during his three nights in captivity, to impress lfr and his court and act as a
head-ransom. Through Sigvatrs mediation, lfr gives ttarr a hearing. Having recited his
now considerably bowdlerized mansngr before the king and his retinue, causing consider-
able uproar, ttarr redeems himself with a suitably impressive display of praise by perform-
ing his Hfl. ttarr puts on such a good performance with his new drpa that he wins not only
his life, but also an arm-ring from the king, and another from strr herself. Duly placated,
the king rewards him further with a place in his retinue.
Entertaining as it is, this anecdote and indeed the title of the poem, which only occurs in
this context (H:689; cf. 703) cannot be regarded as anything other than a medieval inven-
tion (cf. Jesch 2006:253, 265). ttarrs Hfl itself contains only a general plea for patronage,
declaring in the first two stanzas that ttarr has lost his previous lord, Gauta dglingr the
prince of the Gtar (st. 1), and now seeks the favour of the Norwegian king and his retinue
(st. 2). Additionally, the whole narrative design is clearly related to the account of Arinbjrn
hersirs advisory role in the composition of Egill Skalla-Grmssons Hfulausn (Egils saga
chs 612), and to the wider family of head-ransom tales in which a wise ally supports the
skald as he awaits judgement for some misdemeanour, advising him to compose a praise-
poem in order to win his reprieve, and the poet duly meets the challenge of composing his
drpa in the limited time available, his performance inducing the ruler to show clemency
(Nordland 1956; Clunies Ross 2004). Yet even if Sigvatr was not directly involved in the
2
For Vkv, see Fell 1981. See Skj BI:26872 for ttarrs Hfl: more recent editions are followed where indi-
cated. Translations are my own. The 15 sts of Vkv appear in the Separate Saga of St Olfr and the Heimskringla
and Flateyjarbk versions of lfs saga; a few also occur in the Legendary Saga and Fagrskinna (sts 6, 8, 9). Of
the 20 sts and helmingar of Hfl, sts 319 appear in the Separate Saga, Heimskringla, and Flateyjarbk; a few
appear elsewhere: the Oldest Saga (st. 1); the Legendary Saga (sts 1, 19); Fagrskinna (sts 8, 19); Orkneyinga
saga (st. 19); Kntlinga saga (st. 8); Skldskaparml (sts 2, 19, 20); and the First Grammatical Treatise (st. 7).
On the preservation and reconstruction of these remains, see Fidjestl (1982:11718, 12324).
3
See also H:7, and passing references in H(Hkr) chs 70 and 114.
4
See also HLeg:132, Of miok er ort, oc gef ec er at ra, at snuum sumum rndum oc fllum or sum.

328
composition of ttarrs poem, it does seem likely that the story transmitted in the sagas re-
sponds to ttarrs conspicuous debt in the so-called Hfl, which lists lfrs triumphs before
the early 1020s, to his uncles poem, which had carried the tally of lfrs battles as far his
initial successes in Norway in 1015.

Vkingarvsur revisited
The battle-list is a well-used subgenre in skaldic court poetry (Fidjestl 1982:21316). De-
spite the generic form of both Vkv and Hfl, it is immediately evident however that the most
important model for ttarrs poem lay in Sigvatrs Vkv, for Hfl refers to the same sequence
of battles fought by lfr Haraldsson leading up to his victorious return to Norway (Moberg
1941:3349; EE:7682). Each stanza of Sigvatrs Vkv describes a single battle, the first thir-
teen of which are consecutively numbered in the text. The sequence follows lfrs progress
as he proved his credentials as a war-leader, first in the Baltic and Frisia, and England where
he participated in the devastating campaigns of orkell hvi in 100912 , and then in Brit-
tany, Spain and France, before he returned once more to England in 1014 this time as a
mercenary in the service of thelred unrd and then finally sailed home to carve out a
kingdom for himself in Norway. The fact that the stanzas now attributed to Vkv come to an
end with lfrs return to his homeland in 1015, before the pivotal victory at Nesjar in 1016,
suggest that Sigvatr delivered Vkv at an early stage, before lfr began to establish his su-
premacy (Fell 1981:107; Fidjestl 1982:21415).
Although he does not adopt the enumerative style used by Sigvatr, ttarr follows the same
basic structural pattern (Campbell 1971:1011; Fidjestl 1982:214). Accordingly, many stan-
zas in Hfl correspond directly with verses in Vkv: Hfl 56 records battles against Swedes
and Baltic Slavs listed in Vkv 12; Hfl 7 describes lfrs assault on London, recalling Vkv
6; st. 9, on the battle at Ringmere (Hringmaraheii) in East Anglia, corresponds with Sig-
vatrs st. 7; ttarrs st. 10 on the attack on Canterbury corresponds with his uncles st. 8; st.
11 on the expedition to Poitou (Peita) corresponds with Vkv 14; finally, Hfl 15 reports
lfrs encounter with Hkon Eirksson on his return to Norway, repeating the content of Vkv
15. Some victories listed in Vkv do not appear among the scattered stanzas associated with
ttarrs poem, namely the Baltic battles of Vkv 35, and those in Brittany and Spain tallied in
sts 1013. This may reflect the fragmentary preservation of ttarrs poem, or else a tendency
to abbreviate the account of lfrs early career in view of the need to introduce more mate-
rial on his assertion of power in Norway after 1015.
Vkv was apparently composed at an early stage in Sigvatrs career, at a time when lfrs
aspirations to royal authority had also barely begun to be realized. The poem deals in the cli-
chs of martial poetry, but without excessive praise: it comprises a sober list of the range of
experience of a young warlord who has shown his mettle in campaigns overseas. He has sated
the beasts of battle, and defeated many foes, and as the young descendant of kings (enn ungi
jfra kundr, st. 1), and a chieftains son (fylkis nir, st. 3), he has proven himself as a potential
lord of Norway. Composing perhaps seven or eight years later, ttarr looked back on lfrs
career from the high point of his reign. Hfl consequently follows lfrs trajectory from vi-
king warrior to aspiring ruler and eventual overlord in Norway, extending the chronological
sequence and laudatory tone in sts 1519 well beyond that of the surviving verses of Sigvatrs
poem.
Since both poems deal with the same historical material, a certain amount of repetition is
inevitable, such as ttarrs reuse of the place-names associated with lfr Haraldssons victo-
ries: Eyssla, Lundna bryggjur, Hringmaraheii, Kantaraborg, and Peita. There are other
more interesting echoes, however. In Hfl 7, commemorating lfrs participation in the at-
tack on London in 1009, ttarr employs a conventional mythological battle-kenning in a ref-

329
erence to lfr as la linns kennir, Yggs verorinn, tester of showers of the serpent [SWORD
> BATTLES > WARRIOR], daring in Yggrs weather [BATTLE] (for which reading see
H(Hkr):1617). A similar but characteristically plainer formulation occurs in verse in Vkv
(st. 6): engil bau Englum Yggs at, the prince offered Yggrs clash [BATTLE] to the Eng-
lish. The use of the Odinic name Yggr in such simple terms for battle is a skaldic common-
place, and the presence of these poeticisms in congruent expressions would appear accidental
were it not for the fact that they occur in stanzas referring to precisely the same historical
event. The indication that this represents a direct verbal echo might lead us in turn to prick up
our ears at what would otherwise seem like a forgettable piece of padding in Hfl 7: the paral-
lel here between ttarrs arrangement skildir / (gang) en gamlir sprungu, moreover ancient
shields shattered (st. 7), and (gang), ars gamlir sprungu / geirar, where ancient spears shat-
tered, employed by Sigvatr in a much later stanza in Vkv enumerating a battle on the Loire
(st. 14), now seems the more striking.5
One other repeated element in Hfl may provide stronger evidence of deliberate quotation.
ttarrs Hfl opens with lines that echo those purported to have been spoken by Sigvatr in his
Lausavsa 2, which appears to constitute a formal call for hearing connected with Vkv:

Hl mnum brag, meiir

myrkbls, vt kank yrkja[] (Sigvatr, Lausavsa 2/12; Skj BI:260)6
Hear my poem, destroyer of the black [horse of awnings][SAILS > SHIP > SEAMAN], for I know
how to compose verse[]

Hl, manngfugr, minni

myrkbls, vt kank yrkja. (ttarr, Hfl 1/12)7
Hear, noble lord, the commemoration of the dark one [ttarr], for I know how to compose
verse.

Fidjestl has argued that this vsa might be considered a loosely attached introduction to Vkv
(1982:124, 2145). According to the saga accounts, Sigvatr arrived at Niarss in 1015, and
was introduced by his father rr Sigvaldaskld to the aspiring king on his arrival in the
north soon thereafter. Sigvatr had come to perform his praise-poem before lfr, but the king
had stated that he did not want to have panegyrics composed about him and claimed that he
did not even understand skaldic poetry (H ch. 39; H(Hkr) ch. 43). Undaunted, Sigvatr pre-
sents himself before lfr, and speaks what is presented as an improvised lausavsa designed
to win a hearing for his longer poem. The Icelander correctly assumes that his display will
suffice to elicit a change of heart in the suspicious young ruler, allowing the poet to make a
full demonstration of his skill and thereby secure the kings patronage. Finnur Jnsson inter-
prets the echo of Sigvatrs lausavsa in Hfl as the result of textual contamination arguing that
ttarrs second line is uden tvivl indkommen her fra Sigvats Lv. 22; den oprl. linje er sledes
tabt (Skj AI:290). This is certainly possible, although not as engaging as Jan de Vriess more
credulous interpretation, according to which ttarr, caught in a dangerous situation, deliber-
ately referred back to the first tense but ultimately mutually propitious meeting between lfr
and his most beloved retainer, ttarrs own uncle (1999 I:239). The exact circumstances of
5
In each case the repeated rhyming element gang belongs to a separate clause.
6
Myrkbls forms part of a ship-kenning, myrkblr tjalda drasill, constituting the determinant of a warrior-
kenning based on meiir.
7
Finnur finds the opening of Hfl too corrupt to reconstruct, but suggests minni is the possessive my, as part of
an original statement Hear my drpa. Kock (NN 695, 721, 791) interprets manngfugr as a substantive, and
minni as memory, commemorative poem producing an echo of the form as it occurs in Vkv 1 and ingen-
iously reads myrkbls as a play on ttarrs cognomen, svarti the black.

330
composition are unknowable of course, but the suggestion of Fidjestl and de Vries that t-
tarr alluded to his uncles work is supported by a further possible correspondence, for the col-
location of minni and kank in ttarrs opening lines also recalls Sigvatrs kann ek til margs
enn manna / minni, I still know the memory of men about many things in the first stanza of
Vkv proper (st. 1/56).
It is conceivable that the structural and verbal parallels to Sigvatrs poem in Hfl repre-
sented a clever attempt by a younger poet to elicit the kings mercy by allusion to his favoured
uncles maiden performance, as de Vries suspected; but an alternative interpretation is also
possible. In referring directly to Sigvatr, it might simply be argued that ttarr sought to sur-
pass him at his own game, and thus stake his own claim on the kings favour. There are
enough similarities in content, form and expression to indicate the two poems are related, but
it is also clear that ttarr produced a far bolder political statement than Sigvatr. By alluding to
the form of Sigvatrs poem, ttarr leant on the precedent of a successful earlier work, but in
so doing he made his poem represent the perfection of the heroic career described by his un-
cle. In his account of lfrs early battles in eastern Scandinavia and England, ttarr follows
Vkv, but this produces a pattern of direct allusion signalling not only the culmination of
lfrs early aspirations but also the younger poets outdoing of his uncles more cautious
work. Although the head-ransom tale appears to reflect an assumption in later tradition that
Sigvatr was the superior poet, the evidence of the poetry may be witness to a clever profes-
sional strategy on the part of his nephew, bent on securing his social advancement, if not ac-
tually saving his neck.
The kinship between the two poems is represented in the fact that their stanzas are cited
alongside one another in the accounts of the kings early career in the konungasgur (H chs
215, H(Hkr) chs 1315, 19; cf. Fsk ch. 27). This reflects the fact that they describe many
of the same events, but it is also a reminder that both poems were preserved as competing
witnesses in the memorial tradition. In ch. 13 of lfs saga in Heimskringla, for instance,
describing the assault on London, Vkv 6 and Hfl 7 are cited side-by-side, the juxtaposition
inevitably drawing attention to the verbal similarities in the references to Lunduna bryggjur
and the poets mythological allusions to Yggr. The verses on Ringmere in ch. 14 duplicate
narrative content if not actual lexis, but they illustrate the characteristic differences in style
between the two poems, and the means by which ttarr sets out to surpass his model: while
Sigvatr does not record the outcome of the battle (Vkv 7), ttarr relates that lfr piled up a
heap of English corpses and put many to flight (Hfl 9). Sigvatrs stanza constitutes a com-
paratively restrained entry in lfrs curriculum vitae:

Enn lt sjaunda sinni

svering hit vera
endr Ulfkels landi
leifr, sem ek fer mli.
St Hringmaraheii
herfall vas ar alla
Ellu kind, es olli
arfvrr Haralds starfi. (Vkv 7)
Yet a seventh time, as I hear tell, lfr had a sword-thing held in Ulfkells land [East Anglia],
long ago. Ellas offspring [ENGLISHMEN] swarmed all over Ringmere a slaughter of armies
took place there when the heir of Haraldr [lfr] brought about battle.

ttarrs account of the same event establishes a new tone of vivid hyperbole:

engill, frk, at unga

331
 inn herr skipum ferri,
rau Hringmaraheii,
hl valkstu, bli.
Laut fyr yr, r ltti,
landfolk gn randa,
Engla fer, at jru
tt, en mrg fltta. (Hfl 9; H(Hkr):19)
Lord, I heard that your army piled up a heavy heap of the slain far from the ships, reddened
Ringmere with blood. The landsmen, the fyrd of the English, bowed before you, fell to the
ground in the din of rims [SHIELDS > BATTLE] before it diminished, and many took to flight.

The difference in tone and density of imagery is striking (Finnur Jnsson 19204 I:5767;
Ashdown 1930:115). Sigvatr asserts the young lfrs familial credentials as heir of Haraldr
and a veteran of foreign campaigns but ttarr addresses him directly in the second person,
as a ruler whose authority is now undisputed and whose youthful expoits have become the
stuff of skaldic clich. Sigvatr uses direct apostrophe twice in the surviving stanzas of his
poem (sts 5, 11), but for ttarr it becomes a central feature of his work, which not only ex-
tends the sense of the poem developed in the opening stanzas as a plea for patronage, but also
defines more closely the idealized condition of mutual reciprocity between poet and king
(Frank 1994b; Jesch 2006:25861, 2645).
ttarr also enhances the aural effect of his verse by adopting a series of metrical flourishes,
producing more conspicuously ornate effects than appear in Vkv. While ttarr frequently
adopts aalhendingar in the first or third lines of a helmingr instead of skothending (sts 5/1;
7/1, 3; 10/7; 15/5; 17/3, 7; 18/5; 19/1), no such patterning occurs in Vkv. Elsewhere, he uses
the echoic rhyme of dunhenda, in which the last stressed syllable of the first or third line is
repeated as the first stressed syllable of the second or fourth line (sts 1/78; 2/34; 3/78; 4/3
4; 10/78; 15/78); significantly, perhaps, such patterns do also occur in Vkv, but only in two
instances (sts 7/67 and 8/78).
Composing with the advantage of hindsight, ttarr was at liberty to recast the history of
lfrs early career in a more flattering teleological mode. Whereas Sigvatrs verse on the
unsuccessful viking assault on London in 1009 (Vkv 6) describes an inconclusive attack on
London Bridge by men under lfrs command, Hfl 7 claims, rightly or wrongly, that lfr
broke the bridge and quite unhistorically won land (hefr lnd at vinna [] r snnat, it
has turned out for you that you won lands). A key concern in ttarrs poem is to demonstrate
lfrs authority as overlord of Norway by right of conquest. While Sigvatr sets out the im-
pressive range of lfrs experience as a warlord, ttarr constructs images of him as one who
takes, possesses and confers power over land. The claim that lfr won land in England
after the battle of London in 1009 is matched in Hfl 8, which refers to lfrs support of
thelred II in 1014:8 Komt land ok lendir, / lvrr, Aalri, You came into the land,
fief-guardian, and granted thelred land.9 The English king had gone into exile following
Sveinn tjguskeggs conquest of 1013, but when Sveinn died early in 1014, lfr was em-
ployed as a mercenary to fight the Danes during thelreds short-lived restoration. lfrs
efforts to help thelred assert his rightful claims to sovereignty in England are thus presented

8
Hfl 8 is quoted anachronistically in the context of lfrs first battles in England in 1009 in Fsk ch. 27, H ch.
23, and H(Hkr) ch. 13, serving the conceit of the saga accounts which, with the exception of the Legendary
Saga, make lfr the ally of thelred from the time of his first arrival in England. On the connection of Hfl 8
with the wars of 1014 see Moberg 1941:479; EE:789; Campbell 1971:1112; Poole 1987:2745. The st.
should thus be placed just before the account of lfrs voyage to Norway in sts 1314.
9
On the interpretation of l here as fief, cf. CVC:376 s.v.; O:540, s.v. 1. See also Rainford 1996:71.

332
as antecedents to the establishment of lfrs own dominion in Norway, now framed as the
legitimate assertion of power over allodial territories (ttlnd, st. 15)(Rainford 1996:467).
It at this point that ttarrs narrative as we know it extends beyond the historical limits of
the surviving stanzas of Vkv. The transition is marked by a change in tense, from the preterite
verbs employed in sts 315 in the account of lfrs early campaigns also described by Sig-
vatr (Ungr hratt [], When young you launched [], st. 3; ungr sttir [], when
young you laid claim [], st. 15), to the past-perfect forms in sts 1617, describing lfrs
more recent victories over the petty kings of Opplandene (hefr ljtu [] of goldit, you have
rewarded [them] with ill, st. 16; Braut hafi [] r landi [] rekna, You have driven
[them] away from the land, st. 17)(Rainford 1996:28). This culminates in the stirring present-
tense acclamation of lfrs sole kingship in Norway in st. 18:10

N rr fyr eiri
(ik remmir gu miklu)
fold, es forum heldu
fimm bragningar (gagni);
brei eru austr til Eia
ttlnd und r; Gndlar
engr sat elda rngvir
r at slku li.

Now you rule over that land which five princes once held; God strengthens you with a great vic-
tory. Broad hereditary lands lie under you, east towards Eiar; no wielder of Gnduls fires
[SWORDS > WARRIOR] has sat over such a fief before.

If there was ever any such heightened moment of closure in Vkv, it has been forgotten, but
the stanza seems to embody the more sweeping perspective of ttarrs poem, composed at a
point closer to the apogee of lfrs power. This final ideological statement makes imma-
nent the original performance context of the poem (Jesch 2006:263). But it also reflects how
ttarr encapsulated the message of Vkv within a much more ambitious political and poetic
project. Previous rulers may have claimed great domains in Norway, but none who had
proved himself as a warlord in such a litany of foreign campaigns (st. 19). ttarr ratchets up
the level of praise further by allowing his poem to culminate in his pious declaration in st. 18
that lfr holds Norway like a fiefdom (l) from God. In so doing, he recycles the old
skaldic notion that associated successful earthly lordship with divine agency, a topos particu-
larly strongly developed in the poetry of the Hlaajarlar (Frank 2006). Appropriating such a
pagan motif and reworking it to legitimize an unprecedented land-grab in Norway by a zeal-
ous Christian convert was a bold stroke, but it was one that allowed the poet to advertise his
recreation of tradition, and his departure from it.

Conclusion
While it is immediately evident that Vkv and Hfl are closely related, Hfl is more abun-
dantly larded in praise than the other poem, more ambitious in design, and makes more exag-
gerated claims for lfr as ruler of Norway. In every element of structure and expression,
ttarr produces a more elaborate response to the shared material. Yet by his marked structural
and verbal evocations of Sigvatrs poem, ttarr overtly situates his composition in relation to
the work of his older contemporary, the most highly regarded poet in lfrs following, invit-

10
The shift in tense is anticipated proleptically in the intercalary exclamation in Hfl 3/6: n est rkr af hvt
slkri You are now powerful from such prowess (Jesch 2006:263).

333
ing direct comparisons between them. As such he does more than produce an alternate re-
sponse to the conventions of the poetic tradition: Hfl systematically upstages Vkv, structur-
ally, conceptually and even metrically. ttarrs performance represents as clear an example of
poetic one-upmanship as we possess anywhere in the Old Norse skaldic corpus. It is possible,
indeed, that the preservation of Hfl partly reflects ttarrs success in outdoing the early work
of his more prestigious uncle in the eyes of medieval antiquarians, for it is striking that there
is not a single stanza of Vkv in the poetological treatises and connected works, while four
exemplary vsur are quoted from Hfl in Skldskaparml (sts 2, 19, 20) and the First Gram-
matical Treatise (st. 7).
In her recent account of the narrative modality of Hfl, Judith Jesch asks why a court poet
would recount deeds that were already well-known to his listeners. She concludes that the
point was to produce a collective reliving of past experience and fix it in the public memory
(2006:264). But an equally pertinent question here might be to ask why in so doing a skald
would allude to an older poem that must also have been familiar to his audience. As Jesch
points out, performance secured recognition of a poets role as a recorder of events for poster-
ity but it also produced patronage and political advantage. In a conventional form in which
the anxiety of influence must have been keenly registered, the work of successful poets pre-
sented the benchmark against which subsequent virtuosos and arrivistes were measured.
Competitiveness was thus an intrinsic aspect of skaldic verse-making, in which individual
practitioners vied against their fellows, past and present, in the demonstration of their own
primacy as the conduits of tradition.

Bibliography
Ashdown, Margaret, 1930: English and Norse Documents Relating to the Reign of Ethelred the Un-
ready. Cambridge
EE: Encomium Emmae Reginae. Ed. A. Campbell. 1949. London. Rpt 1998. Cambridge
Campbell, Alistair, 1971: Skaldic verse and Anglo-Saxon history. London
CVC: An Icelandic-English Dictionary. Ed. R. Cleasby, Gudbrand Vigfsson, and W. A. Craigie. 2nd
ed. 1957. Oxford
Clunies Ross, Margaret, 2004: Hfulausn and Egils saga. Notes and Queries 51:11418
Fell, Christine, 1981: Vkingarvsur. In: Speculum Norroenum. Norse Studies in Memory of Gabriel
Turville Petre. Ed. U. Dronke. Odense. Pp. 10622
Fidjestl, Bjarne, 1982: Det norrne fyrstediktet. vre Ervik
Fidjestl, Bjarne, 1997: Skaldic poetry and the conversion. In: Bjarne Fidjestl: Selected Papers. Ed.
O. E. Haugen and E. Mundal. Odense. Pp. 1335
Finnur Jnsson, 19204: Den oldnorske og oldislandske litteraturs historie. 3 vols. 2nd ed.
Copenhagen
Fsk: Fagrskinna. In: grip af Nregskonunga sgum. Fagrskinna Nregs konunga tal. Ed. Bjarni
Einarsson. 1984. Reykjavk. (slenzk fornrit 29.) Pp. 55373
Frank, Roberta, 1978: Old Norse Court Poetry: The Drttkvtt Stanza. Ithaca
Frank, Roberta, 1994a: King Cnut in the verse of his skalds. In: The Reign of Cnut, King of England,
Denmark and Norway. Ed. A. F. Rumble. Leicester. Pp. 10624
Frank, Roberta, 1994b: When poets address princes. In: Sagnaing helga Jnasi Kristjnssyni sjtu-
gum, 10. aprl 1994. 2 vols. Ed. Gsli Sigursson. Reykjavk. Pp. 18995
Frank, Roberta, 2006: The lay of the land in skaldic praise poetry. In: Myth in Early Northwest
Europe. Ed. S. O. Glosecki. Turnhout. Pp. 17596
O: slensk orsifjabk. Ed. sgeir Blndal Magnsson. 1989. Reykjavk
Jesch, Judith, 2006: The meaning of the narrative moment: Poets and history in the late Viking Age.
In: Narrative and History in the Early Medieval West. Ed. E. M. Tyler, R. Balzaretti. Turnhout. Pp.
25165
Moberg, Ove, 1941: Olav Haraldsson, Knut den Store och Sverige. Studier i Olav den Heliges frhl-
lande till de Nordiska grannlnderna. Lund
Nordland, Odd, 1956: Hfulausn i Egils Saga. Ein tradisjonskritisk Studie. Oslo

334
NN: Notationes norrn: Anteckningar till Edda och skaldediktning. Ed. E. A. Kock. 192344.
Lund. (Lunds Universitets rsskrift. N.S., sec. 1)
H: Saga Olafs konungs hins helga. Den store saga om Olav den hellige efter Pergamenthndskrift i
Kungliga Biblioteket i Stockholm nr. 2 4to, med varianter fra andre hndskrifter. Ed. O. A. Johnsen
and Jn Helgason. 1941. Oslo
H(Hkr): lfs saga helga. In: Heimskringla. Ed. Bjarni Aalsteinsson. 1979. Reykjavk. (slenzk
fornrit 27)
H: Otte Brudstykker af den ldste Saga om Olav den Hellige. Ed. G. Storm. 1893. Christiania
HLeg: Olafs saga hins helga: Die Legendarische Saga ber Olaf den Heiligen (Hs. Delagard.
saml. nr. 8II). Ed. A. Heinrichs. 1982. Heidelberg
Poole, Russell. 1987. Skaldic verse and Anglo-Saxon history: some aspects of the period 10091016.
Speculum 62: 26598
Rainford, Jessica. 1996. lfr Haraldsson, king and saint of Norway, and the development of skaldic
style (ca. 10151153). Unpubl. PhD. University of Oxford
Skj : Den norsk-islandske skjaldedigtning AIII: Tekst efter hndskrifterne; BIII: Rettet tekst. Ed.
Finnur Jnsson. 191215. Copenhagen
Von See, Klaus, 1977: Polemische Zitate in der Skaldendichtung: Hallfrr vandraskld und
Haldrr kristni. Skandinavistik 7:11519
Vries, Jan de, 1999: Altnordische Literaturgeschichte. 3rd ed. Berlin [1st ed. 1941]

335
 Alternative criteria for the dating of the sagas of Icelanders1
Gurn Nordal, rni Magnsson Institute for Icelandic Studies, Iceland

1
The dating of the sagas of Icelanders has never been an idle pursuit. A date, however vague,
places a saga in a historical context, and therefore has direct bearing on the way it is presented
in a modern edition and ultimately perceived by the scholar and the general reader. Einar l.
Sveinsson in his useful essay, The Dating of the Sagas of Icelanders, enumerated different
criteria for dating individual sagas, of which the date of the earliest manuscript of each saga
(though some are only preserved in late copies) is clearly the most tangible. Other valuable
clues to the time of writing were to him historical evidence, such as references to datable law
codes and social practices, allusions to historical persons and events, as well as literary rela-
tions and linguistic evidence.
One of Einar lafurs categories seems at first more subjective, even though it has as much
footing in a given historical context as references to named persons, that is artistry, the skill
displayed in the sagas (Einar l. Sveinsson 1958:115). In this paper I would like to revisit
this less concrete (Einar l. Sveinsson 1958:40) criterion and consider whether the citation
of skaldic verse and the way the settlement is depicted in the sagas of Icelanders can be used
as guides to the time of writing. These two criteria are intimately connected with the way the
Icelanders relationship with Norway developed over time and is presented in the sources.
The importance of skaldic verse in a courtly milieu in Norway and later in the writing of the
sagas of the kings brought skaldic poetry to the heart of the Christian textual culture, and the
representation of the settlement of Iceland from Norway became the defining moment in
many of the sagas of Icelanders. I have written before on both these aspects of the writing of
the sagas of Icelanders, but have not associated them directly with the issue of dating (Nordal
2007:21937; Nordal 2008:31518). This paper should be regarded as the first tentative step
towards using these two criteria as clues to the date of writing, and rewriting, of the sagas of
Icelanders in the 13th and 14th centuries.

2
Skaldic verse-making was one of the most studied literary forms of the medieval period in
Iceland. Textbooks and treatises on skaldic poetry reveal evolving attitudes to the way skaldic
verse was studied in the 13th and 14th centuries, that is in the period which also fostered the
writing of the sagas of Icelanders. The citation of skaldic poetry underpinned Snorri Sturlu-
sons writing of royal historiography in Heimskringla, and in his explication of the art of the
skaldic poet in his Edda he relied to a large degree on the same corpus of verse as he does in
Heimskringla (though not necessarily the same stanzas). Snorris nephew, lfr rarson,
writing the Third Grammatical Treatise one generation later, was not, however, restricted in
choosing his poetic examples from the historical skaldic canon. It is of note that Snorris
original planning of Skldskaparml is clearly redefined when the work is placed in context
with lfrs treatise in the so-called A- and B-manuscripts of Snorra Edda in the 14th century
(Nordal 2001:836). It is of paramount importance to keep this vibrant reception of skaldic
poetry in mind when we explain the use of skaldic poetry in the different saga genres of the
13th and 14th centuries. Below I will not only endeavour to suggest a link between the applica-

1
Note that a full version will be published by the Center for Medieval Studies in Bergen later this year as part of
a collection of essays on the topic of dating the sagas.

336
tion of skaldic verse in the sagas of Icelanders and the theoretical writings about skaldic verse
in the 13th and 14th centuries, but also propose that the way skaldic verse is used in the sagas
can be suggestive of their time of writing. Some sagas are preserved in more than one version
from the 14th century, and these versions reveal shifting attitudes to skaldic poetry in the 14th
century. The transmission of Njls saga is a case in point (see Nordal 2005).
Given the importance of skaldic verse in the learned milieu and in certain aristocratic and
religious circles we would expect some correlation between the function of skaldic verse in
the poetic treatises at any given time and the role played by verse in different saga genres in
the 13th and 14th centuries, such as the kings sagas, the sagas of Icelanders and the fornal-
darsgur. If we take seriously the change in the appreciation of the historical skaldic canon as
the 13th century wore off, and in the 14th century, we would expect that systematic references
to the respected court poets of the canon, as they are identified in Skldatal, Snorra Edda and
the thirteenth-century kings sagas, such as Heimskringla, would be indicative of the cultural
milieu of the sagas authors of the 13th rather than of the 14th century. In my paper on the use
of verse in the sagas of Icelanders, published in Margaret Clunies Rosss festschrift in 2007, I
approached the question of the importance of verse in the sagas not only from the point of
view of the number of stanzas in each of the c. 40 sagas, but also in light of the cultural iden-
tity of the poets who are cited in the narrative. I proposed the following grouping of the sagas
based on the application of verse in the narrative:
The sagas of the court poets (the skalds sagas): Bjarnar saga Htdlakappa, Egils saga,
Fstbrra saga, Gunnlaugs saga ormstungu, Hallfrear saga and Kormks saga. The poets
in the skalds sagas are acknowledged and respected poets in the skaldic canon and their verse
is cited in Snorra Edda and the kings sagas (Fidjestl 1985:31935; Nordal 2001:789). The
choice of stanzas highlights the generic borderline between the verse in the sagas of Iceland-
ers and the kings sagas, as there is very little overlap between the genres in the transmission
of the poetry. Egils saga is preserved in a fragment from the middle of the 13th century. Kor-
mks saga, Egils saga, Fstbrra saga (also in Hauksbk c. 1300) and Hallfrear saga are
preserved in Mruvallabk, c. 133070, the two last mentioned are also incorporated in the
two great lfs sagas of the fourteenth century. Bjarnar saga is preserved in a fragment from
c. 1300 and Gunnlaugs saga in a manuscript from the first half of the 14th century.
Sagas were the main protagonist is a poet: Gsla saga Srssonar, Grettis saga, Harar
saga ok Hlmverja, Vga-Glms saga, Hvarar saga, Vglundar saga and rar saga
hreu. It is of note that Vga-Glmr is among the acknowledged poets of the canon in Snorra
Edda, cited four times in Skldskaparml in the Codex Regius manuscript, whereas Grettir
smundarson and Hvarr are cited only once (Nordal 2001:789). The authors of the poetic
treatises never refer to Gsli Srssons verse, nor to the poetry by Hrr, Vglundr and rr
hrea. None of the poets in this group, however, is cited in the kings saga corpus. None of
the manuscripts predate c. 133070, or Mruvallabk (where Vga-Glms saga is pre-
served), and their earliest manuscripts span the period to the middle of the 15th century.
Sagas with a strong royal or courtly emphasis: Laxdla saga, Vatnsdla saga and
Finnboga saga ramma. Theodore Andersson pointed out the special thematic relationship
between Laxdla saga and Vatnsdla saga, their common interest in the Norwegian royal
background (as does Sveinsson 1939:xxxvi) and in portraying strong and unassailable chief-
tains in the region, [the sagas betray] the same taste for style and grandeur (Andersson
2006:155). Andersson does not note their disinterest in skaldic verse. Skaldic verse did not fit
into the new courtly framework in Norway in the late 13th century; the last known court poets
are from the late 13th century (Nordal 2001:12730). Laxdla saga is preserved in a fragment
from the late 13th century and Finnboga saga is contained in Mruvallabk. The earliest
manuscripts of Vatnsdla saga, however, is from c. 1390.

337
 Sagas dated to the 14th century and display distinct learned interest in the past: Ban-
damanna saga, Kjalnesinga saga, Flamanna saga, Hnsna ris saga, Krka-Refs saga
and Brar saga Snfellsss, and the two Vinland sagas, Eirks saga raua and Grnlend-
inga saga. These sagas are not concerned primarily with the classic feud pattern, nor are they
interested in unveiling moral or legal dilemmas in a close knit Icelandic farming community
(Hnsa-ris saga is clearly an exception and should perhaps be categorized separately along
with Bandamanna saga). Even though the narrative in the sagas opens in the traditional set-
ting of the Icelandic countryside, a large part of the narrative of some of these sagas is played
out outside the farming community in Iceland, to a large degree abroad, e.g. in Norway,
Greenland (Krka-Refs saga, Flamanna saga, Eirks saga and Grnlendinga saga) and
Vinland (Grnlendinga saga, Eirks saga), or in a quasi-human setting with giants in the
mountains (Kjalnesinga saga and Brar saga). The manuscripts are relatively late (with the
exception of Eirks saga in Hauksbk, which is on the borderline of the sagas of Icelanders),
figuring the fragments of the great slendingasgur-manuscripts of the late 14th and early 15th
century, Vatnshyrna and Pseudo-Vatnshyrna, written in the north of Iceland (Stefn Karlsson
1970; John McKinnell 1970). Kjalnesinga saga and Flamanna saga have been associated
with Haukr Erlendsson the noted lawman and the author of one of the versions of Landnma,
who lived most of his professional life in Norway (see Helgi Gumundsson 1967:7783; Per-
kins 1978:2844).
Sagas relating events in the Eastfjords and the north-east of Iceland: Fljtsdla saga,
Gunnar saga Keldugnpsffls, Hrafnkels saga, Vpnfiringa saga, orsteins saga hvta, or-
steins saga Su-Hallssonar and lkofra saga in the Eastfjords, and Ljsvetninga saga, Reyk-
dla saga (the only stanza in the saga is also cited in Vga-Glms saga) and Valla-Ljts saga
in the north-east of Iceland. The unique character of this region in medieval literary culture is
not limited to the sagas of Icelanders. The same geographical distinctiveness is borne out by
the known origin of poets in the 13th and 14th centuries, in the period when the sagas were put
to writing (Nordal 2001:133), as well as in the poetry cited in the Third Grammatical Treatise
(though lfr rarsons choice of examples may be determined by his own regional knowl-
edge (Gsli Sigursson 2002:11820). The reason for the exclusion of skaldic poetry in these
sagas is possibly regionally determined, rather than aesthetically conditioned. Droplaugar-
sona saga which is the only saga of the Eastfjords to contain verse (6 stanzas), and lkofra
saga are preserved in Mruvallabk, 1330 70, but all the other sagas are preserved in
manuscripts from c. 1400 or later.
Sagas where verse is an integral part of the narrative while there is no principal poet:
Njls saga, Heiarvga saga, Eyrbyggja saga, and Svarfdla saga. Svarfdla saga is clearly
the odd saga out in this company, yet it has narrative themes in common with Njls saga (the
depiction of Yngvildr fagrkinn) and Eyrbyggja saga (the description of the ghost Klaufi). All
four sagas cite verse by known poets of the canon. In Njls saga, Heiarvga saga and Eyr-
byggja saga such verses are used to authenticate sections in the account, and in Svarfdla
saga, two stanzas are attributed to the well-known poet orleifr jarlsskld. A ttr about or-
leifr is preserved in Flateyjarbk. orleifr jarlsskld falls within the literary radar of lfr
rarson and the author of the Fourth Grammatical Treatise, being cited once in each text.
The earliest manuscripts of these sagas are dated to the 13th century, such as the oldest frag-
ments of Eyrbyggja saga and Njls saga. Svarfdla saga, however, is in a manuscript from
the 15th century.
This division of the sagas of Icelanders into these six groups does not take into account the
traditional dating of the sagas, yet interesting patterns emerge when we note the earliest
manuscripts of each group. I discussed these groups at some length in Clunies Rosss fest-
schrift (2007), but there I did not attempt to link the citation of verse and the identity of the
poets to a specific period in time. The canonized poets of Skldatal are found in groups 1 and

338
6, and on the merit of the existing manuscripts alone these groups should be categorized
among the earliest sagas. Even though the stanzas in the kings sagas and Snorra Edda is not
drawn from the same poetic corpus as the verse in the sagas of Icelanders, it is apparent that
some of the poets in the sagas belong to the respected canon of the learned tradition. The divi-
sion of the sagas into different categories is always problematic, particularly when employing
posthumous assessment of them. The speakers of verse in group 2 are not among the poets of
the canon of the kings sagas, even though some of them are cited in the Snorra Edda, yet the
authors of these sagas cast their narrative somewhat in the mould of the skalds sagas.
Three groups are rich in verse (1, 2 and 6), but sagas in the remaining three groups (3, 4, 5)
show little interest in skaldic poetry or omit such poetic references altogether. It is striking
that these sagas have different emphasis from the sagas with verse: these are the 14th-century
sagas with learned interest, some showing particular interest in Greenland, sagas which are
played out in the northeast and east of Iceland, and finally sagas which show particular fasci-
nation with the royal court and/or with courtly romance. One of those sagas is Vatnsdla
saga, which I have categorized with Laxdla saga and Finnboga saga ramma, on the basis of
their common disinterest in skaldic verse and royal fascination. I am currently reconsidering
the association of the exceptional Laxdla with the other two closely related sagas that take
place in Hnavatnsssla.

3
If we can agree that skaldic verse is one of the defining elements behind the writing of the
sagas of Icelanders, at least those sagas which make extensive use of verse (even though we
may disagree on whether these elements are datable), Landnma is certainly the other literary
backbone to the writing of the sagas of Icelanders. The relationship between the sagas and the
three medieval Landnma-versions is complex and difficult to unravel. Some modern scholars
would say that a comparison between the stories of the settlement in Landnma and the sagas
of Icelanders is a futile undertaking. I beg to disagree. Extensive analysis of the relationship
between the sagas and Landnma was undertaken by scholars such as Bjrn M. lsen and Jn
Jhannesson, and even though their insistence on direct textual borrowings is certainly too
strong, they draw attention to noteworthy patterns of literary relations which bring us closer to
drawing up a picture of literary activity in certain areas in the country.
Memories about the pagan past in Iceland and the settlement period were most likely pre-
served in oral memory from the 9th and 10th centuries to the period in which the sagas of Ice-
landers were written. It is therefore illuminating to focus on the seventeen sagas which open
at the time of settlement of Iceland, and thus reflect the authors or the audiences interest in
the migration period and their fascination with the neighbouring countries not only at the time
of the settlement but more importantly at the time of writing. These sagas are, in alphabetical
order: Brar saga Snfellsss, Egils saga, Eyrbyggja saga, Flamanna saga, Gsla saga
Srssonar, Grettis saga, Harar saga, Hrafnkels saga, Kjalnesinga saga, Kormks saga,
Laxdla saga, Reykdla saga, Svarfdla saga, Vatnsdla saga, Vga-Glms saga, rar
saga hreu and orskfiringa saga.
Only one of these seventeen sagas refers to skaldic verse in relation to the settlement narra-
tive. nundr criticizes the cold climate in Iceland in Grettis saga and the writer of the saga
may be using the poetic evidence to introduce a different point of view in the narrative rather
than using the verse to authenticate the account. The motivation behind the construction of the
Landnmabk (Book of Settlement) is contested, but the early settlements of Iceland may
have been set in writing in order to secure land claims by ruling families at the time of writing
in the 12th century. However, the reason why these settlement stories were preserved in a
number of sagas in the 13th and 14th centuries is another matter.

339
 The three known medieval versions of Landnma, Sturlubk, Melabk and Hauksbk, are
all attributed to Icelandic lawmen in the post-Commonwealth period, between c. 1270 and
1310. Sturla rarson (d. 1284), Snorri Marksson (d. 1313) and Haukr Erlendsson (d. 1334)
were aristocratic men who had special relationship with Norway. The prologue to Landnma
in Sturlubk and Hauksbk reflects a Norwegian point of view when framing the description
of the settlement. Fourteen leaves of Haukr Erlendssons Landnma are preserved in the au-
thors own hand in Hauksbk; the remaining part has only come down to us in a 17th-century
copy. Melabk, which has been traced to Snorri Marksson at Melar, is preserved in a frag-
ment from the beginning of the 15th manuscript (Pseudo-Vatnshyrna where it was in company
with many of the sagas of the Icelanders), but the text is thought to be interpolated in rar-
bk from the 17th century. Sturla rarson wrote a version of Landnma which is only pre-
served in a 17th-century copy of a manuscript from c. 1400 (on the manuscripts, see Jakob
Benediktsson 1968:LLIV). The transmission of Landnma is clearly problematic, and it is
time that scholars revisit the question of the relationship between the versions of Landnma
and the sagas of Icelanders, and indeed the versions of some of the sagas (for a recent evalua-
tion, see Sveinbjrn Rafnsson 2001).
Iceland had become part of the Norwegian kingdom (c. 12624) when most of the sagas of
Icelanders are written, with perhaps the exception of Egils saga, the skalds sagas and possi-
bly Laxdla and Eyrbyggja, and some of these sagas may reflect an interest by members of
the Icelandic aristocratic elite to argue for close ties between Iceland and Norway, the old
homeland for many of the settlers, now the seat of the king. But each saga treats the topic of
the settlement independently, and the sagas that begin their narrative in Iceland and omit the
migration story, place less importance on these ties. Moreover, the more than twenty sagas
that begin their narration after the settlement treat the topic with a lack of interest. Four pat-
terns in the sagas depiction of the settlement emerge:
Sagas containing a complex migration story and detailed elucidation of the migrating fam-
ilys relationship with the king. This theme is rehearsed in sagas such as Laxdla saga,
Vatnsdla saga, Eyrbyggja saga, and the sagas of the court poets, such as Egils saga and
Kormaks saga. Some of these sagas, such as Egils saga, Laxdla saga and Eyrbyggja saga,
are preserved in old manuscript fragments from the 13th century, and are probably among the
oldest written sagas of Icelanders.
The emphasis on the history of a fighter-poets family, where the familys life in Norway is
played out in detail, such as in Grettis saga and Gsla saga Srssonar. Other sagas in this
group are Harar saga ok Hlmverja (no migration story), Vga-Glms saga, Vglundar saga
and rar saga hreu. The portrayals of the hero seem to be modelled on the sagas of the
court poets, as we noted above, but in fact these sagas focus on very different themes from the
skalds sagas. All of them deal with personal loss; the loss of land, the loss of freedom, as in
the outlaw sagas, and some end on a very tragic note. There is a strong tendency in all of these
sagas to deepen the portrayal of the hero by linking him to the familys past in Norway. Vga-
Glms saga is not a typical settlement saga, but the familys fylgja in Norway finally settles
in Iceland.
Learned interest in the settlement and in the mythic past of Iceland is attested in sagas such
as Brar saga Snfellsss, Flamanna saga, Kjalnesinga saga (not a migration story; begins
at the time of the settlement), and orskfiringa saga. Some of the settlement stories seem to
be drawn from external written sources such as in Flamanna sagas reference to Landnma.
In this group there is an apparent interest in travels to Greenland.
In some sagas we find a silent reference to the settlement, where there is no migration story
and little importance placed on the settlement. Among those are sagas such as Hrafnkels saga
and Reykdla saga.

340
 The correlation between this grouping and the earlier grouping of the sagas of Icelanders is
no coincidence. I tentatively suggest that there is an evolution from type 1, to type 2 and 3,
with type 4 coexisting at all times, particularly in the later sagas.
Landnma was most probably written in order to secure the present ownership of the ruling
families, and the different versions of the work from the 13th century to the beginning of the
15th century reflect the continguing interest and demand for passing on and redefining the
stories of the migration from Norway. The relationship between the sagas of Icelanders and
Landnma is complex; in some cases there is no textual relationship even though the sagas
recount similar stories as Landnma, but in other examples there are clear borrowings, such as
in the learned sagas.
The shifting focus on the migration to Iceland and the settlement in the sagas of Icelanders
reveals the ambiguity in the sagas deliberation and reconfiguration of the past. The many
Christian writers of the sagas regarded the pagan past in a markedly different way, and some
went as far as to disregard the settlers ties to their old homelands. The sagas of the Eastfjords
reveal a noteworthy indifference to the memories of the settlement. Only two sagas out of ten
begin at the time of settlement (Reykdla saga and Hrafnkels saga), but with no accompany-
ing genealogy that links the families with the old world. The stories of the settlement seem
to be predominantly associated with events and characters in west and north-west of Iceland;
the area where the interest in skaldic poetry and the writing about pagan myth was also most
clearly attested, as in work such as Snorra Edda, the Third Grammatical Treatise and Heim-
skringla. This geographic distinction within the genre can be no coincidence, and throws into
relief the importance of constantly keeping in mind the subtle differences between the sagas
of Icelanders in their depiction of their relationship with Norway and in their portrayal of their
skaldic heritage.

Bibliography
Andersson, Theodore M. 2006. The Growth of Medieval Icelandic Sagas (11801280). Cornell. Cor-
nell University Press.
Einar l. Sveinsson 1958. Dating the Icelandic Sagas. An Essay in Method. Viking Society for
Northern Research. University College London.
Fidjestl, Bjarne. 1985. On a New Edition of Scaldic Poetry. The Sixth International Saga Confer-
ence 28. 7. 2. 8. 1985. Workshop Papers (Copenhagen, 1985), pp. 31935.
Gade, Kari Ellen. 2001
Gsli Sigursson. 2002. Tlkun slendingasagna ljsi munnlegrar hefar: Tilgta um afer.
Reykjavk. Stofnun rna Magnssonar.
Helgi Gumundsson . 1967. Um Kjalnesinga sgu: Nokkrar athuganir (Studia Islandica 26).
Reykjavk. Menningarsjur.
Jakob Benediktsson. 1968. Formli. slendingabk. Landnmabk (slenzk fornrit 1). Reykjavk. Hi
slenzka fornritaflag.
McKinnell, John. 1970. The Reconstruction of Pseudo-Vatnshyrna. Opuscula 4:304338.
Nordal, Gurn. 2001. Tools of Literacy. The Role of Skaldic Verse in the Textual Culture of the 13th
and 14th Centuries. Toronto. University of Toronto Press.
Nordal, Gurn. 2005. Attraction of Opposites: Skaldic Verse in Njls Saga. Literacy in Scandinavia
from Middle Ages to Early Modern Time. Ed. Pernille Hermann (Viking Collection 16), pp. 211
36. Odense. University Press of Southern Denmark.
Nordal, Gurn. 2007. The Art of Poetry and the Sagas of Icelanders. Learning and Understanding
in the Old Norse World. Essays in Honour of Margaret Clunies Ross. Eds. Judy Quinn, Kate
Heslop and Tarrin Wills, pp. 21937. Brepols.
Nordal, Gurn. 2008. The Sagas of Icelanders. The Viking World. Ed. Stefan Brink in collaboration
with Neil Price, pp. 31518. London and New York. Routledge.
Perkins, Richard. 1978. Flamanna saga, Gaulverjabr and Haukr Erlendsson (Studia Islandica 36).
Reykjavk. Menningarsjur.
Stefn Karlsson. 1970. Um Vatnshyrnu. Opuscula 4:279303.

341
Sveinbjrn Rafnsson. 2001. Sguger Landnmabkar. Um slenska sagnaritun 12. og 13. ld (Rit-
safn Sagnfristofnunar 35). Reykjavk. Sagnfristofnun Hskla slands.

342
 Hvernig leit Uppsalabk t ndveru?
Guvarur Mr Gunnlaugsson, Stofnun rna Magnssonar slenskum frum, Iceland
Handriti DG 11 sem varveitt er Hsklabkasafninu Uppslum er merkilegt fyrir margra
hluta sakir. fyrsta lagi er a elsta handrit Snorra-Eddu sem varveitt er og ru lagi hefur
a a geyma srstaka ger af textanum. Handriti er einnig merkilegt fyrir margar teikningar
spssum og auum sum og fyrir sk a v eru stuttar skrr ea tl, sem virist hafa
veri skoti inn Eddutextann. Enn fremur m telja a til athyglisverra atria handritinu
a bkfelli virist ekki hafa veri spara v a margar blasur voru aeins hlfskrifaar
upphafi ea jafnvel hafar alauar. M ar nefna bl. 1v og 26v sem eru alveg au og bl. 22v,
25r, 26r og 37v sem eru ekki fullskrifu.
Hr verur fjalla um samsetningu DG 11, en ekki um textann umfram a sem
nausynlegt er til a tta sig v hvernig a kynni a hafa liti t ndveru.
Til a f betri mynd af Uppsalabk skoai g hana s.l. sumar. Hn er 56 bl sem skiptast
7 kver. Athyglisvert er a fremsta kveri er 10 bl ar sem fremsta og 10. blai eru saman
tvinni. a er ekki algengt a handrit hefjist 10 blaa kveri. Algengast er a kver
handritum su 8 bl me eirri undantekningu a aftasta kveri er oft afbrigilegt. DG 11
er aftasta kveri aeins 6 bl. Anna sem er athyglisvert vi fremsta kveri er a fremsta
blai hefur upphaflega veri skili eftir autt og hefst texti handritsins efst bl. 2r.
Textamevitaur bkavrur ea frimaur hefur liti fremsta blai sem saurbla og
merkt bl. 2r sem fyrstu blasu, sem er rangt. essi ranga blasumerking hefur svo fylgt
handritinu tgfum og umfjllun um a, en hn verur ekki rttari fyrir sk. Fremsta
blai fremsta kverinu er auvita fremsta bla handritsins og saurbl sem eru yfirleitt stk
og/ea fst vi bandi tkuust ekki mildum. Mialdamenn skildu hins vegar fremstu
blasuna handriti oft eftir aua og ltu hana annig jna hlutverki saurblas en hfu
skriftir textans bl. 1v. sumum lgbkum er stundum hf mynd bl. 1v sem veldur
v a textinn byrjar bl. 2r en er myndin hluti af efni handritsins. Auvita mtti hugsa
sr a skrifarinn (ea verkbeiandinn ea verkstjrinn) hafi hugsa sr a bl. 1v yri
teiknu ea mlu mynd eins og lgbk ea handriti heilagra manna sagna, en kemur upp
spurningin af hverjum s mynd hefi tt a vera. Hvaa myndefni hfi kennslubk
skldskaparfrum, sem Snorra-Edda virist fyrst og fremst vera? Varla mynd af konungi ea
heilgum manni. Ef til vill var skrifarinn bara vanur a skrifa lgbkur og byrjai sjlfrtt
a skrifa bl. 2r, en hvorug tilgtan skrir af hverju fremsta kveri er 10 bl.
Ysta tvinni fremsta kverinu er aeins minna en hin tvinnin kverinu. Munurinn er hins
vegar svo ltill a hann gti a sjlfsgu veri tilviljun. Fremsta blai er 197 mm h og
a 10. er 199 mm h mean hin blin kverinu eru bilinu 197204 mm h. Frems-
ta blai er 128137 mm breidd (a mjkkar upp) og 10. blai er 140 mm breidd mean
hin blin kverinu eru 140145 mm breidd. Olof Thorell tekur einnig fram a fremsta
blai s tunt och rynkigt (1977:ix). a er skrti a a skuli vera unnt vegna ess a
a er tvinni me 10. bl. sem Thorell minnist ekki hva varar ykkt. Ekki tk g eftir a
fremsta blai vri ynnra en nnur bl en vissulega ltur a verr t en mrg nnur bl
handritinu. Ysta tvinni er, sem sagt, aeins minna en nnur tvinn kversins en ljst er hvort
eitthva er v byggjandi.
g hef ekki neina ga skringu takteinum af hverju fremsta kveri er 10 bl en mr
hefur dotti hug a upphafi hafi skrifarinn tla a lta binda essa Eddu sna (og e.t.v.
aeins hluta Eddutextans eins og komi verur a hr eftir) inn fyrir aftan ara bk; er
elilegt a hann byrji efst recto-su. En egar hann var kominn leiis me fyrsta kveri s
hann sig um hnd og kva a hann vri kominn me srstaka bk. Ef a er rtt, vantai
hann saurbla v a hann var egar binn a skrifa a bla sem annars tti a vera 1r.

343
Hvernig getur hann leyst a? J, hann skir ntt tvinn og setur utan um ttablaakveri sem
hann er a skrifa og svo egar hann er binn a fullskrifa upphaflegu 8 blin heldur hann
fram aftara blai nja tvinnsins og skrifar a og fr ar me 10 blaa kver ar sem
fremsta blai er alveg autt og textinn hefst bl. 2r og ll nnur bl kversins eru fullskrifu.
San sest hann yfir a skrifa nsta kver og byrjar a sjlfsgu recto-su fremsta
blasins sem verur vi etta bl. 11r sta 9r. E.t.v. er ysta tvinni fremsta kverinu aeins
minna en nnur tvinn af essari stu.
Anna sem vekur athygli vi samsetningu Uppsalabkar er a egar skrifarinn hefur loki
vi Gylfaginningu ( bl. 19r sem er fremsta blai rija kveri) heldur hann fram me 17.
og 18. kafla Skldskaparmla, sem stundum hafa veri kallair Bragarur, n ess a a
s merkt srstaklega (a eru aeins kaflaskil 19r en ekki ttaskil). bl. 22v lkur essum
hluta Skldskaparmla v a Eilfur Gurnarson er nefndur til sgunnar ( 20. lnu): Eptir
essi savgo hevir ort eilifr gvrvnar son i [or]s drapo:-. a sem eftir er af blasunni skil-
di skrifarinn svo eftir aua. v nst skrifar hann skldatal nstu 5 blasur remur
dlkum hverri su (23r25r). Hann fyllir rj dlka 23r og 23v, tvo og hlfan dlk 24r,
en aftur rj dlka 24v og lkur vi tali nearlega midlkinum bl. 25r. Hann skilur
afgangana af blasum 24r og 25r eftir aua. En sta ess a halda fram me
Skldskaparml bl. 25v t.d. me v a skrifa rsdrpu sem virist aldrei hafa veri
skrifu etta handrit snr hann sr a v a skrifa ttartlu Sturlunga (19 lnur) og heldur
fram niur suna me lgsgumannatal sem hann lkur vi nstu su (bl. 26r:12). Hann
skilur svo afganginn af 26r eftir auan sem og allt bl. 26v. Umrtt 26. bl. er aftasta blai
rija kveri handritsins og af eim skum datt mr hug a handriti hefi aldrei tt a vera
strra. a hefi ekki tt a skrifa neitt meira af Snorra-Eddu og skrifarinn hefi aeins veri
a fylla upp auar sur kversins me v a skrifa fyrrnefnd tl.
vaknar spurningin hvort a er mgulegt a skrifarinn hafi hugsa sr a skrifa aeins
hluta Skldskaparmla, .e. Bragarur, og sleppa ru. Snorra-Edda hefi me rum
orum tt a vera lengri handritinu. Finnur Jnsson (1900:iii) er viss sinni sk a vanda
hva etta varar:

Det var skriverens hensigt, efter afslutningen af Gylfaginning, af Skldskaparml kun at skrive
de mytiske kapitler, hvorfor han begyndte med k. 1718 (om Tor og jtterne Hrungne og
Geirrd); dette skete enten fordi disse kapitler srlig interessede ham eller fordi han havde knap
tid, eller snarest bgge dele p engang.

Ef essi tilgta er rtt, hefur skrifarinn (ea verkstjri hans) aeins hugsa sr a skrifa
Gylfaginningu og hluta Skldskaparmla. Gylfaginning og tveir kaflar r Skldskaparmlum
er stuttur texti sem aftur getur skrt hvers vegna skrifarinn gti hafa hugsa sr upphafi a
essi texti vri bundinn inn me ru efni (sj hr a framan).
a kann a vera elilegt a skrifarinn vilji hafa Skldatal eftir Bragarum vegna ess
a arna er hann kominn a fyrsta skldinu sem nefnt er Skldskaparmlum og honum gti
hafa fundist rtt a hafa nst skr yfir ll skld sem skipta mli (Heimir Plsson skaut essari
hugmynd a mr). En sta ess a halda fram me Skldskaparml egar hann hefur loki
vi skldatali, skrifar hann ttartlu Sturlunga og lgsgumannatal og virist svo htta
skriftum.
Hva sem essum vangaveltum lur er ljst a skrifarinn hlt fram a skrifa
Skldskaparml bl. 27r og hefur skriftirnar fremsta blai nju kveri sta ess a halda
fram bl. 26r eftir lgsgumannatali ea efst bl. 26v. Hvers vegna gerir hann a? v
verur auvita ekki svara me neinni vissu og hugsanleg skring er vandfundin.
Fyrirsgn essa hluta Skldskaparmla er athyglisver (27r): her hefr skalldskapar mal ok
heiti margra hlvta. a er sem sagt ekki eins og hann hafi skrifa hluta Skldskaparmla

344
ur sama handriti. Kaflar Skldskaparmla eru a vsu ekki smu r Uppsalabk og
rum handritum, en megni af textanum er skrifa bl. 27r45r nema Bragarurnar sem
ur voru komnar.
Handriti sjlft er dlti merkilegt essum sta. Eins og ur segir hefur skrifarinn skili
bl. 26v eftir autt seinna hafa svo veri dregnar myndir af Ganglera, Hum, Jafnhum og
rija essa su en a sem vekur athygli er hve m hn er og hrein. Hn ltur
nefnilega t fyrir a hafa veri ysta bla handritsins um hr og veri varin (hve lengi er
algjrlega vst). Getur veri a essi hluti handritsins (26 bl. ea rj kver) hafi legi lengi
bundinn ur en afgangnum (30 bl. ea fjrum kverum) var skellt aftan vi og bkin bundin
inn? Fremsta blasa handritsins er einnig dkk og m.
tlit bl. 27r stafestir hins vegar ekki nnefnda tilgtu, og enn sur s stareynd a
textinn er skrifaur strax fremstu su kversins. essi blasa er vissulega ekki vel farin en
hn er ekki a hrein ea m a hn lti t fyrir a hafa veri fremsta (ysta) sa bundins
handrits um langa hr og hn ltur heldur ekki t fyrir a hafa nuddast lengi vi spjald
bandi. Aftasta blasa sama kvers, 34v, er svipu tlits svo a tla m a etta kver hafi
legi vari einhvern tma, en mun styttri tma en aftasta sa nsta kvers undan (26v).
fljtu bragi er heldur ekki a sj neitt skrift handritsins ea stafsetningu sem stafestir a
langur tmi hafi lii milli ess a skrifarinn skrifai fyrstu rj kverin og sari fjgur, en
til a geta dmt um a arf mun meiri rannsknir skriftinni, tknbeitingu skrifarans og
stafsetningunni en g hef haft tk a gera a essu sinni.
g mldi str blaa Uppsalabkar en a kom ekkert t r v sem hnd er festandi.
Vissulega eru blin fjra kveri hennar lgri a mealtali en rum kverum en au eru
alls ekki mjrri a mealtali en bl rum kverum og ekki verur s anna en a
textafltur s svipaur t gegnum allt handriti. Bl bkarinnar lta ekki t fyrir a hafa
veri skorin vi innbindingu vegna ess a textaflturinn situr eim sta sunum sem
elilegt m teljast t fr fagurfrilegu sjnarmii svo a blaskurur hefur a llum
lkindum ekki skekkt mlinguna.
fyrirlestrinum rstefnunni mun g ra essi vandaml nnar og jafnframt sna
nokkrar myndir af handritinu til a heyrendur geti betur gert sr grein fyrir tliti
Uppsalabkar.

Ritaskr
UUB DG 11 (Uppsalabk).

Snorre Sturlasons Edda. Uppsalahandskriften DG 11. Facsimileedition i ljustryck. Historisk inledning

av Anders Grape. 1962. Stockholm.
Snorre Sturlasons Edda. Uppsalahandskriften DG 11. II. Transkriberad text och Paleografisk kommen-
tar av Anders Grape, Gottfrid Kallstenius och Olof Thorell. Inledning och Ordfrrd av Olof Tho-
rell. 1977. Stockholm.
Snorri Sturluson, Edda. Udg. af Finnur Jnsson. 1900. Kbenhavn.
Thorell, Olof, 1977: Inledning. In: Snorre Sturlasons Edda. Uppsalahandskriften DG 11. II. 1977.
Stockholm. Pp. ixxxi.

345
 Ansgars Conversion of Iceland
Terry Gunnell, University of Iceland
In his Conversion of Iceland (1975), Dag Strmbck writes that The Alings decision to
accept Christianity as the state religion of Iceland [in the summer of AD 999/ 1000] leaves us
perplexed, mystified (Strmbck 1975:26). More recently, in The Christianisation of Ice-
land, Orri Vsteinsson describes the same event as one of the really strange events in the
history of Christian missions (Orri Vsteinsson 2000:17). In the paper that follows, I would
like to argue that considering both the nature of religious belief in the Old Nordic (and appar-
ently also much of the Germanic) area in pre-Christian times, and the methodology of some of
the Christian missions issuing from Hamburg-Bremen during this period, there is actually
comparatively little to be perplexed or mystified about. In fact, almost everything about the
process of what happened at the Aling on the days in question seems to have followed Rim-
berts comparatively well-known textbook on how best to convert the Nordic peoples without
bloodshed, in other words, by using insiders, and by making good use of the weaknesses in-
herent in pagan native practices. Indeed, this was an approach Pope Gregory had previously
suggested in his letter to Mellitus in AD 601 (Bede 1965:86 [ch. 30], and something the Ro-
mans seem to have demonstrated even earlier in their effective use of the Bructeri seeress,
Veleda: see below).
To a large extent, I will be following the arguments set forward by Jn Hnefill Aalsteins-
son in his Under the Cloak: A Ritual Turning Point in the Conversion of Iceland (originally
published 1979; revised and extended 1999) about the course of events in Iceland. Indeed, Jn
Hnefills proposal for what happened at the Alingi on this occasion is well grounded in
comparative materials and makes good sense in the light of what we know of pagan religious
practice. The main section of Jn Hnefills outline of the probable passage of events at ing-
vellir is based largely on Ari fris comparatively trustworthy account in slendingabk
(1968:16), but also draws on Kristni saga (1858:2225) from the mid thirteenth century and
lafs saga Tryggvasonar en mesta (19582000:1, 187198), composed soon after 1300,1 es-
pecially concerning the idea of human sacrifice taking place and the idea of the Christians
arranging a comparable symbolic sigr gjf. The key part of the outline runs as follows:

2: Su-Hallr gets the Christians to release him from the official position of lawspeaker and
comes to an agreement with orgeir the Lawspeaker that he will recite those laws that the Chris-
tians should follow[].

3. The pagans sacrifice eight men, two from each quarter, in order to prevent Christianity from
taking over the country. orgeir Ljsvetningagoi, who according to Ari was still pagan when
he made the agreement with Su-Hallr, takes an active part in this sacrifice, probably a leading
role. When the sacrifice has been completed, orgeir lies down under the cloak in his booth, and
gengr til frttar (asks the pagan gods about the future), as usually took place at the end of each
sacrifice.

1
The events are also covered in Theodoricus Monachus Historia De Antiquitate Regum Norwagiensium
(11771188), but only very briefly, suggesting that the whole course of events was essentially the work of God
(the Grace of the Holy Spirit and divine intervention): see Theodoricus Monachus 1998: xxii and 1516.
Another early source which seems to borrow from both slendingabk and Theodoricus is Oddur Snorrasons
Saga lafs Tryggvasonar: See Oddur Snorrason 1932: 128130 (written in the late twelfth century). On the
relationship between all of these sources (including Njls saga), and their comparative values, see further Jn
Hnefill Aalsteinsson 1999: 5562.

346
 4. orgeir, as lawspeaker, had the role of being chairman in legal matters, as well as being the
political leader of the pagans. As such, he had the final say in any conventional dispute.
Through the execution of the sacrifice, he also earned himself the right to have the final say in
matters concerning the pagan religion.

5. The morning after, orgeir awakes, sits up and tells people to come to Lgberg. There he an-
nounces his decision based on what the gods revealed to him during his time beneath the cloak.

6. The pagans accept orgeirs decision, strange as it may sound to them.

(Jn Hnefill Aalsteinsson 1999:214):

In short, Jn Hnefill is arguing that a particular religious performance took place at ing-
vellir, designed to win the peaceful formal acceptance of Christianity by those who were op-
posed to it (and avoid the possibility of civil war).
While he is less convinced than Dag Strmbck (1975:3031) and Bo Almqvist (1974:19)
that any form of bribe took place when Hallr keypti at orgeiri lgsgumanni (slendin-
gabk 1968:16; cf. Kristni saga 1858:22; lafs saga Tryggvasonar in mesta 19582000:1,
191; and Brennu-Njls saga 1954: 271), Jn Hnefill nonetheless believes that orgeirs famed
(and clearly publicly-noted) day and night beneath the cloak (slendingabk 1968:16) was
related to a semi-shamanistic consultation with the gods, an activity paralleled in Irish, Scot-
tish, Smi and Native American tradition, and echoed in a classical account in Geoffery of
Monmouths Historia Regum Britannie (see Jn Hnefill Aalsteinsson 1999:115117, and
213214; Wood 1987:5657; Martin 1999:79; and Turner 1977: 167). According to Jn Hne-
fill, the period under the cloak was the final step of a recognised sacrificial process (a blt), in
other words, that of interpreting the frttir from the gods, the sacrifice and interpretation
being something which orgeir would have been expected to carry out at a time of crisis,
since orgeir in his role as lgsgumaur had the role of chief priest at the ingvellir gath-
ering. Indeed, the law and legal decisions appear to have belonged to a sphere that was closely
associated with the sacred.
The question of whether large-scale human sacrifice actually took place at the Aling on
this occasion is, of course, open to discussion (partly because of the questionable reliability of
Kristni saga and lafs saga Tryggvasonar in mesta). Nonetheless, there is natural logic be-
hind the idea that believers in the gods would have expected some form of sacrifice to be car-
ried out at such a time of decision (if not at all pagan Alingi meetings), and that an outcome,
in the form of a deciding contest or another kind of reading of the frttir should have fol-
lowed this. Parallels might be seen in the sacrifices related to the ritual duels described in
Egils saga (1933:209: the bltnaut which skildi s hggva er sigr hefi) and Kormks saga
(1939:290: another bltnaut), where the same elements are involved, if inverted (the decision
apparently preceding the offering.) There is also natural logic in the idea that a the end result
of a sacred ritual performance of this kind, conducted on the national stage of the time,
would have been accepted without any question by pagan believers, simply because this is
how decisions seem to have been made for centuries among the Germanic and Nordic peo-
ples.
That the Germanic tribes and Nordic peoples trusted lot-casting as a means of making de-
cisions (Latin: sortes; Old Icelandic: hlutan; cf. the throwing of bltspn or perhaps teinlau-
tar) is well documented from the time of Tacitus onwards. As Tacitus notes in Germania, ch.
10, For auspices and the casting of lots, they have the highest possible regard (Goetz &
Welwei 1995:1, 134135; trans. Tacitus 1948:108). Similar behaviour is reflected in Willi-

347
balds Vita S. Bonifacii (from the eighth century) which tells how, even after conversion,
some Germans:

[] continued secretly, others openly, to offer sacrifices to trees and springs, to inspect the en-
trails of victims; some practised divination, legerdemain and incantations; some turned their at-
tention to auguries, auspices and other sacrificial rites; whilst others, of a more reasonable char-
acter, forsook all the profane practices of heathenism and commited none of these crimes (The
Anglo-Saxon Missionaries 1954: 1011).

Similar practices and beliefs are reflected in a number of key works such as the Annals of
Xanten (see Wood 1987:47), where a group of Vikings facing disease in Paris (in 845) cast
lots to find whether the gods would help them; in Alcuins Vita Willibrordi, ch. 11 (from the
late eighth century), which describes how lots are chosen to decide sacrificial victims in For-
siteland/ Heligoland (see The Anglo-Saxon Missionaries 1954:1011); in Rimberts Vita An-
skarii (c. 880: chs. 18, 19, 2627, and 30: see Boken om Ansgar 1986:34; 40; 5455; and 58
60); where lots are used several times to decide a key plan of action; Hervarar saga ok
Heiriks 1943:208 (bltspn) where lots are used to decide an appropriate sacrifice when a
harvest fails; Landnmabk 1968:234 (S 198), where bltspn are used to decide the right
time to take land; Vlusp, st. 62 ( kn Hnir/ hlautvi kisa); and Einar Sklaglamms
Vellekla, st. 30 (possibly from the tenth century, and quoted in Fagrskinna [1984:118]),
where lots (bltspn in the prose, teinlautar in the poem) seem to have been cast in a field
(gekk til frttar/[] velli;/ Tr vildi tna/ teinlautar fjr Gauta: Den norsk-islandske
skjaldedigtning 19121915:A1: 129130; B1: 122). In all of these cases, the reading of the lots
is followed without any question, echoing Tacitus and Alcuins general statements. It is also
hard to attribute all of these examples to hagiographic borrowings. Quite obviously the cast-
ing of lots was a sacred moment, creating its own sacred space. It involved the entrance of the
holy into the everyday, and several levels of participants, most particularly those key inter-
preters/ performers who handled the lots, and the audience that waited to hear their an-
nouncements, trusting their expertise and the fact that the lots represent a line of communica-
tion between this world and that beyond ordinary sight.
A similar trust in using objects to make decisions is seen in the family legends of Icelandic
settlers throwing their high-seat pillars into the sea to decide where to settle in Landnmabk
(1968:4245, 124125, 163164, 232, 302303, 312 and 317 [S 89; 85; 123; 197; 289; 307;
and 310]). This divinatory tradition (directly referred to several times as a fregna til nd-
vegisslna) is almost certainly based on actual practice: it is echoed in a range of oral leg-
ends concerning how people decided church sites from northern Sweden and Norway: see
Campbell and Nyman 1976: 1, 1215, and2, 3742.2 It can also be seen in other accounts
from Landnmabk of people using blt or following the behaviour of horses for a similar
purpose (Landnmabk 1968:4244 and 9697 [S 78 and 68]); and in the practice of seir
being carried out at the beginning of winter (i.e. see Eirks saga raua 1935:206209). Like
those described by Tacitus and Alcuin, these were clearly people who trusted in following
omens. The idea that fate is preordained and can not be avoided is, of course, reflected in Ff-
nirs statement in Ffnisml, st. 11, that vatni druknar/ ef vindi rr/ alt er feigs forr,
and in the comment in Grettis saga (1936:62), that verr hverr at fara, er hann er feigr.
That this view of life was deeply rooted in the Nordic peoples, and difficult to eradicate can
be seen in the range of divinatory folk customs once practised across Scandinavia during

2
That similar legends also existed in northern Norway was confirmed by Reimund Kvideland in an unpublished
lecture held in Tartu, Estonia, in the summer of 2005.

348
Christmas and the New Year period (see, for example, B 1974:99108; and 1985:3540). It
can also be seen in the results of a recent survey of Icelandic belief attitudes (Gunnell 2007).
All in all, it can be argued that anyone who knew the Germanic and Nordic peoples would
have been well aware that this near blind-trust in augury was their key weakness. This is well
demonstrated by the sarcastic comment made by Hrlfr, the son of Helgi hinn magri, when he
started sailing towards the Arctic ocean after a advisory consultation with rr. Hrlfr, the
archetypal member of the younger generation, watching the icebergs approach, wanted to
know hvrt Helgi mundi halda Dumbshaf, ef rr vsai honum angat (Landnmabk
1968:250 [S 218]).
There seems little doubt that some Romans were well aware of the value of dealing di-
rectly with those who did the readings, as can be seen in the accounts of their interactions
with the Bructeri prophetess Veleda, based in the area around Kln in the first century AD. Of
Veleda, and the reason for her respect, Tacitus writes in Book 4 of his Histories that she was:

an unmarried woman who enjoyed wide influence over the tribe of the Bructeri. The Germans
traditionally regard many of the female sex as prophetic, and indeed, by an exercise of supersti-
tion, as divine. [] Veledas prestige stood high, for she had foretold the German successes and
the extermination of the legions (Tacitus, Book 4, ch. 61, in Goetz & Welwei 1995:2, 220221;
trans. Tacitus 1975:247).

Veledas importance is demonstrated by the fact that in AD 69, Civilius, leader of a revolt
against Rome sent her a defeated legionary commander along with other gifts (Tacitus, Book
4, ch. 61, in Goetz & Welwei 1995:II, 220221; trans. Tacitus 1975:247). At a later point, she
was presented with a Roman galley as part of the war spoils taken by her people, who trusted
her to make decisions for them (Tacitus, Book 5, ch. 24, in Goetz & Welwei 1995:II, 252253;
trans. Tacitus 1975:285). Tacitus notes:

[] any personal approach to Veleda or speech with her was forbidden. This refusal to permit
the envoys to see her was intended to enhance the aura of veneration that surrounded the proph-
etess. She remained immured in a high tower, one of her relatives being deputed to transmit
questions and answers as if he were mediating between a god and his worshippers (Tacitus,
Book 4, ch. 65, in Goetz & Welwei 1995:2, 224225; trans. Tacitus 1975:250).

The account brings to mind the Danish Iron Age bog offerings of weapons and ships. It also
offers parallels not only to the figure of orbjrg ltilvlva on her own raised platform, but
also orgeirs isolation beneath his cloak. Of particular interest, however, is Tacitus account
of how a Roman commander contacted Veleda directly via secret messages, aiming to per-
suade her to change the fortunes of war (Tacitus, Book 5, ch. 24, in Goetz & Welwei
1995:2, 252253; trans. Tacitus 1975:285286). Also of interest is Statius revelation that in
AD 77, the Romans seem to have captured Veleda, and instead of killing her, decided to keep
her as hostage (Statius, Silvae, 1, 4, 8890, in Goetz & Welwei 1995:2, 260261; see also
Simek 1993:356). It then seems that she might have been directly involved in getting her peo-
ple to accept a pro-Roman king in AD 83 or 84 (see, for example, Todd 1975:203).
This account leads us directly to other accounts of outsiders apparently getting local Ger-
manic/ Nordic people to change opinion by means of taking the radio station (in other
words, controlling the information reaching the people from the gods). The most obvious ex-
amples in this regard are those contained in Rimberts account of Ansgars attempts to con-
vert Sweden in Vita Anskarii (c. 880) which repeatedly underlines how lot casting seems to
work in Christian favour (by some miraculous means). Most noteworthy is the description of
Ansgars second mission from Hamburg to the Swedes in 851852, which, like the rest of the
book, would surely have been well known by anyone in close contact with the archbishopric

349
in Hamburg-Bremen in the tenth century, and its Nordic missionary activities. Hamburgs
early involvement in the conversion of Iceland is well demonstrated by the mentions of the
failed missionary work both of bishop Frirekr of Saxony (see Kristni saga 1858:1, and
slendingabk 1968:18), and angbrandr, who in both Kristni saga and Njls saga is said to
have been Saxon (see Brennu-Njls saga 1954:256; and Kristni saga 1858:8). Neither man
can have been unaware of Rimberts book, even if they seem to have failed to read (or listen
to readings of) it well enough. Others, however, seem to have paid better attention.
As noted above, Rimbert notes several times how lot casting seems to have played a key
role in how the Swedes gained access to the will of the gods (or perhaps Urr and colleagues).
In Vita Anskarii, ch. 26, Rimbert describes in some detail how Ansgar, newly returned to
Birka in 851852, was facing the slightly difficult situation of a renewal of pagan faith (ap-
parently in ancestors). The bishop visited the local king, bringing him gifts, and offering him
a good meal. The king accepted the presents and the meal, but stressed that he personally had
no final say in religious matters. He says simply:

I neither can nor dare give your mission any promises, without first casting lots to ask our gods
for advice, and getting to know the will of the people. Let your mission follow me to the next
ing and I will put your case before the people, and with the gods permission follow your
wishes, and then your desire will work out. If not I will also let you know.

Rimbert then explains that it was the tradition in that country that any official matter depends
more on the will of the people than the kings power (Boken om Ansgar 1986, 54; original
Latin in Jn Hnefill Aalsteinsson 1999:128). The following day at Birka:

The king first of all assembled his chieftains and began to discuss our Fathers mission with
them. The conclusion was that through the casting of lots [sors] they would find out what the
will of the gods was in this matter. As was their usual custom, they went out into the open air
and threw lots, and the lots showed them that it was the will of the gods that Christianity should
be allowed to be founded there.

The ing then somewhat grudgingly accepts the return of priests to the area (Boken om Ans-
gar 1986, 55; original Latin in Jn Hnefill Aalsteinsson 1999:128).
As Ian Wood (1987:5556) noted, Rimberts account may contain echoes of earlier hagio-
graphic works like Alcuins Vita Willibrordi, but, nonetheless, this account seems to reflect
local practice, if only because of its emphasis on lot casting taking place in the open air (in
campum), something that reflects the comment about teinlautar [] velli in Vellekla, st.
30 (see above). Both comments echo the respect for liminal areas and the wild reflected in
other accounts of religious practice, such as the talk of rskotshelgi (ar sem hvrki er akr n
eng) in Hrafnkels saga 1950:120; Skrnirs rune work involving him going off to a nearby
holt to get hr vi [Skrnisml, st. 32]); and even Sigrdrfas situation on Hindarfiall in Ff-
nisml, st. 42.
Rimberts account of the decisions made via lot casting in Birka is echoed shortly after-
wards by another conversion account involving referral to the gods (ch. 30; Boken om Ansgar
1986, 5860). This account relates of how, in 854, a Swedish raiding party found itself in a
difficult position in Kurland. The warriors, besieging a city, but far away from their ships:

decided to find out by way of throwing lots whether their gods would help them in such a way
that they would either win a victory or escape form there alive. And so they threw lots, but
could not find any of the gods who was prepared to help them.

350
Someone suggests that they next try the Christian god by means of the lots, and it was dis-
covered that Christ would help them. They then convince the city to pay them off with a ran-
som, and return home peacefully. In gratitude to the Christian God, they agree to take up fast-
ing and giving alms to the poor because this was dear to Christ (thereby demonstrating what
they saw as the chief aspects inherent in Christian faith).
As Ian Wood (1987:5657) writes:

The augury could [] be no more than a set of variants on a literary topos. [] For an author
trying to depict the Danes as being only just beyond the pale of civilisation auguries could be a
useful narrative device. They were observed by pagan Germans. [] Despite this there is a case
for thinking that auguries did play a part in the Scandinavians search for divine protection,
even if the particular anecdotes recounted by Rimbert are no more than literary constructs. []
Augury and divination [] may well have been employed by the Scandinavians as a means of
adopting a new deity. Moreover the emphasis on augury in the Vita Anskarii may be taken as an
indication that the Christians were prepared to accept the sortes, at least on those occasions
when they supported the process of evangelization; in such circumstances they might be re-
garded as the manifestation of divine, as opposed to diabolic will. [] This apparent acceptance
of the sortes in Scandinavia is a further indication of the overlap between pagan and Christian
worlds. The missionaries seem to have been content to work within social norms rather than in
opposition to them[].

Wood (1987:56) notes that the conversion activities in Iceland in 999/ 1000 also seem to in-
volve divination, and, in Under the Cloak, Jn Hnefill Aalsteinsson (1999:127128) also
notes parallels between orgeirs behaviour and the role of the oracle in Rimberts account.
Neither, however, notes exactly how close the overall performance (for that is what it seems
to be) that took place at ingvellir is to what Rimbert describes as having happened at Birka.
Both performances involve the following features:

a) Awareness that the acceptance of Christianity (or a new god) can not take place without the
formal general acceptance of the people (or their representatives) at a ing meeting.
b) The passing of gifts to the person in charge of interpreting the will of the gods (via lot-
casting or a blt), a person officially recognised as being a pagan believer, and priest/ leader/
representative of the people.
c) The performance of a recognised ritual ceremony of divination in a sacred space (recog-
nised and known about by those at the ing meeting), carried out by those local people re-
spected by the ing as having the ability to interpret the will of the gods, i.e. by those previ-
ously given gifts.
d) The apparent message from the pagan gods (or other divinatory powers) that Christianity
should be accepted (something interpreted by the person given gifts earlier).
d) The same representatives announcement of the sacred decision of the gods in a different
public space in which legal decisions are taken.
e) The official acceptance of this interpretation by those present.

If we accept Jn Hnefills description of the course of events in Iceland, the parallels between
the two series of events in Sweden and Iceland (linked by the likely involvement of Hamburg-
Bremen in both) are striking (see further Sawyer 1987:82). Whether the conversion accounts
in slendingabk, Kristini saga and lafs saga Tryggvasonar are literary motifs based on
Rimberts work or not (I think it unlikely), they certainly seem to suggest that those responsi-
ble for the Icelandic conversion appear to have known Rimberts work or heard stories from
it and were following the peaceful approach to conversion (via local means of divination)
hammered home there on several occasions (see above). To my mind at least, the parallels are

351
more than a coincidence. Indeed, the approach used in Iceland also reflects the statement later
attributed by Adam of Bremen to a Danish king, who reminded Bishop Aalbert that wild
peoples could more easily be converted by persons who spoke their own language and who
led similar lives than by persons unacquainted and strange to their customs (Sandmark
2004:106)
Over and above this, it is worth remembering that, as in Birka, for most people at the
Alingi the so-called conversion was actually little more than an official performance. As
Orri Vsteinsson (2000:18) argues:

There is more reason to think that an official change of religion held no great significance for
those involved and was probably not expected to affect peoples lives greatly []. That it did
not is supported by the slow development of Christian institutions in the eleventh century.

According to Rimbert, for the Swedes in Kurland, Christianity seems to have centred around
avoiding meat regularly for a few days, and giving alms to the poor. For those at ingvellir in
999/ 1000, the only real consequence was a need to be baptised (and the more sensible seem
to have waited until they found some warm water to do this in at Laugarvatn: Kristni saga
1858:25). There were no priests, hardly any churches and few could read. Everyone was al-
lowed to go on eating horse meat, bearing out children (the banning of which could have
caused practical problems for many farmers), and worshipping in private (at this time, most
autumn, winter and spring worship seems to have taken place indoors) (see slendingabk
1968:17; Kristni saga 1858:25; lafs saga Tryggvasonar en mesta 19582000:1, 197). For
most people there was little immediate change. Nonetheless, the formal performance at ing-
vellir (closely following Rimberts model which encouraged deliberate use of traditional
modes of local religious practice) in the long term paved the way for people to be allowed to
take up and encourage the new belief in comparative peace, as it did in Sweden. It also al-
lowed Christians from home and abroad (and most noteworthy from Hamburg where sleifr
Gizurarson later went to study) to enter into influential positions at the Aling. This was to
have much greater cultural and political consequences.

Bibliography
Almqvist, Bo, 1965: Norrn niddiktning 1. Nid mot furstar. Nordiska texter och underskningar, 21.
Uppsala: Almqvist and Wiksell.
The Anglo-Saxon missionaries in Germany, 1954: Trans. by C. H. Talbot. New York: Sheed and
Ward.
Bede, 1965: A history of the English church and people. Trans. by L. Sherley-Price. Revised. Har-
mondsworth: Penguin.
Boken om Ansgar: Rimbert: Angars liv, 1986. Trans. into Swedish by E. Odelman. Stockholm: Pro-
prius frlag.
Brennu-Njls saga 1954: Ed. by Einar lafur Sveinsson. Reykjavk: Hi slenska bkmenntaflag.
(slensk fornrit, 12.)
B, Olav, 1985: Hgtider og minnedagar. Oslo: Det norske samlaget.
B, Olav, 1974: Vr norske jul. Oslo: Det norske samlaget.
Campbell, , & Nyman, . (eds.), 1976: Atlas ver folkkultur, 2: Sgen, tro och hgtidssed (2 vols:
kartor and kommentar). Uppsala: AB Lundeqvistska bokhandeln. Pp. 1215.
Den norsk-islandske skjaldedigtning, 19121915: Ed. by Finnur Jnsson, 4 vols (A.12, B.12.) Kben-
havn: Gyldendal.
Eddadigte I: Vlusp. Hvaml, 1964. Ed. by Jn Helgason. 2nd ed. Kbenhavn: Munksgaards forlag.
(Nordisk filologi, serie A: tekster.)
Eddadigte III: Heltedigte, frste del, 1961. Ed. by Jn Helgason. Kbenhavn: Munksgaards Forlag. (Nor-
disk filologi, serie A: tekster.)
Eirks saga raua, 1935: Ed. by Einar lafur Sveinsson & Matthas orarson. Reykjavk: Hi slenska
bkmenntaflag. (slensk fornrit, 4.) Pp. 193237.

352
Fagrskinna, 1984: Ed. by Bjarni Einarsson. Reykjavk: Hi slenska bkmenntaflag. (slensk fornrit, 29.)
Pp. 55364.
Goetz, H.-W., & Welwei, K.-W. (eds.), 1995: Altes Germanien: auszge aus den antiken Quellen ber
die Germanen und ihre beziehungen zum Rmischen Reich: quellen der alten Geschichte bis zum
jahre 238, III. Darmstadt: Wissenschaftliche Buchgesellschaft.
Gunnell, Terry 2007: a er til fleira himni og jru, Hras: Kannanir slenskri jtr og trar-
vihorfum 20062007. In Rannsknir Flagsvsindum, 8 (Flagsvsindadeild): Erindi flutt
rstefnu desember 2007, ed. by Gunnar r Jhannesson. Reykjavk: Flagsvsindastofnun. Pp.
801812.
Hervarar saga ok Heidriks, 1943: In Fornaldarsgur Norurlanda, I. Ed. by Guni Jnsson and Bjarni
Vilhjlmsson. Reykjavk: Bkatgfan forni. Pp. 191242.
Hrafnkels saga Freysgoa, 1950: Ed. by Jn Jhannesson. Reykjavk: Hi slenska bkmenntaflag.
(slensk fornrit, 11.) Pp. 95133.
slendingabk, 1968: In slendingabk. Landnmabk. Ed. by Jakob Benediktsson Reykjavk: Hi
slenska bkmenntaflag. (slensk fornrit, 1.) Pp. 128.
Jn Hnefill Aalsteinsson, 1999: Under the cloak: A pagan ritual turning point in the conversion of
Iceland (Kristnitakan slandi). Trans. by Jn Hannesson and T. Gunnell. Reykjavk: Hsklat-
gfan; Flagsvsindastofnun.
Kormks saga, 1939: Ed. by Einar lafur Sveinsson. Reykjavk: Hi slenska bkmenntaflag. (slensk
fornrit, 8.) Pp. 201302.
Kristni saga, 1858: In Biskupa sgur, I. Ed. by Jn Sigursson and Gubrandur Vgfsson, 2 vols.
Kaupmannahfn: Hi slenska bkmenntaflag. Pp. 132.
Landnmabk, 1968: In slendingabk. Landnmabk. Ed. by Jakob Benediktsson. Reykjavk: Hi slen-
ska bkmenntaflag. (slensk fornrit, 1.) Pp. 29397.
Martin, Martin, 1999: A description of the western islands of Scotland circa 1695; A voyage to St
Kilda. Edinburgh: Birlinn.
Oddur Snorrason, Saga lafs Tryggvasonar, ed. Finnur Jnsson (Kbenhavn: Gad, 1932).
lafs saga Tryggvasonar en mesta 13, 19582000: Ed. by lafur Halldrsonar. Kbenhavn: Munks-
gaard.
Orri Vsteinsson, 2000: The christianisation of Iceland: Priests, power and social change 10001300.
Oxford: Oxford University Press.
Sandmark, Alexandra, 2004: Power and conversion: A comparative study. Uppsala: Uppsala Univer-
sitet. (Occasional Papers in Archaeology, 34.)
Sawyer, Peter, 1987: The process of Scandinavian christianization in the tenth and eleventh centuries.
In The christianization of Scandinavia: Report of a symposium held at Kunglv, Sweden, 49 Au-
gust 1985, ed. by B. Sawyer, P. Sawyer & I. Wood. Alingss: Viktoria Bokfrlag. Pp. 6887.
Simek, Rudolf, 1993: Dictionary of northern mythology. Trans. by Angela Hall. Cambridge: D. S.
Brewer.
Strmbck, Dag, 1975: The conversion of Iceland: A survey. London: Viking Society for Northern
Research.
Tacitus, Cornelius, 1948: On Britain and Germany. Trans. by H. Mattingly. Harmondsworth: Penguin.
Tacitus, Cornelius, 1975: The histories. Trans. by K. Wellesley. Harmondsworth: Penguin.
Todd, Malcolm, 1975: The northern barbarians 100 BCAD 300. London: Hutchinson.
Theodoricus Monachus, 1998: Historia De Antiquitate Regum Norwagiensium: The ancient history of
the Norwegian kings. Trans. by D. and I. McDougall. Viking Society for Northern Research text
series. London: Viking Society for Northern Research.
Turner, F. W. (ed.), 1977: The Portable North American Reader. Harmondsworth: Penguin.
Wood, Ian, 1987: Christians and pagans in ninth-century Scandinavia. In The christianization of Scan-
dinavia: Report of a symposium held at Kunglv, Sweden, 49 August 1985, ed. by B. Sawyer, P.
Sawyer & I. Wood. Alingss: Viktoria Bokfrlag. Pp. 3667.

353
 Egill Skalla-Grmssonr on the Library Site in Trondheim?
Jan Ragnar Hagland, Norges teknisk-naturvitenskapelige universitet, Norway
The runic inscription currently identified as N-829 in the corpus edition of medieval Norwe-
gian runic inscriptions (Niyr)1 was excavated in 1975 on the municipal library site in Trond-
heim, Norway. It was not satisfactorily interpreted till 1994, when James E. Knirk in a saga-
cious little study solved the difficulties of reading and established a close textual connection
between the form and content of the inscription and the first half of stanza 48 in the slenzk
fornrit-edition of Egils saga Skalla-Grmssonar. Even if somewhat damaged and incomplete
the runic half-stanza can, according to Knirk, and in all probability rightly so, be read as fol-
lows:

S skyli rnar rsta

er ra vel kunni
at verr mrgum manni at []

The corresponding part of the stanza in Egils saga Skalla-Grmssonar reads as follows:

Skalat mar rnar rsta

nema ra vel kunni
at verr mrgum manni
es of myrkvan staf villisk

It is quoted as number four of six drottkvtt stanzas in the part of the saga that deals with
Egils journey to Vrmland in Sweden (ch. 7076) a half stanza that has been dealt with in
various ways in scholarly work over the years (cf. e.g. Olsen 1948).
The focus of Knirks article is, apart from the runological interpretation, the importance of
the runic inscription to the long-lasting discussion of the authenticity of the stanza in Egils
saga. He concludes on this point that most likely the runic verse preserves an older half-
stanza that was remoulded by tradition or by the author of Egils saga into the first half of the
stanza skalat mar rnar rsta (str. 48) and that the inscription provides insight into the
composition of this verse in Egils saga and a confirmation of its status as non-authentic
(Knirk 1994, 41819). By looking upon the runic verse as a source Knirk, then, both explic-
itly and implicitly dismisses the possibility that the runic inscription should be seen simply as
a quotation of the verse in the Egils saga. There is no reason, in my opinion, to disagree with
a position such as this and the arguments used to support it need not be reiterated here, except
that we should be reminded of the fact that the inscribed object in question has an archaeo-
logical dating based on the stratigraphy of the excavation area. The layer in which the stick
was found is dated to the period 11751225. Thus it is slightly older than or roughly contem-
porary with the composition of Egils saga, to quote Knirk (1994a, 419) once again. It should
be added here that the existence of a runic parallel text to the first half of stanza 48 in Egils
saga has, since Knirks discovery of it, been pointed out elsewhere without much further
elaboration, however, on its significance to the composition of the saga text (cf. Hagland
1998, 626f., MacLeod 2001, 419).
The question of authenticity concerning skaldic verse preserved in saga texts in general
and in Egils saga in particular has, moreover, been much debated also in quite recent years

1
Preliminary web-edition http://www.hf.ntnu.no/nor/Publik/RUNER/runer-N774-N894.htm Museum nr N-
25803 in NTNU Museum of Natural History and Archaeology, Trondheim.

354
(for an overview, cf. Jnas Kristjnsson 2006). Even so the aim of the present contribution is
not to continue that line of discussion.
There are, as I see it, questions other than those concerning authenticity of skaldic verse
and of written literary sources that can be asked about the runic verse from Trondheim and the
half-stanza in Egils saga: How are we to understand the relationship between an independent
text, previously unknown to us, such as the runic half-stanza and that of the corresponding
part of a stanza in the saga with which we are occupied here? Does it in any way affect our
understanding of the sagas composition on this particular point or perhaps of saga literature
more in general? The aim of the present contribution is to explore somewhat more in detail
questions such as these.
The existence of a half-stanza about the carving of runes evidenced by the inscription from
Trondheim does not per se need to have anything to do with the use of an almost identical text
incorporated in Egils saga. The runic inscription, as we can all see, contains a not completely
preserved half-stanza in a fairly simple linguistic and metrical form (cf. Knirk 1994, 417). We
do not know why the half-stanza was carved on this inconspicuous wooden stick in medieval
Nidaros. The inscribed object does not seem to have been intended for any other purpose than
that of carrying the inscription that is to say that we have to deal with a so-called rnakefli
in Old Norse terms. We know many runic inscriptions that must have served didactic pur-
poses of various kinds, first and foremost to contribute to the learning and teaching of how to
read and write this particular script (cf. e.g. Knirk 1994 and Seim 1998). Judging from its con-
tent there is potentially a didactic purpose of some sort underlying even the inscription with
which we are occupied here.
From other sources we can, vaguely, see the contours of a lore in metrical form connected
to runic literacy. The so-called rune-poems such as the Norwegian and the Icelandic (cf.
Finnur Jnsson 1915, 2489 and Page 1999) form part of a runic mnemonic tradition about
which we do not know very much. As far as the Norwegian rune-poem is concerned, the age
of the preserved text is uncertain and the transmission of it problematic. It is a fairly simple
set of so-called runhent kvilings, obviously there to help memorize the name of the runes
in the futhark, e. g. as in (F) vldr frnda rge/ fesk ulfr skge etc. As has been pointed
out by Jonna Louis-Jensen (1994, 41) no definite text of the Norwegian rune-poem has yet
been established. The poem as we know it has been tentatively dated in the thirteenth century,
but, as suggested by Aslak Liestl (1977, 308) there are reasons to believe that the Norwegian
rune-poem was preceded by an older one about which little is known. There is, however, evi-
dence for an interrelation between a Norwegian and an Icelandic tradition when runic lore
such as the rune-poems is concerned. Suffice it here just to mention Jonna Louis-Jensens
convincing solution to a name riddle in a runic inscription from the old church at B in Tele-
mark, Norway, dated in the late twelfth or early thirteenth century. The solution to the riddle
rests on the fact that the carver knew a tradition of rune-poems which was closer to the Ice-
landic than to the Norwegian poem as we know them at the present state of research (cf.
Louis-Jensen 1994, 42). This kind of cross-reference seems to indicate the existence of a
common Norwegian-Icelandic lore or tradition of verse connected to runic literacy as far as
the learning and teaching of the futhark were concerned at least (cf. also Hagland 2005).
Whether that also applied to popular knowledge or beliefs about the use of runes, we do not
know very much. There are runic inscriptions that seem to contain references to magico-
medical activities or knowledge (cf. e.g. Dillmann 2006, 55; Gustavson 2002, 4345), the
most well-known of which is no doubt B-257 from Bryggen, Bergen, Norway. The inscription
is dated in the late 14th century and opens with the following two verses: rst ek btrunar/rst
ek bjargrnar[] obviously relating in some way or other to the eddic poem of Sigrdrfuml
(st. 19), other parts of the inscription to other eddic poems (cf. Liestl 1964, 4143). Whether
an inscription such as this should be regarded as an act of exercising magic per se, or whether

355
it should be understood as a reference to or mediating of lore pertaining to the use of runes, is
difficult to decide with any degree of certainty. There has been, I think it is fair to say, a ten-
dency among scholars to perceive of it as a magic act in itself. That, however, is beyond the
scope of our present concern.
Even if quite different in content as well as in form the N-829 from Trondheim and the B-
257 from Bergen may be understood as having some sort of didactic purpose didactic in the
sense of transmitting lore about the use of runes for specific purposes. The N-829 seems to be
a kind of definition of what it takes to be a (successful?) rune-carver, even if we do not have
the entire text preserved: In order to be allowed to carve runes the carver should be able to
interpret them successfully etc. A statement such as this might even be part of a didactic con-
text in the more narrow sense of the term the context of learning how to read and write with
runes. There are a few runological points to consider here. The unusual form of the yr-rune
() in N-829 may be vaguely indicative of an Icelandic runological connection. The inscrip-
tion contains, however, runological and linguistic features that seem to point in several direc-
tions as far as the question of a possible provenance is concerned (for further details cf. Knirk
1994, 416). The runological features of the inscription as such, then, do not seem to point in
any particular direction as far as the geographical origin is concerned. In an international port
such as Nidaros towards the end of the 12th, or beginning of the 13th century a runologically
and linguistically heterogeneous inscription like this may be taken, nonetheless, to represent
and allude to knowledge shared by people from a greater part of the Norse-speaking area than
the local community in which the inscribed object was found.
The B-257, on the other hand, seems to be an attempt at conveying lore about beliefs asso-
ciated with the art of writing in runes, even if, of course, interpretations other than this are
feasible. In the case of B-257 the inscription, regardless of how we understand it, relates to
texts that existed at the time of carving, we must assume when taking the archaeological con-
text into consideration. The intertextual dimension of the runic inscription, as I see it, depends
on texts that existed primarily in oral form, sufficiently well known to the carver and suffi-
ciently known to be recognized and appreciated by an audience of which we know very
little. In the case of N-829 we cannot point out any particular text, written or oral, other than
the version we know from Egils saga. The inscription seems to communicate something
which was widely known or commonly accepted, we may assume: This is what it takes to be
allowed to carve runes etc. There is every reason to understand this as a text that was equally
well known by a Norwegian and an Icelandic audience as part of some popular lore concern-
ing the use of runes more or less on the same level as the rune-poems we may think.
How are we, then, to understand the Skalat mar rnar rsta half-stanza in Egils saga? Ac-
cording to Bjarni Einarsson, years before the Trondheim inscription was know to the interna-
tional community of saga scholars, none of the six drottkvtt stanzas quoted in the Vrmland
episode of the saga should be looked upon as literary sources. They are all to be considered
integral parts of the narrative rather than footnotes as in Snorri Sturlusons kings sagas,
again according to Bjarni Einarsson (1975, 265). The stanza 48 in particular presupposes that
the audience was familiar with the context in which it belonged or was spoken, Bjarni says
(1975, 258). As we have seen already, Knirk is not at variance with this point of view. He
points out, however, that the first half of the stanza is kept in general terms and could have
been used in various contexts, whereas the second half of it corresponds specifically to the
immediate narrative context (Knirk 1994, 418). For this reason Knirk assumes, as we have
seen, that the first half of stanza 48 may transmit an older half stanza remoulded by tradition
or perhaps by the author of the saga.
There is, however, in all probability more to it than that. If the half-stanza as evidenced by
the runic inscription from Trondheim, can be taken to convey a piece of commonly known
lore transmitted in rhymed style, the use of a similar, but not quite identical half-stanza in

356
Egils saga seems to acquire a somewhat different significance. It cannot, as I see it, simply be
the insertion in the saga narrative of a certain stanza a stanza that was remoulded by tradi-
tion or by the saga author, as suggested by Knirk. In the composition of a saga narrative an
insertion like that would as such not significantly surpass the level of mechanical compilation.
Having a commonly known piece of rhymed lore about the art of writing in runes as a hind
carpet, we may, rather than anything else, understand the version given in Egils saga as an
artistically creative allusion to something with which a 13th century audience was well ac-
quainted. Within the immediate context of the saga the negation of this particular item of lore
would be well understood and appreciated, we must assume. The function of this quote
should, then, in my opinion be understood as a deliberate pun made on a piece of lore, com-
monly known, or known at least by the initiated few in an Icelandic saga audience. In addition
the quoted half-stanza represent an elegant and pointed comment almost an ironical aside
we may think to the surrounding prose narrative. The first half of stanza 48 in Egils saga
can, in consequence, be compared to the kind of intertextuality we often find in modern litera-
ture, e. g. as in Henrik Ibsens Peer Gynt: When Peer in act IV, on the occasion of his excited
encounter with the exotic Anitra, feels the urge to quote an esteemed author in order to
comment upon the excellency of the situation, he notoriously does so, as we all know, in the
following way: das ewig weibliche ziehet uns an (Ibsen 1931, 158). The initiated part of
any audience, will, of course, immediately catch the humour and irony behind this rather
crude rendering of Goethes Faust Das Ewig-Weibliche zieht uns hinan, and so on.
Ibsens text, in this particular case, quite explicitly signals the intertextual reference, so to
speak. The first half of stanza 48 in Egils saga, however, does it much more subtly or implic-
itly. It does it to the extent that a modern reader cannot possibly detect the intertextuality at
work until an independent text pops up in this case the runic inscription from Trondheim
revealing to us that there was, in the high Middle Ages, a previously unknown literary context
at play here a context that was in all probability not restricted to the written word alone.
Texts like this must have existed also as part of an orally based culture. The independent runic
text from Trondheim helps us discover that the saga author deliberately makes use of a more
widely distributed literary context than the immediate narrative prose context surrounding the
half-stanza in Egils saga. He does so in a way that certainly adds to the artistic quality of the
saga it may certainly be seen as part of the saga authors art. Bjarni Einarsson (1975, 258)
presupposes, as we have seen already, that the saga audience knew the circumstance in which
stanza 48 was spoken.2 This is to say, if I read him correctly, that the understanding of the
content of the stanza depends entirely on the narrative context in which the stanza is quoted.
Taking the lore indicated by the runic inscription of N-829 into consideration Bjarnis as-
sumption on this point can be justified only in part: The first half of the stanza exploits com-
mon knowledge about runic writing by harping on the particular intertextuality that N-829
reveals to us. This is obviously something that exceeds the framework of the immediate narra-
tive context by universalizing the occasion for a poetic statement. The second half-stanza
brings, however, everything back into the flow of narrative that immediately precedes the
stanza as a whole and continues after it:

[].
sk telgu tlkni
tiu launstafi ristna,
at hefr lauka lindi
langs ofrtrega fengit.

2
Ogs her forudsttes det at tilhrerne ved i hvilken anledning strofen blev kvdet.

357
In a case like this, then, we get a tiny glimpse into the saga authors workshop on a very spe-
cific point. This glimpse certainly shows us a nuance of the stuff of which sagas are made. It
does not, however, seem to add any significant arguments to the long standing discussion of
saga genesis. It cannot, for instance, contribute much either in support or in contradiction to
Per Wieselgrens self-confidant statement that, according to his investigations, Egs., first and
foremost, is the outcome of old oral tradition (1927, 232)3 nor does it affect significantly
other more general statements on these matters. We shall, in consequence, for the present pur-
pose restrict ourselves to keeping in mind the intertextual aspect in the broadest sense of the
term of a small fraction only of the long narrative we know as Egils saga Skalla-
Grmssonar as one of several aspects of saga composition.

Bibliography
Dillmann, Franois-Xavier. 2006. Les magiciens dans lIslande ancienne. Uppsala.
Finnur Jnsson. 1915. Den norsk-islandske Skjaldedigtning II B. Kbenhavn og Kristiania
(Gyldendalske Boghandel-Nordisk Forlag).
Gustavson, Helmer. 2002. [rsberttelse frn] Runverket. Kungl. Vitterhets Historie och Antikvitets
Akademiens rsbok 2002. Stockholm, 4345.
Hagland, Jan Ragnar. 1998. Runes as Sources for the Middle Ages. In: Dwel, Klaus (hrsg.), Runenin-
schriften als Quellen interdisziplinrer Forschung. Proceedings of the Fourth International Sym-
posium on Runes and Runic Inscriptions in Gttingen 49 August 1995. Berlin New York (Wal-
ter de Gruyter), 619629.
2005. On scalds and runes. In: Papers on Scandinavian and Germanic Language and Culture. Pub-
lished in honour of Michael Barnes on his sixty-fifth birthday 28 June 2005. = Nowele vol. 46/47,
pp. 7785.
Ibsen, Henrik. 1931. Samlede verker. Sjette bind. Oslo (Gyldendal Norsk Forlag).
Jnas Kristjnsson. 2006. Kveskapur Egils Skallagrmssonar. Gripla XVII, 735.
Knirk, James E. 1994a. Runes from Trondheim and a Stanza by Egill Skalla-Grmsson. In: Uecker,
Heiko (hrsg.) Studien zum Altgermanischen. Festschrift fr Heinrich Beck. Berlin New York
(Walter de Gruyter), 41120.
1994b. Learning to Write with Runes in Medieval Norway. In: Lindell, I. (ed.), Medeltida skrift och
sprkkultur. Nordisk medeltidsliteracy i ett diglossiskt och digrafiskt perspektiv. II. Nio frels-
ningar frn ett symposium i Stockholm vren 1992. Stockholm (Sllskapet Runica et Medivalia),
169212.
Liestl, Aslak. 1964. Runer fr Bryggen. Bergen [Offprint from Viking 1963].
1977. Runedikt. In: Kulturhistorisk leksikon for nordisk middelalder XXI, sp. 30809.
Louis-Jensen, Jonna. 1994. Norrne navnegder. Nordica Bergensia 4, 3552
MacLeod, Mindy. 2001. Bandrnir in Icelandic sagas. Scripta Islandica 52, 3551.
Olsen, Magnus. 1948. Ti lnnruner. Maal og minne, s. 12224.
Page, R.I. 1999. The Icelandic Rune-Poem. London (Viking Society for Northern research/University
College London).
Seim, Karin Fjellhammer. 1998. Runes and latin Script: Runic Syllables. In: Dwel, Klaus (hrsg.),
Runeninschriften als Quellen interdisziplinrer Forschung. Proceedings of the Fourth Interna-
tional Symposium on Runes and Runic Inscriptions in Gttingen 49 August 1995. Berlin New
York (Walter de Gruyter), 50812.
Wieselgren, Per. 1927. Frfattarskapet till Eigla. Lund.

3
att Egs. vervgande r en produkt av gammal muntlig tradition.

358
 Frheste bersetzungen im Norden:
Konzepte Ziele Traditionen
Agathe M. Hahn, Den Arnamagnanske Samling, Kbenhavns Universitet, Denmark
bersetzungen lateinischer religiser Texte zhlen zu den ltesten berlieferten Produktionen
altnordischer Literatur. Sowohl ihr Medium, die lateinische Schrift, wie auch ihre Ausgangs-
texte sind Kulturimporte, die im Zuge der Christianisierung in den Norden kamen und we-
sentlich zum Anschlu Skandinaviens an Europa beitragen sollten.
Die Texte, die als lteste didaktische Literatur den religis-philosophischen berbau im
Norden vermittelten, zeigen die Aufnahme und Transmission dieser fremden Ideen und Kon-
zepte in ihrem Kern: dort, wo Sprache und Inhalt gelernt wurde.
Mein Vortrag soll dies exemplarisch anhand von zwei Schultexten zeigen: Spakmli
Prospers, einer Prosabersetzung der Epigrammata des Tiro Prosper Aquitanus und dem alt-
nordischen Elucidarius. Es soll ein Einblick in die mitunter recht eigenwilligen bersetzungs-
strategien gegeben werden und der mgliche Zweck der bersetzungen, das anzunehmende
Zielpublikum wie auch der sptere Gebrauch der Werke errtert werden. Ein Vergleich dieser
bersetzungen ist interessant, da der eine der Texte mglicherweise fr die Verwendung in
einer Unterrichtssituation geschaffen wurde, sich auf jeden Fall an ein gebildetes, vorzugs-
weise geistliches Publikum wendet, whrend der andere auf eine breitere, volkstmlichere
Verbreitung zu zielen scheint.
Beide Texte waren im mittelalterlichen Europa als Material fr den Lateinunterricht weit
verbreitet und wurden auch in die Volkssprachen bersetzt. Die altnordischen bertragungen
zeichnen sich jedoch dadurch aus, da sie erstaunlich frh entstanden: Die lteste Handschrift
des altnordischen Elucidarius stammt von ca. 11501200 und ist damit eine der ersten volks-
sprachigen Versionen in Europa und eines der ltesten erhaltenen altnordischen Manuskripte
berhaupt; Spakmli Prospers ist die einzig berlieferte vernakulre bertragung der
Epigrammata, bewahrt in zwei islndischen Fragmenten (ca. 1200 und 1300). Dieser somit
recht begrenzten berlieferung steht der Elucidarius mit einer wesentlich reicheren Tradition
bis ins beginnende 16. Jh. gegenber, wie auch der Tatsache, da die Spuren dieses Textes
auch in der autochthonen altnordischen Literatur zu sehen sind. Die verschiedenen Funktio-
nen der bersetzungen, die sich anhand der berlieferung erahnen lassen, will ich durch ei-
nen Vergleich der bersetzungsstrategien dieser beiden Texte untermauern.

359
 The Phraseological Matrix of the Vlsung-Niflung Cycle
Richard L. Harris, English Department, University of Saskatchewan, Saskatoon, Canada
Extant texts bearing witness to the lengthy oral traditions of the Vlsung-Niflung Cycle range
widely in genre and style, as well as in date and place of origin. Nearly 1000 years separate
the extended digression on Sigemund in Beowulf and the Vlksbuch vom gehrnten Sigfrid, a
chap-book first published in 1726. Texts more significant to our understanding of the story
are 1. in the North, the variously dated Old Icelandic legendary poems of the Edda and the
13th-century Vlsunga saga, derivative of them and in part also of iriks saga, and 2. in the
South, the Nibelungenlied (c. 1200) and later narratives primarily dependent upon it.
The nature of those pre-literary traditions upon which such works are ultimately based
arising in part from 5th-century figures and events of the Germanic Migration Period, but
much embellished with folkloric attenuations must remain forever a mystery. And while
much has been written attempting to clarify the relationships between extant texts, such en-
deavours also lack conclusive usefulness, grounded as they are on complex conjecture and
inflated by widely conflicting arguments and theories.
In this paper I will consider the stories of Sigurd and his world using a point of view quite
different from those of earlier approaches, examining the proverbial corpus of the Vlsung
tradition as it is manifest in those various texts to which we have access. Gubrandur Vigfs-
son, commenting on the significance of proverbs in the sagas, wrote (in his discussion of
Hrafnkatla, Origines Islandicae, 1905, II, 492): These saws are to a Saga what the gnomic
element is to a Greek play. In compiling my on-line Concordance to the Proverbs and Pro-
verbial Materials of the Old Icelandic Sagas [http://www.usask.ca/english/icelanders/] I am
trying to provide systematically presented access to the paroemial record of the Old Icelandic
corpus for use in literary critical studies and textual research. By identifying the proverbial
materials of the Vlsung-Niflung Cycle in this paper and analyzing their uses where they oc-
cur, I hope to identify those patterns of paroemial wisdom inherent in the underlying oral tra-
dition, their development in the extant versions, and the ways in which they affirm the values
and enhance our understanding of the Sigurd stories in their various contexts.

360
 More inroads to pre-Christian notions, after all?
The potential of late evidence
Eldar Heide, Centre for Medieval Studies, University of Bergen, Norway
The Old Norse written sources give us a rich impression of many aspects of pre-Christian
Scandinavian culture and religion. But of some topics they do not give us enough evidence for
a meaningful image. In such cases we seem to have two options: To give up or try to expand
our material. I will argue in favour of the latter. I believe that 19th and 20th century material
can provide breakthroughs in the study of pre-Christian notions. It is deeply problematic, and
it is difficult to use such material but it is possible, and the alternative in many cases is less
satisfactory. I mean all kinds of late evidence, not least lexical material, although the focus
in research has been on folklore.

Background / research history

It seems that most scholars today reject the use of late evidence in the reconstruction of pre-
Christian culture and religion. But in the early 20th century central experts on Old Scandina-
vian religion based their interpretations heavily on 19th and 20th century folklore like Olrik
(e.g. 1901) Celander (e.g. 1911), and Lid (e.g. 1928). Jan de Vries more or less put an end to
this in the 1930s (1931, 1932, 1933), arguing that a thousand years of Christianity had
changed the popular traditions too much by mixing them up with Christian notions and prac-
tices.
Then followed a rejection of the Old Norse prose evidence, too, peaking in the 1950s and
60s (e. g. Sigurur Nordal 1940, Baetke 1951, and Olsen 1966). This hypercriticism was re-
jected during the 1980s and 90s (e. g. Schjdt 1988, 2000, Meulengracht Srensen 1991a,
1991b, Bagge 2002). The late evidence, however, has not yet been reinstated, although there
is a growing tendency to make use of it (e.g. Strm 1991, Bertell 2003, Ahola 2004, Bek-
Pedersen 2006, 2007, Heide 2006a, 2006b, 2006c). Such works are still exceptions and oc-
curred even during the hypercritical period (e.g. Rooth 1961). What is accepted within todays
paradigm is, generally speaking, only Old Norse evidence and earlier ethnographic, archaeo-
logical and iconographic evidence.

Objections to the objections

The continued rejection of late evidence is somewhat peculiar, for several reasons:
The arguments for rejecting the late evidence are in principle the same as those used for
rejecting the Old Norse prose evidence: Because they are far younger than the pagan period
they are untrustworthy evidence of pagan traditions. If one in spite of this accepts the Old
Norse evidence it seems inconsistent to completely reject later evidence. There is a big differ-
ence of degree but no difference in principle. What we do is deeply problematic no matter
what. There is no safe ground to seek refuge on.
The rejection of late evidence also seems inconsistent with the broad acceptance of Indo-
European studies. If it is possible to reconstruct parts of an Indo-European tradition after
thousands of years of contamination, how can it be impossible to do something of the same
after less than one thousand years?
Finno-Ugric pre-Christian studies have always been dependent upon late evidence and
remains a respected field in spite of this. Why are the same methods impossible in our field?

361
 Within Medieval Scandinavian history most scholars again accept the retrospective
method that based upon 17th century evidence makes assumptions about farms and land dis-
tribution half a millennium earlier.
Place-names are again accepted as evidence for pre-Christian religious conditions, even
though many of them are not attested from the Middle Ages.
Many scholars occasionally use late evidence although they seem to accept the ruling
paradigm (e.g. Clunies Ross 1981: 379, Lindow 1990, Drobin 1991: 118 ff.). One might ex-
pect this to produce dissatisfaction with the paradigm or at least an explanation of why the
late evidence is acceptable in those cases but this rarely happens.
Some works relying on late evidence remain widely accepted or influential, e.g. Magnus
Olsens explanation of the proverb reynir er bjrg rs in Skaldskaparml (Edda Snorra St.:
106), his explanation of the word varlok(k)ur in Eirks saga raua (: 412), and his interpreta-
tion of Skrnisml (Olsen 1909, 1916, 1940).
Some scholars recognize the potential of folklore but ignore it because they believe it is
impossible to filter out the valuable information from the rest (e.g. Strm 1985: 8,
Steinsland 2005: 64). This view may be shared by many.
But a few scholars claim to reject late evidence on principle and completely. This position
is an illusion, however, because there is so much in our understanding of the Old Norse texts
that is based upon late evidence. Examples of this are hrll m., which is mentioned in Dar-
raarlj and refers to a part in the warp-weighed loom; gagl n., which refers to a wild goose;
flannfluga f., which refers to a woman running away from her fianc, literally from the male
member; the verb rbenda, which means to bind together from both sides and is derived
from the neuter *rband, which refers to robands (that attach the sail to the yard) but which is
not attested in Old Norse, and many bynames, like bellingr. Their meanings do not emerge
from the Old Norse occurrences but are easy to reconstruct on the basis of the modern Scan-
dinavian languages, and are accepted by everyone. Examples like these are abundant although
it is often not realized because the explanations are seen as so obvious that the reasoning is
not explicit. Mythological examples can be mentioned as well, e.g. the giantess Skais epi-
thets ndurds and ndurgo (Edda Snorra St.: 31). Everyone agrees that ndurr m. means
ski and refers to Skais skiing. But this does not emerge from any Old Norse source. The
meaning of ndurr is taken from the Modern Scandinavian languages (Fritzner 188396 I: 54,
III: 1088). Other mytho-religious examples are the word skeggbroddr m., which has to do
with the god rrs ability to raise a gale by blowing in his beard (Perkins 2001, Heide 2006a:
284 ff.), and vrr in varlok(k)kur, which refers to some kind of guardian spirit.
I hope this is enough to demonstrate that in some cases all of us accept late evidence. Then
I think we should leave behind the question of whether it is possible to use such evidence and
concentrate on how and to what extent it is advisable and desirable.

Why is it necessary and possible? The nature of our evidence

Late evidence is valuable because lack of information is an even bigger problem for us than
unreliable information. Our oldest evidence represent only a small piece of past reality. Even
in Iceland only a small part of the traditions that existed in the 13th and 14th centuries were
written down in manuscripts that have reached us. From the rest of Scandinavia virtually
nothing has reached us. Accordingly our most reliable sources are completely insufficient. Of
most past notions they give us no image at all; of others they give us an image that we cannot
understand because it is too scant. But parts of what was not written down may have survived
in popular traditions until it was collected by folklorists, lexicographers, and others in modern
times. Of course most of the pre-Christian traditions were gone by then but it appears that

362
fragments have survived, especially of simple and basic notions. Such fragments may help us
a lot.
I will give an example. According to the Old Norse evidence one of the names of Lokis
mother was Nl (also Laufey. Edda Snorra St.: 34, 100, Srla ttr: 275). It apparently means
needle but there is nothing whatsoever in the Old Norse evidence nor archaeology nor ico-
nography that can tell us how this is to be understood. But in 19th and 20th century traditions
there is. In Sigfs Blndals Icelandic dictionary (1920: 511) there is a saying which connects
Loki to the sewing needle: If one licks the end of a thread in order to thread a needle, one
licks Lokis backside (sleikja rassinn honum Loka) as if Loki could be conceived of as a
thread. In that case his mother being a needle makes sense. In popular notions we also find a
link between sewing needles and motherhood: One should never give a woman a needle with-
out the thread in it because that would cause her never to have any children (Scotland, Mac-
Culloch 1936: 255), or: A woman should never let someone else thread the needle for her
because this would give her difficult deliveries (Sweden, ULMA 10071, p. 37). In both cases
there is apparently an idea that the thread is the baby and the needle the mother. This makes
sense because the needle is a natural symbol of women and their traditional work, and the eye
of the needle has a shape very suggestive of the female genitals. This again corresponds to the
comparison between threading a needle and sexual intercourse, which is widespread. Ice-
landic tradition also has an explicit connection between Nl as the name of Lokis mother,
and threads and sewing. In Icelandic, the appellative loki m. among other things means a
knot or tangle (on a thread), which could be personified and identified with the person Loki.
When a knot appeared on a thread during sewing or spinning, a verse was pronounced, during
the disentangling of the knot:

Styr heitir hann fair inn. Spearhead your father is called.

Sknl heitir hn mir n, Shoe needle your mother is called.
au skulu bi stinga rassinn r, They should both prick you in the arse
ef u fer ekki upp af rinum. if you will not leave the thread.
(Guni Jnsson 1954: 189)

In one version of this verse Lokis mother is called just Nl needle, like in the Old Icelandic
accounts (Recording SM 85/585, Sigrur Gsladttir, 1970, Hlmavk.).1
Are we really better off without material like this? Without it we seem to have no chance
of success but with it there is at least a theoretical chance. I prefer uncertain possibility to cer-
tain impossibility.
To this some may object that we do not usually face a situation like this. In most cases the
Old Norse sources, perhaps supplemented with other early evidence, give enough information
for a meaningful image. Therefore, in most cases late evidence is not needed. Certainly but
there are also many problems that have not been solved during more than a century of ingen-
ious study of the Old Norse sources like the god Loki and the sorcery form seir. The Old
Norse sources give a lot of information about these things but apparently not enough. Sup-
plementary information from late evidence might help us.
Some scholars fear that the acceptance of late evidence will lead to speculative interpreta-
tions because we will be lead astray by false evidence. This certainly is a risk but a strict
source criticism may lead to the same because the smaller the amount of evidence, the fewer
interpretations are contradicted by it and the smaller the chance that it contains the neces-
sary material (cf. Meulengracht Srensen 1991b: 243). My experience is that I am guided by
the material as it accumulates, even if parts of it come from late evidence. The low reliability
of 19th and 20th century information does not mean that it is worthless. It only means that we
1
Thanks to Rsa orsteinsdttir at Icelands Arnamagnan Institute for this information.

363
need more pieces of information in order to build something that we can believe in. But that is
OK because there are ten times more of them available.
We should not ignore the difficulties of late material but neither should we ignore the pos-
sibilities. There is so much evidence out there that we do not even know about but which may
hold keys to our problems like the notions of needles and threads, which have not been
taken into account in any discussion of Loki. Why should we not even look for this kind of
material?

How can we do it?

In order to find useful late material, we have to look for it in dictionaries, folklore collec-
tions, archives, and so on. Because of the long-lasting negative attitude towards late evidence
it is easy to find new and interesting material.
The problem is to determine if the material is of value for a certain interpretation. One
cannot just assume that a certain piece of late evidence is valid for ancient times. That will
always have to be demonstrated in some way or other. I will discuss some of those ways. De
Vries was willing to accept folklore when it gave the same picture as the Old Norse sources
because in that case it could support interpretations of the Old Norse material (de Vries 1931:
60 ff.). This may be so, but in my opinion different information is even more valuable because
we are most in need of information that can give more parts of the image. Such information is
more difficult to use, of course, but its potential is bigger. Schjdt (2000) points out a crite-
rion that seems fruitful: If individual pieces of information that we do not trust one by one,
together form a pattern that corresponds to something we know from reliable sources, then we
may believe in the information. Schjdt discusses the legendary sagas but the reasoning
should be universally valid. But in 19th and 20th century traditions, most of the potentially
valuable material does not belong to such patterns. Therefore, what we need most is criteria
for validating isolated evidence. I will mention some ways to do this; first some ways to infer
from late evidence alone:
It explains the evidence. This may seem like a problematic criterion but in some cases is
not, for instance in ndurr in ndurds and ndurgo. The meaning ski is accepted because
it fits perfectly; it explains all the occurrences in a consistent and plausible way.
Occurrences geographically far apart, for example: The word gand- may mean staff,
pole in Iceland as well as in Swedish Ostrobothnia in Finland but not in the areas between.
This common meaning can hardly be borrowed in recent times because there never was much
contact between Iceland and the Gulf of Bothnia. Therefore the gand- meaning staff, pole
probably is a common heritage, i.e. from Proto-Nordic times or earlier even if it is not at-
tested in Old Scandinavian manuscripts (Heide 2006a: 124).
Widespread motifs. In other cases the widespreadness of a motif can tell us that it is an-
cient. One example of this is the notion that the soul or spirit is the same as a persons breath
(Heide 2006b). We have no evidence telling us that in the Viking Age there was a notion of a
breath soul. But still we can be quite sure that it was there because there is abundant evi-
dence of the connection between soul and breath in late traditions and languages from north-
ern Europe, the Classical world and most of the rest of the world. The derivation of the soul
from breath seems to be nearly universal. In that case it is farfetched to claim that this notion
is borrowed from somewhere in recent times.
Differing forms of a word, for example: A word *alfskot is not attested in Old Norse
manuscripts but in all probability it existed, because the modern Norwegian dialect forms of
this word differ so much: alvskot, algskot, lmskot, lskot etc. (ibid: 229).
Cultural fossils may be utilized with the help of etymology. Place names are the best
known example. For instance, farm names like Ullevi and Torshov are petrified fragments

364
from pre-Christian times, handed down to us through the centuries and still telling us about
religious conditions back then (cf. Olsen 1915). In the same way names of plants, birds, in-
sects, stars, stellar constellations etc., may be cultural fossils from ancient times. One example
is terms like Norwegian solulv, Swedish solvarg, English sundog, Norwegian (sol)gil, Ice-
landic gll / gll, lfur, etc., which refer to parhelia or mock suns to the left and right of the
sun under certain atmospheric conditions. These terms seem to be reflections of the same an-
cient traditions as we see in Grmnismls account (st. 39) of the wolves trying to swallow the
sun (Heide 2006a: 206 ff., 220 f.).
Motif cannot be derived from Christianity. One example of this is dialect-Swedish month
names like jultungel and dis(tings)tungel (Dalarna), which seem to be relics of the pre-
Christian calendar and give us information of that (Nordberg 2006: 15 ff.). This we can know
because they cannot be derived from the Christian calendar.
Freezer in neighbouring culture, for example: Scandinavian names of gods are re-
corded in the 18th century evidence for Saami religion, e.g. Hovrengaellies (Horagalles)
from *rkall. This may give information of the Scandinavian gods.
Motif is not tradition dominant. If a motif appears to be dominant in a certain tradition,
there is a good chance that it has spread to more characters and narratives than were attributed
to it in earlier times (Eskerd 1947: 79 ff.). If, on the other hand, the tradition only contains
scattered information about a motif, the chance is high that that information is a relic of some-
thing old. It is the lectio difficilior of the tradition, so to speak.
In most cases, however, one will combine late and early evidence in a way that seeks to
anchor the late evidence in the past. For instance, the name Nl of Lokis mother indicates
that notions of needles and thread and motherhood existed at least as early as in the 13th cen-
tury; probably earlier. Another example is the staff etymology of gand-, which may be an-
chored by place-names containing the element gand-. They are strikingly associated with
staff-shaped fjords and lakes, and of course their shape was the same in the past. In most cases
the anchor will be evidence from early written evidence and archaeological finds but it may
be other things, like topography in this case. Distinctiveness is a criterion for making such
connections: The more distinctive a motif is, the less likely is it that it is found in separate
areas or periods without connection. One can also use late evidence as an idea bank in the
interpretation of the oldest written evidence. Once one has seen a certain pattern in the richer
late evidence, one may discover it even in the scanter, early evidence; details that have es-
caped ones notice may get a new meaning and fit into a pattern that one has seen in the late
evidence (e.g. yawning in Old Norse evidence, see Heide 2006b; cf. Schjdt 2000: 38).
With the help of criteria like these it is possible to extract probable data about pre-Christian
times from late evidence. But still, of course, such data are less reliable regarding pre-
Christian times than a clear statement from e.g. an Eddic poem. Therefore, many pieces of
evidence of this type are needed to support each other, preferably in combination with early,
more reliable evidence. Because of this one should seek to scan through large amounts of data
when utilizing late evidence. But then success seems possible. If a pattern that one sees is
confirmed again and again, even by unreliable evidence, one might be on to something.

The culture-etymological approach

Olriks generation produced many works that made extensive use of late evidence. Some of
them are influential today, but most of them are not. This may be because the scholars were
led astray by the late evidence. But it may also be because the late material was not used in
the best way. It was not common to seek to validate late information with the help of criteria
like those listed above. But more importantly, one did not demand that all the information of a

365
phenomenon be seen in relation to each other. This is not customary today, either, but in my
opinion this demand is essential.
As an illustration, I will take gandr, or gand-(ur), which was the topic of my PhD disserta-
tion, and which is found in Old Norse sources as well as modern Norwegian, Faroese and
Icelandic traditions, and in the evidence for Saami religion. It struck me how much the results
on gand- differed and how this seemed to result from the scholars studying different parts of
the evidence. Because of this I decided to study all the forms of gand- in relation to each other
(cf. Rooth 1961: 8), as they all ought to be connected in some way. There ought to be a link
between the variants, close or distant, because they all contain the word gand-, which can
hardly have arisen by mutation.
In this perspective the variants of gand- may be compared to the branches of a tree, or the
preserved manuscripts of a textual critic, or the ramifications of languages in a language fam-
ily, or the ramifications of a word in etymology. Some traits probably are snagged on the
gand- tradition secondarily but still there must be a reason why just that trait got snagged on
gand- in particular rather than some other phenomenon. We can assume that the early gand-
tradition had similar traits, making it natural to associate the new trait with gand- in particu-
lar. Accordingly, even corruptions have some information value. Therefore I tried to find a
model that could explain the total evidence of gand-, the relationship between the different
forms. This meant an essential cluster of notions from which the variants may be understood
as representations or derivations, similar to the archetype in text criticism or the reconstructed
forms marked with *asterisks in etymology.
As an illustration I will take the excerpt from my dissertation that I presented in Durham
(Heide 2006b). The main meaning of gandr in the medieval evidence is soul or spirit sent
forth (in shape). But in 19th century Shetlandic, gander has meanings that appear to deviate
completely: strong gust of wind, sudden powerlessness, and nausea, vomiting. I sug-
gested that these meanings are derived from a cluster of notions that can be reconstructed as
follows:
Spirit = breath = air in motion, wind.
Spirits (= breath) can leave and enter body through respiratory passages / throat.
Entering spirit replaces bodys own spirit => powerlessness.
Spirit entering down throat forces out stomach content; = vomiting.
This is all logical if one conceives of a persons spirit as the same as the persons breath,
and this conception can be supported by a lot of material from Old Norse evidence as well as
Northern European folklore and Eurasian comparative material.
This approach is not using late evidence in the interpretation of Old Norse evidence. One
should rather conceive of all the evidence, including the Old Norse evidence, as reflections of
an original or essential cluster of notions which is the objective of the reconstruction. There-
fore, this may be called a culture-etymological approach. One should only try to reconstruct
notions not narratives so this is not the same as the historic-geographic method of classic folk-
loristics.
The outlined approach does not imply a claim that all evidence really is ancient or that
all evidence is equally valuable. It just recognizes late attestations and late forms as ade-
quate input in calculations, and demands that all forms are accounted for, placed in relation to
the rest of the material. Some forms may be connected more closely to the essential cluster of
notions, others may be identified as digressions or corrupted variants just like some of the
word forms, manuscripts and languages in etymology, textual criticism or language history
are more distant from the origin than others. But they, too, have to be accounted for and
placed in relation to the others as part of the total reconstruction.
In the mentioned disciplines, this goes without saying. It should not be the other way
around in the reconstruction of pre-Christian Scandinavian religion and culture. We should

366
base our interpretations upon the total evidence for the phenomena we are studying, also the
late examples. This not only will give us more evidence like the Shetlandic forms of gand-.
It may also make it easier for us to break out of the limitations of our presuppositions and our
21st century western middle class backgrounds. We should not allow ourselves simply to ig-
nore the evidence that appears not to fit in.
A cluster of notions reconstructed in this way cannot be placed very accurately in time or
space. It is hypothetical and it is not always certain that it ever existed. This may seem like a
serious objection to the approach but the same is usually the case with the reconstructions of
etymology and textual criticism, too. (For example, the Hrafnkels saga Freysgoa that we
read is reconstructed on the basis of late manuscripts.) If it is acceptable there, it should be
acceptable here. A hypothetical archetype may be methodologically necessary and the best
we can aim at. But of course it is important to be aware of the status of our results. It is also
important to be aware that this kind of reconstruction only gives one part of the image. Still,
that part can be interesting.
I use the comparisons with etymology and textual criticism to illustrate a way of thinking
and as examples: It is not impossible to reconstruct something from a remote past on the basis
of material that is far later and in addition is scattered geographically and chronologically. But
of course there are significant differences. What one seeks to reconstruct in those disciplines
is more focused than what can be reconstructed with a culture-etymological approach. It is a
specific word(form) or text rather than a loose cluster of notions. Moreover, religious or cul-
tural reconstructions cannot be based upon sound-laws and common errors, although sound-
law reasoning can be useful in such studies. Instead, association and common sense will have
to do, alongside with the criterion that the model of understanding that explains most of the
evidence should be preferred. This may sound flimsy but the semantic half of etymological
reasoning is in exactly the same situation. (The change in the meanings of word forms is as
arbitrary as that of cultural change.) However, the decisive criterion is the inter-subjective
judgement of the scholarly community. Can a culture-etymological approach produce inter-
pretations that many competent scholars find interesting and plausible?
The utilization of late evidence is not something that has been tried once and for all and
proven impossible. There are better ways to use it and there is a lot of material that has yet to
be discovered. I would like to invite those of you who recognize the potential of late evidence
to join me in the forming of a network that can develop and stimulate the utilization of such
evidence. The time and place for a meeting will be announced at the conference.
Thanks to Karen Bek-Pedersen, Odd Einar Haugen, Judith Jesch, Hkan Rydving, and Jens
Peter Schjdt for comments on a draft of this paper.

Bibliography
Ahola, Joonas, 2004: North European Popular Heroes: Representing the past?. In Jn Haukur Ingi-
mundarson (ed.): Northern Veche. Akureyri. 15153.
Baetke, Walter, 1951: Christliches Lehngut in der Sagareligion. Berlin.
Bagge, Sverre, 2002: Mellom kildekritikk og historisk antropologi. Historisk Tidsskrift 81. 173212.
Bek-Pedersen, Karen, 2006: Oppositions and Cooperations in the Baldr Myth, with Irish and Welsh
Parallels. Journal of Indo-European Studies 34. 526.
, 2007: A Myth in Folktale Clothing?. Journal of Indo-European Studies 353/4. 111.
Bertell, Maths, 2003: Tor och den nordiska skan. PhD thesis, University of Stockholm.
Celander, Hilding, 1911: Lokes mytiska ursprung. Uppsala.
Clunies Ross, Margaret, 1981: An interpretation of the Myth of rrs Encounter [] Speculum
Norrnum. 37091.
Drobin, Ulf, 1991: Mjdet och offersymboliken []. In L. Bckman (ed.): Studier i religionshisto-
ria. Lberd. 97141.
Edda Snorra Sturlusonar, 1931. Ed. Finnur Jnsson. Copenhagen.

367
Eirks saga raua, 1935. In Matthas rdarson (ed.): Eyrbyggja saga. Reykjavk. 193237.
Eskerd, Albert, 1947: rets ring. Etnologiska studier i skrdens och julens tro och sed. Stockholm.
Fritzner, Johan, 188396: Ordbog over Det gamle norske Sprog 13. Kristiania.
Guni Jnsson, 1954: slenzkir sagnattir og jsgur X. Reykjavk.
Heide, Eldar, 2006a: Gand, seid og ndevind. PhD thesis, University of Bergen.
, 2006b: Spirits through respiratory passages. In J. McKinnel (ed.): The Fantastic in Old Norse /
Icelandic Literature. University of Durham. 35058.
, 2006c: Spinning seir. In A. Andrn (ed.): Old Norse religion in long-term perspectives. Lund.
16470.
Lid, Nils, 1928: Joleband og vegetasjonsguddom. Oslo.
Lindow, John, 1990: Loki and Skai. In Snorrastefna. Reykjavk. 13042.
MacCulloch, J. A., 1936: The misty isle of Skye. Stirling.
Meulengracht Srensen, P., 1991a: Om eddadigtenes alder. In G. Steinsland (ed.): Nordisk
hedendom. Odense. 21728.
, 1991b: Hkon den gode og guderne. In P. Mortensen (ed.): Fra stamme til stat i Danmark 2.
rhus. 23544.
Nordberg, Andreas, 2006: Jul, disting och frkyrklig tiderkning. Uppsala.
Olrik, Axel, 1901: Odinsjgeren i Jylland. Dania VIII. 13973.
Olsen, Magnus, 1909: Fra gammelnorsk myte og kultus. Maal og Minne. 1736.
, 1915: Hedenske kultminder i norske stedsnavne. Kristiania.
, 1916: Varlokur. Et bidrag til kundskap om gammelnorsk trolddom. Maal og Minne. 121.
, 1940: Reynir er bjrg rs. Maal og Minne. 14546.
Olsen, Olaf, 1966: Hrg, hov og kirke. PhD thesis, Univ. of Copenh.
Rooth, Anna Birgitta, 1961: Loki in Scandinavian mythology. Lund.
Schjdt, Jens Peter., 1988: Forskningsoversigt. Hovedtendenser []. Danske studier. 13353.
, 2000: Forskelle i anvendelsen af islndingesagaer og fornaldarsagaer []. Artikler udgivet i
anledning af Preben Meulengracht Srensens 60 rs fdselsdag. rhus. 3446.
Sigfs Blndal, 1920: Islandsk-dansk Ordbog. Reykjavk.
Sigurur Nordal, 1940: Hrafnkatla. Reykjavik.
Steinsland, Gro, 2005: Norrn religion. Oslo.
Strm, Folke, 1985: Nordisk hedendom. [Gteborg].
Strm, ke, 1991: Folklore och nordisk myt. In L. Bckman (ed.): Studier i religionshistoria.
Lberd. 34348.
Srla ttr, 1860. In Flateyjarbk I. Christiania. 27583.
Vries, Jan de, 1931: Contributions to the study of Othin []. FF Communications. 94. 179.
, 1932: Die Bedeutung der Volkskunde fr mythologische []. Germanisch-romanische
Monatsschrift 20. 2739.
, 1933: The problem of Loki. FF Communications 110. Helsingfors.

368
 A Short Report from the Project on Codex Upsaliensis of
Snorra Edda
Heimir Plsson, Dept. of Scandinavian Languages, Uppsala University, Sweden
In my capacity as one of the group of researchers at present working on the Uppsala-Edda,
Codex Uppsaliensis, MS DG 11 4to in The University Library in Uppsala, I am now quite
convinced that when we talk about The Prose Edda, we are actually dealing with more than
one mediaeval edition or version of the Edda. It is common to talk about MSS Gks. 2367 4to
(Codex Regius, referred to as R), AM 242 fol. (Codex Wormianus, referred to as W) and
Utrecht MS nr. 1374 (Codex Trajectinus) as one version (referred to as the RTW version);
and DG 11 (from Uppsala) as another, although, as many scholars have earlier pointed out, it
is probable that the Codex Wormianus should also be treated as a special edition. Indeed, the
same may be said about certain other mediaeval mss. not discussed in this short article.
My main question today is that of whether these two main versions of the Edda, that is to
say the RTW version and the U version really have the same origin or whether they can even
be treated as copies of the same manuscript, as some have suggested.
The key point here is that, as Anthony Faulkes has earlier stated, The version in the Upp-
sala manuscript differs from the others in all parts of the Edda, but particularly in Skldska-
parml, where various passages and verses are absent, the material is very differently ordered
and the whole structure of the work is different (Faulkes 1998:xii). These differences suggest
that there is good reason to consider the U as being quite unique and thus worth close investi-
gation.
The relationship between the U and the RTW versions has been long discussed from vari-
ous different viewpoints. Unfortunately most of these discussions start and finish with the
question of which of the versions is closest to Snorri Sturlusons original work. In short, there
are two main research viewpoints here, one (let us call it the German school, from Mogk to
Mller) suggesting that the Uppsala Edda is a near copy of Snorris draft of the work, while
the other group (Finnur Jnsson, Sigurur Nordal and many others) claims that U is a tasteless
cut version of the geniuss masterpiece As Anne Holtsmark put it (Holtsmark 1950:vii) Tek-
sten er atskillig kortere enn den i RWT, sannsynligvis er den dradd sammen av den som skrev
U eller dens nrmeste forelegg. Faulkes (2005:xxviii) had a similar opinion arguing: It has
been subject to extensive verbal shortening, with the result that in many places the text hardly
makes sense. Various passages that are in the other manuscripts are lacking, and much of the
material that is included is in a different order. Since Faulkes is nowadays seen as a leading
figure in research into The Prose Edda, his words carry a lot of weight. There is however, a
third view, namely that in the thirteenth century, the Edda was regarded as a work in progress.
In this connection, we can quote Sverrir Tmassons statement that:

[] skum ess hve frvikin milli einstakra handrita eru sums staar mikil, m einnig gera r
fyrir a egar 13. ld hafi handbkin Edda veri til mrgum gerum; texti eirra handrita
sem vi eigum n af bkinni s a vsu upphafleg sm Snorra, en hafi veri aukinn me
athugagreinum annarra frimanna 13. ld, eins og reyndar er algengt um verk klassskra
hfunda og frimanna alls staar annars staar Evrpu (Sverrir Tmasson 1992:534).

Elsewhere, Sverrir has reflected that:

[] en af athugunum frimanna er ljst, a textahef verksins virist vera af tveim rtum

runnin og vands hvort rekja megi r til eins sameiginlegs upphafs, verks Snorra Sturlusonar
ea hvort a verk hefur fr ndveru veri ein heild. Niurskipan efnis hfuhandritunum er

369
 mismunandi og gti veri vsbending um a ritstjrar byrjun 14. aldar hafi sett verki saman
eftir tveimur ea fleiri handritum (1996:3).

Faulkes (and Holtsmarks) approach when considering the possible cuts found in the U
manuscript centres on the basic assumption that once upon a time there was only one original
Edda, and that this Edda was then copied, distorted and miscopied, resulting in (at least) two
different versions, a theory that fits well into the classical approach adopted with regard to the
discussion of the Icelandic Sagas. In other words, in the beginning there was a more or less
perfect original which was later copied and recopied, leading eventually to the shape of the
extant texts we know today.
All of the opinions noted above encourage us to do our very best to get as definite an an-
swer as possible to the question of whether the two main extant versions of Snorris Prose
Edda can actually be traced back to the same textual source. My intention here is to try and
present some new arguments that have relevance to this question.
As early as 1992, Franois-Xavier Dillman (in Textafri og goafri: Um rfina be-
tri tgfu Snorra-Eddu) pointed out that U obviously contains the original version of the
Njrr-Skai myth, and that in this case, the U text can certainly not be explained as a correc-
tion of the RTW text. It must thus go back to another original. In Scripta Islandica, 59/2008,
Lasse Mrtensson and the present author wrote another article on the suspensions pointing out
that in the cases of some of the quotations from The Poetic Edda (especially from Vlusp
and Vafrnisml/ Grmnisml) and from jlfr Arnrssons Sexstefja, it again seems
clear that the scribe of U must be copying written sources different from those which were
behind the original of the RTW version. Maja Bckvalls studies of other Eddic quotations in
Gylfaginning (Bckwall 2008), point in the same direction. The time seems right to take an-
other close look at the overall discrepancies between the two main versions of the Prose
Edda. In the following, I will only present the results of a close study of the vocabulary used
in the U and R manuscripts, more precisely that used in Gylfaginning and the main myths.
Sigurur Nordal, in his introduction to Egils saga published in slenzk fornrit, II (1933), af-
ter having pointed out that a close examination of the differences between the text of the saga
contained in Mruvallabk (M) and in the fragment usually called the eta-fragment of
Egils saga reveals that the younger text (M) was some 510% shorter, states that Snorra-
Edda hefur, svo sem kunnugt er, stt svipari mefer Uppsalabk, a styttingin s ar
ger af miklu minni varfrni og smekk, enda myndi enginn kannast vi stl Snorra eirri
mynd. Here Nordal is referring to his own research published in a book on Snorri Sturluson
written several years earlier (1920). D. O. Zetterholm in his study, Studier i en Snorre-text
(1949:73 ff.) examines the texts usually mentioned as shortenings, comparable to Edda (all in
all parallels found in nine so-called kings sagas and sagas of Icelanders.1 From his own
studies of the two versions of Fr rs til tgaraloka in the Prose Edda, Zetterholm drew
the conclusion that U was a shortened version, while R represented an expanded text.
Both Nordal and Zetterholm, like the other scholars noted above, based their conclusions
on close comparisons of the texts studies of limited sections of the Edda. Since it is obvious
that the differences vary quite a bit, I have limited myself to comparing digitalized texts of the
whole of the extant Gylfaginning as it is in the Uppsala version and the actual texts in Regius
(mostly in Gylfaginning, but some in Skldskaparml), along with the texts of the myths con-
cerning the dwarfs blacksmith work and the Andvari-gold in Skldskaparml. Instead of
comparing the average differences of the whole, I have compared the manuscripts piece by

1
In this connection scholars tend to ignore that they really are comparing an apple to an orange. Edda represents
a totally different genre than do the sagas.

370
piece, isolating the myths and the longer epic tales on one hand, and the informative texts
about the gods and the cosmos on the other. As Table 1 reveals, the so called U shortenings of
the text vary quite a bit.

Table 1: The length of seven individual episodes counted in words2

Contents Main characteristics 1 DG 11 Reg DG %
4 Words Words of R
01 The main crea- Very similar originals, but if the same then a 2 1693 2562 66%
tion myths very conscious redaction in U
02 Lokis off- Same original. 1 1153 1158 99,5%
spring
03 rr and Almost exactly the same. 1 1063 1109 95,8%
Hrungnir
04 tgara-Loki Too different to make the idea of a common 3 2066 3594 57%
original convincing. Maybe different oral ver-
sions?
05 Freyr and U does not know of Skrnisml. 4 150 454 33%
Gerr
06 The mead of Far too different to have been copied from the 4 392 924 42%
poetry same original.
07 Iunn and jazi Far too different to have been copied from the 4 452 996 45%
same original.

I have limited myself here to just seven episodes of some 16 that I have been working on, in
order to show some of the most important features. My own conclusions are given on a four-
point scale (1-4), where 1 means almost certainly the same original and 4 hardly possible
to imagine the use of the same original. As one can see in the column farthest to right, word-
ing of the episodes in U vary in length from being 33% of the length in the Regius-version, to
100%. This, of course, does not prove that the U-version can not be the end result of a com-
pletely thoughtless (or to use Finnur Jnssons favorite word, vilkrlig) shortening carried out
by a tasteless scribe or editor (as Nordal suggests), although (as an old teacher) I must note
that the perfect handwriting of the U scribe certainly displays little signs of insanity or delib-
erate attempts to destroy the texts.
In order to get deeper into the problem of comparing the two versions, I developed a
method that turned out to be of great help. In order to compare the two texts, I first of all
lemmatized them, thus ignoring whether a word appeared just once and in only one gram-
matical form or many times and in different tenses or cases. I then threw out the system-
words, that is to say the prepositions, conjunctions, many pronouns, and so on, reckoning
that these are the kind of words we add most deliberately when telling or retelling a story. The
remaining content words were then compared word by word. The picture that came out of
this (shown in Table 2) seems to provide concrete support for the feeling that arose from the
comparison given in Table 1.

Table 2 Content words in seven episodes in Regius and Upsaliensis

2
I owe my thanks to Eysteinn Bjrnsson, who most kindly allowed me to use his web-edition of Gylfaginning
(http://www3.hi.is/~eybjorn/gg/index.html) the texts from Skldskaparml I have digitalized myself. The
episodes in Table 1 and 2 can be found in 1931 and 1977 years editions as follows: Episode 01 1977:3.208.2;
1931:8.1719.12 Episode 02 1977:16.719.4; 1931:34.638.6 Episode 03 1977:38.1041.3; 1931:100.18
104.15 Episode 04 1977:23.2629.18; 1931:47.2061.4 Episode 05 1977:20.1420.24; 1931:40.1541.30
Episode 06 1977:36.2037.22; 1931:82.285.12 Episode 07 1977:35.1236.17; 1931:78.182.1

371
 Episode R U R not U U not R U as % of R
01 The main creation myths 480 362 150 30 75%
02 Lokis offspring 325 325 100%
03 rr and Hrungnir 257 254 10 7 99%
04 tgara-Loki 480 376 147 43 78%
05 Freyr and Gerr 108 49 65 6 45%
06 The mead of poetry 251 137 126 12 55%
07 Iunn and jazi 248 156 117 25 63%

The most tempting conclusions to draw from this are that when we are facing almost exactly
the same content words in both versions, the original source is likely to be the same. At the
same time, it seems clear that the suggestion that the text of DG 11 is an abbreviation of an
earlier original that was more or less the same as the original of the RTW version is at least in
some parts out of question. The explanations of the backgrounds of the different versions are
clearly far more complicated. One of our main tasks in the coming years is going to be that of
coming closer to the real process that lies behind these manuscripts of the Prose Edda.

Bibliography
Edda
Finnur Jnsson 1931 (ed.): Edda Snorra Sturlusonar. Udgivet efter hndskrifterne af Kommissionen
for det Arnamagnandske Legat. Kbenhavn.
Grape, Anders & al. 1977 (ed.): Snorre Sturlassons Edda. Uppsala-handskriften DG 11. 2. Transkribe-
rad text och paleografisk kommentar. Uppsala: Uppsala universitetsbibliotek.

Other sources
Dillman, Franois-Xavier, 1992: Textafri og goafri : Um rfina betri tgfu Snorra-Eddu.
In: lfar Bragarson (ed.): Snorrastefna : 25.27. jl 1990. Pp. 918. Reykjavk.
Faulkes, Anthony (ed.), 1998: Introduction. In: Snorri Sturluson: Edda : Skldskaparml : 1. Introduc-
tion, Text and Notes. Pp. viilxxii. London.
Holtsmark, Anne, 1950: Innledning. In: Snorri Sturluson : Edda : Gylfaginning og prosafortellingene
av Skldskaparml. Utg. av Anne Holtsmark og Jn Helgason. Pp. vxiv. Kbenhavn.
Lasse Mrtensson och Heimir Plsson, 2008. Anmrkningsvrda suspensioner i DG 11 4to (Codex
Upsaliensis av Snorra Edda spren av en skriven frlaga? Scripta Islandica 58. Pp. 135155.
Sigurur Nordal, 1920: Snorri Sturluson. Reykjavk.
Sigurur Nordal, 1933: Formli. In: Egils saga Skalla-Grmssonar, slenzk fornrit II. Pp. vcv. Reyk-
javk.
Sverrir Tmasson, 1992: Snorra Edda. In: Vsteinn lason (ed.): slensk bkmenntasaga I. Pp. 534
542. Reykjavk.
Sverrir Tmasson, 1996: Nskpun ea endurtekning? slensk skldmennt og Snorra Edda fram til
1609. In: Sverrir Tmasson (ed.) Guamjur og arnarleir: Safn ritgera um eddulist. Pp. 164.
Reykjavk.
Zetterholm, D. O., 1949: Studier i en Snorre-text : Tors frd till Utgrd i Codices Upsaliensis DG 11
4 och Regius Hafn. 2367 4.

372
 Law recital according to Old Icelandic law:
Written evidence of oral transmission?
Helgi Skli Kjartansson, University of Iceland
In pre-literate Scandinavia, law must have been an important body of oral lore, shared and
developed by acknowledged experts who deliberately passed it on to their successors.
Just how fixed or how flexible the transmission of law may have been is a perennial ques-
tion. Similarities in written law within Scandinavia and even beyond may suggest an un-
broken chain of transmission over many centuries (see Amira 1960:222 for an example of
verbal similarities in Scandinavian, Frisian and Langobardian law). Often, however, it is diffi-
cult to determine whether similarities arise from common origin, parallel development or bor-
rowing. Or whether possible borrowing could represent one Germanic tradition following
another or both borrowing from the same Latin sources. Similarly, differences between re-
corded legal traditions may reflect independent development over a long period or more re-
cent legal creation again including borrowings from written sources.
Because the answer is probably All of the above, these are questions of proportion, an-
swered in widely different ways by individual scholars and generations of scholars. The cur-
rent generation is torn between a resurgent interest in oral culture and a critical aversion to
conjectures not directly supported by the sources at hand.
Whatever the proportions, recorded Scandinavian law surely contains some traditional
matter of very old origin, reflecting painstaking efforts by generations of aspiring young men
to acquire the legal wisdom of their elders. While the form of that wisdom and the mechanism
of its transmission belongs largely to the realm of conjecture, one particular mechanism,
known from the written sources, may have been crucial for passing down the legal knowledge
later codified, i.e. the official law recital by the lawspeaker at the assembly.

Law recital: the explicit rules

For Viking Age Scandinavians the regional or provincial assembly (including the general as-
sembly aling of an entire island: Gotland, Man, Iceland) was a key social institution. The
region sharing an assembly also shared a legal tradition, with the assembly serving as the
venue to clarify and standardise its law.
The lawspeaker was an official of the province or region, the leader of its assembly and re-
sponsible for its law. Written provincial laws make no mention of a lawspeaker in Denmark,
while in 13th century Norway and to some extent Sweden the office is under royal control. It
is tempting, however, to assume that before the development of the three centralised king-
doms the office of lawspeaker had been a common Scandinavian tradition, his function more
like what we know from Gotland and especially Iceland, perhaps (at least in Norway and
Sweden) including public law recital.
Law recital as practiced in Iceland is described in the older (ca. 1250) of the two main
manuscripts of the law collection Grgs:

It is also prescribed that a Lawspeaker is required to recite all the sections of the law over three
summers and the assembly procedure every summer. (Grgs 1:187) [] It is also prescribed
that the Lawspeaker shall recite all the sections so extensively that no one knows them much
more extensively. And if his knowledge does not stretch so far, then before reciting each section
he is to arrange a meeting []with five or more legal experts, those from whom he can learn
most [] (Grgs 1:188) All men with seats on the Law Council are also required to be always
present at the reciting whenever the Lawspeaker wishes to recite the laws, whether that is at

373
 Lgberg or in the Law Council or in the church if the weather out of doors is unpleasant. []
the Lawspeaker [] is also to recite the assembly procedure every summer and all the other
sections so that they are recited every three summers if the majority wish to hear them. Assem-
bly procedure is always to be recited on the first Friday of the assembly if men have time to hear
it. (Grgs 1:193)

In the same manuscript a stray provision is inserted at the end of the Christian Laws Section:

All the laws are to be recited over three summers. [] No new law is to have effect for more
than three summers and it is to be announced at Lgberg the first summer [] All new laws be-
come void if they are not included in the recital every third summer. (Grgs 1:51)

This provision assumes a law recital not only regular but so keenly observed that it would be
immediately noticed if a given piece of legislation had gone unrecited for three consecutive
years. Although it is hard to imagine such a rule being very effective in practice, this does not
lessen its value as evidence of the reality of law recital: whoever framed the rule was in no
doubt that law would be regularly recited in the foreseeable future.
Law recital is also implicitly assumed in a number of expressions in the written law such as
here i.e. at the general assembly, or today/tomorrow i.e. the same day or following day,
counting from the day the assembly procedure was supposed to be recited. Such expressions
are not used consistently, however, and the lawspeaker himself is always referred to in the
third person,1 also in the chapters which use today/tomorrow (e.g. Grgs 1:54). All the
same, the written law seems to contain remnants of text that was written either to be delivered
as law recital or to imitate the recital word for word. In either case it serves as evidence that
law recital actually took place.
One possible assumption is that laws were written to imitate the recital as closely as possi-
ble because that was how people always tried to learn them. Each legal provision, then, was
felt to exist in one precise verbal form, much like modern statute law, with the lawspeakers
recital its correct manifestation. In such case the law would not be unlike church hymns where
terms like we (the congregation), here (in church) and now (during service) belong to
the text in any circumstances, whether it is being sung in church or, for instance, being copied
in a solitary monastic cell. Considering how often the same substatial rules appear in very
different verbal form in the two preserved Grgs versions, I consider it rather unlikely that
verbal fidelity had been much more highly valued at an earlier stage of transmission.
The other possibility, that those clauses were originally penned to be recited, would put
them in a class with a rather stern letter received by the Icelandic aristocracy from their out-
raged archbishop, probably in 1180 (DI 1:262264). The letter was written in Norway, cer-
tainly in Latin but preserved in a copy of a contemporary Icelandic translation. The translator
uses the terms here and this country to mean Iceland, obviously not following the Latin
but because it was in Iceland that the letter was going to be presented to the recipients per-
haps by reading aloud at the general assembly. A similar concern might have directed the pen
of some legal codifiers.

How late?
The above provisions on law recital were, when copied in the preserved manuscript, clearly
not active law. The phrasing is anything but confident: if his knowledge does not stretch so
far; or in the church if the weather out of doors is unpleasant; if the majority wish to hear

1
Danish provincial law occasionally speaks in the first person, which is sometimes interpreted as evidence of
law recital in Denmark as well.

374
them; if men have time to hear it. What in fact replaced the law recital is described in be-
tween the provisions already cited:

It is also prescribed that in this country what is found in books is to be law. And if books differ,
then what is found in the books which the bishops own is to be accepted. If their books also dif-
fer, then that one is to prevail which says it at greater length [] (Grgs 1:190; cf. Foote 1984)

When this provision was drafted, law was subject to written transmission which already has
resulted in a bewildering variety of legal texts a stage reached, perhaps, around 1200. By
then the codification of Icelandic law had gone through several stages, one of them well
documented in Ari fris slendingabk. According to his contemporary (1133 or earlier)
account a decision had been adopted in 1117:

[] that our laws should be written in a book at Haflii Mssons in the coming winter, accord-
ing to what Bergrr [the lawspeaker] and other wise men who were selected for it said was law
and after consultation between them. They were to introduce all such new laws as seemed to
them better than the old laws. They were to announce [segja upp] them next summer in the Law
Council and keep all those which the majority did not oppose. And the outcome was that the
Treatment of Homicide and much else in the laws were written and read out [sagt upp] by
clergy in the Law Council the following summer. (cited in Grgs 1:4, Introduction)

Before this official decision to codify Icelandic law, some legal matter had presumably been
written down, on private or ecclesiastical initiative, suggesting the feasibility of a larger-scale
codification.
The codification effort of 11171118 was not continued, and failed to produce the intended
official law code. Instead it ushered in a period of unco-ordinated even competitive com-
pilation of different legal texts, the resulting mess eventually accepted by the already cited
provision on what is found in books. During that period, law recital appears to have ceased
because written law made it redundant.
The question remains, however, as to whether law recital in its final stage was done from
memory or from a written text.2 In his account just cited Ari uses the term for recite (segja
upp) when describing the reading aloud by clergy of proposed legislation drafted by commit-
tee during the preceding winter. If the lawspeaker was, at some stage, supposed to recite the
law from a book, and if the preserved provisions on law recital were drafted at that stage, we
could not expect them to sound any different from what they do. The wording in Old Ice-
landic does not distinguish between recital from memory and recital from a written copy.
If law recital was, at a certain stage, intended to entail the reading out of a written text,
how, then, did written law make it redundant? Looking at the sheer volume of the preserved
Grgs text, the answer is obvious. Written law simply outgrew the stamina of the lawspeaker
and the patience of his intended audience. Not only did it grow in magnitude but also, as wit-
nessed by the two preserved versions, in variety. Given the lawspeakers duty to recite all the
sections so extensively that no one knows them much more extensively his five or more
legal experts would be able to refer him to an ever increasing variety of legal texts each of
which contained some extra material and not necessarily in agreement as to what belonged to
each section.

How early?
If law recital in Iceland came to an end sometime during the 12th century, when did it begin?

2
Or a bit of both, if the lawspeakers memory was supported by written notes as suggested by Karlsson 2007:
222234; an illiterate lawspeaker might even have had an assistant (a suffleur of sorts) managing his notes.

375
 In one sense, law had always been recited. The functioning of assemblies, courts and the
like might entail the reciting of procedural rules similar to the lawspeakers annual reciting
of the assembly procedure. When people needed to change, fix or clarify the law, one or more
versions would be stated and the accepted law publicly recited just like the written law at
the Icelandic general assembly in 1118. But this question specifically concerns law recital in
the narrow sense of repeated official recital by the lawspeaker of (ideally) the entire body of
accepted law.
There is no direct evidence of such procedure prior to the 12th century. On the other hand,
there is precious little evidence for almost anything occurring before the appearance of written
sources, which, of course, does not mean that nothing happened in prehistory. Assuming, for
the sake of caution, nothing not directly supported by the sources, could mean dating most
developments as late as possible, which would obviously distort our entire take on the past.
One potential reason to assume the practice of reiterative law recital to be old is that it
would have been more useful in a preliterate society (counting Scandinavian society as prelit-
erate for the present purposes because runes seem not to have been an option for the preserva-
tion of law prior to the adoption of the Latin alphabet). On the other hand, there have been
illiterate societies all over the world managing their legal traditions without this particular
device. I would suggest that the idea was much more likely to occur to people who knew how
the church used its fixed texts, not only in rituals but, for instance, proclaiming the charter of
the church or reading the story of its patron saint.
Another argument for an early origin of law recital is the assumption that without it there
would have been no lawspeaker. In Iceland there is every reason to accept the tradition, re-
lated by Ari, that the office of lawspeaker is as old as the general assembly, ca. 930. However,
the lawspeaker had other important functions to which law recital could be a later addition.
Even if the term lawspeaker (cf.German Gesetzsprecher) may suggest the particular func-
tion of law recital it is an accident of translation. In Old Norse the term lgsgumar relates
to his central duty of segja lg (mnnum): tell (people) what is the law, further defined as
his duty to respond to questions about valid law.
A final argument for the assumption of law recital as an early practice at the Icelandic gen-
eral assembly is the evidence for law recital in continental Scandinavia, at least in Norway
and Sweden.3 Here as in Iceland the main evidence is in 13th century manuscripts, or even
later ones, perhaps reflecting 12th century practice. If the practice is common to the three
countries, however, it points to a common inheritance which would date it prior to the settle-
ment of Iceland, i.e. not later than the 9th century. Alternatively, if it was an innovation which
spread from one country to others, Iceland would have been more likely to copy the mainland
(Norway) than vice versa, dating the innovation to ca. 900 at the latest.
Norwegian and Swedish provincial law is much less explicit about law recital than the Ice-
landic provisions cited above. It is, in fact, largely confined to the indirect evidence of vo-
cabulary.4
The lawspeaker is there called lgmar/laghmaer rather than lgsgumar, but the differ-
ence is irrelevant as his office or province is known as lgsaga/laghsagha, corresponding to
Icelandic lgsgumar. As in Iceland, the lawspeakers central duty (most clearly reflected in
narrative sources) is to know the law and pronounce in legal disputes. His existence is thus no
evidence of law recital.

3
I have earlier (Kjartansson 1989) discussed the Icelandic evidence for law recital and been justly criticised
(most recently by Karlsson 2004:53) for disregarding the evidence from mainland Scandinavia.
4
See Fritzner for definitions and examples of Icelandic and Norwegian terms and Fornsvensk lexikalisk databas
for Swedish ones.

376
 Then there are examples of the law referring to itself as recital: Norw. uppsaga/lgtala,
Sw. laghsagha. In Sweden the stgtalagen concludes with the words: Now your recital is
finished and brought to the end.5 In Norway one part of the older Frostaingslg begins with
a phrase which similarly seems to refer to the law (or that particular part of it) as recital.6
This may be compared to a clause in Grgs (found in both manuscripts in this case) which
contains the following reference to law recital:

When men are to publish assault or injury, then it is lawful to do so on the same day as the ac-
tion, during the following night, and during the two days and two nights thereafter. Then the
publishing has been done before the third sunrise, as the law recitals phrase has it if what it
says there is rightly understood. (Grgs 1:143)

Here, the law recital (uppsaga) is a given text, written or remembered, containing the exact
phrase before the third sunrise which the author7 is trying to clarify.
The three cases, from Sweden, Norway and Iceland, all imply law being recited. But there
is no indication if it had been recited, or was supposed to be recited, as a part of the repeated
and comprehensive recital of the entire body of law. The reference could just as well be to a
one-off recital, such as Ari describes in the Icelandic Law Council in 1118. In the Swedish
and Norwegian cases, as possibly in the Icelandic one, it is a question of recital from a written
book, with no indication if the law in question had been recited from memory at an earlier
stage.
Finally there is the Norwegian term lgtala, apparently meaning recital. It occurs in a
variant reading where the bishop is supposed to come to the assembly and listen to the lgtala
instead of the lgbk (law book) of the main manuscript (Norges gamle Love 1:378). Here,
finally, we have a recital which is definitely reiterative, the same old law being read out at the
assembly every year. But the rule is contained in the ecclesiastical part of the law and its in-
tent is to enable the bishop to control provincial law provisions on church matters. Such law,
originally adopted at the initiative of the church, surely existed in writing from the beginning
and is unlikely to have ever been recited from memory.
While there can be few rock-solid conclusions here, my suggestion is that law recital the
repeated official recital of the entire body of recognised law was an isolated Icelandic ex-
periment, commencing perhaps either shortly before or shortly after the codification effort of
11171118 and probably reflecting the same motivation as that effort to develop and stan-
dardise the countrys law.

Fixed or flexible?
My doubts about the reiterative law recital as a transmission mechanism of oral law should
not be taken to imply that no oral law was ever transmitted. Private tutoring of aspiring law-
yers by acknowledged experts, together with the in-service education of experts conferring
with each other, would in any case be more effective to ensure the preservation of legal
knowledge than recital at the assembly.
How fixed or how flexible was the oral transmission of law? And how accurately did the
original codification of law reflect an earlier oral tradition? The answers if we only knew
them must have depended on social preferences rather than the arrangement of transmission.
5
Nu r laghsagha iur lykta ok ut sagh.. stgtalagen, final chapter.
6
at er uppsaga laga vrra lgum manna at engi skal [] (Norges gamle Love 1:217). The phrase seems
corrupt, perhaps conflated from two variants, but the meaning of uppsaga is reasonably clear.
7
A legal expert rather than the legislature itself; see, however, Finsen (1883:684) for a contrary view. Finsen
accepts the possibility of the text being written though disputing Maurers earlier suggestion as to the identity of
the text.

377
If the literate 12th century experts contributing to the codification of Icelandic law probably
[] felt [] free to vary and extend the wording in response to the casuistical queries that
might occur to them or be put to them by other people (Grgs 1:11, Introduction), surely the
preliterate expert felt no less free, whether he was the lawspeaker in his official capacity or
the lawyer training his apprentice.
People are perfectly capable of learning entire volumes of even complicated prose by heart.
But it is literate culture that prompts them to do so, whether it is Avicenna reciting the Koran
from memory at the age of ten, or a modern actor learning his part in play after play, or the
opera singer able at any time to perform any of dozens of memorised roles. For such learning
the written text is not only an indispensible tool for student and teacher alike; it is also the
model which in the first place suggests that even a prolonged stream of prose ought to exist in
a permanent form. While preliterate culture may operate with any amount of poetry, more or
less permanently memorised, and any number of fixed phrases, we should not expect it to
aspire to the word-for-word fixation of long pieces of prose. Nor should we expect it to adopt
a mechanism, like the official law recital, to transmit such fixed prose. Not unless the sources
definitely point in the opposite direction which they, in the present case, do not.

Bibliography
Amira, Karl von, 1960: Germanisches Recht. Vol. 1, Rechtsdenkmler. 4. Aufl. bearb. von Karl Au-
gust Eckhardt. Berlin. (Grundriss der germanischen Philologie 5:1.)
DI 1 = Diplomatarium Islandicum: slenzkt fornbrfasafn [], Vol. 1, 1857. Ed. Jn Sigursson.
Kaupmannahfn.
Finsen, Vilhjlmur (ed.), 1883: Grgs: Stykker som findes i [] Sklholtsbk [] Kbenhavn.
Foote, Peter, 1984: Some lines in Lgrttuttr. In: Peter Foote: Aurvandilst. Norse Studies. Odense.
(The Viking Collection: Studies in Northern Civilisation 2.)
Fornsvensk lexikalisk databas. Sprkbanken, University of Gothenburg. http://spraakbanken.gu.se/
fsvldb/.
Fritzner = J. Fritzners ordbok over Det gamle norske sprog, dvs. norrn ordbok. Dokumentasjonspros-
jektet, Oslo University. http://www.dokpro.uio.no/.
Grgs 1 = Laws of Early Iceland: Grgs: The Codex Regius of Grgs with material from other
manuscripts. Vol. 1, 1980. Transl. by A Dennis, P. Foote & R. Perkins. Winnepeg. (University of
Manitoba Icelandic studies 3.)
Karlsson, Gunnar, 2004: Goamenning: Staa og hrif goorsmanna jveldi slendinga. Reyk-
javk.
Karlsson, Gunnar, 2007: Inngangur a mildum: Handbk slenskri mialdasgu. Vol. 1.
Reykjavk.
Kjartansson, Helgi Skli, 1989: Lagauppsaga lgsgumanns: Erindi flutt hj Flagi hugamanna um
rttarsgu [] Reykjavk. (Erindi og greinar 23.)
Norges gamle Love 15 = Norges gamle Love indtil 1387. Udg. ved R. Keyser og P.A. Munch,
Gustav Storm og Ebbe Hertzberg. Vol. 15. 18461895. Christiania.
stgtalagen, at the website of the University of Lund. http://www.nordlund.lu.se/Fornsvenska
/Fsv%20Folder/05_Filer/OgL-A.rtf.

378
The Construction of Memory in Medieval Icelandic Literature
Pernille Hermann, Scandinavian Institute, Aarhus University, Denmark
In recent decades there has been increased focus on memory studies in the humanities. In Old
Norse studies, the concept of memory is often used in research literature. Often, however, the
concept is used contextually, thus being meaningful for the immediate topic under discussion,
whereas it is relatively rare that the concept is qualified theoretically. Works by J. Glauser
(e.g. Glauser 2000 and 2007), though, are examples in saga studies of a more theoretical ap-
proach to the concept of memory. Inspired by these examples, I will examine how memory
studies, especially notions of cultural memory, as first introduced by J. Assmann (e.g. Ass-
mann 2005), can possibly provide us with a useful theoretical framework for understanding
medieval Icelandic literature dealing with the past. In so doing, I will be concerned with the
representation of the past in medieval literature, focusing on selected references to the settle-
ment of Iceland. The relation between literature and memory, especially the notion of mem-
ory in literature (see e.g. Erll and Nnning 2006) will be dealt with, as will as well the
founding function and the mythic status of the literature about the settlement.

Bibliography
Assmann, Jan, 2005: Das kulturelle Gedchtnis. Schrift, Erinnerung und politische Identitt in frhen
Hochkulturen. Mnchen.
Erll, Astrid and Ansgar Nnning, 2006: Concepts and Methods for the Study of Literature and/as Cul-
tural Memory. In: Literature and Memory. Theoretical Paradigms. Genres. Functions. Ed. by A.
Nnning, M. Gymnich & Roy Sommer. Tbingen. Pp. 1128.
Glauser, Jrg, 2000: Sagas of the Icelanders (slendinga sgur) and ttir as the literary representation
of a new social space. In: Old Icelandic Literature and Society. Ed. by M. Clunies Ross. Cam-
bridge. Pp. 203220.
Glauser, Jrg, 2007: The Speaking Bodies of Saga Texts. In: Learning and Understanding in the Old
Norse World. Ed. by J. Quinn, K. Heslop, T. Wills. Turnhout. Pp. 1326.

379
 Hjarta sjnir. Ekphrasis and medium in Lknarbraut
Kate Heslop, Deutsches Seminar, Universitt Zrich, Switzerland
The skaldic poems on Christian subjects from thirteenth and fourteenth century Iceland offer
fascinating source material for the study of mediality in medieval societies. Of course, earlier
skaldic poetry is also transmitted to us via writing, whether in runic or Latin script, and we
need to keep in mind that all that skaldic poetry we have which is dated later than c.1000 was
also composed not in an environment of pristine primary orality, but rather in one in which
written and oral cultures co-existed and writing was increasingly taking over important func-
tions (the writing down of the laws in the winter of 111718, as reported in slendingabk ch.
10, is the classic index of this). However, the significant social relationships and mythic pat-
terns encoded in the traditional skaldic practice of performance before a listening audience
probably ensured that poetry composed early in the literate period was not much affected by
the advent of writing.
At some point this situation changes. To take an example from the very end of the period,
Ktrnardrpa, usually dated to the second half of the fourteenth century, is full of references
to books and writing. Its highly self-referential last stanza exhorts: Lttu, mar, letri etta
Look, man, at this writing, before going on to compare a punning reference to the poets
name to a lst brief (closed letter, a calque on littera clausa, a private, sealed letter), as a
figure for the obscurity of the wordplay. Ktrnardrpa, then, has what we might call a me-
dial self-consciousness (Medienwissen). It is highly aware of its own status as written text
but in the late fourteenth century, this is perhaps not surprising. More interesting would be an
earlier example, one in which a nascent medial self-consciousness can be observed arising out
of the media change (Medienwechsel) associated with the arrival of writing.1 In this earlier
phase we can speak of processes of interference between written and embodied media. Tak-
ing medium in a broad sense as that which makes perceptible (Sieber 2004), interference
between different media could be said to make perception perceptible, to draw attention to the
process of perceiving, and in so doing, to make a self-reflective Medienwissen possible.
George Tate (1978:35) has observed the tendency of the Lknarbraut poet to experiment with
the drpa form. In what follows I will show that this poem, long recognised as occupying a
key position in the development of Christian skaldic poetry (Schottmann 1973:82), is also
innovative in other ways, which bear witness to its medial self-reflexivity.

1. Frames
One obvious locus of medial interference in Old Icelandic textuality is saga prosimetrum.
Performances of skaldic verses are part of the narrative in many slendinga sgur, and also
appear (mixed with other modes of verse citation) in genres such as the konungasgur. The
way in which stanzas are cited as part of the prose narrative means that the sagas reader or
hearer experiences not a poem as such, but a performance of a poem. The provision of a frame
focuses our gaze on what is inside it, heightening the contrast between what is inside (a fic-
tional poetic performance) and what is outside (written prose).2 I will return to a particularly
telling example of saga prose as frame at the end of this paper. The Christian long poems, on
the other hand, are for the most part not transmitted in narrative settings, but rather as self-
contained textual entities. Here we search in vain for the meta-level of a fictional perform-
ance situation, with its more or less explicit commentary on the mediality of the poetic text.
But a frame is present it is just in a different location, within the poem. The introductory and
1
For the terminology used here, see Kiening (2007:30515).
2
For further discussion of framing in saga prosimetrum, see Heslop (in press).

380
concluding verses (upphaf and slmr), the stef and the poets parenthetical comments, are
places where the three (or four)-way relationship poet-dedicatee-subject-audience is defined,
where the title and sometimes even, as in Ktrnardrpa, the authors name are given. Per-
haps paradoxically, this self-enclosing framing does not work to close the text off. Rather it is
on this meta-level that various kinds of reference outside or across the frame deixis, for
example (cf. Hausendorf 2003), or intertextuality (especially common in the Christian skaldic
corpus, cf. Attwood 1996) become possible.
Lknarbraut Way of Grace, the poem I will be discussing in this paper, is not among the
much-anthologised skaldic classics. The most recent and useful edition, from which the fol-
lowing introductory comments and all my quotations and translations from the poem are
drawn, is George Tates for the new Skaldic Poetry of the Scandinavian Middle Ages project
(Clunies Ross et al. 2007). Lknarbraut is transmitted complete in the collection of poetry on
Christian subjects in ms. B of Snorra Edda (AM 757a 4to, c. 1400, probably from northern
Iceland), where it comes between Leiarvsan and Harmsl. Unlike Harmsl, Lknarbraut is
anonymous, though its title is medieval, appearing both in the right margin of the ms. at the
beginning of the poem and in the penultimate stanza. It consists of 52 drttkvtt stanzas with
a single stef. The stef occurs in the first part of the poem only (sts. 1029), which depicts key
events of salvific history: the incarnation, passion, harrowing of hell, resurrection, and last
judgement. The second, stef-less part (sts. 30-46) is devoted to an adoratio crucis Adoration
of the Cross and draws extensively on various Latin and Old Icelandic texts about the cross
from the liturgy, hymns, and exegetical and homiletic literature; Tate suggests that its close
connection to Good Friday liturgy suggests that it may have functioned [] as a verse ser-
mon (Clunies Ross et al. 2007:229). Features such as its formally correct but infrequent ken-
nings, (a single occurrence of) full rhyme of and , borrowings from poems in the reasona-
bly well-established twelfth-century group (Geisli, Harmsl, Leiarvsan, Plcitusdrpa),
lendings to the fourteenth-century Gumundr poetry, and finally a general similarity to Fran-
ciscan passion poetry (first suggested by Holtsmark 1965) lead Tate to date Lknarbraut to the
late thirteenth century.3
The poem begins with a nine-stanza upphaf, which invokes complex processes of circula-
tion and mediation between three actors: the skald, speaking in the first person; God, apostro-
phised in the second person in almost every stanza; and a third-person collective, in one in-
stance (st. 8) particularised as the skalds brr ok systr brothers and sisters. God is, as con-
ventional in Christian poetry, the source of poetic inspiration. He gives orgntt word-
abundance (1), munnshfn mouth-content (2), gipt sarar gzku gift of true grace (3),
albjart starljs wholly radiant light of love (4), and himneskt s heavenly seed (5), and
he also receives hugubnir loving prayers (3), mn ml my utterances (8), essi blr
hrr this joyful encomium (9). The skald processes Gods gift of inspiration into poetry,
though his subordinate position is emphasised: he merely supplies lyndis l minds land for
Gods seed (5), and his blindi mns ms munar blindness of my despondent mind must be
driven out by Gods light (4). The brr ok systr, finally, are summoned to the poem (kve
ek bi brr ok systr at kvi, 8) and asked to help the skald with their prayers. In st. 6, a
third-person group, who may or may not be the same as the brr ok systr (they are denoted
by a warrior-kenning, but this is of course no barrier to their being clerics) are said to benefit
from the sannr vxtr true fruit (i.e. the poem) for their souls salvation.

3
Proposed datings in the literature, as is often the case for Christian skaldic poetry, vary considerably: Clunies
Ross (2009:66) appears to prefer a relatively early date, c. 1260; Tate: late C13th (Clunies Ross et al.
2007:228); Schottmann: wohl um 1300 entstanden[] (1973:82); Holtsmark: vi gjr neppe galt i sette
L[knarbraut] [] etter 1300 (1965:col. 554). In light of the evidence available this range is not surprising (and
could perhaps even be widened).

381
 The relations between skald, dedicatee, and audience are worth pausing over here. God is
not only both dedicatee and subject (as usual for a skaldic praise poem, with some excep-
tions), he has also taken over the Odinic role of supplying inspiration. The gift-exchange that
traditional skaldic poetry proclaims as its pragmatic context is also active here, but it is
marked by collapses and doublings. God supplies not only material for composition, as
earthly princes do (leifs hefnir, fengu yrkisefni Avenger of lfr, you furnished matter for
verse, Arnrr tells Magns gi in Hrynhenda 14) but also the essence of the poem itself
which the skald offers promptly back to him as hrr praise(-poem). The role of the third-
person audience is also quite different to that of the traditional skaldic audience-within-the-
poem (essentially, to listen quietly and be impressed). The audience of Lknarbraut is ex-
pected to actively participate both in the poetic process, by helping the skald with their
prayers (st. 8), and in the gift-exchange, as by means of the poem (af mnum munni from my
mouth, st. 6) they obtain benefit for their souls. There is no trace of such traditional topoi as
the call for a hearing, and in general the upphaf is, apart from the two references to the mouth
cited above, dominated by the sense of sight, and metaphors of light and blindness, rather than
hearing. The abstraction of kve ek [] at kvi is surely no accident. God, the poems ad-
dressee, is not physically present, and the poem takes place in an imagined, visionary space.
The task of the audience, as we shall see, is precisely that of envisioning.

2. Lta seggja sveitir

The first section of Lknarbraut, as already mentioned, narrates the important events of
Christs life, interspersed with the stef-stanza with its reference to krossmark ins hsta Krists
the cross-sign of the most high Christ, which the skald offers to Christ (ek vilda bja htt
stef, st. 13) as if present, as Tate observes (Clunies Ross et al. 2007:244). The passion de-
scription has several traits which could indicate the Franciscan influence suggested by
Holtsmark. The descriptions of the nails (st. 16) and wounds (st. 20) evoke the stigmata, key
to Franciscan spirituality;4 and the normative response to these events is represented as imagi-
native sympathy leading to compassion (Hvat megi heldr of grta hvern mann, er at kannar
What could be more able to make weep each man who ponders it, st. 19). This imaginative
sympathy is aroused in literally textbook fashion by the use of enargeia, vividly sensual de-
scription. Quintilian writes that enargeia is a quality which makes us seem not so much to be
talking [dicere] about something as exhibiting [ostendere] it. Emotions will ensue just as if
we were present at the event itself. (Quintilian, Institutio oratoria [The orators education]
VI.2:29, cf. VI.2.29: Quintilian 2002:60-61). Pseudo-Longinus agrees: Another thing which
is extremely productive of grandeur, magnificence and urgency, my young friend, is visuali-
zation (phantasia) [] the situation in which enthusiasm and emotion make the speaker see
what he is saying and bring it visually before his audience. (Pseudo-Longinus, Peri hupsous
[On the sublime] 15.1: Russell and Winterbottom 1989:159) At first the senses of hearing
(Glymr var hr af hmrum heyrr High clanging was heard from hammers, st. 16), and
touch (Vst bar vf it hsta vtar kir af grti Certainly the highest woman [Mary] bore
cheeks wet from weeping, st 18) are appealed to, but the climax comes in the appeal to sight
in sts. 258, the middle stanzas of the poem.
These stanzas set themselves apart in several ways. In them, the tense of narration switches
from the preterite, used consistently until this point, to a mixture of the future (hardly surpris-
ing, seeing as future events are being described) and, more significantly, the present. Their

4
Caution is necessary here: Anne Derbes warns that the emphasis on the imitatio crucis was not confined to the
Franciscans. The origins of this sensibility predate Francis by decades, perhaps by more than a century. A grow-
ing empathy for Christs suffering during the passion emerged as early as the late eleventh century and gained
momentum in the twelfth. (1996:17).

382
idiom is strikingly visual. Not only are they full of iconographic details familiar from the vis-
ual arts (though the same could be said of the stanzas on the crucifixion, Marys weeping,
etc.), but their rhetoric is one of seeing, so much so that it seems justified to regard these stan-
zas as an ekphrastic interlude. Christ is represented as Pantocrator in st. 25 (he geypnir styrk-
liga sna skepnu alla saman holds in his hand mightily his creation all at once), the general
resurrection is depicted in st. 26 (hverr meir hringmts skal skunda hvatliga r hauri each
tree of the sword-meeting [BATTLE > WARRIOR] shall hasten quickly from out of the ground)
and in st. 27, the cross and the arma Christi, or instruments of the passion, appear (kross
sndr jum me bli ok saumi the cross will be shown to the people with blood and
nails, viir Mistar sj mti sr svipur ok spjt trees of Mist [WARRIORS] see before them
the whips and spear). In st. 27, the climax of the ekphrasis, verbs of seeing cluster: sndr, sj
and the direct statement in the first line of 27b: Lta seggja sveitir Hosts of men look.
Ekphrasis is presently the focus of intense scholarly interest, both within5 and without6 Old
Norse studies. Most interest in the Old Norse field has focused on the shield poem genre and
associated poems, usually thought to be early, which depict scenes from Old Norse myth and
legend, though Margaret Clunies Ross proposes a Christian ekphrasis tradition related to the
titulus (2007) and recently suggested (2009) that the two surviving verses of a Mruflokkr
Poem about Mary transmitted in Codex Wormianus could be a piet. Ancient ekphraseis
almost always describe fictional objects (Laird 1996:96; famous examples include the shield
of Achilles in Book 18 of the Iliad, and the paintings of the Trojan war in the Temple of Juno
at Carthage in Book 1 of the Aeneid), not ones which are present before the audiences eyes,
and in fact the thing described need not be an object at all. Most discussions of the skaldic
shield poems, however, assume that real, present objects are being described (as Edith Marold
trenchantly observes, sehr wenige Forscher haben zu der Annahme gefunden, da die
Schildbeschreibung eine Fiktion sein knnte (1976:452)) and elements in some poems, such
as their use of deixis, could suggest this. In accordance with this tradition, I will therefore
present a work of art which could fit the bill for the ekphrastic interlude in Lknarbraut, al-
though I am agnostic as to whether it was necessarily present at the moment of composition,
or, a fortiori, that of reception (estranged, thanks to written transmission, from any proposed
origin of the poem in a real visual experience.

5
A roundtable discussion of ekphrasis held at the 13th Saga Conference in Durham in 2006 has recently been
printed in the 2007 issue of the journal Viking and Medieval Scandinavia; cf. also Clunies Ross 2006 and 2009.
6
There is a vast bibliography here: cf. e.g. the classic article of Fowler (1991), the complementary collections
edited by Goldhill and Osborne (1994) and Elsner (1996), a recent special issue of the journal Classical Philol-
ogy (Bartsch and Elsner 2007), and in German Wandhoff 2003.

383
 Fig. 1. Panels from Bjarnastaarhl (from Gubjrg Kristjnsdttir 2000:275).

This object is the carved wainscoting depicting the last judgement taken from the farm Bjar-
nastaarhl to the National Museum of Iceland in 1924 (see Fig. 1).
The panels are now thought to have originally been made for the cathedral at Hlar (the
only building big enough to accommodate a work measuring at least 7.2 m wide x. 2.8 m
high) and to date from the mid-twelfth century (Gubjrg Kristjnsdttir 2000:27477). Only
thirteen fragments have survived, but Selma Jnsdttir (1959) was able to demonstrate from
them that the Hlar panels are an example of the Byzantine style of last judgement. The clas-
sic Western example of this iconographic tradition is the late twelfth-century mosaic from the
west wall of the basilica at Torcello in the Venetian lagoon (see Fig. 2). Selma presents a
number of other examples in various media (painting, fresco, ivory relief, manuscript illumi-
nation) dating from the eleventh to the fourteenth centuries.

384
Figure 2. Mosaic from the basilica of Torcello (from campus.belmont.edu/ hon-
ors/byzitaly/Torcello.html). Cf. esp. 1. Christ as Pantocrator, 2. the dead rising from their graves, 3.
the cross and 4. the arma Christi (lying on the cloth on the throne).

If we compare the Torcello image to the description in sts. 2527, we see that the images de-
scribed in the poem (Christ as Pantocrator, the dead rising from their graves, the cross and the
arma Christi) are all present, lying on the central axis of the mosaic approximately from top
to bottom. These elements appear in almost the same arrangement (only the location of the
dead rising from their graves varies) in all the examples of last judgements analysed by Selma
Jnsdttir, so if she is right in her claim that the Hlar panel is also an example of this type it
seems reasonable to suppose that they were also present there.7

7
Hrur gstssons reconstruction of the Hlar panel (Hrur gstsson 1989: Fig. 24), while usefully indi-
cating where the surviving fragments would fit, omits the uppermost, fifth, field (an Anastasis) containing the

385
 If this identification is correct, it may suggest avenues for further research into Lknar-
brauts possible connections with Hlar. In the present context I am more interested in how
the ekphrasis functions in the rhetoric of the poem; that is, in asking, in Simon Goldhills
words, what is [this] ekphrasis for? (Goldhill 2007:1). I would make two observations here.
The first, which I will return to in the last part of this paper, is the vertiginous uncertainty
about what exactly is being pointed at in sts. 258: the stanzas depict an intradiegetic act of
looking (resurrected mankind witnessing the last judgement), which may or may not mirror an
actual act of looking being engaged in by the poems audience (the congregation at Hlar con-
templating the last judgement panel), and which is also meant to correspond to an image in
their imaginations. This uncertainty is typical of ekphrasis, especially antique ekphrasis,
whose deictic force signal[s] a space for a further and different play of fictions (Bartsch and
Elsner 2007:iv). The second is Goldhills contention that ekphrasis is designed to produce a
viewing subject (2007, 2, his emph.). His argument how and why this is done is rather rari-
fied, perhaps thanks to his recalcitrant material (a collection of 36 Greek epigrams about a
small sculpture of a cow), but in Lknarbraut the production of a viewing subject is quite
overt:

Kross mun ingi essu

jum sndr me bli
uggs fyllaz allir
aumir menn ok saumi.
Lta seggja sveitir,
svpur ok spjt mti
sr ok sj me dreyra
sjlfs Krists viir Mistar (27)

At this assembly the Cross will be shown to the people with blood and nails; all wretched men
will then be filled with terror.
Hosts of men look, and the trees of Mist [WARRIORS] see before them the whips and spear with
the blood of Christ himself.

The switch into the present tense, while it certainly serves the end of vividness (cf. Poole
1991:2456) and has been seen as typical of Old Norse ekphrasis (Clunies Ross 2007), further
blurs the boundaries between the viewers depicted in the poem and the poems readers or
hearers, who, like the aumir menn, seggja sveitir and viir Mistar here, are referred to in the
present tense and by collective heiti or warrior-kennings in the upphaf and stef. The present-
tense verb lta then refers not only to the intradiegetic witnesses of the last judgement, but
also to the audience now (who for the believer, of course, are one and the same, just at differ-
ent points in eschatological time), looking with their hjarta sjnir hearts eyes (46) and en-
acting the imaginative identification which is the poems aim. The intercalary clause in ll. 3
4, echoed by similar clauses in sts. 26 (kemr tti fear comes then) and 28 (gipt rtrat
grace will not fail then), then supplies an exemplary affective content for their visualisation.
Ekphrasis reveals itself here as a powerful tool for awakening compunction and faith in the
audience, the skalds part in the circulation of intellectual and spiritual gifts invoked in the
upphaf.

dead rising from their tombs; nor does Selma Jnsdttir mention the Anastasis in her detailed description of the
mosaic, whose fields she describes as four in number (1959:16). This may be why the risen dead are missing
from the reconstructions of the Hlar panel, but there seems to be no good reason for this: the Anastasis in Tor-
cello was clumsily restored in the nineteenth century, but the fifth field is part of the original twelfth-century
composition (cf. Demus 1944), and so could also have been part of the Icelandic carvers model.

386
 The second half of Lknarbraut consists of a series of figures of the cross (sts. 317), a list
of its virtues and powers (sts. 3841), and Christs reproaches from the cross (sts. 4345).
There is no space here to discuss the fascinating issues of intertextuality (relationship to litur-
gical and other texts) and performance (the Good Friday liturgy as a possible performance
context) that Tate raises in his commentary on these stanzas. It is worth briefly noting, how-
ever, that the rhetorical and affective strategies identified in the first part of the poem continue
to dominate here. The (largely traditional) figures key, blossom, ship, ladder, bridge, scales,
altar are presented in a series of narrative vignettes in a mixture of past and present tense,
and references to visualisation are conspicuously absent, perhaps due to their intellectual
rather than emotional appeal. But the rhetoric of visualisation and visibility returns with a
vengeance in the following stanzas, especially (but not only, cf. st. 39, 42) in Christs re-
proaches, where he speaks directly to mar hverr hauri each man on earth (43), com-
manding them to look upon his suffering (sts. 43, 44), and the final stanza before the slmr
hammers the point home:

Leium hr hauri
hjarta vrs me trum,
systkin mn, fyr sjnir
sigtis meinlti. (46)

My brothers and sisters, let us bring the hard torments of the faith-guardian [= God (= Christ)]
on earth before our hearts eyes with tears.

3. The Medienwissen of ekphrasis

Lknarbraut concludes with the conventional gestures the twelfth-century group of poems had
established for the genre of Christian drpur: a prayer and, in the penultimate stanza, an act of
self-naming. The skald also asks God, the poems dedicatee, for a reward, gjld me leigum
fyr ger mna recompense with interest for my poetry-making (49), very much (with the
exception of the rather baffling me leigum) in the traditional manner. But this request is im-
mediately qualified: the poet has already received other good things (gi) from God, more
than he can repay. As in the upphaf, the skaldic quid pro quo is complicated for the Christian
poet, here by his awareness that Christ has already bought his salvation pretio magno with a
great price (1 Cor. 6:20).
Can we draw any preliminary conclusions about media change and medial self-
consciousness in skaldic textuality based on this analysis of Lknarbraut? Lines of question-
ing which go to medieval Icelandic realia (do sts. 258 really describe the Hlar last judge-
ment panels? could the poem have been performed in front of this image? could this have
happened during Good Friday celebrations of which Lknarbraut formed a part?) are both
fascinating and frustrating. What we undeniably can see are the changes in the fictive per-
formance situation that the text itself posits. It constructs an audience within the text, one
which actively participates, via prayer and imaginative visualisation, in a shared emotional
experience. The performance takes place in an imagined space one in which the skald ad-
dresses God, and Christ addresses the audience filled with vividly evoked images, and the
gi the skald hopes to obtain are spiritual benefits rather than gold rings. The visual domi-
nates the aural. Not only is the audience repeatedly encouraged to visualise the events de-
scribed, but Gods words are bjartari ok fegri gulli ok gimsteinum r vllum brighter and
fairer than gold and gems from the fields (7), and most of the verbs describing the skalds
performance are abstract (fra, stofna, bja, inna, framm bera) with only kve (8, in the odd
construction kve ek at kvi) referring specifically to speaking. The subject-matter obviously

387
has some bearing on this, determining, for example, the poems relentless focus on compunc-
tion as the emotional payload of the poems enargeia. But the emphasis on imaginative visu-
alisation is surely also connected to the liberation, effected by writing, of the poem from any
particular performance context. And while the preponderance of light imagery certainly has
much to do with Christian ideals of claritas, describing Gods words as bjartari ok fegri gulli
ok gimsteinum indicates that text is conceived of in visual that is, written terms.
The following brief passage from Orkneyinga saga ch. 85, discussed by Russell Poole in
the ekphrasis roundtable in 2006, suggests that the tensions between visual and linguistic art-
forms were quite differently located in the oral context:

at var einn dag um jlin, at menn hugu at tjldum. mlti jarl vi Odda in litla: Geru
vsu um athfn ess manns, er ar er tjaldinu, ok haf eigi sarr lokit inni vsu en ek minni.
Haf ok eigi au or inni vsu, er ek hefi minni vsu.

One day during Yule-tide, people were seeing to the wall-hangings. The Earl spoke to Oddi inn
litli. Make a verse about the behaviour of the man who is there on the hanging, and have your
verse completed no later than I have mine. Also, dont have the words in your verse that I have
in my verse (cited from Poole 2007:245).

The possibility of endless verbal variation in the description of one and the same object which
lies at the heart of skaldic poetics is neatly encapsulated in the last line of this quotation. This,
I would suggest, was the ruling contrast betwen verbal and visual modes of representation in
the oral world in which the skaldic poetic came into being. In the increasingly literate world
of the thirteenth and fourteenth centuries, ekphrasis offered the skalds a chance to stage the
act of looking within the poetic text and so to dramatise the exemplary work of envisioning,
carried out within the poem by an intradiegetic audience, and taken up by the hjarta sjnir of
the texts readers.

Bibliography
Attwood, Katrina, 1996: Intertextual Aspects of the Twelfth-Century Christian Drpur. Saga-Book 24:
22139.
Bartsch, S., and J. Elsner, eds., 2007: Special Issue on Ekphrasis. Introduction: Eight Ways of Looking
at an Ekphrasis. Classical Philology 102: ivi.
Carruthers, Mary, 2000: The Craft of Thought: Meditation, Rhetoric, and the Making of Images, 400
1200. Cambridge.
Clunies Ross, M. et al., eds., 2007: Poetry on Christian Subjects. Turnhout. (Skaldic Poetry of the
Scandinavian Middle Ages 7.)
Clunies Ross, Margaret, 2006: The Cultural Politics of the Skaldic Ekphrasis Poem in Medieval Nor-
way and Iceland. In H. Fulton et al., eds. Medieval Cultural Studies: Essays in Honour of Stephen
Knight: 22740. Cardiff.
Clunies Ross, Margaret, 2007: Stylistic and Generic Identifiers of the Old Norse Skaldic Ekphrasis.
Viking and Medieval Scandinavia 3: 15992.
Clunies Ross, Margaret, 2009: Mruflokkr an Icelandic piet? In Margrt Eggertsdttir et al., eds.
Greppaminni. Rit til heiurs Vsteini lasyni sjtugum: 6370. Reykjavik.
Demus, O., 1944: Studies Among the Torcello Mosaics, II. The Burlington Magazine for Connois-
seurs 84:4139.
Derbes, Anne, 1998: Picturing the Passion in Late Medieval Italy: Narrative Painting, Franciscan Ide-
ologies, and the Levant. Cambridge.
Elsner, J., ed., 1996: Art and Text in Roman Culture. Cambridge.
Fowler, D. P., 1991: Narrate and Describe: The Problem of Ekphrasis. Journal of Roman Studies 81:
2535.
Goldhill, S. and R. Osborne, eds., 1994: Art and Text in Ancient Greek Culture. Cambridge.
Goldhill, Simon, 2007: What is ekphrasis for? Classical Philology 102: 119.

388
Gubjrg Kristjnsdttir, 2000. Dmsdagsmynd fr Bjarnastaahl. In Hjalti Hugason, ed.
Frumkristni og upphaf kirkju: 27477. Reykjavik. (Kristni slandi 1).
Hausendorf, Heiko, 2003: Deixis and speech situation revisited: the mechanism of perceived percep-
tion. In F. Lenz, ed. Deictic Conceptualisation of Space, Time, and Person: 24969. Amsterdam.
Heslop, Kate, in press: Hearing Voices. Uncanny Moments in the slendingasgur. Gripla 19.
Holtsmark, Anne, 1965: Lknarbraut. Kulturhistorisk leksikon fr nordisk middelalder 10: cols. 5534.
Hrur gstsson, 1989: Dmsdagur og helgir menn Hlum. Reykjavik.
Kiening, Christian, 2007. Medialitt in medivistischer Perspektive. Poetica. Zeitschrift fr Sprach-
und Literaturwissenschaft 39: 285352.
Laird, Andrew, 1996: Vt figura poesis. Writing Art and the Art of Writing in Augustan Poetry. In Els-
ner 1996: 75102.
Marold, Edith, 1976: Skaldendichtung und bildende Kunst der Wikingerzeit. In H. Birkhan, ed. Fest-
gabe fr Otto Hfler zum 75. Geburtstag: 44978. Vienna.
Poole, Russell, 1991: Viking Poems on War and Peace: A Study in Skaldic Narrative. Toronto.
Poole, Russell, 2007: Ekphrasis: its Prolonged Echoes in Scandinavia. Viking and Medieval Scandi-
navia 3: 24567.
Quintilian, Marcus Fabius, 2002: Institutio oratoria. The Orators Education. Ed. and trans. D. A. Rus-
sell. Cambridge, MA. (Loeb Classical Library 126).
Russell, D. A. and Winterbottom, M., eds., 1989: Classical Literary Criticism. Oxford.
Schottmann, Hans, 1973: Die Islndische Mariendichtung: Untersuchungen zur Volkssprachigen Ma-
riendichtung des Mittelalters. Mnchen.
Selma Jnsdttir, 1959: An 11th Century Byzantine Last Judgement in Iceland. Reykjavik.
Sieber, A., 2004: (Un)Erwnschte Effekte. Mediengebrauch, Synergie und Strung im hfischen Ro-
man. Das Mittelalter 9: 5563.
Tate, George, 1978: Good Friday Liturgy and the Structure of Lknarbraut. Scandinavian Studies 50:
3138.
Wandhoff, Haiko, 2003: Ekphrasis: Kunstbeschreibungen und virtuelle Rume in der Literatur des
Mittelalters. Berlin.

389
The Herjlfr Legend from Hrjedalen and Its Resemblances to
the Stories of Landnmabk
Olof Holm, Dept. of History, Stockholm University, Sweden

Introduction
Landnmabk is a well-known work in the Old Icelandic literature. It contains hundreds of
short stories describing the first settlers in Iceland were, the places they settled and built upon,
and who their descendants were.
Less well known is a similar story of how the province of Hrjedalen in present-day Swe-
den was settled by a man called Herjlfr hornbrjtr. This story is found in the Norwegian
manuscript AM 114 a qv., which is dated to c. 131525 and now preserved at the Arna-
magnan Institute in Copenhagen.
In recent research it has been argued that the Herjlfr story is a learned product, formed af-
ter and dependent upon Landnmabk. I would like to scrutinize the arguments for this opin-
ion and see if they really are valid.

The stories of Landnmabk

To exemplify what Landnmabk is about, I would like to summarize one of its stories, told
in chapter 344 in the Sturlubk version of Landnmabk. Here we meet a man called Ketill
hingr, who lived in Namdalen in Norway in the time of King Haraldr hrfagri. He was the
son of orkell, Earl (jarl) of Namdalen. We are told that he called together his retinue in order
to help his relative rolfr Kveld-lfsson, whom King Haraldr wanted dead. This attempt
failed and rolfr was executed. After Ketill had taken his revenge, by killing the men who
had slain rolfr, he had to flee together with his family. He then came to Iceland and colo-
nized land there. He lived at Hov. His first son was Hrafn, who first told the law in Iceland.
His other son was Helgi, who married Valds Jlgeirsdttir and had children. We then hear
about their descendants and of other children of Ketill and their descendants in several gen-
erations.
Unfortunately, the oldest version of Landnmabk has not survived. It was probably edited
in Iceland as early as c. 1100 or in the early 12th century. What we do have are versions from
the 13th and early 14th century as well as a couple from later time. These versions are ex-
panded in relation to their originals and contain material borrowed from sagas and other liter-
ary sources. The oldest version of Landnmabk, however, cannot possibly have been based
on literary sources, since such sources simply were not at hand in Iceland before c. 1100 the
editing of Landnmabk was in fact one of the very earliest literary achievements in Iceland.
Instead, the oldest Landnmabk must have been based on oral tradition on legends. In light
of the large number of freestanding stories included in Landnmabk and the enormous
amount of topographical and prosopographical details, there must have been a great number
of local informants, living in all regions of Iceland, who transmitted these legends. By whom
and for what purpose all this material was collected and compiled in the earliest version of
Landnmabk is another question, which, on the whole, is shrouded in mystery (Benediktsson
1969 and 1978; Rafnsson 1974; Meulengracht Srensen 1993: 826).

The Herjlfr story

The manuscript AM 114 a qv. consists of several texts of diverse nature (Table 1). The first
text, where the Herjlfr story is included, is a compilation of different sources concerning the

390
border between Norway and Sweden (NgL 2, pp. 48791). This compilation can be divided
into several sections, going back to different sources. I will follow the division made by Nils-
Erik Eriksson (1984). He divides the compilation into seven main sections lettered AG (Ta-
ble 2). The Herjlfr story is found in section E, which reads as follows:

Sua bar ordr i Trosavik vitni ok XII ellimenn med honum at Noregs menn bygdu fyrstunni He-
riardal med eim atburd at madr het Heriulfuer hornbriotr; hann var merkismadr Halfdanar ko-
nongs suarta; hann vard firir konongs reidi ok flydi austr i Sviariki till Anundar konongs; var
ar vl med honum tekin en o missfll honum sua at hann lagdezt med frenkono konongs
eirri, er Hlga het, ok flydi aftr vestr i Noregs kon[on]gs riki i dall, an er a var alaudr, er nu
er kalladr Heriardall. a settuzst ar Hlga ok bygd gerdu ar sem nu heita Sliarosvellir. eirra
son var Hakon valr; hans son Frode; hans son Herlaugr hornstigi; hans son orer droge; hans
son orbiorn makarkalfuer; hans son orbiorn gamle; son hans Dagr; hans son Liotr, er fyrst let
gera kirkiu i Heriardall; hans son orallde; hans son Rafn; hans [son] Eilifur tppimadr; hans
son Biorn; hans son ordr; hans son Haforer; hans son Asulfuer. (NgL 2, p. 490.)

Scholarly comments
The text about Herjlfr hornbrjtr has been discussed by Sveinbjrn Rafnsson in his disserta-
tion Studier i Landnmabk, published in 1974. He concludes that this text is clearly depend-
ent upon Landnmabk (klart beroende av Landnama). One of his arguments for this con-
clusion is that Herjlfr hornbrjtr is also mentioned in Landnmabk. Another of his argu-
ments is that there are, according to him, several learned constructions ([]tskilliga lrda
knepigheter) in the text that are similar to what he finds in Landnmabk. He exemplifies
these learned constructions by mentioning 1) the combining of the name Herjlfr with the
place-name Herjardalr (Hrjedalen) which gives the latter an ethymology, and 2) the empha-
sizing of Herjlfrs link to King Hlfdan svarti, which gives the landnmsmar Herjlfr a
relation to the Norwegian royal family (Rafnsson 1974, pp. 1967, fn. 2).
Nils-Erik Eriksson has also commented on the text about Herjlfr. In his article published
in the Norwegian Historisk tidsskrift in 1984 he draws the same conclusion as Rafnsson,
namely that the text is dependent upon Landnmabk: Texten i avnitt E r sprungen ur
samma politiska och litterra milj som skapat Landnmabk. He repeats the arguments of
Rafnsson and he emphasizes that both the Herjlfr story and the stories in Landnmabk fol-
low a pattern, which can be formalized in four points: 1) the landnmsmar, the settler, is
mentioned by name, 2) he is introduced with rank, kinship, descent etc., 3) his landnm, the
land which he settled or colonized, is described, and 4) finally, his descendants are mentioned
by name (N.-E. Eriksson 1984, p. 403).
These statements by Rafnsson and Eriksson have, to my knowledge, gone unchallenged in
later research despite the fact that they can be seen as controversial. For two distinguished
scholars belonging to an older generation, Nils Ahnlund and Knut Liestl, it was quite obvi-
ous that the text describing how Hrjedalen was settled went back to local traditions (Ahnlund
1924, p. 120; 1926, p. 353; 1948, pp. 4357; Liestl 1929, p. 24). Ahnlund and Liestl have
simply taken this for granted, and it is easy to see why when we read the introductory lines of
section E, quoted above: Sua bar ordr i Trosavik [i.e., Trasviken in Hede parish in northern
Hrjedalen] vitni ok xii ellimenn med honum at Noregs menn bygdu fyrstunni Heriardal med
eim atburd at [] Ahnlund (1948, p. 44) pointed out that the fact that twelve ellimenn eld-
erly men confirmed rdrs oral testimony may indicate the story was a hereditary legend, a
common property for many settlement districts in Hrjedalen.
In order to evaluate these different opinions, it is necessary to look closer at the manuscript
AM 114 a qv. and the compilation of sources concerning the border therein.

391
Table 1. Texts included in AM 114 a qv.
Page(s) Text
1 (Originally a blank page, intended as a cover)
25 The compilation of sources concerning the border between Norway and Sweden (see Table 2)
617 A copy of the polemical pamphlet A Speech against the Bishops from the time of King Sverrir, c. 1200
(Holtsmark 1931, pp. 120)
18a113 An arithmetical passage in Latin (facsimile in Holtsmark 1931, pl. XIV, 4)
18a1430 An account of the border between Norway and Russia (NgL 2, pp. 1523)
18b20 Homilia on the Imbrudagar

Table 2. The compilation of sources concerning the Norwegian-Swedish border in AM 114 a qv. (sec-
tions lettered according to N.-E. Eriksson 1984).
Section Contents
A An adapted text containing testimonies concerning the border between Gtaland and Norway, probably
excerpted from several written sources (NgL 2, pp. 48782)
B A compilation of several testimonies concerning the border between Norway and Gtaland (NgL 2, pp.
4883911)
C Excerpts from a Norwegian vitnebrev (letter of witness) drawn up at a thing assembly (aling) in
Sveg in Hrjedalen between 1267 and 1282, dealing with the border between Norway and Sweden,
primarily alongside Hrjedalen (NgL 2, pp. 48912905; cf. Holm 2003, pp. 15160)
D An account of the border between Norway and Sweden alongside Jmtland and Hrjedalen, after an
anonymous source (NgL 2, p. 490617)
E The story of how Hrjedalen was settled by Herjlfr hornbrjtr (NgL 2, p. 4901829)
F An excerpt from the list of the borders of Jmtland recorded in the medieval law book of Jmtland; this
law book (now lost) was probably an adjusted version of the national Norwegian legal codex of King
Magns lagabtir from 1274 (NgL 2, pp. 49030116; cf. Holm 2003, pp. 1608)
G Excerpts from a treaty drawn up between c. 1260 and 1283, ratifying a number of boundary markers
along the Swedish-Norwegian border, primarily alongside Vrmland and Dalecarlia (NgL 2, p. 49117
34
)

Textual context
The compilation of sources concerning the Norwegian-Swedish border in AM 114 a qv.,
where the Herjlfr story is included, has served as a handy overview of valuable information
about the border, excerpted from different sources of different ages. Its scribe uses, as shown
by Anne Holtsmark (1933, pp. 305), a handwriting and an orthography typical for scribes
working at the royal Norwegian council in Oslo around 131525. The royal Norwegian coun-
cil is obviously the milieu where all the texts in the manuscript were written. However, we do
not know for sure if the extant version of the compilation is the original version or just a copy
of a now lost original. The compilation cannot therefore be dated more closely than to c.
12751325.1
The original sources excerpted in the compilation were at that time preserved all across
Norway. The law book of Jmtland, for example, from which an excerpt of a border descrip-
tion is taken in section F in the compilation, was preserved in Jmtland (Holm 2003, pp. 160
2). The letter drawn up in Sveg between 1267 and 1282, from which excerpts are taken in
section C, must have been preserved in Hrjedalen probably in Sveg since it concerned a
local dispute (Holm 2003, pp. 1515). Other sources, which are excerpted in sections A, B
and G, must have been preserved further to the south in the Norwegian kingdom since they
concern the southern part of the borderline (cf. Table 2).

1
C. 1275 is the earliest possible date due to the dates of the sources excerpted in sections C, F and G of the com-
pilation. Cf. Table 2.

392
 To put together this compilation of geographically very widespread sources must have im-
plied quite a large undertaking. The more precise background to why this undertaking was
done at that time some time between c. 1275 and c. 1325 is not known.
Even if we do not know the exact background to the production of the compilation, we can
at least state that the compiler had considered rdrs testimony, confirmed by oath, to have
sufficient value to be included along with the different accounts of the Swedish border. The
value lay in that rdr claimed that Norwegians had settled Hrjedalen. This is underlined by
the compiler in the introductory lines in section E, where he states that the testimony of rr
shows at Noregs menn bygdu fyrstunni Heriardal. Obviously, the reason he chose to include
rrs story was that the information in it could be used for political purposes. It could be
used for claiming that the province of Hrjedalen always had belonged to Norway. I will re-
turn to this political aspect later on.

The oral background

There is no reason to doubt what the compiler explicity states in the introduction of section E,
namely that the text about Herjlfr hornbrjtr really is based on testimonies by rr in Tras-
viken and twelve elderly men in the province of Hrjedalen. The compiler was faithful to his
sources. This can be seen, for instance, in section F, where he accurately quotes an account of
the Swedish border also known from another source (JHD 2, no. 159; cf. Holm 2003, pp.
1604). How, then, do Rafnsson and Eriksson explain that the text, based on testimonies by
farmers in Hrjedalen, could be a learned product, formed after and dependent upon Land-
nmabk? Actually, they do not explain this, and that is a weakness in their treatments of the
Herjlfr story. In my opinion, their conclusions are wrong. We have no information of Land-
nmabk ever being read by others than Icelanders (and some Icelandic immigrants living in
Norway) during the Middle Ages, and even if it were, it is unreasonable to think that farmers
of Hrjedalen had read it and then formulated a story in the same pattern as the stories in
Landnmabk. Instead, the resemblances between the Herjlfr story and the stories in Land-
nmabk are much easier explained by being exponents of a similar oral narrative tradition
spread in the Norse world.
It is understandable that popular traditions shew similar features over great geographical
distances. Brynjulf Alver has pointed out that in our old, oral society there was always great
competition among the legends kept alive. If a legend was going to survive in this competi-
tion, it had to be conventionalized and schematized until it received such a form that the tell-
ers wanted to tell it and the listeners wanted to listen to it (Alver 1962, p. 111). In my opinion,
in the story of how Hrjedalen was first settled and in Landnmabks stories we see exam-
ples of legends that have yielded to standardizing rules in the Norse oral, narrative tradition in
a similar way.
Concerning the combining of the name Herjlfr with the place-name Herjardalr in the
story, I would say that this is not a learned construction, as Rafnsson and Eriksson presume,
but instead a common feature found in popular stories in Scandinavia. Per Vikstrand (2008)
has observed that in popular stories explaining place-names, it is very common that a place-
name is explained from the name of a fictitious figure that is assumed to have had a connec-
tion with the place in its establishing phase.2

2
Another example of such a popular story, recorded before c. 1325 in the same region, is the story about
Blafinnungs tiorn, Blafinnugs tiorn, Blafinz tiorn, a small lake that served as a boundary marker between
Jmtland (Norway) and ngermanland (Sweden) in the 13th century. This story explains the place-name of the
lake from the name of a fictitious Smi, Bl-Finnr, who is said to have been drowned in the lake by the first
Christian, rni illi from Hllesj. The story was recorded in the above-mentioned border description in the me-
dieval law book of Jmtland, excerpted in section F of the compilation (cf. Table 2). On the oral background of

393
 What about Herjlfr hornbrjtr then? How do we explain that he is mentioned both in the
story of how Hrjedalen was settled and in Landnmabk? In the Melabk version of Land-
nmabk Herjlfr is said to have been the father of rolfr and the grandfather of rasi
rolfsson, who travelled from Hordaland in Norway to Iceland and settled there:

Madr het rasi orolfs son Heriolfs sonar horna briotz. Hann for af Horda landi til Islandz ok
nam land [] (Landnmabk, ed. F. Jnsson, 1900, p. 235.)

In the Hauksbk version of Landnmabk it is also said that his son rolfr and another son,
lfr, were kings in Oppland in Norway:

orolfr svn Heriolfs horna briotz ok Olafr broir hans varv konvngar at Vpplondvm [] rasi
var svn orolfs. Hann for af Horalandi til Islandz ok nam land [] (Landnmabk, ed. F.
Jnsson, 1900, p. 103.)

The most likely explanation to why Herjlfr occurs in several stories, and to why his family in
one case is connected with Hrjedalen, and in another with Oppland, is obviously that he for a
long time had been so well-known in legends that in these stories he was associated with (al-
leged) events he initially had nothing to do with.
We then may ask ourselves: Why did people in Iceland and Hrjedalen in the Early Middle
Ages keep such settlement stories in mind? Why were these stories important? I do not be-
lieve it is a coincidence that we find this type of story in Iceland and in Hrjedalen. At that
time Iceland and Hrjedalen had one thing in common besides being part of the Norse world:
they had both been colonized by farmers in the Viking period and were, thus, relatively young
societies.3 Obviously farmers living in such colonization areas had reason to keep in mind
stories about the origin of their settlements and then pass them on to the next generation. We
also find examples of this in the colonization areas of Swedish Lapland, where peasants in the
early 20th century often knew the names of the first settlers from some 150 years ago and
could tell stories about them (see, e.g., Pettersson 1946, pp. 1945). An example recorded in a
17th century court record from Jmtland, the province north of Hrjedalen, can be quoted. In
this example an old man, testifying in a trial, tells the story of how Stugun an isolated vil-
lage of medieval origin surrounded by huge forests was settled six generations back in time:

Hemming Bengtsson i Stugun om sine ttatijo hr frhrdes, och berttade effter gammal her-
melsse, man effter man sllunda, nembligen att twenne pijgor skolle suttit der p plattzen i een
stugu, som wore barnfdde i Gewgs by, och een man wid nambn Phl befriat sigh med den
ena, och aflade tillhopa tree snner, Ifvar, Oloff och Bengt som begynte oppbruka iorden. Hust-
ru Brita hr kommen aff Ifwars tt och barn, ssom frst, Ifwar Phlsson 2) Erich Ifwarson 3)
Iwar Ersson och 4) Erich Ifwarson som var hustru Britas fader [] (Court record of Ragunda
1687, quoted after K. G. Eriksson 1998.)

Clearly history was important for all these descendants of the first settlers, probably because it
gave them an identity in areas where their way of life was still quite new.

this story, see Holm 2003, pp. 1646.

3
Of course, people had lived in Hrjedalen since the Stone Age, but the historical villages and farms in Hrjeda-
len cannot be traced further back in time than the Viking Age or the Early Middle Ages. Hence, in Hrjedalen
there are no settlement names typical for the Iron Age (e.g., hem-, hov-, sta(d)-, vi- and vin-names, as well as
theoforic names). Brink 1990; Brink & al. 1994, pp. 137, 1401. The oldest traces of farming (fossil fields in
Tordalshgen and Annflon, Sveg parish, and Hedningsgrdet, Tnns parish) in Hrjedalen are dated to c. 900
1200. Norrman & Robertson-kerlund 1979; Hansson 1997, p. 130.

394
 Both in the stories of Landnmabk such as the story of Ketill hingr, quoted above
and in the Herjlfr legend great attention is paid to the honourable deeds of the first settlers.
Ketill hingr was the earls son who stood up against King Haraldr hrfagri, while Herjlfr
hornbrjtr was the merkismar of King Hlfdan svarti who won the heart of a member of the
Swedish royal family. Such ingredients in the stories were important, if not necessary, for
keeping the stories alive in oral tradition (cf. above). But these ingredients, together with the
genealogies in the stories, were probably also important because they lent lustre to the de-
scendants of the settlers (cf. Mundal 2001, p. 62). If there were deeds worth telling about any
descendant of a later generation, these were included too, as when we hear that Ketills son
Hrafn first told the law in Iceland, and that Herjlfrs descendant in the ninth generation,
Ljtr, first had a church built in Hrjedalen.
I have drawn the conclusion that the story of how Hrjedalen was settled, found in AM
114 a qv., has its origin in the popular culture of the peasants of Hrjedalen and is unaffected
by literary learning. In Hrjedalen this story had been retold and memorized again and again,
by one generation after the other. By describing the kinship with the first settler and the hon-
ourable biography of this settler, the narrator was able 1) to bridge the distance between the
past and the present, demonstrating the rightness of the story, and 2) to lend lustre to his own
family. Although it is not explicitly stated in the text, the narrator witnessing in this case,
rr from Trasviken, must be assumed to be the son of slfr Hafrisson, the last descen-
dant mentioned in the legend otherwise the genealogy does not make sense.

The recording
Details concerning the testimonies sworn by rr in Trasviken and the twelve elderly men
are not known. In the preserved compilation it is not said on what occasion these testimonies
were given and recorded. It is not even clear whether the testimonies were originally recorded
in a letter, although this seems probable. All such information has been excluded by the com-
piler and this is not only the case in section E, where the Herjlfr legend is quoted, but in all
the other sections of the compilation as well. In no case does the the compiler mention, for
example, the date of a certain source he is quoting, though at least the letter excerpted in sec-
tion C and the treaty excerpted in section G must have been dated in the original.
As I pointed out earlier, it must have been quite a large undertaking to collect information
from the geographically widespread sources on which the compilation is based. This work
also must have involved local officials in the border provinces of the Norwegian kingdom,
officials who possessed knowledge of local sources of interest in the matter. Maybe the per-
son who made the compiler aware of the letter concerning the Norwegian-Swedish border,
which was drawn up in Sveg in Hrjedalen between 1267 and 1282, also knew of a living
tradition in Hrjedalen, one that described how this province at the border had been settled
from Norway by Herjlfr hornbritr. Perhaps this led to the recording of rrs story in a
now lost letter. In that case section E in the compilation is excerpted from this supposed letter.
The compiler has chosen to quote rrs sworn and confirmed story seemingly literally and
has only added a minimum of background information in the introductory lines.
One reason to why rr in Trasviken and nobody else was appointed the main witness
may have been that he was considered to possess extraordinary knowledge acquired by
memorization. He may have been what we, using a contemporary term, could call a minnin-
garmar, a person who enjoyed a certain authority in this basically oral society by being old
and well informed (cf. Brink 2005, p. 94). Also the twelve ellimenn elderly men, who testi-
fied with him, must have possessed some kind of authority.

Historicity

395
rr and the twelve men testifying must have been convinced that the Herjlfr legend was
true otherwise they would not have sworn to it. Those who recorded the testimonies con-
cerning the Herjlfr legend, including the compiler, must also have been convinced of its ve-
racity. And the story certainly appears to be trustworthy, as rr was able to account for a
lineage of 15 generations descending from Herjlfr hornbrjtr and his mistress Helga. rr
was also able to give the correct name of a Norwegian king ruling about 15 generations ago at
this time, namely Hlfdan svarti in the 9th century, the father of King Haraldr hrfagri.
But to what extent can we trust this legend, recorded c. 1325 at the latest? We know
through modern research that a characteristic of oral tradition is that it is variable. After some
150200 years, historical legends have often changed so much that rather little of the facts
told in the legends are in accordance with the real events that once took place. We know this
from studies where the contents of historical legends have been checked against good histori-
cal sources such as court records (cf., e.g., Nordb 1928; Hodne 1973). There are various rea-
sons for this. Those telling a legend and those who retell it afterwards always possess a living
tradition of such matters as the legend is dealing with. From this tradition they pick out details
and motifs which fill out the legend. As Alver (1962, p. 111) has pointed out, this process
probably often occurs subconsciously. A historical legend then often tends to, so to say, be
improved over the course of time and become more and more unhistorical. Ahnlund once ex-
pressed this in the following way: Folksgnen var enligt all erfarenhet ingen som helst kri-
tik, nr den verkstller sina kombinationer (Ahnlund 1948, p. 58). Against this background
we cannot expect the Herjlfr legend to be a reliable historical source.
I have already mentioned that the similarity between the names Herjlfr and Herjardalr is
typical for unhistorical popular stories explaining place-names (cf. Ahnlund 1948, p. 54). But
there are other features of the legend that seem to be unhistorical as well, for instance the
statement that Herjlfr and Helga, as the first settlers of Hrjedalen, had settled down and
cleared land at a place sem nu heita Sliarosvellir. Sliarsvellir, the meadows at the river
Slyans outlet into the river Hrjn, near the south-western border of Hrjedalen, would have
been a rather dismal place to start a farm, at least if the intention was to clear land and grow
cereals (cf. Andersson 1987, p. 22). This place lies at the bottom of a valley, at high altitude,
where the risk of severe frost was much greater than, for example, close to the place where the
church was built in Lillhrdal. However, the area around Slyans outlet, with its lush grass
vegetation, was excellent for haymaking and grazing.
Why does the legend tell us that this was the place where Herjlfr cleared land and settled?
I suspect, as Ahnlund (1948, pp. 534, 57) also did, that the legend has been fixed to this
place, because there is a large mound that is strikingly similar to a burial mound (see picture
in Hemmendorff 2002, p. 135). The mound is actually a natural formation, formed by erosion,
but the farmers in Hrjedalen have in all times considered it to be a real burial mound. In the
records from the Swedish inquiries into the ancient monuments, made in the 1680s, this
mound holds a prominent place and it even has a name: Kungshgen the kings mound
(Rannsakningar efter antikviteter 1, p. 243).

Bibliography
Ahnlund, Nils, 1924: Oljoberget och Ladugrdsgrde. Svensk sgen och hvd. Stockholm.
1926: Folktraditionen som historisk klla. In: Historisk tidskrift (Stockholm) 46. Pp. 34263.
1948: Jmtlands och Hrjedalens historia 1. Intill 1537. Stockholm.
Alver, Brynjulf, 1962: Historiske segner og historisk sanning. In: Norveg 9. Pp. 89116.
Andersson, Kerstin, 1987: Sgnen om Hrjulf Hornbrytare och Sliarosvellir. In: Oknytt 8:12. Pp. 17
23.
Benediktsson, Jakob, 1969: Landnmabk. Some remarks on its value as a historical source. In: Saga-
book 17. Pp. 27592.

396
 1978: Some problems in the history of the settlement of Iceland. In: The Vikings. Proceedings of
the Symposium of the Faculty of arts of Uppsala university June 69, 1977. Ed. by Th. Andersson
& K. I. Sandred. UppsalaStockholm. (Acta Universitatis Upsaliensis. Symposia Universitatis Up-
saliensis annum quingentesimum celebrantis 8.) Pp. 1615.
Brink, Stefan, 1990: Cult sites in northern Sweden. In: Old Norse and Finnish religions and cultic
place-names. Based on papers read at the Symposium on encounters between religions in Old Nor-
dic times and on cultic place-names held at bo, Finland, on the 19th21st of August 1987. Ed. by
T. Ahlbck. boStockholm. (Scripta Instituti Donneriani Aboensis 13.) Pp. 45889.
2005: Verba volant, scripta manent? Aspects of early Scandinavian oral society. In: Literacy in
medieval and early modern Scandinavian culture. Ed. by P. Hermann. Odense. (The Viking collec-
tion 16.) Pp. 59117.
Brink, Stefan, Korhonen, Olavi & Wahlberg, Mats, 1994: Bebyggelsenamn. In: Kulturminnen och
kulturmiljvrd. Temared.: K.-G. Selinge. Stockholm. (Sveriges nationalatlas.) Pp. 13445.
Eriksson, Karl Gran, 1998: Stuguslkter ett verk med frtjnster och brister. In: JLS-nytt. Med-
lemsblad fr Jmtlands lns slktforskare 75. Pp. 68.
Eriksson, Nils-Erik, 1984: Det ldsta grnsdokumentet om Hrjedalen och Jmtland AM 114a, 4:o. In:
Historisk tidsskrift (Oslo) 63. Pp. 384415.
Hansson, Anders, 1997: Jakt och jordbruk i Jmtlandsfjllen under yngre jrnlder och tidig medeltid.
In: Arkeologi i Mittnorden. Ett symposium kring nya arkeologiska forskningsrn. Red.: K. Gull-
berg. Vasa. (Acta antiqua Ostrobotniensia. Studier i sterbottens frhistoria 4.) Pp. 12735.
Hemmendorff, Ove, 2002: Arkeologiska utflykter. stersund. (Upptck Jmtland/Hrjedalen.)
Hodne, Bjarne, 1973: Personalhistoriske sagn. En studie i kildeverde. Oslo.
Holm, Olof, 2003: Den norsk-svenska riksgrnsens lder och hvd. En studie av rikssamlingsprocesser
och grnsbildning i mellersta Skandinavien. In: Collegium medievale 16. Pp. 135237.
Holtsmark, Anne, 1931: En tale mot biskopene. En sproglig-historisk underskelse. Oslo. (Skrifter
utg. av Det Norske Videnskaps-Akademi i Oslo 2. Hist.-filos. klasse 1930:9.)
JHD = Jmtlands och Hrjedalens diplomatarium. Utg. av Jmtlands (lns) biblioteks diplomatarie-
kommitt & al. 1. Under red. av R. Swedlund & al. 1943. stersund.
Landnmabk IIII. Hauksbk. Sturlubk. Melabk m.m. Udg. ved Finnur Jnsson. 1900.
Kbenhavn.
Liestl, Knut, 1929: Upphavet til den islendske ttesaga. Oslo etc. (Instituttet for sammenlignende
kulturforskning. Ser. A. Forelesninger 10 a.)
Meulengracht Srensen, Preben, 1993: Fortlling og re. Studier i islndingesagaerne. rhus.
Mundal, Else, 2001: Genealogiane i norrne kjelder ideologi og funksjon. In: Norsk slektshistorisk
tidsskrift 38. Pp. 5764.
NgL = Norges gamle Love indtil 1387. Udg. ved R. Keyser & al. 15. 184695. Christiania.
Nordb, Halvor, 1928: ttesogor fr Telemark. Oslo. (Skrifter utg. av Det Norske Videnskaps-
Akademi i Oslo 2. Hist.-filos. klasse 1928:1.)
Norrman, Jan & Robertson-kerlund, Agneta, 1979: Kulturhistorisk inventering och arkeologisk un-
derskning inom dmningsomrdena fr Svegs och Byarforsens kraftverk, Svegs sn, Hrjedalen.
Stockholm. (Riksantikvariembetet och Statens historiska museer. Uppdragsverksamheten. Rapport
1979:22.)
Pettersson, Olof, 1946: Gamla byar i Vilhelmina 3. Skogstrakterna i ster och Volgsjtrakten. Utg. av
H. Geijer & K.-H. Dahlstedt. Stockholm. (Etnologiska kllskrifter 1.)
Rafnsson, Sveinbjrn, 1974: Studier i Landnmabk. Kritiska bidrag till den islndska fristatstidens
historia. Lund. (Bibliotheca historica Lundensis 31.)
Rannsakningar efter antikviteter. P Akademiens uppdrag utg. under redaktion av C. I. Sthle 1. Upp-
land, Vstmanland, Dalarna, Norrland, Finland 1. Text. Utg. av I. Olsson, N.-G. Stahre & C. I.
Sthle. 1960. Stockholm. (Kungl. Vitterhets Historie och Antikvitets Akademien.)
Vikstrand, Per, 2008: Folketymologier. Om ortnamnens meningsskapande funktion. In: Nordiska
namn Namn i Norden. Tradition och frnyelse. Handlingar frn Den fjortonde nordiska namn-
forskarkongressen i Borgarnes 1114 augusti 2007. Red. av G. Kvaran & al. Reykjavk & Uppsala.
(NORNA-rapporter 84.) Pp. 4618.

397
 Srla saga sterka and Rafns edition
Silvia Hufnagel, Arnamagnan Institute, University of Copenhagen, Denmark
Srla saga sterka is a young fornaldarsaga which exists in thirty-two paper manuscripts dat-
ing from the seventeenth to the twentieth centuries. After a short definition of the fornal-
darsgur, the content and different versions of Srla saga sterka will be presented. Srla
saga sterka exists in two versions, a shorter edited by Rafn on the basis of Rask 32 and a
longer edited by Birner on the basis of Stock. papp. fol. nr 56. Other manuscripts seem to
follow a mixed version. The manuscripts which were accessible to Rafn when he prepared his
edition will be described, and the possible reasons for Rafns choice of the base text follow. In
the end the consequences of this choice will be highlighted.

Fornaldarsgur norurlanda
The Fornaldarsgur norurlanda are sagas that are set in the Northern countries (other than
Iceland) in the period before Icelands settlement. Their main features include quests, bridal
quests, fights of all sorts, viking raids, revenge, and supernatural figures and objects.
The term fornaldarsgur norurlanda was coined by Rafn in his three-volume edition
from 182930, and literary means stories of ancient times (fornld) of the northern lands
(norurlnd). Today scholars emphasise the difference between the fornaldarsgur as a cor-
pus and as a genre. As a corpus, the term refers clearly to the thirty-one stories in Rafns edi-
tion, but as a genre, the term is less clearly defined.
Fornaldarsgur as a genre are defined in two different ways: Firstly, as a genre that differs
from other genres, and secondly, as a genre with subgroups that differ from each other. As a
genre, the difference between them and the slendingasgur is the setting of time and place,
and the difference between them and the konungasgur is the occupation of the main hero and
the time setting. Doubtless, scholars would also add the difference in the degree of historicity
and the sense of reality as a dividing factor.
There is a long tradition of distinguishing three subgenres within the fornaldarsgur.
Reuschel (1933) identified a heroic, a viking and a popular circle. Einar l. Sveinsson (1959)
and Schier (1970) divide the sagas into a heroic, a viking and an adventure tradition. Jnas
Kristjnsson (1988) characterises them as heroic tales, viking tales and romances. Hermann
Plsson (1985), however, distinguishes only between two groups, hero legends and adventure
tales/viking romances. Common to these groupings is that there are the heroic stories on one
end of the spectrum, and the adventure stories, or popular romances, on the other. The heroic
stories are in the tragic mode, with the hero dying at the conclusion of the saga, whereas the
adventure stories have a happy ending. Viking stories seem to be situated between the two
other subgenres, and are, depending on the scholar, either characterised by having a tragic
ending (cf. Reuschel 1933) or by having a happy ending (cf. Plsson 1985). Futhermore, some
of these subgroups share features with other saga-genres, e.g. the bridal quest is a common
feature of the riddarasgur (cf. Kalinke 1990). This shows that the subgenres are more like
general traits in the stories, rather than fixed groups with strict boundaries, as Reuschel (1933)
and OConnor (2006) have argued.
The distinction between fornaldarsgur and other genres has also sparked discussion, and
some texts seem to belong to more than one genre. Yngvars saga vfrla, for example, is
sometimes considered a fornaldarsaga (cf. Plsson 1985:38), and sometimes not: Einar l.
Sveinsson categorises the story as only beslgtede med f[ornaldarsgur] (Sveinsson
1959:501). Sveinsson also regards a few other ttir, published in Guni Jnssons and
Bjarni Vilhjlmssons edition (194344), and in Guni Jnssons edition (1950, 195459),

398
as related to the fornaldarsgur. These are, like Yngvars saga vfrla, very similar to the
fornaldarsgur, but are set in a historic past after Icelands settlement instead of the ahistoric
time before ca. 8701. Sveinsson seems to use the criteria of the ahistoric past as relevant for a
distinction between fornaldarsgur and other genres, whereas Plsson does not. Plsson
counts the sagas set in the historic past as exceptions within the group of fornaldarsgur, but
for him there is no doubt about Yngvars saga vfrla being a true fornaldarsaga. This
discussion shows again the difficulty of defining fornaldarsgur and their subgenres and
distinguishing them from other literary genres.

Srla saga sterka

Srla saga sterka is considered to be one of the younger fornaldarsgur, dating from the fif-
teenth century (cf. Plsson 1988 or Simek/Plsson 2007). The story is transmitted in thirty-
two manuscripts, all of which are post-medieval, and is also found in eight sets of rmur, one
of them from medieval times (cf. Driscoll/Hufnagel). Rafn claims in his edition (1829
30:3:11) that although there are textual differences, they are so minor that one cannot talk of
more than one version.
Srli, son of King Erlingur in Uppland, sets out on a Viking voyage at the age of fifteen.
On his way home after a successful season of raiding, he becomes stranded in Blland, where
he kills blmenn. Srli fights a troll woman, Mna, whose life he spares in change for her
loyalty, an armour that no weapon can destroy, and a sword that can cut stone as well as steel.
Later, Srli and his father and brother help King Haraldur of Norway against berserks who
want to marry Haraldurs daughter. Srli kills the berserks, but refuses the hand of King Har-
aldurs daughter as reward. He sets out on a Viking voyage again and encounters King Hlf-
dan Brnufstri, whom Srli kills and whose beautiful dragon-headed ship and all other be-
longings he takes. Srlis father predicts revenge from King Hlfdan Brnufstris sons Sig-
mundur and Hgni, and Srli sets out to offer them wergild. At the same time Hlfdans sons
are whetted to avenge their father by their sister Marcibil, and the younger son, Hgni, sails to
Uppland to kill Srli. However, Hgni and Srli do not encounter each other on their jour-
neys.
Hgni meets only Srlis father and King Haraldur of Norway, who came for support; he
kills them both and tries to plunder Ingibjrgs, Srlis sister, bower, but is prevented from
doing so by a magic fog that the foster-mother of Ingibjrg conjured. He sees Ingibjrg and
falls in love with her, but does not try to abduct the princess for fear of the magical powers of
her foster-mother, and returns home instead. The story switches to Srli, who meets King
Hlfdan Brnufstris older son, Sigmundur, and offers wergild, which is refused. A fight
between them results in Srlis victory, and the few survivors of King Hlfdans army barri-
cade themselves in the town, where they are joined by Hgnis brother-in-law. The fight re-
sumes, and again Srli is victorious but only until Hgni returns from his victory over
Srlis father. A duel between Hgni and Srli ends with a draw, and they become sworn
brothers and marry each others sisters.
Srla saga sterka was first edited by Birner in his magnificent Nordiska Kmpa Dater
from 1737. Birner based his edition on Stock. papp. fol. nr 56, which is a copy of Stock.
papp. 8vo nr 7. Rafn used Rask 32 as the base text in his edition from 182930.

Biography of Carl Christian Rafn

1
Einar l. Sveinsson most probably refers to Helga ttr rissonar, Tka ttr Tkasonar and orsteins ttr
bjarmagns, all of which are set at King lafr Tryggvasons (9951000) court in Norway. These ttir are
indeed not necessarily regarded as fornaldarsgur.

399
Carl Christian Rafn was born on the 16th of January 1795 in Brahesborg on the island of Fyn,
Denmark, and died on the 20th of October 1864 in Copenhagen, Denmark. His parents were
tenants on a dairy farm. He went to grammar school in Odense, and in 1814 he began study-
ing law at the University in Copenhagen, where got his first degree in 1816. In 1818 he initi-
ated the foundation of the National Library of Iceland (Landsbkasafn, then stiftsbibliotek),
and in 1821 Rafn gave up his teaching position at the military school in Copenhagen to take
up a new position at the Arnamagnan Collection/University Library of Copenhagen, even
though that post was unpaid. There he reviewed the Arnamagnan Collection and gained in-
valuable insight into the collection and the manuscripts contents.
One of Rafns first publications was Nordiske Kmpe-Historier, a three-volume collec-
tion of Danish translations of fornaldarsgur norurlanda (in vol. 1 and 3) and ireks saga
af Bern (in vol. 2), published in 182126 by Popp in Copenhagen. He re-published most of
these sagas and others of the same genre under the title Nordiske fortids historier in 1828
30, again by Popp in Copenhagen. He then published the stories of vol. 1 and 3 of Nordiske
Kmpe-Historier and more fornaldarsgur in normalized Old Norse under the title Fornal-
dar sgur nordrlanda in 182930, again in Copenhagen in three volumes.
Rafns other well-known publications include Jmsvkinga saga (1824), Freyinga
saga (1832), Antiqvitates American (1837), Antiquits Russes (185052) and, together
with Finn Magnusson, Grnlands historiske Mindesmrker (183845). He also published a
multitude of articles concerning runes and historic monuments.
Jmsvkinga saga was published as a test for the newly founded Nordiske
Oldskriftselskabet with the aim to publish and translate Old Norse literature and to further the
studies of Old Norse literature, history and culture, especially archaeology. The society was
founded in 1825 by Rafn, Rasmus Rask and others, and came under the presidency of the
monarch, King Frederik VI. Rafn was the first secretary of the society.
(http://www.oldskriftselskabet.dk). Jmsvkinga saga turned out to be very popular and had
more than 1,000 subscribers in Iceland alone.
Antiqvitates American, published in 1837 in Copenhagen by Rafn and others, is called
Rafns bermteste Vrk (Grndal 1869:30) and caused a great deal of attention to and
interest for the North. Its contents are sagas, annals, geographical studies etc. that refer to
North America. The texts are presented both in the original language and a Latin translation
with some excerpts in English. Antiquits Russes followed a similar pattern, although
focusing on the East.
During his life, Rafn was a member of a large number of societies and received many
honorary degrees. He established connections with foreign scholars and antiquarians to
further knowledge about the North, and published and translated many Old Norse texts. His
scholarly work, though, is said to be lacking Originalitet og Skarpsind (Dansk biografisk
Leksikon 1899:353), and the translations and editions are sometimes not considered careful
enough. Nevertheless, his edition of the fornaldarsgur norurlanda has proved to have had a
great deal of influence.

Manuscripts in Copenhagen
At the time when Rafn prepared his edition of Fornaldar sgur nordrlanda, there were five
manuscripts containing Srla saga sterka in Copenhagen: AM 168 fol., AM 171 a fol., AM
560 d 4to, Rask 32, and NKS 1806 4to. Rafn was most likely also aware of other manuscripts
in Iceland containing the saga, as there are manuscript catalogues of the National Library of
Iceland, then stiftsbibliotek, in the collection of the Nordiske Oldskriftselskabet, of which he
was the secretary. Furthermore, he was in contact with Icelandic scholars who might have
given him information about manuscripts contents.

400
 a) AM 168 fol. is a paper manuscript from the end of the seventeenth century. It comprises
nineteen leaves: Egils saga einhenda ok smunda berserkjabana is on fols 1r10v, and
Srla saga sterka on fols 11r19v. This codex was once part of a larger codex, together
with AM 101 fol. and AM 147 fol., but it was taken apart by rni Magnsson. It once
also contained a now lost text of Hlfdanar saga Eysteinssonar after fol. 10v.
b) AM 171 a fol. is a paper manuscript from the second half of the seventeenth century and
comprises eighteen leaves: Srla saga sterka is on fols 1r11r, a Faroese verse about
Srla saga sterka on fol. 11r, and Sturlaugs saga starfsama on fols 11v18v.
c) AM 560 d 4to is a paper manuscript, written on the farm Geirrareyri in 1707. It com-
prises one hundred and twenty-two leaves: Srla saga sterka is on fols 1r18r, and
Egils saga Skallagrmssonar on fols 19r122v. The manuscript has several lacunae and
was most probably part of a larger codex comprising AM 560 ad 4to. These manu-
scripts contain slendingasgur, some of which are influenced by riddara- and fornal-
darsgur.
d) Rask 32 is a paper manuscript from the second half of the eighteenth century. It con-
tains a collection of fornaldar- and riddarasgur on two hundred and twenty-two
leaves. Srla saga sterka is written on fols 8v19r. Fol. 2 was written by the manu-
scripts owner Benedikt Bogason (17491819), a wealthy farmer from Staarfell in
western Iceland, who was interested in poetry, philosophy, theology and agriculture.
His father, Bogi Benediksson, was the owner of the printing press at Hrappsey, which
was run from 1773 to 1795. Fols 94v:25100r and 175r188v were written by Gsli
Jnsson (16991781), priest in the Saurbr ing in western Iceland. His son, the priest
lafur Gslason (17271801, Mla-lafur), wrote fols 3r94v:24, 100v174v and
189r222v. There is an older foliation on fols 13r (300) and 113r (400), and a note in
the margin of fol. 154v (last leaf of a quire) states that Hlfs saga ok Hlfsrekkum is
supposed to follow. This leads to the conclusion that the manuscript was once bound
differently, containing other stories.
e) NKS 1806 4to was written in the second half of the eighteenth century by the younger
Olavius, Jn lafsson jun. (ca. 173875). The paper manuscript contains only Srla
saga sterka on twenty-four leaves.

Versions of Srla saga sterka in the manuscripts

Rafn states in his edition (182930:3:11) that even though there are textual differences in the
manuscripts, they hardly change the meaning of the story and are therefore of little interest or
value and not extensive enough to justify speaking of different versions. Busch (2002:167),
however, disagrees and distinguishes between a shorter and a longer version.
The five manuscripts that were in Copenhagen in Rafns time can be divided into three
groups. The first group comprises AM 560 4to, Rask 32 and NKS 1806 4to. The second
group comprises only AM 168 fol., and the third group consists only of AM 171 fol.
Birners edition, a copy of which Rafn had, would constitute a fourth group.
Birner presents the longer version according to Stock. papp. fol. nr 56 in his edition from
1737 and follows his base text closely. Compared with the shorter version, this narrative is
more detailed in battle scenes, and has some additional characters mostly in combat scenes
as e.g. Tjrfi hinn finnski in the battle between King Haraldur of Norway, and Garar and
Tfi from Morland. This version sometimes gives more explanations, e.g. the reason why
Mnas sister fought against her and what relationship King Sigurur of Denmark had with
King Hlfdan Brnufstri. Women are depicted slightly differently; less favourable, level-

401
headed or wisely counselling than women in other fornaldarsgur2, for example Brana in
Hlfdanar saga Brnufstra. Steinvr, King Haraldurs daughter, is fascinated by Srli and
cannot stop looking at him. Duke Casto tells Marcibil off when she, full of rage and sorrow,
tries to egg him and her brother Hgni on to avenge her father, King Hlfdan Brnufstri. At
the end of the story, Marcibil invokes her mothers anger in the grave, if she, the mother,
knew that Hgni made peace with Srli instead of killing him to avenge that Srli killed
Hgnis and Marcibils father, King Hlfdan Brnufstri. Hgni offers Srli peace on the
condition that he acquires Srlis beautiful sister, Ingibjrg, as a bride. The longer version
thus has a stronger focus on the adventure tale and the bridal quest than the shorter version.
The version in AM 171 a fol. is even longer and more detailed than that in Birners edi-
tion and also differs greatly in wording. King Erlingur is a descendant of inn, whereas
Dagny, King Erlingurs wife, is komin af sum. Although a few scenes are shorter than in
Rask 32, there are usually longer fights with more characters. Some of the battle scenes have
slightly different sequences and action. In the battle between King Haraldur of Norway,
fought with the participation of Srli and his family, against Garar and Tfi from Morland,
Loinn is killed earlier and supplanted by Smur. Furthermore, Tfi does not become a
dragon. The furious Marcibil eggs Duke Casto and her brother Hgni on to avenge her father,
King Hlfdan Brnufstri, before Srli sets out to offer wergild for her father. Here, the cli-
max is built up in a different and more dramatic way, as the audience knows earlier of the
imminent danger than the hero. Some details explain later actions, e.g. King Hlfdan
Brnufstris insults to Srli when they meet give a strong reason for Srlis anger. Women
are again depicted meaner, weaker and less self-ruling or wise than in the shorter version.
Marcibil threatens Duke Casto with sexual deprivation if he does not avenge her father. Ingib-
jrg leaves the decision as to whether she should accept Hgnis offer of marriage to her
brother and mother. This version has an even stronger focus on the adventure tale than the
version in Birners edition.
AM 168 fol. is the shortest version of all. Fewer details are given, for example in AM 168
fol., fol. 13v it says:

ok hleypti me a borgina. Kongr hlt nu eina rstefnu hvat til skilldi gjra slkum vanda,

whereas the same section in Rask 32, fol. 11v reads:

ok hleypti konngr undan me etta li borgina, en eir brr eltu flttan allt at borgarportum,
ok lauk sv striinu. Haraldr konngr hlt n rstefnu vi li sitt, hvat til brags skyldi taka
eim vanda ok nausynjum, er eim voru n at hndum komnar.

In this way, AM 168 fol. seems to be an extract of the version in Rask 32, although there are
some differences in some details and sequences of the story. Some details agree with the
longer version, e.g. Srli and Sigvaldi chase and kill all of Tfis and Garars army.
NKS 1806 4to is a copy of AM 560 d 4to, which has two lacunae. Rafn might have there-
fore excluded these manuscripts as possible base texts. His choice for a base manuscript for
his edition of Srla saga sterka was eventually Rask 32, cited as Nr. 82b 4blf. Handrita-
Vibtirnum (Additamentis) hsklans bkasafni (Rafn 182930:3:11). When comparing
the fornaldarsgur that exist in more than one version and are printed in both Birners and
Rafns editions, it seems that Rafn prefers to edit a different version than Birner, if possible.
However, in this case it was by necessity, as there was no manuscript containing the longer
version in Copenhagen. AM 171 a fol. seems to be too close to the longer version, and AM

2
See also rmann Jakobsson, Annette Lassen and Agneta Ney (2003:10).

402
168 fol. seems to be an exctract of the version in Rask 32, and AM 560 d 4to and NKS 1806
4to are defective. The choice to use Rask 32 as base text might have therefore been fairly easy
for Rafn.

Consequences of Rafns choice of base text

There are three subsequent editions of the fornaldarsgur norurlanda, each one following
Rafns edition closely.
Valdimar smundarsons three-volume edition (Reykjavk, 188591) is mainly based on
Rafns edition from 182930, but with a slightly different orthography and without a variant
apparatus. Srla saga sterka, which is printed in the third volume on pp. 30943 without any
information, seems to follow Rafns edition without changes.
Guni Jnsson and Bjarni Vilhjlmsson also based their three-volume edition (Reykjavk,
194344) on Rafns edition, but take into account more recent editions containing
fornaldarsgur or their vsur. The editors added four other fornaldarsgur, normalised the
spelling further and decided against a variant apparatus. Srla saga sterka, contained in the
third volume on pp. 191228 is again printed following Rafns edition with no more
information than the shelf mark of the base text, Addit. 82 b 4to in the University Library
(now Rask 32).
Guni Jnsson was the sole editor of the most recent edition (Reykjavk, 1950 and 1954
59) and followed his and Bjarni Vilhjlmssons edition closely, but changed the order of the
stories sometimes and spread them over four volumes. There is no variant apparatus. Srla
saga sterka is in the third volume on pp. 367410 without any information.
The editions from 194344 and 195459 in the slendingasaga-series proved to be very
popular not only amongst Icelands population, but also amongst scholars. These editions
are fairly cheap and readily available, and are, with their normalised texts, not cluttered by
variant apparatus or notes, and thus easily accessible and readable.
Birners and Rafns editions are impossible or nearly impossible to purchase and not
available in every library, and the manuscripts containing Srla saga sterka are even more
difficult to access. There is no modern-standard, critical edition of the saga yet. Therefore,
very often it is only the later editions without a variant apparatus that are available to the
scholar, and when only consulting these editions, the scholar will miss a lot of valuable
information. When the scholar does not know of the other versions of Srla saga sterka,
especially the longer versions, he or she might get a different impression of Srla saga sterka
from the work as it is contained in the manuscripts.

Bibliography
rmann Jakobsson, Annette Lassen and Agneta Ney, 2003: Inledning. In: Fornaldarsagornas struktur
och ideologi. Handlingar frn ett symposium i Uppsala 31.82.9 2001. Ed. by rmann Jakobsson,
Annette Lassen and Agneta Ney. Uppsala. (Nordiska texter och underskningar 28.)
Breve fra og til Carl Christian Rafn, med en Biographi. Ed. by B. Grndahl. 1869. Kjbenhavn.
Busch, Kay, 2002: Grossmachtstatus & Sagainterpretation die schwedischen Vorzeitsagaeditionen
des 17. und 18. Jahrhunderts. Dissertation an der Friedrich-Alexander-Universitt Erlangen-
Nrnberg. Available at: http://www.opus.ub.uni-erlangen.de/opus/volltexte/2004/51/, accessed 14
April 2009.
Dansk biografisk Leksikon. Tillige omfattende Norge for Tidsrummet 15371814. Ed. by C.F. Bricka.
18871905. Kjbenhavn.
Driscoll, M.J., and Silvia Hufnagel: Fornaldarsgur norurlanda: A bibliography of manuscripts,
editions, translations and secondary literature. Available at: http://www.staff.hum.ku.dk/
mjd/fornaldarsagas/index.html, accessed 25 April 2009.

403
Einar l. Sveinsson, 1959: Fornaldarsgur Norurlanda. In: Kulturhistorisk Leksikon for nordisk
middelalder. Fra vikingetid til reformationstid. Ed. by J. Brndsted et al. Vol. 4. Kbenhavn. Pp.
50007.
Fornaldar sgur nordrlanda. Eptir gmlum handritum. Ed. by C. C. Rafn. 182930. Kjbenhavn.
Fornaldarsgur norrlanda. Ed. by Valdimar smundarson. 1st volume reprinted 1891. 188591.
Reykjavk.
Fornaldarsgur norurlanda. Ed. by Guni Jnsson and Bjarni Vilhjlmsson. 194344. Reykjavk.
Fornaldar sgur norurlanda. Ed. by Guni Jnsson. 1950, 195459. Reykjavk.
Hermann Plsson, 1985: Fornaldarsgur. In: Dictionary of the Middle Ages. Ed. by J. R. Strayer. Vol.
5. New York. Pp. 13743.
Hermann Plsson, 1988: Srla saga sterka. In: Dictionary of the Middle Ages. Ed. by J. R. Strayer.
Vol. 11. New York. P. 373.
http://www.oldskriftselskabet.dk, accessed 14 April 2009.
Jnas Kristjnsson, 1988: Eddas and Sagas. Icelands medieval literature. Translated by Peter Foote.
Reykjavk.
Kalinke, Marianne, 1990: Bridal-Quest Romance in Medieval Iceland. Ithaca. (Islandica 46.)
Nordiska Kmpa Dater. Ed. by E. J. Birner. 1737. Stockholm.
OConnor, Ralph, 2006: Interrogating Genre in the Fornaldarasgur. In: Viking and Medieval
Scandinavia 2/2006. Pp. 27595.
Reuschel, Helga, 1933: Untersuchungen zu Stoff und Stil der Fornaldarsaga. Bhl-Baden. (Bausteine
zur Volkskunde und Religionswissenschaft 7.)
Schier, Kurt, 1970: Sagaliteratur. Stuttgart. (Sammlung Metzler. Realienbcher fr Germanisten.
Literaturgeschichte 78.)
Simek, Rudolf, and Hermann Plsson, 2007: Lexikon der altnordischen Literatur. 2., wesentlich
vermehrte und berarbeitete Aufl. Stuttgart. (Krners Taschenausgabe 490.)

404
 Odin an immigrant in Scandinavia?
Anders Hultgrd, Uppsala, Sweden

Introduction
The origins of the god Odin and his cult in Scandinavia has captured the interest of scholar-
ship for almost two hundred years. Many different interpretations have been proposed on
more or less convincing grounds. My paper aims at reconsidering the main lines of interpreta-
tion and critically review the arguments brought forward (the use of two forms Wodan and
Woden is no slip of the pen, see Hultgrd 2007).
The issue includes a complex of problems, the most significant being the following ones:
The geographical aspect. The identification of an area of origin for the cult of Odin
(within or outside the Germanic speaking world) and the way it spread further.
The genesis of the Odin figure proper as we know it from Viking age and medieval
sources.
A common Germanic god who preserved a continuity in character or underwent a thor-
oughgoing transformation.
We may roughly distinguish two main lines of interpretation:
A. The first one considers Odin as a deity having arrived to Scandinavia from the south in
a relatively late period. His rise to pre-eminence was a gradual one. This line of interpretation
was dominant in the late 19th and early 20th centuries but has been revived in recent times.
B. The second line of interpretation regards Odin as an ancient deity common to all Ger-
manic peoples. He had a prominent position from the very beginning and his character re-
mained essentially the same through the centuries.

Odin as an incomer to Scandinavia

The first line of interpretation (A) is found in several variants depending on three different
factors and the way they are combined:
the geographic area where the origins of the god are sought.
the circumstances under which the cult of the god emerged and the way it was diffused.
what parts of the character of the god were changed during the spread of his cult.
A. 1 Several scholars place the origins of the Odin cult in the lower Rhine area from
where it spread to the north. Some of them also emphasize the connection with the Gallo-
Roman cult of Mercury.
In the first half of the 20th century the most prominent representative of that line of inter-
pretation was Karl Helm. He stated that the notice of Tacitus in his Germania that the Ger-
manic tribes worship Mercury as the foremost god (deorum maxime Mercurium colunt) re-
ferred to western Germania, more specifically the lower Rhine area where the cult of Wodan
existed already in the first century C.E. Helm suggested that Herules returning home to Scan-
dinavia brought with them the cult of the god in the 5th century. The Odin cult was thought to
have spread over the entire area of Germanic speaking peoples in varying ways but usually
linked to movements of warlike elite groups. Alois Closs (1934) thus referred to I quote-
eine kriegerische Kulturbewegung whicht brought the worship of Odin to Scandinavia.
In more recent research the idea that the worship of Odin emerged in the lower Rhine area
has been revived by Michael Enright (1996), Richard North (1997) and Ludwig Rbekeil
(2002 and 2003).
According to Enright the cult typical of Woden emerged in that area under Celtic influ-
ence as a warband religion inspired by the leader of the Batavians, Civilis. The success of this

405
new form of religion depended largely on the prestige conveyed by the cult of a prominent
Germanic warlord.
North suggests that the Germanic cult of Odin originated as a worship of Mercury in Ro-
man Gallia. Through warfare and trade the cult spread over the Rhine into northern Germania
and along the coasts of the North Sea. By the end of the 5th century it had reached England
and southern Scandinavia and the god was worshipped with his Germanic name Woden.
Rbekeil maintains that the form of the gods name as reflected by Scandinavian tradition
is the original one: proto-Germanic *Winaz > inn. Germanic tribes settled along the
lower Rhine were, according to Rbekeil, impressed by the sacrificial cult of Celtic prophetic
priests. These priests were called by a term cognate with Latin vates, Celtic *atis > Old Irish
fith. The basic word underlying inn is *wi which is a Germanic reflection of Celtic
*atis. The Germanic tribes began to address their worship to the same god to whom the
Celtic priests dedicated their sacrifices but applied the name of the vates to the god. It was a
Germanic reception on a superficial ritual level of the cult of the Celtic god Lug.
2 Other scholars point instead at a more southern and southeastern provenance of the cult
of Odin. From these areas, the central and eastern parts of the Roman empire, as well as the
territories north of the Black Sea, cultural influences that partly can be attested by archaeo-
logical evidence, reached Scandinavia and brought with them also the worship of Odin. This
interpretation includes the assumption that the runic script was part of the same cultural proc-
ess, it is suggested that the runes came to Scandinavia through the intermediary of the Goths
when settled in southeastern Europe and this would explain the close connection of Odin with
the runes.
A number of scholars have followed this line of interpretation. In the beginning of the 20th
century two Scandinavian scholars made influential contributions to the problem of the ori-
gins of the Odin cult. Bernard Salin assumed more than one cultural wave reaching Scandina-
via from the south and southeast. One of the later brought the belief in Odin and the other
sir and at the same time also the runes. He seems to have been the first one to interpret the
iconography of the C-bracteates as representing the god Odin. Sune Ambrosiani referred also
to the bracteates and their Roman models suggesting that the cult of Odin was nothing else
than the cult of the Roman emperor transposed to a Scandinavian setting where it had blended
with a native worship of an animal deity. Elias Wessn (1924) emphasized the role of the
Goths in southern Russia in transmitting the Odin cult together with the runes to Scandinavia.
A Roman influence through Mithriacism on the development of the Odin figure has also
been suggested by some scholars. Hilda Ellis Davidson (1964) suggests that Odin developed
from an earlier Germanic war god: In Twaz we have an early Germanic war god, an ances-
tor of Odin (p. 60) but sees also a connection with the Roman Mithras (Ellis Davidsson
1978; Kaliff & Sundqvist 2004)
We may classify the view of Lotte Motz as a third variant of the immigration theory. The
Odin figure has emerged after the image of an ecstatic visionary and magician with shaman-
istic features. Eastern tribes penetrating into the territory of Germanic peoples brought with
them this image of an itinerant shaman deity who probably also borrowed traits taken from
the worship of a native Germanic god. At a later stage, according to Motz, Odin was then
transformed into a mounted warrior god and magician.

Odin as an ancient and prominent god of all Germanic peoples

B. Looking at the second main line of interpretation, i.e. the assumption that Odin represents a
very ancient god among all the Germanic peoples is intimately bound up with the Indo-
European question.

406
 Here we are faced with two different types of interpretations (1 and 2). One combines the
myth of the war between the sir and the vanir with archaeological evidence. The other is
based on the comparative study of Indo-European mythologies and deities.
1 The myth of the war between the sir and the vanir (Vlusp; Snorris Edda) as well as
the story of the immigration of the sir into Scandinavia (Snorris Edda; Ynglingasaga) is
considered to reflect a pre-historic immigration wave into Scandinavia by a warlike people
speaking an Indo-European language and worshipping the gods who later became known as
the sir. The conflict between the peaceful native population and their deities, the vanir , and
the Indo-European intruders ended with a fusion of the two population groups in which the
Indo-European element became dominant. This event is thought to correspond in archaeologi-
cal terms to the arrival of the Battle Axe culture around 2000 B.C.E.
This interpretation has long been a favourite theme in the history of ancient Scandinavian
religion. As representatives we may mention Philippson (1953), Henrik Schck, Eugen Mogk.
In the broader perspective of Indo-European studies the archaeologist Marija Gimbutas has
elaborated the sharp contrast between the peaceful agriculturalists of Old Europe, who mainly
worshipped mother goddesses, and the warlike patriarchal Indo-Europeans who coming from
the plains of southern Russia and Kazakhstan penetrated much of Europe. Being worshippers
of male sky gods they adhered to a different type of religion (e. g. Gimbutas 1988).
A. 2 Without involving themselves in the complicated interpretation of the archaeological
evidence for Indo-European origins and diffusion, other scholars prefer to concentrate on the
comparative study of Indo-European mythologies. They deny any historical reality behind the
myth of the war between the sir and the vanir and the euhemeristic migration stories in me-
dieval Old Norse literature.
In the first half of the 20th century scholars like Hermann Gntert made broad comparisons
covering the entire field of Indo-European languages and cultures. Gntert (1923:151153)
emphasized the similarities between Odin and the Indic god Varuna and concluded that both
reflect the Indo-European type of a binding world ruler and king of the gods. Having a
common background these two gods acquired new features as time passed on, however. Odin
was no doubt a god common to all Germanic groups but he attained his supremacy gradually
his worship spreading from certain important cult centres.
In the first place we have the French comparatist Georges Dumzil whose three-functional
system is too well known to be presented here in detail. As to the divine world it implies that
the main deities of the early Indo-Europeans can be classified in a tripartite system. The
common characteristics found between the Indo-European deities of the same functional cate-
gory are according to Dumzil best explained as based on inherited tradition. The first func-
tion is shared between two deities representing different aspects of the sovereignty. Odin who
together with Tr belongs to the first function stands for the religious-magical aspect, and Tr
for the judicial one. Odin/Woden as head of the Germanic pantheon constitutes thus for
Dumzil a very ancient feature.
Many scholars have accepted the tripartite system as elaborated by Dumzil but others use
the comparative Indo-European approach without building on Dumzils theory. Jan de Vries
(195657 406) points to striking affinities in character between the Indic god Rudra and
Odin and to a certain extent also between the Greek Hermes and Odin. De Vries speaks of
Odin as a high god and seems implicitly to accept Dumzils classification of the Indo-
European deities when referring to Odin as die dunkle Seite des Hochgottes (410; cp. also
412). Also taking the precise similarities between Lug and Odin into account(de Vries 1961)
we may state that for de Vries Odin represents from the very beginning a prominent deity part
of the Indo-European legacy (412).
Jaan Puhvel refers likewise to the Indic Rudra as a counterpart to the Celtic Esus-Lugus
and to the Germanic Odin but comes to a different conclusion than Jan de Vries. For Puhvel

407
Odin seems to be in origin the semidemonic patron of the warriors. In warlike Celtic and
Germanic society a homologue of Rudra ascended to the pinnacle of the pantheon assimilat-
ing or supplanting whatever pristine god may have kept company there (1987:201). While
admitting that Odin was an ancient god among Germanic peoples Puhvel appears nevertheless
to assume a gradual elevation of the Odin figure to the chief god of the pantheon. The roots of
Odin in the religious and social context of the Mnnerbnde means for Kershaw (2000) that
the god has an Indo-European background.
In his reconstruction of an Indo-European family of deities Peter Jackson (2002) proceeds
independently of Dumzil and suggests that Odin belongs with the proto-Indo-European god
who is reflected in the Indic Varuna and the Greek Ouranos.

The validity of the arguments

The first of the two main lines of interpretation (A) includes different types of argumentation.
Since the opinion of Karl Helm influenced much of the subsequent discussion on the origins
of the Odin cult (most recently North 1997) it is appropriate to start with the reasons he ad-
duced to support his view. Helm chose to interpret the notice of Tacitus on the worship of the
Germanic Mercury (see above) as referring only to western Germania more specifically the
lower Rhine area. There but not elsewhere, according to Helm, was Wodan worshipped as the
chief deity in the 1st century C. E.
Admittedly most of the information on Germanic culture and religion that reached the
Romans came from (or passed through) the Rhineland area but this fact in no way excludes
the possibility that other Germanic (including Scandinavian) tribes also worshipped Odin in
the same early period.
More weight should be put on Helms second main argument. In the entire source material
from Scandinavia (archeological findings, rock carvings, bracteates, early runic inscriptions)
there is no evidence of a cult of Woden/Odin. If the bracteates depict the god an assumption
of which Helm is skeptical we dont reach any farther back in time as the early 6th century.
However, the lack of references to Woden/Odin in Scandinavian sources prior to the 5th
and 6th centuries is not a convincing argument in view of the almost total absence of written
sources and the uncertainty of identifying deities known from the Viking period in the rock
carvings and other early iconographic material. I agree with Helm that the presence of the
Woden/Odin figure on the rock carvings cannot be demonstrated. On the other hand there is
no convincing evidence to prove his absence either.
Among modern proponents of the Rhineland hypothesis North (1997) follows the reason-
ing of Helm but Enright (1996) sees the typical Woden/Odin cult (not necessarily the origin
of the god, however) as intimately bound up with the emergence of the Celtic-Germanic war-
band (or comitatus) institution. Focus is on the Rhineland area with its mixture of Gaulish and
Germanic elements. He elaborates an extensive parallelism between Sertorius, leader of the
Celtiberians in their wars against Rome, and the chieftain of the Batavians, Civilis, who took
Sertorius as a model for his political agenda. Both men developed a similar religious propa-
ganda which in the case of Civilis shaped the typical image of Woden/Odin . The Scandina-
vian Odins involvement with warfare and prophecy, his attributes such as the one-eyedness
and the spear, correspond in a surprising manner with what is known about Civilis and the
rising phenomenon of the warband. This correspondence shows according to Enright that the
god and his cult ultimately received its characteristics from the Gallo-Roman and Germanic
worship of Mercury in the form it was propagated by Civilis.
The principal objection to this hypothesis is that the similarities adduced do not compel us
to accept the conclusion reached by Enright. The similarities may be explained differently and
gods are not necessarily shaped by specific political and cultural circumstances. Thus, an al-

408
ternative explanation would be that Woden/Odin already possessed the characteristics that
suited a warlord and his warband and therefore became their particular deity.
Rbekeils theory rests on a linguistic argument which is combined with the particular
ethnic and cultural situation prevailing in the Rhineland area where Celtic and Germanic
groups were in close contact. His linguistic reasoning is complicated and presupposes some
stages that cannot be directly attested. The assumption that the Germanic groups called the
god they took over from Celtic Rhineland tribes after the name of the priests that served him
is too speculative and is clearly inspired from the linguistic argument. If this does not hold
good, the theory as a whole cannot stand.
For those who favour a more southern and southeastern origin of the Odin cult (A. 2) the
main argument is the cultural impact of the Roman empire and the influences that reached
Scandinavia from southeastern Europe. One must ask why precisely Odin among all the dei-
ties was imported. His connection with the runes is the only plausible argument for this ex-
ception and is also put forward by the adherents of a southeastern origin of Odin. Again there
is a presupposition that appears to be doubtful: the runes were invented among the Goths in
southern Russia on the base of the Greek script (e.g. Wessn 1924:26). Not to speak of the
idea that a core of historical truth underlies the euhemeristic story of Odin and his followers
migrating into Scandinavia from Asia Minor (Salin and others) . This idea still influenced
much of the early 20th century scholarship.
As to the second main line of interpretation (B.) comparisons with other Indo-European
gods serve the purpose of demonstrating the presence of Odin among all Germanic peoples
already from the settlement of Indo-European speaking groups in northern Europe. The point
is that the study of the homologues of Odin in Indo-European mythologies show him to be
part of an ancient religious legacy. This in turn implies some sort of a common Indo-
European pantheon which must be put far back in time. Here we are confronted with the main
difficulty that faces the proponents of an Indo-European heritage. Could myths and images of
deities be preserved in oral tradition throughout two or three millennia and still be recognized
as deriving from a common source? We would be on safer ground when using the compara-
tive etymological-onomastic approach which does not yield much result for the Odin figure
except for some of his heiti.
The attempt to link the origin of Odin and the sir with the appearance of the Corded
Ware/Battle Axe culture in Scandinavia is based on uncertain parameters. It is assumed that
this culture a) was brought to northern Europe by invading population groups and b) that
these invaders were Indo-Europeans bringing a new form of religion with them. Both supposi-
tions are far from certain. To identify ethnic and language groups solely on archaeological
grounds is hazardous and there is evidence to suggest that the Battle Axe culture to a large
extent had a local origin.

Conclusions
The present survey and discussion has demonstrated the difficulties in ascertaining the hy-
pothesis of Odin being a late incomer in Scandinavia. Conversely we have no clear evidence
to show the presence of his cult before the middle of the first millennium C. E. To judge from
the information given by Roman writers the god was worshipped by Germanic tribes on the
continent as one of the main deities already in the 1st century C.E. (cf. also Simek 2003:110
111). In my view there is nothing to suggest that Woden/Odin was not worshipped in Scandi-
navia during the same period (cf. also Hultgrd 2007 and Schjdt 2008:451).
We have to assume a continuity in the cult of this god in Scandinavia and the Germanic
continent from a very early time as long as there is no conclusive evidence for the opposite.

409
 Bibliography
Ambrosiani, Sune 1907. Odinskultens hrkomst.
Closs, Alois 1934. Neue Problemstellungen in der germanischen Religionsgeschichte. Anthropos
29:477 496.
Dumzil, Georges 1959. Les dieux des Germains. Paris.
1986. Les dieux souverains des Indo-Europens. 3 d. revue et corrige. Paris.
Ellis Davidson, Hilda R. 1964. Gods and Myths of Northern Europe.Middlesex.
1978. Mithras and Wodan. tudes mithriaques, ed. by J. Duchesne-Guillemin
Enright, Michael 1996. Lady with a Mead Cup. Four Courts Press.
Gimbutas, Marija 1988. Accounting for a great change. Review of Colin Renfrew, Archaeology and
Language 1987. Times Literary Supplement June 2430 1988.
Helm, Karl 1913. Altgermanische Religionsgeschichte Bd. 1. Heidelberg.
1946. Wodan. Ausbreitung und Wanderung seines Kultes. Giessen.
Hultgrd, Anders 2007. Wotan-Odin. Reallexikon der Germanischen Altertumskunde 35: 759785.
Jackson, Peter 2002. Light from distant asterisks. Towards a description of the Indo-European reli-
gious heritage. Numen 49:61102.
Kaliff, Anders & Sundqvist, Olof 2004. Oden och Mithraskulten: Religis ackulturation under ro-
mersk jrnlder och folkvandringstid. Uppsala.
Kershaw, Kris 2000. The One-eyed god. Odin and the (Indo-)Germanic Mnnerbnde. Washington
D.C.
Motz, Lotte 1993. The king, the champion and the sorcerer. Wien.
North, Richard 1997. Heathen Gods in Old English Literature.
Petersen, Henry 1876. Om nordboernes gudedyrkelse og gudetro i hedenold: En antikvarisk
undersgelse. Kbenhavn (deutsche bersetzung: ber den Gottesdienst und den Gtterglauben
des Nordens whrend der Heidenzeit, Gardelegen 1882).
Philipsson, Ernst Alfred 1953. Die Genealogie der Gtter in germanischer Religion. Mythologie und
Theologie. Urbana.
Puhvel, Jaan 1987. Comparative Mythology. Baltimore and London.
Rbekeil, Ludwig 2002. Diachrone Studien zur Kontaktzone zwischen Kelten und Germanen.
2003. Wodan und andere forschungsgeschichtliche Leiche, exhumiert. Beitrge zur Namenfor-
schung 38:2541.
Salin, Bernard 1903. Heimskringlas tradition om asarnes invandring. Studier tillgnade O. Monteli-
us, 133134.
1904. Die altgermanische Thierornamentik: typologische Studie ber germanische Metallgegens-
tnde aus dem IV. Bis IX. Jahrhundert. Stockholm.
Schjdt, Jens Peter 2008. Initiation between two worlds. Structure and symbolism in pre-Christian
Scandinavian religion. University Press of Southern Denmark.
Simek, Rudolf 2003. Religion und Mythologie der Germanen. Darmstadt.
Vries, Jan de 1956157. Altgermanische Religionsgeschichte. Berlin.
1961. Keltische Religion. Stuttgart.
Wessn, Elias 1924. Studier i Sveriges hedna mytologi och fornhistoria. Uppsala.

410
 Saga-intertekstualitet i Einar Mr Gumundssons Fodspor p
himlen
Lise Hvarregaard, Institut for Nordiske studier og Sprogvidenskab, Kbenhavns Universitet
Romanerne: Ftspor himnum 1997, p dansk Fodspor p himlen 1999 (herefter FPH),
Draumar jru 2000, p dansk Drmme p jorden 2001 (herefter DPJ) og Nafnlausir vegir
2002, p dansk Navnlse veje 2003 (NV) der er udgivet under et som Fodspor p himlen-
trilogien, udgr en epoke i Einar Mr Gumundssons forfatterskab.
De tre romaner har som udgangspunkt Olaf og Gudny og deres brns videre skbne, og vi
fr ogs som i en rigtig saga Olafs og Gudnys slgtshistorier1 helt tilbage til Gudnys
tipoldemor. De tre bgers handlingsgang ligger kronologisk i forlngelse af hverandre,
personer og topografi gr igen men man kan godt lse hver roman som en selvstndig
historie. Samtidig udgr de tre romaner sammen med Universets engle, (1993, dansk 1995)
der delvist er personfllesskab med, en samlet afdeling i forfatterskabet.2 Nvnes skal det at
Paul fra UE, der ogs nvnes i novellesamlingen Mske er posten sulten er Olafs sn og
sledes barnebarn af Olaf Vrager og Gudny.
I det flgende fokuseres p dialogen med sagaerne i Fodspor p himlen, dog inddrages
DPJ og NV nr det er relevant.
Som i Einar Mr Gumundssons vrige romaner benyttes Reykjavik-kronotopen,3 men
tidsmssigt er vi tilbage i tiden fra frste tredjedel af det tyvende rhundrede. Det er et billede
af et historisk Reykjavik, hvor store dele af de nu byggede kvarterer4 ikke er kommet til.
Slgtshistorien spejler sledes den strre fortlling: byens og nationens.
FPH rummer et persongalleri s stort som i en saga,5 og ogs slgtstavlerne bagerst i
bogen medgiver romaner sagaprg.6 Vi mder her billeder af mennesker, som ingen kendte
andetsteds fra, personer fra ubetydelige steder.7 Parateksten som motto: Digtet af Jhannes
r Ktlum (18991972),8 forudgriber og anslr tonen til samme tema: At give de fattige, de
sm og de mange stemme.9
Fodspor p himlen rummer iflge anmelderne strke referencer til sagaerne.10 Sren
Vinterberg definerer11 FPH som en 1930er-saga, hvor sagatonen ikke er til at tage fejl af:
Opregningen af slgten og dens greml, inklusive den respektfulde omtale af personer, som
straks igen er ude af sagaen. De gammeltestamentlige telefonbogsopremsninger er forbundet
med bevidstheden om, at hver liv har sin vrdighed og er omtale vrd og Det er svrt at
give udtryk af en stor slgts- og kollektivromans personer, fromme slidere og fyrige elskere,
1
Slgtstavle findes bagerst s. 213 i FPH som det ogs er sdvane med slgtstavler i flere af de trykte sagaer.
2
Jf. Jn Yngvi Jhannsson 2006, s. 609 ser fra Universets engle og frem et nyt kapitel i EMGs forfatterskab,
baseret p beretninger fra egen slgt og personfllesskab ogs med personer fra novellesamlingen Mske er
posten sulten (2001, dansk 2002)
3
In these family sagas he continues to use the history of Reykjavik as a basis (strur Eysteinsson og lf-
hildur Dagsdttir 2006, s. 460).
4
Eksempelvis Grafarvogur og Grafarholt.
5
I Laksdla saga nvnes eksempelvis 350400 navngivne personer (Bredsdorff (1971) 1995, s. 39).
6
EMG lgger ikke skjul p at romanen bygger p hans egen slgts historie, hvad dedikationen: Til minde om
min bedstemoder, Ingibjrg Gsladttir, f.04.11.1882 25.06.1965 vidner om. P samme mde er Universets
engle dedikeret til broderen Plmi.
7
Brostrm citerer Laxness, 1999, s. 437, for at vise sammenhng mellem sagafortllerne og EMG.
8
Se ogs slensk bkmenntasaga IV, 2006, Silja Aalsteinsdttir s. 365.
9
I modstning til historiens fokus p de magtfulde, de f og de store.
10
Erik Skyum-Nielsen 2005, s. 54 pviser med udgangspunkt i romanen en rkke centrale saga-ln: Nye og
gamle fortlleformer i moderne vestnordisk prosa i postkolonialt perspektiv.
11
Kmpeliv i Nord Politiken, 23.04 1999.

411
seje sm koner og kmper. Vinterberg fremhver ligeledes den sproglige enkelhed og
pointerer at romanen, som sagaen, er fuld af skarpskren komik og vrdighed. Foreningen
af realisme og poesi i det Nordatlantiske, er en linie som Vinterberg finder gr via Laxness
direkte tilbage til Gunlg, Gunnar og Njl.12 Bent Vinn Nielsen13 fremhver mere generelt at
fortllestilen er umiskendeligt islandsk og refererer til Laxness og Gudbergur Bergsson, der
har samme legende forhold til tiden, og uanstrengt springer mellem fortid og nutid. Iflge
Vinn Nielsen er det fremherskende hvordan slgten og enkeltskbnerne i romanen, nok
beskrives som almindelige mennesker og samfundsborgere, men aldrig bliver ydmyge af sind.
Personerne er som sagapersonerne stolte og lader sig ikke kngte af skbnens ugunst. Vinn
Nielsen definerer Gudmundssons almue som vild og blodig, hvilket man ogs med rette kan
sige om de kompromilse karakterer fra sagaerne.
Et andet karakteristika romanen har til flles med sagaerne er personprsentationerne, der
foregr efter samme mnster. Nr nye personer prsenteres i sagaen, sker det i reglen i
begyndelsen af et kapitel, og i sammenhng med genealogi og tilknytning til grd, egn og
social status (Meulengracht Srensen 2006, s. 228). Derefter flger en mere eller mindre
knap karakteristik, af ydre, frdigheder og karakter. I disse oplysninger ligger gemt kimen
til og et forudgreb om, hvilke forventninger man kan have til denne person. Meulengracht
Srensen slr fast, at man kan stole p denne karakteristik, den er autoritativ, og de
sparsomme oplysninger vi fr, kommer senere til at betyde noget vsentligt i forhold til
personens rolle i plottet (Ibid, s. 228). Personens vsentlighed i fortllingen er proportional
med de ord der bruges p beskrivelsen, centrale personer fr sledes flere ord med p vejen i
prsentationen, medens de mindre vsentlige knap fr et navn. Den frste person vi
prsenteres for i FPH, er (farmor) Gudny, frst som lille munter pige der ser et varselssyn om
fattigdom; senere trder de to aspekter: fattigdom og lyst sind momentvis frem hele vejen
gennem romanen. Senere fremhves Gudnys smukke hr og hendes pleje af det flere gange;
smukt langt hr er et sknhedsideal velkendt fra sagaerne.
I afsnit 2 trder Sigrun14 frem som et citeret vidne og vi fr Gudnys genealogi15 og fde-
og opvkststed p plads. I andet kapitel prsenteres Olaf. Hans fortlling begynder
retrospektivt, idet der startes med hans gravsted uden afmrkning. Netop oplysningen om
denne manglende afmrkning er central, for Olaf fik et eftermle, om end han ikke
umiddelbart huskes for noget positivt Den mde, han var p over for mor, og alt hvad vi
mtte lide p grund af hans opfrsel, det bliver aldrig tilgivet. s. 13. Allerede i afsnit 2 fr
han dog en forsonlig prsentation af slgt og fdested, og opvksten ved og forbindelsen til
havet understreges. Gudny og Olaf, de to mest centrale personer mder vi alts som de frste i
romanen. De reprsenterer ogs kontraster idet beskrivelsen af dem viser hvordan deres sind
og forhold til verden er prget af en henholdsvis mrk og lys grundstemning.
Harald, oldefaderen, beskrives med slgt, bopl og social status af husmand i tredje
kapitel, afsnit 1. Oldemoderen prsenteres med genealogi og fdested, i afsnit 2 og ogs det
faktum at hun er stor og arbejdsduelig fremhves i den sparsomme karakteristik. Onkel Halli
prsenteres i indledningen til afsnit 2. i femte kapitel. Og sdan bliver det ved fremad i
bogen: personerne prsenteres efter samme struktur, som den der bruges i sagaerne.

12
Ppegningen af sagaparallellerne fortsttes i Vinterbergs anmeldelse af DPJ i Politiken 30.10. 2001
Mesterlig nutidssaga og i anmeldelsen af NV, med den prgnante titel Underklassens islndingesaga
Politiken, 4.11 2003
13
Stjerneruser og himmelgarn Information, 23.04. 1999.
14
S. 73 afslres at Sigrun er opkaldt p sagamaner efter farmoderen Sigrun.
15
Nr nye personer skal introduceres, gennemgs ikke sjldent deres slgtstavle i adskillige led Erik Skyum
Nielsen 2005, s. 54.

412
 Vi mder ogs sagaprget i brugen af de mange indforstede lokale ge/tilnavne,16 i
beskrivelsen af de almindelige, samt de mere excentriske personer: Almindeligst er kategorien
erhvervsbetegnelse som tilnavn: Vrager Olli og Bjssi-Blik eller efter sport: Bokse-grimur,
geografisk tilnavn: Amerikaner Olli, desuden, erhvervet er manden: fattiginspektr Simon.
Modellen klenavn efter udseende ses, idet Ragnar kaldes for Big boy og navneforkortelser:
hypokorismer der signalerer familir status: Som eks. Valla (Valborg), onkel Halli (Harald)
ogs kaldet FlynderHalli og Gummi (Gudmund) er almindelige.
Kompositorisk er FPH inddelt i 5 dele, med 26 kapitler, der igen er underinddelt i
nummererede afsnit. Frste afdeling af bogen er overvejende centreret om prsentationerne af
Gudny og Olaf og deres slgter i fortiden. Lngere fremme flger vi brnene og specifikt
de ldste brns videre skbne. Samtidigt er fortid og nutid vvet sammen ved hjlp af en
pause- og fragmentstruktur,17 der muliggr spring i tid, rum og personkonstellation. Let og
elegant veksles mellem flash backs og flash forwards, ligesom der skiftes stil nr der pendles
mellem den store gennemgende fortlling og anekdoter, sagn og sm sidehistorier.
Generelt indgr i FPH, (DPJ, samt i NV) en mosaik af genrer. I FPH, indgr en rkke
referencer til folketro, eventyr, sagn og skrner. Erik Skyum-Nielsen (1991, s. 33) mener at
denne legende genrecollage er med til at indstte EMGs og Einar Krasons romaner i en
nordisk fantastisk tradition fra Halldr Laxness og William Heinesen, der kan ses som
nordiske sidestykker til Latinamerikas magiske eller fantastiske realisme (Ibid., s. 33).
Genrekonglomeratet udgr ogs en gammel manuskripttradition fra fr en sdan
benvnelse blev kendt. I ldre manuskripter er det ikke ualmindeligt at mde forskellige
genrer i samme skrift, kogebger og salmer, skrevet som prosa (for at spare pergament) flger
uformidlet efter hverandre. Beretninger om historiske begivenheder findes side om side med
fiktive fortllinger og der er tradition for ikke at gre opmrksom p genreskift.18 I sagaerne
er der ogs eksempler p indlejrede sidehistorier og i skjaldesagaerne digte, som vi ogs
trffer p i FPH.
Herefter skal fremdrages et par eksempler p fortllinger i fortllingen, der er med til at
fremhve det mundtlige prg, idet det er fortllinger der bliver fortalt.19 Sidehistorierne har
samme funktion som mundtligt overleverede vittigheder, anekdoter og vandrehistorier, der
eksempelvis tilfrer humor og giver plads til inddragelse af overnaturligt fantastiske aspekter.
Vi trffer anekdoten i fortllingen om hvordan Olaf i fuldskab slger Kopavogur for en
flaske sprit og som en middelalderlig ridder tager ud for at krve bydelen tilbage, her l
nemlig forfdrenes trvegrd.
Humoristisk er den eventyragtige beretning om de tre ugifte brdre, Halli opholder sig
hos. Deres berring med kvinder gr kun s langt som til at anvende ordet kvinde som
skldsord. s.136, og de mener at brnde sig p kvindetj.
Poetisk er fortllingen om Ragnars og Annas forelskelse p Ingimundur og Lauras grd,
der kan medregnes under forelskelsen og grdens kronotop (FPH s. 115), og som ogs kan ses
i UE. Der er paralleller til beskrivelsen af tiltrkningen mellem Gunlg og Helga i Gunlg
Ormstunges saga. Intertekstualiteten ses ogs i beskrivelsen af Anna: Hun var en lysende

16
Ibid s. 54.
17
strur Eysteinsson og lfhildur Dagsdttir (2006, s. 460) peger p en dynamisk postmoderne attitude i den
narrativt fragmentariske form, der opstr nr traditionel historiefortlling cuttes op i mindre enheder og mixes i
en collage.
18
Skriveren Gsli Konrsson (17871877) kan ses som et mellemled mellem sagaerne og EMGs prosa, idet han
nedskriver anekdoter, sagn, sagaer og skrner i en saganr stil. Jf. Magns Hauksson: slensk bkmenntasaga
IV, 2006, s. 317.
19
Jn Yngvi Jhansson (2006) mener EMG i FPH trilogien udforsker romanens grnser ved at blande folkelige
fortllekunst sammen med erindringer og selvbiografiske trk og at den nationale epik indlemmes i romanens
polyfoni.

413
sknhed, lyshret med klare jne s. 115. Som et ekko af sagaernes antydningskunst (Erik
Skyum-Nielsen 2005, s. 54) er det fint og diskret underdrevet hvordan: De fulgtes ad til
dagens arbejde og satte sig ved aftenunderholdningen s tt ved hinanden, som de kunne s.
115. Fortllingen ender dog med, at de unge ikke fr hinanden, da Ingimundur sender Anna
bort. Gunlg fr heller ikke Helga; omstndighederne omkring dette faktum er forskellige,
men der synes at vre parallelle trk i de to unge mnds psykologi, som bevirker udfaldet.
Det overnaturlige indgr som en naturlig del i de indlagte fortllinger, sagn og
anekdoter.20
Det glder eksempelvis for den poetisk vemodige fortlling i fortllingen om Olafs farfar:
Magnus der lever med slkvinden Erla og fr 4 brn med hende, denne historie bygger p et
velkendt islandsk folkesagn Slhammen.21
Barok humor mder vi i den lille fortlling om Magnus der gr igen i taxien s. 148,
Ragnar ved, det er hans fars genfrd og bliver ikke skrmt.
Ogs dramatiske og voldsomme ddsfald indgr i fortllingerne. Eksempelvis den lille
beretning om Bjarne p Grimsstadir der angribes af en tyr, stanges og dr, og som
kontrasterer Olafs ret fredelige dd p hospitalet.
Gudny og Olaf er vsentlige personer i romanerne og derfor er beskrivelsen af mdet
mellem dem badet i et srligt magisk lys. Det frste mde mellem Gudny og Olaf kan bedst
betegnes med det tilfldige mdes kronotop, det kan ogs rubriceres under vejens kronotop
og under kronotopen for dreng mder pige. Vi er ogs som tidligere nvnt under Reykjavik-
kronotopen. Mdet mellem Gudny og Olaf beskrives sledes:

[] i slutningen af rhundredets frste tir kommer en ung pige gende, alene og forladt, men
med kraftigt hr og optimistisk mine, gennem byens tilslede gader[].S pludselig kommer en
ung fyr hen til hende[].De spadserer rundt i byen, og pludselig er min farmor ikke alene
mere. FPH s. 39.

Olaf citerer i den tidlige forelskelse kvad for Gudny, p en direkte pasticherende mde, der
m betegnes som stilistisk intertekstualitet. Stroferne har mindelser om uddrag fra Hjsangen,
desuden kan der drages paralleller til de hyldestdigte Kormak digter til Stengerde, da han ser
hende frste gang i Kormaks saga (Meulengracht Srensen 2006, s. 254, Bredsdorff
1971/1995, s. 58).
I FPH er der eksempler p forskellige former for gteskaber, ligesom i sagaerne.22
Bredsdorff nvner hvordan Njals saga rummer flere forskellige typer af gteskaber end
andre sagaer, ogs det forhastede (Bredsdorff 1971/1995, s. 83) som man kan sige, vi mder
her.
Gudny beskrives refuldt som den, der holder sammen p familien. Hun oplever modgang
undervejs, men det er hos hende, sympatien er placeret. Til den religise Gudny knytter sig
flere syner, varsler og drmme, der som anticiperende spor gr som en rd trd igennem
romanen. Det glder hendes syn om fattigdom og brn der m fjernes fra hjem og forldre.
Og p den anden side er drmmene om Ragnars rejse til Spanien ogs kodede profetier.
Allerede fr Ragnars fdsel har Gudny drmt om hans rejse og deltagelse i krigen i Spanien:

Farmor drmte tit underlige drmme, mens hun ventede Ragnar. Nogle gange befandt hun sig i
fremmede diamantminer, hvor soldater gik rundt gennem salene, eller hun kurede p enden ned
i mrke kulminer. Hun forstod ikke, hun kunne vre s vidtberejst, mens hun sov FPH s. 63.

20
Vinn Nielsen 2001 fremhver ud fra disse sidehistorier EMG som den strste nordiske magiske realist.
21
Islandske folkesagn og eventyr, 1990. P dansk ved Peter Sby Kristensen, s.104.
22
I Gisle Surssns saga er der eksempler p flere former for gteskaber. Erik Skyum-Nielsen 1988, s. 114.

414
Drmmens funktion her kan ses som et forudgreb om, hvad der senere skal ske, vi fr dog i
modstning til sagaerne allerede p samme side at vide hvad drmmen betyder og at den
kommer til at g i opfyldelse. Gudny drmmer om Ragnar, da han deltager i den spanske
borgerkrig og man ingen efterretninger fr om hans skbne: Ragnar kom til hende i drmme.
Han stod badet i et kraftigt lysskr og sagde stninger p fremmede sprog. Nogle gange var
der rde pletter i lysskret s. 210. Senere viser det sig at drmmen gr i opfyldelse som i
sagaerne. Ragnar bliver sret, overlever og vender hjem, hvorefter Hans moder s sine
drmme g i opfyldelse s. 210.
Drmmene er med til at karakterisere Gudny, der har ndelige prferencer til drm, vision,
varsel og forbn. Hun er synsk, hvilket det ikke er ualmindeligt, at isr kvinder i sagaerne er.
Motivet med social deklassering, hvor personer opdrages i et andet milj og under lavere
forhold end dem, de er fdt til, udgr et velkendt sagamotiv (Meulengracht Srensen 2006, s.
278). Af og til er den rette herkomst ogs skjult. Efter trngsler og modgang genetableres den
sociale orden, og personen indtrder igen i den rette sociale sammenhng. Det sker
eksempelvis i Gunlg Ormstunges saga23 hvor Helga den fagre lever sine frste opvkstr i
Thorgerd Egilsdatters varetgt, indtil hun 6 r gammel indsttes p sin rette plads som
Thorstein og Jofrids datter. Et andet eksempel er fra Laksdla saga hvor trlkvinden
Melkorka afslrer sin herkomst som irsk prinsesse. Senere rejser snnen Olaf til Norge og til
Irland, hvor han mder morfaderen kong Myrkjartan og behandles som en prins og et
barnebarn.24 Meulengracht Srensen nvner som eksempel Sigurd Fafnersbane, der vokser op
som forldrels i lavere sociale kr end dem, han er fdt til, og som til sidst indsttes i den
rette position, der svarer til hans hje t.25 Ogs i slaugs26 historie fra Ragnars saga viser
samme mnster sig. Som spd gemmes slaug i en harpe, senere findes hun af
fosterforldre, hos hvem hun vokser op, og i kraft af sknhed og klogskab genvindes den
sociale position.
Dette motiv er gennemgende i FPH, hvor de ni brn sttes i pleje ude p landet, da
forldrene ikke kan srge for dem. Brnene er overladt til uvisse kr, der er ingen forldre
eller myndighed, der har kontrol med om de behandles ordentligt.27 Det er specifikt Sigruns,
Ragnars og Hallis historier, vi flger. Som en anticipering husker vi Gudnys syn fra hun var
lille pige. Hun s en sogneflytning af en hel familie, der flyttes og skilles p grund af
fattigdom. Underforstet er det at en lille dreng dr, som flge af plejefaderens vold og at
faderens klager over bonden blev overhrt (FPH s. 10). Slgtsflelsen er strk og det synes
som det vrste der kan overg en familie er at blive skilt ad. Gudny fastholder gteskabet
med Olaf, selv efter at han slr hende og drikker hendes ln op.
Det store lyspunkt i romanen oplever man efter at et nulpunkt er net: Familiens oplsning
er total, alle er borte fra hjemmet, kun Gudny er tilbage efter Olafs dd p Epidemihospitalet.
Gradvist genetableres familien centreret omkring Gudnys lejlighed. Sun kommer hjem for at
blive konfirmeret, og de ldste brdre kommer til byen for at arbejde, alle bor de hos
moderen. Frst Halli og Ragnar, siden Ivar og Kre, s flger Sun og Gummi. De 3 yngste
brdre: Finn, Olaf og Einar vedbliver med at vre p grde p landet nogen tid endnu (FPH s.
150). Vi ser, hvordan familien sger sammen og at brnene her kommer hjem og p rette
plads i en genopretning af slgtspositionen. Selv om det her er en fattig slgt, udtrykkes der
megen slgtsfornemmelse og samhrighed.
23
Gunlg Ormstunges saga s. 161.
24
Laksdla saga s. 211.
25
Skjult identitet og alt kommer p rette plads motivet er sledes langt ldre end fremstillingen i H.C
Andersens Den grimme lling fra 1843, det udgr ogs et velkendt motiv i folkeeventyret.
26
Ogs kaldet Krka. Iflge Meulengracht Srensen 2006, s. 278.
27
Meulengracht Srensen 1993, s. 163164 nvner hvordan de laveste i hierarkiet i sagaerne er de ttelse.
Eksempelvis trllene der lever udenfor deres slgts beskyttelse og hvor tten i vrste fald er ukendt.

415
 Det synes som om et andet karakteristisk EMG fnomen ogs her gr sig gldende,
nemlig at en forventningshorisont brydes.

Det er srt. Sigrun, der grd hjt i bilen og drog af med sindet fuldt af skuffelse, fik sig et hjem
p landet, hvorimod Ragnar, der tog imod det nye med forventning, rejste til et ddsrige og frst
steg op, idet han stak af. FPH s. 107

Gudny og Ragnar regner med, at han vil f det godt og f nogen skolegang, da han skal vre
hos en prst p den egn, Gudny stammer fra. Men sdan gr det langtfra. Han sttes til hrdt
fysisk arbejde og er for trt til at lre om aftenen. Sarkastisk beskrives prstens lunkenhed
og underforstet er det hans griskhed efter at udnytte plejedrengenes arbejdskraft, der udgr
den egentlige rsag til den manglende skolegang. Denne prst er i vrigt gennemfrt
beskrevet som en grisk og usympatisk person; ligesom i sagaerne understreges hans
usympatiske vremde af et passende udseende. Han er laskefed, med mundvige som smalle
streger og oppustede kinder (FPH s. 104) og dertil hovmodig. Anderledes gr det med
Sigrun, der savner sit hjem og sin familie og som ingen forventninger har til opholdet p
landet. Da den vrste hjemve har lagt sig, erfarer hun at parret, hun er kommet til, er
forstende og krlige. Gudny har tidligere i nostalgiens skr beskrevet hvor dejligt der er ude
p landet om sommeren og nu erfarer Sigrun at det passer. I modstning til dette str Ragnars
oplevelser, han fr en kraftig lektion i social uretfrdighed og opnr mistillid til landlivet.
Han lrer kun en ting af sine ophold p landet, og det er ikke at lade sig udnytte og kngte af
dem der er strkere.
Hallis ophold hos de tre kvindeforskrkkede brdre i Kambar afstedkommer, ret
forudsigeligt, at han aldrig selv gifter sig, men bor hos sin mor til hendes dd, hvor han er 55.
Som et paradoks spr spkonen Nanna ham, at han vil blive omgivet af kvinder og den
spdom gr p en lidt anden mde i opfyldelse, da han testamenterer lejligheden til
voldsramte kvinder.
Vi ser i FPH hvordan personer med ansvarsflelse over for slgtens mindre heldigt stillede
medlemmer fremstilles positivt. Det er sledes af stor betydning hvem man er i slgt med,
ligesom i sagaerne.
De som p nogen mde har rd og overskud til slgtsflelse eller hjlper drligere stillede,
beskrives i sympatiske vendinger.28 Det glder eksempelvis Magnus, Olafs bror, der ser Olafs
drikkeri, og holder op med at drikke, hvorefter han hjlper og retter op p broderens svigt
over for brnene. s.135. Senere anbefaler Magnus, der er en del af Djvleslnget: de hrdeste
havnearbejdere, den generte Halli, s han fr job. Det ses hvordan der er held ved at have
indflydelsesrige slgtninge, nr job skal sges som havnearbejdere. Magnus srger ogs for
at de tre brdre Ragnar, Halli og Kre bliver udvalgt til arbejde, da det er smt med det og
senere belnnede Kre Magnus for hans tjeneste ved at invitere ham med til
bokseturneringer. s. 152. Brdrene gr gengld. Dog fr Ragnar ikke arbejde dagen efter,
fordi han har brugt arbejdstiden p at agitere. Her gr det ogs ud over brdrene Ivar og Kre,
der heller ikke tilbydes at komme igen nste dag. s.154. Slgtsforbindelser kan sledes vre
en fordel men kan i andre tilflde hmme den enkeltes muligheder. Der er her referencer til
NS, hvor det er centralt, at hvad der vedrrer en af snnerne uvilkrligt falder tilbage p de
vrige.
Brdrene Halli og Ragnar er forskellige som nat og dag, som Egil og broren Thorolf i Egils
saga var det. Efter at drengene er blevet voksne fremgr det at Halli er genert og utilpas i
byen, medens Ragnar derimod er udadvendt, veltalende og veltilpas i byen. Halli er halvt s
stor som sine brdre s. 150. Kontrasten mellem de to meget forskellige brdre bliver ekstra

28
I sagauniverset medfrer slgtsskab forpligtelser og rettigheder. Jf. Meulengracht Srensen 1993, s. 167.

416
tydelig, da de begge bor hos moderen. Halli arbejder trofast det samme sted i mange r,
medens Ragnar ofte skifter job.
Ragnar figuren har paralleller til Gunnar fra Hlidarende, fra Njals saga. Selve
navneligheden bestr i 6 bogstaver, 2 stavelser. endelsen -nar er ens i begge navne. Ragnar
beskrives som en idrtsmand og en kmpe, ligesom forbilledet Gunnar, om hvem der i Njals
saga str: Han var en Mand af hj Vkst og stor Styrke og uden Lige i Vaabenfrdighed.
Han kunde, naar han vilde, haandtere Svrd og Spyd med Kejten ligesaa godt som med hjre
Haand [].I fuld Hrkldning sprang han hjere end sin egen Hjde og ikke lavere baglns
end fremad [] Njals saga s. 55. Gunnar er en sagaperson, der i forhold til de almindeligvis
knappe personkarakteristikker ofres en lngere beskrivelse p og hans mor og slgtsforhold
p den side prsenteres fr faderen. Ragnars strrelse understreges af at ingen almindelig
uniform passer ham, da han skal bruge en sdan til kampen i den spanske borgerkrig. FPH s.
63.
Ragnars fortlling udgr en parallel til sagaudviklingshistorierne parallelt med
eksempelvis Egil, eller Gretter, hvor vi flger Ragnar29 fra ung utilpasset lmmel til voksen
helt.
Ragnar er som ung dreng hidsig og rastls, han er som Egil tidlig udviklet og ligner som 10
rig en dreng p 14.

Alle ved at Ragnar var hj af vkst, men han var ikke alene hj, han var ogs strk, og han var
tidligt udviklet. Ragnar har lige rundet de ti, da han sidder i bilen og ligner en prins i en fornem
karet, men han ser ud til at vre fire r ldre FPH s. 102.

Han kan arbejde som en voksen mand og det udnyttes. Beskrivelse og fremhvelse gennem
sammenligning med andre brn, der er velkendt fra sagaerne, er brugt her.
Mellem Gudny og Ragnar er der strke bnd, det ses i hendes drmme om ham medens
han er i fare i Spanien. Senere sidder de ofte i samtale og drikker kaffe. I flere sagaer indgr
venskaber mellem mdre og snner.
Ragnars ven Grimur, er en slagsbroder og i beskrivelsen af ham i slagsml er der tydelige
referencer til beskrivelserne af Egils stridslyst som dreng i Egils saga. Direkte intertekstualitet
findes Folk sammenlignede Grimur med Egil Skallagrimsson. s. 113. Grimur kan ikke blive
viking, men passionen for kamp og tummel kanaliseres ud i at han bliver bokser og fr
tilnavnet Bokse-Grimur.
Vikingetidens kamplege og hestekampe udmyntes her i boksning, hammerkast, hkkelb
og lngdespring for Kres vedkommende, og senere bliver det til et vertikalt slag mellem
kommunisterne p den ene side og politiet p den anden side, efter byrdets planer om at
nedstte timelnnen for de hrdest arbejdende arbejdsmnd. Ragnar gr nrmest bersrk. En
anden af Ragnars venner: Olli, afvbner under et sammenstd 6 politimnd, man m
uvilkrligt tnke p eksempelvis Gunnars i kamp fra NS.30 Her er der mange srede p begge
sider, hvor der i sagaerne er mange faldne efter et slag. Heldigvis er kampen en styrkeprve,
men ikke med faldne. Olli appellerer generelt til brydeglade politibetjente.
Som frivillige i den spanske borgerkrig oplever Ragnar og Olli slag og krigsfrelse for
alvor i Spanien. Frst opleves rejsen og krigen som eventyr, men til slut fremgr det at Ragnar
oplever lede ved krigen og kampene. I denne beskrivelse er EMG sarkastisk i trd med
Laxness i Gerpla. Dog er der specifik litterr intertekstualitet til Thormod Kolbrunsskjald,31
hvor Ragnar som en sand helt hjlper de andre srede og brer sine lidelser stoisk, da han
29
Ragnar deler navn med helten Ragnar Lodbrog, Meulengracht-Srensen 2006, s. 279 opridser en rkke sagaer
og digte som beretter hans historie, der ogs findes hos Saxo og i senmiddelalderens folkeviser.
30
Njals saga s. 111, og s. 116, hvor Gunnar i den sidste kamp srer 16 mand, heraf flere ddeligt, og drber to.
31
Sagaen om fostbrdrene s. 114

417
sres og ironisk nok mister heldigvis kun n finger. Efter hjemkomsten bliver trfningerne
og hndelserne i Spanien, fangenskabet og den vanskelige hjemrejse som i en gte saga
transformeret 32til eventyrlige beretninger, der fortlles igen og igen.
Beretningerne om Olli har karakter af skrner og her er det sted i romanen, hvor der er flest
overdrivelser; dette tilfrer Olli et eventyrligt skr. Han har ogs som en farverig person flere
tilnavne.
Ollis rejser ud i verden. Han snyder sig som Egil med. Senere oplever han eventyr og
bliver afholdt i det fremmede, her s meget at et par i USA vil adoptere ham. Der er
referencer til Njals saga, hvor det indirekte fremgik at dronning Gunhild gerne ville have
beholdt Hrut.33 Olli beslutter, som Hrut, at vende tilbage til Island og den kvinde der venter
der.34 Olli kommer forandret hjem, hvor slgten der nrmest regner ham for dd, overraskes
over hans ankomst. s.122. Senere fr Olli arbejde p maskinfabrikken Odin, og nsten
selvflgeligt har han magiske evner. Maskiner fordobler deres afkast og vi m tnke p
momentvis magisk heldige sagapersoner, som Ravnkel Frjsgode, der eksempelvis fr lam
der nsten fdes med to hoveder.35
Olli lider momentvist af udlngsel og drager s til ss, vi husker Gunlg Ormstunge og
andre sagahelte, der ogs har udlngsel. Han er en kvindebedrer, og gifter sig tre gange36
med piger der hedder Unnar. Dog er det kun en af dem, der kaldes for Unnar den fagre
s.159. Fra Njals saga kendes historien om Unn, der var gift med Hrut og blev skilt, og det
bliver Olli og Unn ligeledes her. Ogs i Laxdla saga indgr en landnamskvinde ved navn
Unn (Jf. Bredsdorff 1971/1995, s. 53). Unnar i FPH beskrives som smuk: hj og slank, hvilket
er parallelt med sagaernes sknhedsidealer.
Det klassiske trekantmotiv kendt fra flere sagaer, eksempelvis LS og GOS, findes i FPH,
hvor Olli tager af sted til den spanske borgerkrig og forlader den gravide Unn. Efter krigens
slutning vender han omsider hjem og finder Unn med et nyt barn p armen og en ny mand.

Da Olli vendte hjem fra krigen, havde han en datter p t r, men ingen kone. Unnar den fagre
havde opgivet at vente og havde ladet en anden mand flytte ind.

Jeg regnede ikke med, at du ville komme tilbage, sagde hun med grden stende i halsen, da
Olli dukkede op i dren med en fornjet mine, som om han var taget af sted i gr eller lige var
smuttet ned i en kiosk for at kbe tndstikker FPH s. 203.

I sagaerne flges erfaringen med at den ventende har giftet sig med en anden af jalousi og
drab. Her sker en intervention i forhold til det klassiske forlg, der behandles med
humoristisk distance. Olli trkker p skulderen og accepterer tingenes tilstand uden at hvne
sig p den nye mand.
Rkken af direkte og indirekte intertekstuelle referencer der er opridset her, viser
forbindelsen og dialogen mellem Fodspor p himlen og sagaerne. Sagaparalleliteten kommer
til udtryk i slgtsmotivet og i personprsentationerne, hvor samme struktur som i sagaernes
benyttes og igennem interventioner af klassiske sagamotiver.
Desuden i det store persongalleri, ved direkte allusioner, ved personer der flges fra vugge
til grav og fremstilles som moderne helte, ogs genremosaik er et sagaparallelt trk.

Litteratur
32
Amerikaner Olli og Bokse Olli.
33
Njals saga s. 35.
34
Hruts fstem Unn, venter ham p Island, ligesom Unnar venter her. Ibid s. 33.
35
Man kunne sledes sige, at det var lige ved, der var to hoveder p hvert dyr Ravnkel Frjgodes saga s. 33.
36
En parallel til Gudrun der gifter sig 4 gange i Laksdla saga.

418
Bakhtin, Mikhail, 2000: Tidens og kronotopens former i romanen 193738 Oversttelse ved Harald
Hartvig Jepsen: i Rum, tid & historie. Kulturklassiker, Klim.
Bjrnsson, rni, 2003: Wagner and the Volsungs. Icelandic sources of Der Ring des Nibelun-
gen.Viking Society for Northern Research, University College, London.
Bredsdorff, Thomas, 1995: Kaos og krlighed. En studie i islndingesagaers livsbillede. Gyldendal,
(1971) Efterskrift til 2. udgave.
De Islandske sagaer 13. (1930) Red. Gunnar Gunnarsson og Hans Kyrre. Gyldendal, Danmark,
1960, 2000.
Eysteinsson, strdur and lfhildur Dagsdttir, 2006: Icelandic Prose Literature 19402000 i A
History of Icelandic Literature, Ed. By Daisy Neijmann, University of Nebraska Press, Lincoln &
London.
Gumundsson, Einar Mr, 1995: Universets engle, fra Englar alheimsins, 1993. Oversat af Erik
Skyum-Nielsen, Vindrose.
Gumundsson, Einar Mr, 1999: Fodspor p himlen, fra Ftspor himnum, 1997. Oversat af Erik
Skyum-Nielsen, Vindrose.
Gumundsson, Einar Mr, 2001: Drmme p jorden, fra Draumar jru, 2000. Oversat af Erik
Skyum-Nielsen, Vindrose.
Gumundsson, Einar Mr, 2003: Navnlse veje. fra Nafnlausir vegir, 2002. Oversat af Erik Skyum-
Nielsen, Vindrose.
Hvarregaard, Lise, 2006: Sagatrk i Einar Mr Gumundssons Universets engle i The fantastic in
old Norse/Icelandic literature. Sagas and the british isles. Vol.1. The thirteenth international saga
conference. Durham and York. Ed. John Mckinnell, et al. Durham University, England.
Hvarregaard, Lise, 2007: The Streaming Nature in the Works of Einar Mr Gumundsson 97th
SASS Annual meeting: Quad Cities area of Davenport, Iowa/Rock Island, Illinois, April 2628.
Hvarregaard, Lise, 2009: Ironi og intertekstualitet. Sagatrk i Einar Mr Gumundssons romaner
Kbenhavns Universitet.
Jhannsson, Jn Yngvi, 2006: I: slensk bkmenntasaga V, Ritstjrn: Gumundur Andri Thorsson,
Ml og menning, Reykjavik.
Jhannsson, Jn Yngvi, 2006: I fuldt flor? Om islandsk litteratur fra rtusindeskiftet i: Nordisk
litteratur 2006, Red. Jgvan Isaksen, NORDBOK, Nordisk Ministerrd, Kbenhavn.
Meulengracht Srensen, Preben, 1993: Fortlling og re. Studier i islndingesagaerne. Aarhus
Universitetsforlag.
Meulengracht Srensen, Preben, 2006: Kapitler af Nordens litteratur i oldtid og middelalder. Aarhus
Universitetsforlag.
lason, Vsteinn, 1998: Dialogues with the Viking Age. Heimskringla. Ml og menning, Reykjavk.
Plsson, Heimir, 2000: Lykill a slendingasgum. Ml og menning, Reykjavik, 1998 og norsk Nkkel
til islendingesagaene, oversat af Gro Tove Sandsmark, Cappelen akademisk forlag, Oslo.
Rubow, Paul V., 1923: Saga og Pastiche. Gyldendal, Danmark.
Skyum-Nielsen, Erik, 2002: Center og periferi i Fortllere og steder i Norden. 11 nordiske
forfattere p Hald hovedgrd. Dansk forfatterforening og Nordbok.
Skyum-Nielsen, Erik, 2005: Nye og gamle fortlleformer i moderne vestnordisk prosa i postkolonialt
perspektiv i Ordenes slotte. Red. Auur Hauksdttir, et al. Stofnun Vigdsar Finnbogadttur,
Danmark.

419
 Die deutschsprachigen Fassungen und Verarbeitungen der
Jmsvkinga saga von den 1920er bis zu den 1940er Jahren
Michael Irlenbusch-Reynard, Stavanger, Norway
Die deutschsprachige Rezeption der Jmsvkinga saga, im 19. Jahrhundert einsetzend im Um-
feld der pommerschen Geschichtsforschung,1 verlagerte sich zu Beginn des 20. Jahrhunderts
hin zur Romantisierung und Mystifizierung mit einem Hhepunkt in den 1930er Jahren, der
sich in zahlreichen Verdeutschungen unterschiedlichster Originaltreue, aber auch regelrechter
Nutzanwendung niederschlug.2 Dominierte im ersteren Falle durchaus der Unterhaltungswert
einer Wikingergeschichte mit prgnanten Charakteren und Hhepunkten, so bediente sich im
letzteren die Germanenideologie der Saga als Beleg fr eine Tradition von Mnnerbundtum
und Gefolgschaft. Die folgende Untersuchung widmet sich insbesondere der Methodik zur
Vereinnahmung der J. s. im ideologischen und propagandistischen Kontext.

Deutschsprachige bersetzungen
Die Grundlage der bersetzungen stellen die schwedischen und dnischen Editionen des 19.
Jahrhunderts dar, wiewohl offensichtlich auch im Rahmen der Altnordischen Saga-Bibliothek
ein Textband mit den reihenblichen deutschen Kommentaren und bersetzungshilfen vorge-
sehen war, der von Carl af Petersens, dem frheren Herausgeber der Versionen AM 510, 4to
(1879) und AM 291, 4to (1882), besorgt werden sollte,3 jedoch nie zustande kam.
Auf Deutsch erschienen zunchst drei bersetzungen im 19. Jahrhundert: 1827 von Lud-
wig Giesebrecht hchst textgetreu4 und 1875 von Anton Edmund Wollheim mit glttendem
Satzbau5 beide mit kleinen Auslassungsfehlern6 nach SKB 7, 4to (1824) und 1892 von
Ferdinand Khull in frei formulierendem und dabei krzendem Umgang mit AM 510, 4to
(1879).7
1
Z. B. Giesebrecht 1840 bzw. 1843 und Klempin 1847, dabei anknpfend an bersetzte dnische Vorstudien,
namentlich Vedel Simonsen 1827.
2
Als wertvolle Grundlage diente die von Julia Zernack zusammengestellte Bibliographie deutschsprachiger
Sagabersetzungen (Zernack 1997). Zur notwendigen Abgrenzung wurde lediglich Literatur einbezogen, welche
die J. s. unabhngig vom Umfang als eigenstndigen Text fhrt, sich also nicht mit einer Inhaltsangabe oder
einer reinen Referenz begngt.
3
Vgl. Briefe Hugo Gerings an Eugen Mogk vom 23. 5. und 15. 8. 1890, wiedergegeben bei Fix 2004, S. 315 und
324. Die Handschriftenfrage wurde in diesem frhen Planungsstadium noch nicht errtert.
4
Dies geht so weit, da frndi grundstzlich archaisierend mit Freund wiedergegeben wird, dazu einmal mit
Vetter (Giesebrecht 1827, S. 101) eigentlich Onkel, etymologisch jedoch korrekt als Vaterbruder. Auf-
fllige bersetzungsfehler sind dreihundert Mann (S. 106) fr tv hundru manna (SKB 7, 4to [1824], S. 18),
was Rechtes an Hlfe (S. 120) fr ltinn styrk (SKB 7, 4to [1824], S. 32) und es wog jedes Korn einen Pfeil
(S. 132) fr v eyri eitt kornit (SKB 7, 4to [1824], S. 44).
5
Es werden Stze teilweise anders verbunden, hierbei wird die Syntax gelegentlich vereinfacht und zudem aus-
schlielich Prteritum verwendet. Gleichzeitig werden jedoch zahlreiche erluternde Zustze und Funoten mit
Originaltext geboten. Falsch bersetzt sind: ok er n hlfo kafari, geingr n aldri af eim (SKB 7, 4to [1824],
S. 7) und war nun doppelt so eifrig auf den Schiffen, die er nicht verlie (Wollheim 1875, S. 157); bakaz
(SKB 7, 4to [1824], S. 9) hockte sich nieder (S. 157); eir sna skipum snum (SKB 7, 4to [1824], S. 11)
Sie banden ihre Schiffe fest (S. 158); jrnhurir (SKB 7, 4to [1824], S. 14) Eichenpforten (S. 159). Diese
bersetzung, die lediglich den Teil bis zur Grndung der Jomsburg durch Plna-Tki abdeckt, befindet sich
zwar in einer Anthologie, ist aber dennoch originr, weshalb ich sie an dieser Stelle auffhren mchte.
6
Fehlstellen: Giesebrecht 1827, S. 106: Haraldr jarl qvez vera hrddr fyrir Veseta ok sonum hans (SKB 7, 4to
[1824], S. 18), S. 121: Noregi (SKB 7, 4to [1824], S. 33), S. 133: deyr orleifr (SKB 7, 4to [1824], S. 45);
Wollheim 1875, S. 155: oc dreymir hana (SKB 7, 4to [1824], S. 3) und ok ickir n frgr mar (SKB 7, 4to
[1824], S. 4).
7
Es fehlen zudem smtliche lausavsur. Khulls darber hinaus von Satzumstellungen geprgte Textbehandlung
macht es unmglich, in diesem Rahmen detailliert auf Abweichungen und zweifelhafte Umformulierungen ein-

420
 Walter Baetkes bersetzung der J. s. nach SKB 7, 4to (1875) schlielich, 1924 im Band 19
der Reihe Thule vorgelegt, erschien, nebst einer bertragung der Jmsvkinga drpa von Fe-
lix Niedner, in einem Band zusammen mit solchen der Orkneyinga saga und der Kntlinga
saga. Die Absicht, islndische Darstellungen von Geschehnissen auerhalb Islands und Nor-
wegens zu vereinen, fhrt hierbei zu dem herausgeberisch interessanten Nebeneffekt, eine
Quelle mitsamt ihrer wenngleich sehr knappen Auswertung vorzustellen; zugleich ergeht
frh ein Hinweis8 auf die fragwrdige Historizitt der J. s., deren Text im Thule-Band auf die
Kntlinga saga folgt.
Der Anspruch, bei den bersetzungen mglichste Treue an[zustreben],9 trifft allerdings
nur teilweise zu: Baetke entfernt sich nicht unntig vom Sagatext, ist jedoch uneinheitlich
hinsichtlich der bei der Thule-Reihe immer wieder kritisierten Behandlung von Eigennamen,10
ignoriert die blichen Tempussprnge des Originals und erzhlt fast durchgehend im Prteri-
tum. rgerlich sind Verstndnis- und Auslassungsfehler: In der Seeschlacht im Hjrungavgr
lt Baetke gegen orkell hfi Yriarskeggi den Andern11 antreten und gegen Sigurr kpa
Arnmod und seinen Sohn Arni und Finn [sic!]12. Im weiteren Verlauf scheint Baetke bei der
Berichtigung der kontextuell fehlplazierten Episode von Bis Verlust seiner Hnde die ber-
sicht verloren zu haben die Passage ok hlepr sian fyrir bor me kisturnar13 fehlt
schlichtweg.
Sowohl Baetke wie auch Giesebrecht geben lediglich den mit Tki beginnenden zweiten
Teil der J. s. wieder; Khull fgt in einem Anhang die Vorgeschichten ber die dnischen K-
nige anhand der lateinischen Version von Arngrmr Jnsson (Edition 1877) sowie eine Auf-
stellung der textlichen Unterschiede zur Redaktion AM 510, 4to, die erst bei Tki einsetzt, bei.
Baetkes bersetzung steht nicht allein fr sich selbst, und in geringerem Umfang trifft dies
ebenso auf jene Giesebrechts und Khulls zu,14 weshalb sie mit besprochen wurden sie alle
lieferten die Grundlage fr zahlreiche Anthologien, Zitatmaterial fr Abhandlungen und Aus-
gangspunkt fr Nacherzhlungen im hier behandelten Zeitraum.

Deutschsprachige Nacherzhlungen und Bearbeitungen

zugehen. Klare bersetzungsfehler sind: Die Vter nahmen (Khull 1892, S. 19) fr eir fegar toku (AM 512,
4to [1879], S. 38) und hundertundvierzig (S. 19) fr .c. (AM 512, 4to [1879], S. 39).
8
Baetke 1924, S. 10.
9
Baetke 1924, S. 454.
10
Strt-Haraldr: Stutz-Harald (durchaus elegant, aber kommentarlos); Gautland: Gautland (mit Erluterung in
Funote); Bretland: Wales; Sigurr kpa: Sigurd Mantel (Baetke 1924, S. 406, 395, 397, 407). Giesebrecht 1827
tappt hier mehr als einmal in die Falsche-Freunde-Falle: Gotland, Brittenland, Sigurdr Kappe (S. 91, 93, 104).
Wollheim 1875 behlt Bretland wie auch Borgundarhlm [sic!] erklrend bei (S. 155 und 157), schreibt dafr
abwechselnd Got- und Gautland (S. 155) und schiebt einmalig einen Pseudo-Dativ ein: Birne Brezki (S. 158),
der spter stets Birn [sic!] Brezki heit und zuvor Bjrn von Bretland genannt wurde. Khull 1892 bewahrt zwar
Bretland (S. 7), verfhrt aber sonst oft chaotisch, z. B.: Bjorn [sic!] der Bretische (S. 7)/der Bretlnder
(S. 13)/Breski (S. 22); Borgundar-Holm/Borgundarholm (S. 11); Wendenland (S. 14)/Windland (S. 22); Haward
der Hauende (S. 27)/Hggwandi (S. 35).
11
Baetke 1924, S. 423; SKB 7, 4to (1875), S. 27: moti orkatli brour hans. Yriarskei anna Sigurr steklingr
af Hlga lanndi orir hirtr Giesebrecht 1827, S. 124, korrekt: Yrjarskeggi, zum anderen []
12
Baetke 1924, S. 423; SKB 7, 4to (1875), S. 27: moti Siguri kapu eir fegar Arnmr ok rni ok Fir Gie-
sebrecht 1827, S. 125, korrekt: Arnmodr und Arni und Fidr, Vater und Shne.
13
SKB 7, 4to (1875), S. 30, vgl. Baetke 1924, S. 428429. Giesebrecht 1827, S. 130131, korrigiert die Plazie-
rung (nicht jedoch die Redundanz) und weist in einer Funote darauf hin, was Baetke unterlt.
14
In den Nacherzhlungen ist dies teilweise indirekt erkennbar: Beispielsweise findet sich Baetkes Verstndnis-
fehler der Schlachtaufstellung bei Ball [1941], S. 44, wieder; Kath [1941] orientiert sich sowohl an Baetke wie
an Khull. Vorlage fr Lawrenz [1938?] war nicht nur wie angegeben Giesebrecht 1827, sondern zustzlich
fr den Abschnitt bis zu Plna-Tkis Tod Vedel Simonsen 1827 (speziell S. 7486, 9198, 110114, 127130,
138151), der sich in weiten Teilen auf Saxo sttzt.

421
Die J. s. ist keine der klassischen Islndersagas; ihr gesamter Inhalt, ihr Schauplatz, ihre Pro-
tagonisten entsprechen wenig den auch bei deutschsprachigen Lesern populren Geschichten
von islndischen Bauern und Helden, die sich vorwiegend gegenseitig befehden, viel dich-
ten, aber stets bodenstndig bleiben. Genau dies vereint, eine Geschichte von Berufswikin-
gern in einer mystischen Burg, ergibt jedoch eine ideale Vorlage fr literarische Bettigungen,
die von gegltteten Nacherzhlungen fr eine jugendliche Leserschaft bis hin zu frei Erfunde-
nem rund um die von der Saga, einem Quellborn germanischer Ehre und heidnischer
Treue,15 gelieferten Motive reichen, zumal sie zum Teil auf deutschem Boden spielt.16 Zu
derartiger Einvernahme trgt die Wahl suggestiver Bezeichnungen wie Polen und
Pommern noch bei.17 Deutlich sichtbar sind vlkische Einflsse, deren Ausrichtung in den
erzhlerischen Werken variiert, u. a. rassenpflegerisch, lebensreformistisch, antislavisch, anti-
rmisch. Einige kurze Einzelbetrachtungen sind daher naheliegend.
Auf den Identifikationsfaktor baut Werner Heider auf, dessen Erzhlung voller geflgelter
Helme18 am weitesten vom Original entfernt ist. Sein Protagonist ist ein Sachse, also ein
Deutscher, der an der Seite Plna-Tkis die Jomswikinger mitbegrndet. Gleichzeitig wird
ein pathetischer Kontrast zu den Slawen, die [i]n die harte Rede der Nordleute zischel[n],19
hergestellt, was zusammen mit dem Motiv der Blutsbrderschaft zwischen einem wendischen
edlen Wilden und Plna-Tki eine geradezu Karl-May-hafte Stimmung erzeugt.
Das mit 1923 bereits relativ frh erschienene Zeitbild von Elisabeth Hersen zeichnet sich
durch ein impliziertes Frauenbild der Jomswikinger als Zentralthema aus: Dessen Antipoden
sind Sigvaldi, der Plna-Tki ein altersbedingtes Desinteresse unterstellt, und Bjrn inn bre-
zki als geschworene[r] Frauenhasser20, der dies sogar seinem Erbbierschwur hinzufgt.21
Das Eigentmliche dabei ist der erfundene Beschlu, Frauenerlaubnisse als Belohnung fr
eine Tat ,verwegener Art zu verleihen22 als offensichtliches Prestigesymbol nach sexu-
elle[r] Kampfauslese.23 Hersen gibt sich merklich Mhe, weiblichen Einflu als zersetzend
zu kennzeichnen; romantisch wird so weniger der liebestolle Vagn, sondern der misogyne,
nostalgierende Mnnerbndler Bjrn.
Bei Gerhard Ramlow erffnet sich eine ungewhnliche Diskrepanz zwischen Ductus und
Absicht: Seine ausschmckende Nacherzhlung entbehrt des schwlstigen Pathos, wie er den
brigen hier besprochenen meist zu eigen ist, und berrascht dagegen stellenweise mit saga-
typischem, trockenem Humor und natrlichem Benehmen.24 Erst das Nachwort verfllt in den
blichen Sprachstil und benennt das Anliegen des Buches, bei der Selbstfindung, in der sich
heute das deutsche Volk zu seinem blutmigen Schicksal bekennt,25 mithelfen zu wollen
die unterhaltsame Unaufdringlichkeit auch des sonstigen Buchinhalts jedenfalls wirkt ein-
gnglicher als die ideologische Schwere anderer Werke.

15
Ball [1941], S. 6.
16
Jantzen 1928, S. 486, vgl. auch Fischer 1934, S. 575.
17
Vgl. z. B. Ball [1941], passim, und Ramlow 1936, passim.
18
Vgl. Heider [1937] S. 14, 21 und 80.
19
Heider [1937], S. 32, vgl. auch S. 61 und 99.
20
Hersen [1923], S. 49.
21
Vgl. Hersen [1923], S. 141.
22
Vgl. Hersen [1923], S. 29.
23
Hermand 1995, S. 80 (vgl. auch S. 290), vgl. ebenso Bruns 2001, S. 80. Die ansonsten unauffllige Nacher-
zhlung von Lydia Kath scheint dies direkt von Hersen bernommen zu haben, vgl. Kath [1941], S. 33.
24
Vgl. Ramlow 1936, S. 125 und 141142. Schlichtheit gilt wiederum fr Genzmer 1944, der im wesentlichen
Baetke 1924 mit kleinen, erklrenden Erweiterungen wiedergibt. Ich fhre Ramlow und Genzmer in diesem
Kapitel wegen ihrer Vollstndigkeit mit auf, obwohl sie die J. s. in Textsammlungen einbeziehen.
25
Ramlow 1936, S. 314.

422
 Kurt Herwarth Ball26 bringt den germanischglubigen, antiklerikalen Aspekt ins Spiel: Fr
ihn zeichnen sich die Jomswikinger und ebenso die norwegischen Jarle! dadurch aus,
noch nicht unter Kreuz und Krummstab gezwungen27 worden zu sein. berdies sei durch
die gewaltsame Christianisierung seitens des Frankenknigs Karl die lebendige Urkraft der
Wikinger, die bis in das Herz des Mittelmeeres hinein [] nordischen Raum [schufen],28
vertrieben worden, wie Ball einen sich im Weltschmerz ergehenden Vagn sinnieren lt. Balls
Nacherzhlung berschreitet eindeutig die Schwelle von der Unterhaltungsliteratur mit Unter-
tnen hin zu ideologischer Benutzung eines altnordischen Stoffes, wie sie im folgenden Ab-
schnitt besprochen wird.

Thematische und ideologische Einbindungen

Die Rezeption der J. s. nimmt teils einen propagandistischen Ansatz, der auf den Vorherr-
schaftsanspruch Deutschlands und die Kriegerherrlichkeit der Deutschen abhebt, berwiegend
jedoch einen ideologischen, nmlich die Belegfhrung der germanischen Arteigenheit von
Gefolgschaft und Mnner(kampf)bund.29
Die Indienststellung der Seegermanen30 und mit ihnen der Jomswikinger fr die Kriegs-
marine geschieht bereits kaiserzeitlich mit Heims 1904 und setzt sich fort bis zu
Busch/Ramlow 1940, Seeherrschaft als Bedingung zur Weltherrschaft postulierend mit histo-
rischen Wurzeln in den berlegenen germanischen Seefahrern der Wikingerzeit.31 Zu Beginn
des Ersten Weltkrieges setzt sich das Vorwort zu Neckel 1915 vehement der Verkennung des
deutschen, also germanischen Charakters entgegen;32 der germanische Trotz zum einen,
eine doppelte Ethik zum Wert des Lebens fr Frieden und Krieg zum anderen33 sei der Weg
nicht nur zum Verstndnis des Heldentodes der Jomswikinger, sondern vor allem zur rechten
Einstellung des deutschen Soldaten zum Tten und Sterben.34
Die ideologische Einbindung der J. s. vollzieht sich in erster Linie durch manipulative Pr-
sentation des Sagatextes. Eine Gesamtbetrachtung der hier untersuchten Literatur zeigt, da
gewisse Schlsselstellen (Burg, Gesetze, Trunkenheit & Gelbde, Seeschlacht, Zauberei, Hin-
richtung) und Aspekte (Gefolgschaft, Heidentum) hchst unterschiedlich prioritiert werden.
Dies ist gleichwohl keine Frage der prinzipiellen Vollstndigkeit: Lawrenz [1938?] wie Fah-
nemann [1937] komprimieren die gesamte Saga fr den Schulgebrauch zur Herausstellung
des Heldenhaften und geben gezielt Raum fr Plna-Tki und das Kollektiv, aus dem einzig
Vagn herausragen darf. Im Vergleich von 24 Titeln, die sich der J. s. in ihrer Argumentation

26
Ball hatte 1933 die Leitung der antisemitischen Zeitschrift Hammer bernommen; im Vorwort zu den
Jomswikingern findet sich denn auch ein Verweis auf den Zeitzeugen al-Tartushi als einen Sklavenhndler,
den spanische[n] Jude[n] Ibrahim ibn Jacub (Ball [1941], S. 6).
27
Ball [1941], S. 6, vgl. auch S. 39 und 54.
28
Ball [1941], S. 55.
29
Von der hier eingesehenen Literatur kann lediglich Fischer 1934 als reines Lesebuch gelten; als auffllig sach-
lich erweist sich Genzmer 1944. Ansonsten knnen die Grenzen zwischen Unterhaltung und Propaganda durch-
aus flieend sein, zumal in den maritimen Sammlungen.
30
Busch/Ramlow 1940, S. 1725 passim.
31
Vgl. Heims 1904, S. V; Busch/Docter 1935, S. 8; Busch/Ramlow 1940, S. 14; morsch gewordene Welten
[wurden] zertrmmert (Plamann 1929, S. 3; hnlich Vogel 1942, S. 152).
32
Vgl. dazu ausfhrlich Zernack 1994, S. 210213. Die Neuausgabe verzichtet auf diese Apologien, lt aller-
dings auch die Rechtfertigung fr Hkons Brutalitt und den Respekt Besiegten gegenber fort (vgl. Neckel
1940, S. 50, und Neckel 1915, S. 7273).
33
Neckel 1915, S. 23 und 14.
34
So lt sich Hkon als hingebungsvoller Landesvater interpretieren: Man mu das Menschenopfer verstehen
aus der Haltung der Mutter, die ihren Sohn ,dem Vaterland opfert, indem sie ihn auffordert, sich freiwillig zum
Kriegsdienst zu melden, und nicht aus der Menschenschlchterei menschenfressender Barbaren. (Bohne 1937,
S. 34.)

423
bedienen,35 lt sich feststellen, da generell die Darstellungstiefe schwankt: Der Wortlaut der
Gesetze, der Verlauf der Seeschlacht und die Aburteilungszene werden vorwiegend detailliert
geschildert, wohingegen die Umstnde des fatalen Gelbdes mehrheitlich nur kurz erwhnt
werden. Die Mittel sind mitunter plumpe Umdichtungen und wesentlich subtiler das
gezielte Auslassen von Sagageschehen als ein weniger pointierendes als selektierendes Prin-
zip, das heit das gezielte Unterdrcken von kleinen, aber wichtigen Umstnden, sowohl in
der Handlung selbst als auch in der Ein- und Hinfhrung. Die schwerwiegendsten Manipula-
tionen betreffen den Grund der Fahrt nach Norwegen und die Wende in der Seeschlacht.
Einige Autoren sind bestrebt, eine Formulierung zu finden, nach welcher die Jomswikinger
nicht Opfer mangelnder Selbstkontrolle werden. So wird eine fixe Idee Sigvaldis zur Erobe-
rung Norwegens erfunden, um Sveinns List erzhlerisch zu umgehen36 oder die gesamte Epi-
sode einfach verschwiegen.37 Alternativ wird eine urschliche, unprovozierte Heimtcke
Sveinns suggeriert oder zumindest die Katerstimmung am Folgetage ausgelassen.38 Wenn-
gleich eine Situation ausgenutzt wurde, der Gegner sich also unehrenhaft verhielt,39 sollen
demnach doch stets die Gebote der eigenen Ehre und die unbedingte Erfllung des Gelbdes
selbstverstndlich sein. Gleichzeitig wird sptestens hier in unterstreichender Wortwahl auf
die Disqualifizierung Sigvaldis hingearbeitet. Indem seine Fhrerqualitt angezweifelt wird,
indem sich in seinen bermtigen Schwur seine Mannen verketten und wohl oder bel
losziehen,40 wird eine latente Vakanz angedeutet, und nicht zufllig wird Vagns Gelbde mit-
unter auf den kriegerischen Aspekt, die Ttung orkell leiras, reduziert41 ohne Erwhnung
Ingibjrgs und somit ohne Einlassung auf den Sittenverfall bei den weiberlose[n] Schwert-
brder[n]42.
Der Wendepunkt der Schlacht im Hjrungavgr wird mehrfach auf ein heftiges, aber letzt-
lich natrliches Unwetter vereinfacht oder gekrzt,43 wodurch die Flucht Sigvaldis eindeutige
Feigheit darstellt. Eine Umdeutung als geglckter Rckzug trotz Hkons Gtterbeschwrung
indes ermglicht den Trotz der Gefangenen aus einer vllig neuen Perspektive: Hauptsache,
der Fhrer ist gerettet.44 Seine tatschliche Ttung45 wiederum ist konsequent fr einen kla-
ren bergang zum einem neuen Fhrer, denn ein junger Bursch, Vagn, bernahm den Ober-
befehl.46
Diese beiden nderungen sind bezeichnend fr die kontextuelle Einordnung der
J.-s.-Exzerpte und -Paraphrasen: Gefolgschaft und Kriegerehre. Die Freude ber den entron-
35
Ohne die reinen Nacherzhlungen von Ball, Heider, Hersen und Kath.
36
Die Jomswikinger [] saen [] in ihrer Halle beim Trunke. [] Sigwaldi schwor, er werde niemals die
Norweger ber seinen kleinen Staat, der sich selbst Gesetze gegeben hatte, herrschen lassen. Er wolle selbst
einmal ber Norwegen regieren oder im Kampfe fallen. [] Kurze Zeit darauf hatten die Jomswikinger ihrem
Erbfeind [Hkon ] eine Seeschlacht zu liefern. (Wagenfhr 1935, S. 16.)
37
Die Jomswikinger unternehmen nach Palnatokis Tode einen Kriegszug gegen den Jarl Hakon von Norwegen;
in der Hjrungabucht [] treffen die Flotten aufeinander. (Dauch 1940, S. 176.)
38
Svend [], dem die Macht des Bundes schon lange ein Dorn im Auge war. (Fahnemann [1937], S. 8; vgl.
Wikingerfahrten, S. 9.) [Svend hatte] ein groes Gelage auf Seeland hergerichtet, zu dem er auch den Bund
der Jomswikinger eingeladen hatte. [] Nun nahm das Gelage ein Ende, und die Jomswikinger machten sich
sofort von dem Gelage auf, und danach fuhren sie aus []. (Wllenweber 1938, S. 2627.)
39
Auf einem anderen Blatte steht, da abgesehen von der tatschlichen Konstellation pragmatische List und
Ausnutzung von Situationen vielmehr eine sagatypische Tugend ist.
40
Naumann 1939, S. 107.
41
Vgl. Fahnemann [1937], S. 10 und 14; Wikingerfahrten [1937], S. 9 und 11.
42
Strasser 1938, S. 110, und Strasser 1943, S. 81.
43
Vgl. Wagenfhr 1935, S. 17; Wllenweber 1938, S. 29; Fahnemann [1937], S. 13 (nach Stogebet Hkons);
Wikingerfahrten [1937], S. 10.
44
Busch/Docter 1935, S. 187 (vgl. auch S. 185187), und Busch/Ramlow 1940, S. 83 (vgl. auch S. 8083). Zur
Abnderung gehrt, da Sigvaldi unbehelligt von Vagn entkommt.
45
Vgl. Wikingerfahrten [1937], S. 10.
46
Wagenfhr 1935, S. 17.

424
nenen Fhrer bleibt ein Einzelfall, der Tenor ist vielmehr: Heldentum des Fhrers [ist]
nichts, wenn es sich nicht mit Sorge fr Leben und Rettung seiner Leute verbindet.47 Bjrn
inn brezki, der seinen Mann aus dem Tumult nach Plna-Tkis Ttungsbekenntnis rettet, wre
ein geeignetes Beispiel gewesen, doch er wird hchstens als ergrauter Nebendarsteller in der
Schluszene zugelassen. Dafr wird Vagn als Lichtgestalt im Sinne des Gemeinschaftsgeistes
einer Gefolgschaft herausgestellt, gegebenfalls durch berdramatisierung der Situation der
Gefangenen vor Hkon.48
Hier liegt, neben der Auflistung der Gesetze, der Schlssel zur Verwertung der J. s.: Plna-
Tki verbleibt der mystische Grndungsvater49 der Burg und des Mnnerbundes, gerne ver-
glichen mit einem Ritterordensbund unter einem Hochmeister,50 doch letztlich ist dieser
Urzustand vom vlkischen Standpunkt aus gesehen etwas Ungesundes.51 Vagn steht fr die
Verjngung und immerwhrende notwendige Erneuerung52 der Mnner(kampf)bnde als Er-
gnzung der Sippenbnde,53 aber auch vor allem Hochschule [] germanischer Jungman-
nen- und Kriegererziehung.54 Die Qualifikation des Knaben geschieht bereits innerhalb der
Sippe; Vagn zeigt frh seine Unbndigkeit,55 und auch die brutalen Beweise kriegerischer
Gesinnung und Tchtigkeit, die Svend seinem Vater lieferte, zwangen diesen also anzuerken-
nen, da Svend von rechter Kriegerart sei.56 Die Todesverachtung der Jomswikinger, die als
Ausdruck des gefolgschaftlichen Zusammenhaltes im Geiste der gemeinsamen Idee so betont
wird, bedingt nicht nur die Hingabe zum Fhrer, sondern auch die entsprechende Haltung des
Fhrers selbst eine Widmung eingedenk des Hitlerjungen Quex und Horst Wessels an jeden
Pimpf, der schon Gefolgschaftsfhrer [ist] oder es gern werden [mchte]57 und die Einrah-
mung in einschlgige Hitler-Zitate58 schlielich ist die Kulmination der Instrumentalisierung
der J. s.59

Nachbemerkung
Die Botschaft, welche die ideologische Literatur anhand der J. s. zu vermitteln sucht, ist nicht
nur germanisches Heldentum schlechthin, das sich ebenso durch die slendingasgur illu-
strieren lt, sondern auch ein bestimmtes Verstndnis von Gefolgschaft und Fhrer: Zu-

47
Naumann 1939, S. 88.
48
Nach heldenmtiger Verteidigung unterlagen schlielich auch Wagn und seine Mannschaft dem starken
Gegner. Alle anderen Gefangenen sollten hingerichtet werden. Wagn allein durfte frei abziehen. Doch der Wi-
kinger lehnte das Angebot ab: er wolle sich das Leben nur schenken lassen, wenn auch seine Mannen befreit
wrden. (Vogel 1942, S. 149.) Fahnemann [1937] lt alle vor Vagn, so unausgesprochen auch Sveinn Bason,
hingerichtet werden (vgl. S. 14, ebenso Wikingerfahrten [1937], S. 11).
49
Lawrenz [1938?] interpoliert die Episoden von Apfelschu und Skilaufen nach Saxo, vgl. S. 56.
50
Naumann 1939, S. 112, nmlich dem Deutschen Orden (vgl. Strasser 1943, S. 86).
51
Ott 1940, S. 120.
52
Wir mgen oft und im vlkischen Sinne durchaus mit Recht die frhe Lebensverschwendung beklagen,
die dem Mann hier bevorsteht, aber wir mssen uns sagen: zeugt er nicht leiblich, weil er zu frh stirbt, so zeugt
er doch hundertfltig im Geiste. (Naumann 1939, S. 30.) Der Normalfall ist eine permanente Fluktuation von
Jnglingen aus den Sippenverbnden, die als stahlharte, wetterfeste Mnner wieder aus der Kameradschaft
ausscheiden (Vogel 1942, S. 61), wie sie durch Vagns Verbindung mit Ingibjrg illustriert wird.
53
Als artgeme Zuchtform, die sich ein wehrhaftes Freibauerntum geschaffen hat (Wllenweber 1938, S. 5).
54
Wllenweber 1938, S. 6. Eine vornehme Abkunft sicherte unter anderem die rassische Auslese als Vorbe-
dingung jeder Leistung (S. 19).
55
Vgl. Wllenweber 1937, S. 10, und Wllenweber 1935, S. 119.
56
Wllenweber 1937, S. 129; Plna-Tkis Hintergedanken werden so ignoriert.
57
Wagenfhr 1935, S. 9.
58
Die Welt ist nicht da fr feige Vlker. Die Ehre ist genau so wenig ein leerer Wahn, wie es die Treue ist.
[] (Dauch 1940, S. 164 und 180.)
59
Die Anthologie Meyn 1938, die Vagns Aufnahme wiedergibt, ist in Zusammenarbeit mit der HJ herausgege-
ben.

425
nchst ist es ein elitistisches Bild einer Parallelgesellschaft zur Welt der Familien, erreichbar
nur durch besondere Qualifikation und geprgt durch stndigen inneren Legitimierungsdruck.
Hierzu tritt, bei aller Unterordnung, jedoch Gegenseitigkeit: Gefolgsleute wie Fhrer haben
freinander einzustehen. Die Mehrzahl der hier untersuchten Titel wurde zwischen 1934 und
1939, insbesondere 1937, (erstmals) herausgegeben jenen Jahren, in denen der Fhrerstaat
mitsamt seinen Organisationen sich fr viele zu bewhren schien.

Bibliographie
Textausgaben der Jmsvkinga saga
1824. Jmsvkinga saga tgefin eptir gamalli klfskinnsbk hinu konngliga bkasafni Stockhlmi.
[Transkription: Rasmus Christian Rask.] Kaupmannahfn. (Fornmanna sgur.)
1875. Jmsvkinga saga efter skinnboken 7, 4to Kungl. Biblioteket i Stockholm. Utg. af Gustav
Cederschild. Lund. (Lunds universitets rsskrift, Afdelningen fr philosophi, sprkvetenskap och
historia 11 [1874].)
1877. Jmsvkinga-saga i latinsk Oversttelse af Arngrim Jonsson. Udg. af [Gustav] A[ntonio] Gjes-
sing. Kristianssand.
1879. Jmsvkinga saga (efter Cod. AM. 510, 4:to) samt Jmsvkinga drpa. Utg. af Carl af Petersens.
Lund.
1882. Jmsvkinga saga efter Arnamagnanska handskriften N:o 291. 4:to i diplomatariskt aftryck.
Utg. af Carl af Petersens. Kbenhavn. (SUGNL 7.)

Deutschsprachige bersetzungen
Baetke, Walter (bertragen von []), 1924: Die Geschichte von den Seekriegern auf Jomsburg. In:
Die Geschichten von den Orkaden, Dnemark und der Jomsburg. 1.3. Tsd. Jena. (Thule 19.)
S. 392436. <SKB 7, 4to>
Giesebrecht, [Heinrich] Ludwig [Theodor] (Aus dem Islndischen von []), 1827: Geschichte der
Freibeuter von Jom. In: Neue Pommersche Provinzialbltter 1. S. 90139. <SKB 7, 4to>
Khull, Ferdinand (Erzhlt von []), 1892: Die Geschichte Palnatokis und der Jomsburger nach der
jngsten altnordischen Bearbeitung. Graz. (Separatabdruck aus den Jahresberichten des k. k. zwei-
ten Staats-Gymnasiums pro 1891 und 1892.) <AM 510, 4to>
Wollheim, A[nton] E[dmund] (Herausgegeben von []), 1875: Die Jmsvknger [sic!] Saga. In: Die
National-Literatur der Skandinavier. Eine prosaische und poetische Anthologie aus den besten nor-
dischen Schriftstellern mit erluternden, kritischen und biographischen Notizen. 1: Die altnordische
Literatur. Berlin. (Die Classiker aller Zeiten und Nationen 5; 1. Abt.) S. 154159. <SKB 7, 4to>

Deutschsprachige Nacherzhlungen, Be- und Verarbeitungen60

Ball, Kurt Herwarth, [1941]/ 1936: Die Jomsburgwikinger. Der Geschichte und den alten Sagen
nacherzhlt. 4.8. Tsd. Karlsbad-DrahowitzLeipzig. (Volksdeutsche Reihe 4.) <B>
Bohne, Gerhard (Fr den Unterricht ausgewhlt und erlutert von []), 1937: Zeugnisse altnordischen
Glaubens. Leipzig. <B>
Busch, Fritz Otto & Docter, Heinz, 1935: Germanische Seefahrt. 1.5. Tsd. Berlin. <B>
Busch, Fritz Otto & Ramlow, Gerhard, 1940: Deutsche Seekriegsgeschichte. Fahrten und Taten in
zwei Jahrtausenden. 1.10. Tsd. Gtersloh. <B & Busch/Docter 1935>
Dauch, Bruno (Bearbeitet von []), 1940/[ 1939]: Von deutscher Art. Ein Lesebuch fr hhere
Schulen. 3. Teil. 2. Aufl. Frankfurt am Main. <B>
Fahnemann, Franz, [1937]: Die Wikinger von der Jomsburg. Die Saga vom Heldenkampf einer ger-
manischen Gefolgschaft. Saarbrcken (et al.). (Erlebte deutsche Welt. Arbeitsstoffe und Erzhlun-
gen fr die deutsche Schule 18.) <B>
Fischer, Hans W[aldemar], 1934: Gtter und Helden. Germanisch-deutscher Sagenschatz aus einem
Jahrtausend. Berlin. <B>
Genzmer, Felix, 1944: Germanische Seefahrt und Seegeltung. Mnchen. <B>

60
Erschlossene Ausgabejahre nach der Deutschen Nationalbibliographie. <B/G/K> kennzeichnet die berset-
zungsquelle, soweit feststellbar.

426
Heider, Werner, [1937]: Die Helden der Jomsburg. Eine Erzhlung aus der Zeit der Wikinger. Reut-
lingen. <?>
Heims, P[aul] G[erhard], 1904/[ 1903]: Auf blauem Wasser. Ein Buch von der See fr die deutsche
Jugend. 2. verm. Aufl. Braunschweig. <?>
Hersen, Elisabeth (Nordischen Sagen nacherzhlt von []), [1923]: Die Wikinger von Jomsburg.
Zeitbild aus dem zehnten Jahrhundert. Berlin. <?>
Kath, Lydia, [1941]/ 1934: Jomsburg. Eine Wikingergeschichte. 33.38. Tsd. Berlin. (Trommler-
buch.) <B & K>
Lawrenz, Hans (Nach der bersetzung Giesebrechts aus dem Islndischen fr die Jugend bearbeitet
von []), [1938?/ 1935]: Die Jomswikinger-Helden von Wollin. 2. Aufl. Breslau. (Schriften zu
Deutschlands Erneuerung 64.) Eingebunden in: Die Edda und drei islndische Sagas. Eine Aus-
wahl fr den Unterrichtsgebrauch. [1938.] BreslauBerlin. (Schriften zu Deutschlands Erneuerung.
Sachleseheft 2.) <G & Vedel Simonsen 1827>
Meyn, Ludwig, 1938: Altgermanisches Leben: [3.] Kampf und Kurzweil. Frankfurt am Main. (Natio-
nalpolitische Sammlung [338].) <B>
Naumann, Hans, 1939: Germanisches Gefolgschaftswesen. Leipzig. (Meyers kleine Handbcher 22.)
<K>
Neckel, Gustav, 1915: Germanisches Heldentum. Quellensammlung altgermanischer Lebenszeugnis-
se. 13.17. Tsd. Jena. (Tat-Bcher fr Feldpost [7].) <K>
, 1940/ 1934: Germanisches Heldentum. 21.25. Tsd. Jena. (Deutsche Reihe 21.) <K>
Ott, Stefan, 1940: Die altnordische Dichtung in der Schule. Einfhrung und Mglichkeiten der schuli-
schen Behandlung. Elingen a. N. (Schriftenreihe der Hochschule fr Lehrerbildung Elingen
a. N., Schulpraktische Reihe 5.) <B>
Plamann, J[oseph] O[tto] (Dargestellt von []), 1929: Wikingerfahrten und Normannenreiche. 1.
10. Tsd. Jena. (Deutsche Volkheit [67].) <B>
Ramlow, Gerhard (Nach altnordischen Texten von []), 1936: Mnner des Nordens. Wikinger Le-
ben, Fahrten und Kmpfe. Berlin. <B>
Strasser, Karl Theodor, 1938/ 1933: Die Nordgermanen. 3. Aufl. Hamburg. <B>
, 1943/ 1928: Wikinger und Normannen. 5. Aufl. Hamburg. <B>
Vogel, Paul (Bearbeitet von []), 1942/[(c) 1939]: Die Germanen. 3. Aufl. Frankfurt am Main. (Volk
und Fhrer. Deutsche Geschichte fr Schulen; Ausgabe fr Oberschulen und Gymnasien, Kl. 2.)
<K>
Wagenfhr, Horst (Herausgegeben von []), 1935: Gefolgschaft. Der germanische Kampfbund.
Hamburg. <B & G>
Wikingerfahrten, [1937]. Paderborn. (Schninghs Arbeitsbogen, Germanentum A 80.) <B & Fahne-
mann [1937]?>
Wllenweber, Fritz, 1935: Altgermanische Erziehung. Dargestellt auf Grund der Islandsagas und an-
derer Quellen zum Frhgermanentum. Hamburg. <B>
(Nach den Quellen bearbeitet von []), 1937: Germanische Jungmannschaftszucht. 1: Die Sippen-
gefolgschaft. Hamburg. <B>
(Nach den Quellen bearbeitet von []), 1938: Germanische Jungmannschaftszucht. 2: Der Wiking-
bund. Hamburg. <B>

Zeitgenssische Literatur
Giesebrecht, [Heinrich] Ludwig [Theodor], 1840: Wendische Geschichten aus der Zeit der ersten Lu-
dolfinger. In: Baltische Studien 7.1. S. 1110.
, 1843: Wendische Geschichten aus den Jahren 780 bis 1182, 1. Berlin.
Jantzen, H[ermann], 1928: Thule. Altnordische Dichtung und Prosa. 2. Reihe, Bd. 18 bis 20. [].
[Rezension.] In: Die Literatur, 30. S. 486.
Klempin, Robert, 1847: Die Lage der Jomsburg. In: Baltische Studien, 13.1 S. 1107.
Vedel Simonsen, Lauritz Schebye, 1827: Vedel Simonsens geschichtliche Untersuchung ber Joms-
burg im Wendenlande. Aus dem Dnischen von Ludwig Giesebrecht. Stettin.

Forschungsliteratur
Bruns, Karin, 2001: Zeit der Frauen? Vlkisch-nationale Konfigurationen des Geschlechterverhltnis-
ses in Politik, Literatur und Film. In: Vlkische Religion und Krisen der Moderne. Entwrfe artei-

427
 gener Glaubenssysteme seit der Jahrhundertwende. Hrsg. v. St. von Schnurbein & J. H. Ulbricht.
Wrzburg. S. 7598.
Fix, Hans, 2004: Die Anfnge der Altnordischen Saga-Bibliothek. In: Verschrnkung der Kulturen.
Der Sprach- und Literaturaustausch zwischen Skandinavien und den deutschsprachigen Lndern.
Zum 65. Geburtstag von Hans-Peter Naumann. Hrsg. v. O. Bandle, J. Glauser & St. Wrth. Tbin-
genBasel. (Beitrge zur Nordischen Philologie 37.) S. 305330.
Hermand, Jost, 1995: Der alte Traum vom neuen Reich. Vlkische Utopien und Nationalsozialismus.
2. Aufl. Frankfurt am Main. (Edition Zeitgeschichte.)
Zernack, Julia, 1994: Geschichten aus Thule. slendingasgur in bersetzungen deutscher Germanis-
ten. Berlin. (Berliner Beitrge zur Skandinavistik 3.)
, 1997: Bibliographie der deutschsprachigen Sagabersetzungen 17911995. Berlin. (Berliner Bei-
trge zur Skandinavistik 4.)

428
 The Gosforth Fishing-Stone and Hymiskvia: An Example of
Inter-Communicability between the Old English and Old
Norse Speakers
Tsukusu It, Faculty of Arts, Shinshu university, Japan

Introduction
When the scholars try visualizing the flag-ship monument of the Anglo-Scandinavian society
in Danelaw during the Viking age, they would select either Hogbacks in Brompton, Yorkshire
or the slimly designed cross in Gosforth, Cumbria. How much the difference of the two areas
of production affected their artistic variations in the whole Danelaw artistry is not this papers
topic. How much the remains of each area can tell us about their days of Anglo-Scandinavian
amalgamation is what this paper attempts to reveal. I should like to use the Fishing-Stone in
Gosforth as a specimen in which the mutual communication between the Old English [hence
OE] and the Old Norse [hence ON] speakers could be found through the comparing the picto-
rial narrative in Danelaw with the poetic narrative preserved in Iceland, i.e. Hymiskvia.
For whom was the Fishing Stone in Gosforth made in the first place? Was it for an indi-
vidual converted pagan, or for the laymen as a whole in the parish? That the Fishing-Stone is
not an individual monument is proposed by Richard N. Bailey in his exhaustive study on
sculpture in Viking Age Danelaw. In his examination, Bailey considers that [t]he dimensions
of this carving suggest that it was not part of a cross or slab but originally formed a section of
an architectural frieze or wall-panel (Bailey 1980: 131). According to Baileys interpretation,
the stone consists of the upper and lower panels both of which symbolize the Christian strug-
gle with the devilish enemy. The upper panel shows a hart keeping a serpent under control by
blowing water through its nostrils at it, if it accords with Plinys History. The iconographic
significance of the upper panel harmonizes the fishing scene on the lower panel, in which an
angler tries to catch a huge snake, namely Leviathan (JobXLI,1). Bailey concludes that the
Gosforth church of the tenth century may have been decorated with a continuous line of such
narrative sculpture of both Christian and pagan symbols in concord which enlightened the
people in the parish with Christian doctrine (Bailey 1980: 13132).
We may also assume that the Gosforth parish church belonged to the English ecclesiastical
organisation. It is the place-name Gosforth, which is unequivocally an English name, that
informs of the English characteristic of the parish. Rollarson, though with some rather dispa-
rate evidence, attempts to prove that the churches in Cumbria had long held the connection
with the Northumbrian church since the eighth century: it would appear likely that all the
lands west of the Pennines were also the part of Northumbria and that the western frontier
extended to the coast (Rollarson 2003: 28). We may well then be tempted to think between
the lines that it is likely that the laymen in the Gosforth parish were OE speakers before the
time of the Viking settlement.
With regard to the linguistic question, Fellows-Jensen finds that Gosforth is in the area
where Scandinavian inflexions survive in place-names and which are most remote from, or
inaccessible to, the influence of Standard English (Fellows-Jensen 1985, 412). It is very
likely, therefore, that the Viking settlement affected the speech of the inhabitants of Gosforth
area. The Fishing-Stone must have been appreciated by the parish laymen who spoke either
OE or ON, or both, or at least an OE dialect peculiar to this region.
As Bailey concludes, the picture on the Fishing-Stone can be interpreted in both pagan and
Christian ways: rs fishing expedition to catch the migarsormr, and the hart of the sym-

429
bol of Christ fighting against the devilish serpent (Bailey 1980: 13132). It is true that the
parish folk could interpret the picture in Christian ways since the church with the Northum-
brian connection had taught the scripture, its anecdotes and symbolism to them. How were
they, then, disposed to Scandinavian mythology unless they heard the narrative on the pagan
god rr in expedition with Hymir the Giant? Today, we know the story thanks to Snorris
prose version retelling the whole story in detail. Yet, the people in Cumbria may not, or can-
not, have been able to read Snorris version for two reasons: Firstly, Snorri was not yet born
at the time, and secondly, Snorris version is not exactly the same as the one on the Fishing
Stone in Gosforth.
Two pieces of sculpture that can be interpreted as telling the story of rs expedition to
catch migarsormr are left in Sweden and Denmark: The former in Altuna and the latter
Thy. Significantly enough, in both pictures, r is depicted as the one whose foot being put
out below the boats floor. In Snorris version, on the other hand, r is said to have forced
both of his feet against the boat and kicked the bottom (hljp bum ftum ggnum skipit ok
spyrndi vi gurunni) (Falkes 1988: 45)1. Despite the difference in number of rs feet, both
the Scandinavian picture stones show that rs foot crushes through the bottom of the boat,
and, in this respect, accord with Snorris narrative.
In the Fishing-Stone in Gosforth, however, there is no trace of rs foot under the boat,
which suggests that the narrative version of which the Gosforth sculptor made use in his
working was different from those for the Altuna and Thy sculptors. In fact, we have only an-
other piece of narrative that tells the story of rs attempt to catch the migarsormr:
Hymiskvia in so-called Elder Edda.2 In this version r is not told to have pushed either his
foot or feet through the bottom of the boat. In this sense, it is quite feasible that the Gosforth
sculptor heard a narrative about rs expedition in the same group as that of Hymisqvia.
Whether the parish folk in Gosforth could not only hear Hymiskvia but also understand the
contents of the verse is wholly hypothetical. Even if the examination of this possibility does
not aspire to proof of this hypothesis, it should suggest intercommunicability between the
speakers of OE and ON during the amalgamation process in the Viking Age England.

rs Hunt in Hymiskvia
Terry Gunnell, in his introduction to Eddic Poetry, draws our attention not merely to the con-
tent and the metre of the poetic works but to the way of performance in recital of the Eddic
Poems / Songs / Chants / Dramas (Gunnell 2005: 9597). The metre of Hymiskvia is
fornyrislag and the content is how r brings Hymirs cauldron to brew mead for sir. This
song consists three parts: The first part tells the story of Tr coming home with r to see his
mother, the second part deals with the contest between Hymir and r, and the third part de-
scribes how r breaks Hymirs treasury cup bringing the cauldron after killing all the giants,
the last of which ends with the narrators enquiry for another narrative story about r and his
servants, jalfi and Rskva, and summoning the beginning of the brewery feast. The whole
structure is, however, loose and not well fit. The extent form of the poem may not be the
original. Furthermore, I should suggest that the part of rs hunting with Hymir must have
been engrafted in later version. We can never know whether the whole story of Hymiskvia

1
While Falkes interprets the bottom as that of the sea, the picture stones in Altuna and Thy do not seem to indi-
cate that rs foot or feet reach(es) the sea-bed. I, therefore, would interpret that the sculptors of either stones, if
they heard the Snorris version of the narrative, might have thought of the grunni as the ship-bottom. Still it
needs further study.
2
Terminus ante quem is suggested in Kommentar zu den Liedern der Edda, 277, while some assumption of later
composition based on the fact that Snorri does not make use of Hymiskvia is mentioned in KL, s.v. Gud-
ediktning.

430
was told in Scandinavianized Cumbria, but at least the story of rs rowing the boat with
Hymir to catch migarsormr is known in sundry lands of Scandinavian culture.
The examination of the possibility that the OE speakers in Cumbria could understand the
meaning of Hymiskvia if they heard it allows us to enframe only the episode of rs
launching the boat with Hymir, because the rest of the song is not surely known to the English
people during the Viking period.

Old English version of Hymiskvia

The following hypothetical stanzas are the part of Hymiskvia which the OE speakers in
Cumbria might feasibly have heard and understood. The possibility of mutual oral communi-
cation between the OE and ON speakers in Danelaw is persuasively argued by Matthew
Townend in his monograph. I owe much to his theory in constructing my hypothesis in trans-
literating the ON text into OE.
The basis on which my transliteration from ON to OE was made is the theories of switch-
ing-code and dialect congruity (Townend 2002: 4346). As is often stated, the lexical sub-
stitution of cognate (Townend 2002: 108) is the very key to open the ears of both the OE and
ON speakers. The text, and context, of this papers material is what the laypeople in Cumbria,
where the parish churches had been secured for hundreds of years, would have understood
when they first heard the song about rs attempt of fishing migarsormr. When newcom-
ers came to their parish, they accepted them if the strangers wanted to join their community.
The linguistic medium between the two peoples, thanks to the dialect congruity, must have
been equipped with inherent learnability (Townend 2002: 4546). Both languages have
cognate vocabularies as well as systematic sound correspondences. Nevertheless, in my ar-
gument, I will show some of the elements that may have caused slight misunderstandings and
may have made either side of the speakers of the two languages aware of sets of pun causing
ironical allusions in communication.
The episode of rs expedition starts from the 15th stanza and ends at the 27th. The 36th
stanza has an interesting kenning for giants, and it might have caused a certain effect if the
song had been sung among the Anglo-Saxons, though the limitation of the space in this paper
allows me to examine the story only until the 25th.

Hypothetical Hymiskvia Heard by the Ears of the OE speakers

OE version ON version
15.
hwa leton hie hvern lto eir
heofude scortran hfi skemra
ac on seae ok seyi
sian bron. san bro.
t Sife wer, t Sifiar verr
dre swefan geng r sofa gengi.
an mid lle einn me llo
oxum twm Hymes xn tv Hymis

16.
ohte harum Hrungnes spillere tti hrom Hrungnis spialla
weorung [wrau]Hlowriddan uerr Hlrria
well fullmicel: vel fullmikill:
Munon t apan orum we weoran Munom at apni rom uera
wi wamt we rie libban vi veiimat vr rr lifa.

17.

431
Wih-weorung cw sic willan wg rowan Vorr kvazk vilja vg roa,
gif beald eoten bte gefe. ef ballr jtonn beitor gfe.
[]
[]

18.
Hymir cw: Hymir kva:
Hweorf to hiord, gif hyge trywe Hverf til hjarar, ef hug truer,
breotere / brytere beorg-Dana, bte gesecan! brjtr bergdana, beitor skja!
eos wene ic at e mynege ess vnter mik at er myne
egenu of oxan ead-fong wre gn af uxa aufeng vesa.

19.
Sven seoslig swaf to sceagan Sveinn sslega sveif til skgar,
r se oxa allsweart forestod3 ars uxe st alsvartr fyrer:
ofbreat *dior [ir?] yrses rdbana braut af jre urs rbane
heah-tun ufan horna twegra. heah-tun ufan horna twegra.

20.
Hymir cw: Hymir kva:
werk ynca in wirs micle, Verk ykkja n verre miklo,
ceolwalda, anne *cusc sitst. kjla valde, an kyrr siter.

21.
Bd hlynn-gat hfera drihten Ba hlunngota hafra drttenn
*tren apan ut-faran trunn apa tar fra;
ond se eoten self talde en s jtonn sna tale
lytel fus lengra rowan. ltla fse lengra at roa.

22.
Drog mre Hymer modig hwalas Dr mrr Hymer mogr hvale
an on angle up sona twegen einn ngle upp senn tvaa;
and fter in *skut Wodene sibling en aptr skut ne sifjar
Wih-ward wi wilas wae gierwede sealf. Vorr vi vlar va gre sr.

23.
Agnade on angel se e ealdrum beorgare Egnde ngol ss ldom bergr
wurmes anbana oxan heafode; orms einbane uxa hfe:
gan wi angel seo e is goda fiond gein vi ngli ss go fi
ymbgyrd benian allra landa. ubgjr nean allra landa.

24.
Drog deorlice ddranc urr Dr djarflega drakr rr
wurm attorfag up at borde orm eitrfaan upp at bore;
hamere cnuwode heah-fell *skare hamre kne hfjall skarar
ymblyt ufan wolfes hnit-broor. ofljtt ofan ulfs hnitbror.

25.
Hran-galle hlemmade, and holc uton. Hreinglkn hlumo, en hlkn uto,
for seo fyrn folde allu samen: fr en forna fold ll saman:
[] []
sanc sealf sian se fisc in mere. skesk san s fiskr mar.

In this transliteration, the correspondence between the vowel systems of OE and ON is not
always consistent. In fact in the last line of the 25th stanza, we can see how ON weak verb

3
As to the treatment of ON phrasal verbs with adverbial as transliteration into affixation, see Ito 1993.

432
skkva in middle voice could have been understood by the OE speakers. There need to be two
syllables in the first half of the long line in order to switch the code of ON middle voice into
an OE reflexive phrase. The problem is not only limited to the problem of the number of syl-
lables, but a set of phonemically hypothetical vowels for OE transliteration cannot always be
discovered by a simple search for cognate words in the phonological set of vowels in ON,
even though the importance of the existence of cognate words in both languages is crucial.
Nevertheless, at length hearing process could allow the audience to understand the narrative
story that the singer/reciter chanted. It is not unthinkable to assume the existence of those
people who could provide some comments on the text and taught some of the younger genera-
tions how to understand certain specific poetic devices such as kennings and alliteration dur-
ing the recital.
The following section will examine how the mutual intelligibility may have occurred
through this hypothetical transliteration.

Commentary
In the 15th stanza, we can easily find the different 3rd person plural pronouns in the two ver-
sions. Interestingly enough, the OE speakers eventually invited ON pronouns in the course of
time. It is a natural process, for but for the pronoun, the first line may show how similar the
two languages had been even before the amalgamation. Note the nasal sounds appear differ-
ently in each version. In OE it is the 3rd person plural preterite inflectional suffix of ltan that
has a nasal sound, while in ON it is the indefinite pronominal adjective indicating accusative.
We can see why much the English language eventually lost their inflectional suffixes.
line 6, dre in OE is the cognate of ON r, but the meaning of the OE word is adverbial
immediately, at once. In this context, however, either will fit, and the whole story will not be
affected by the differences in meaning of both words, i. e. their mutual communication may
have been thought established. Actually, in North dialect of OE, infinitive suffix n had dis-
appeared when the Scandinavian settlers started living in Danelaw. So the correspondent line
would be more like dre swefa geng and it would sound more like the ON original.
In the 16th stanza, the cognate pair spillere and spialla may have been understood as al-
most the same meaning as interlocutor, though ON spialla may allude more intimate friend-
ship. An example of more provoking interpretation is on the second line. ON verr food has
its cognate weor in OE, but the latter had become different in meaning, worth. OE
weorung celebration, feast must have been more similar in meaning, but it would sound too
much differently. OE wrau maintenance, support does not have long syllable, and it would
not discord the tone as well as the intent of the poem.
There is one problem in the 16th stanza. It contains an ON idiomatic phrase, vera + in-
finitive. OE does not seem to have such an expression so far. We need further investigation
upon it.
In the 17th stanza, as for the first word of ON version, Kommentar provides with several
etymological hypotheses, most of which elucidates, regardless the different nuances among
the assumed etymologies, that the meaning of the word would be the protector of the sanctu-
ary (Kommentar: 30102). It has its cognate word in OE: wih. Yet, in the Christian context
for the Anglo-Saxons, the first element of the word wih is an idol (Bosworth-Toller), and
the quick inference from it would have been a transliteration into wih-weorung bearing a
meaning of honour to idol, while in the pagan context it would have borne the meaning of
glory of holiness, which is close in meaning to ON Vorr. Nevertheless, the ON Vorr
might have sounded like OE Weorr, which in turn would have signified as an adjective bad,
grievous (Bosworth-Toller). If a West-Saxon had heard this phrase, they would unequivo-
cally have decided the new comers god as of evil nature.

433
 The 18th stanza shows an interesting example of showing a prejudice of the locals towards
the Danish people (Kommentar: 310). r is here regarded as the destroyer of the Mountain
Danes, which must signify the giants. It must be either the Norwegian poet / recitalist, as is
said in Kommentar, or the English audience / interlocutor that composed this phrase. At least,
when the people in Cumbria heard this phrase, they would not have been displeased if they
had long been living with Norwegian settlers among them. Fellows-Jensen asserts that the
Danish settlements into the west of the Pennines started at a comparatively early date, yet the
settlement of western and central Cumberland seems likely to have been basically the work of
Norwegians (Fellow-Jensen 1985: 412). The fact that the extent text of Hymiskvia has a
Norwegian taste does not discord with my hypothesis that the Gosforth parish-folk might
have heard a narrative of rs expedition very similar to the Hymiskvia version.
In the 18th stanza, I should put another comment on the usage of impersonal phrase using
ON vntar mik. No equivalent phrase can be found even though OE has several impersonal
phrases. I substitute it with wenan in the personal phrase. I also substitute ON munu with OE
mynegian, even though the cognate of munu in OE is munan, for the difference in meaning is
too large. On the other hand, ON infinitive vesa has its correspondence with OE subjunctive
wre, because of its similarity in phonetic sound as well as its fitness in the subordinate
clause with mynegian. Furthermore, ON beitor, pl. would be transliterated into OE bt, a
feminine word, though it holds the meaning of what can be bitten, food. In this context, its
meaning can be either a lunch for r on the sea, or the bait for fishing. After all, the modern
English word bait derives from a Scandinavian loan word: we could simply borrow this ON
word for it. It is not considerable that in this point of the story Hymir would expect that an
oxs head should be a bait for rs fishing.
In the 19th stanza, OE seoslig has meanings of afflicted, vexed, not being the same as of
ON ssliga busily, smartly, speedily. Still it may be appropriate in the context, and the OE
speakers might have understood in their ways.
Another crux is ON jrr. In OE we find no cognate of it, and maybe OE dior / deor ani-
mal will be a sufficient substitution. A Scottish word tirr is an ambiguous word with regard
to its etymology, and I presume it may have derived from ON jrr. According to The Scot-
tish National Dictionary, Tirr v3, n3, adj. has its meanings as:

1. v. To snarl, to speak in an irritable, bad-tempered way.

II. 1. A passion, a fit of bad temper or rage, an excited angry condition; a quarrel. 2. An irritable,
quarrelsome child.

III. adj. Bad-tempered, quarrelsome, of a passionate, irritable nature [] . [Orig. somewhat un-
certain, poss. different words having fallen together. For I. Sc. forms cf. Norw., Dan. tirre, to
tease, irritate, tirren, terren, peevish, fractious [] .]

The characteristics of jrr a bull can be deduced as Scottish tirr, though without more solid
evidence.
The 20th stanza is a syntactically debatable example, showing the genitive pronoun stylis-
tically separated from its object (Kommentar: 315). Whether this word order can be under-
stood in OE context is to wait for further argument. There is a kenning of r which coinci-
dently the same as Ceolwald, son of Cuthwulf, an ancestor of West-Saxon kings.
Another crux exists in the 21st stanza. ON hlunngota would sound like OE hlynn-gat. ON
hlunnr + goti Rolling + Horse is a famous kenning. Yet in OE gat would have sounded to
mean goat or she-goat, while hfera drihten would have almost the same meaning as that
in ON. Whether the OE speakers would have misunderstood or not is a question, and, fur-

434
thermore, how they misunderstood is another one. The lord of he-goat requests a rolling-
she-goat to go out trunn of a monkey does not make sense, or does it? ON trunn does not
have an exact equivalent in OE, but it would sound like OE tren poisonous.
The 22nd stanza is the most crucial one. It tells us that Hmir drew two whales. In the
Fishing-Stone, there seems to be three huge fish under the boat, and the two of them are try-
ing to bite the bait of the oxs head. The sculptor of the stone must have known the number of
the whales that should be caught by one of the sailors on the boat he curved.

The ON adverb senn is transliterated into OE sona, only the consonants of which correspond
to the original. Nevertheless, the switching-code can be applied when the mutual communi-
cation progresses.
Another example of such code-switching function can be observed in the first line of the
24th stanza. ON djarflega would have its etymological cognate in OE as earflice. However,
the meaning of the latter, necessarily, usefully, profitablly, is again so different from that of
ON, and it would be out of context if the OE speakers heard that way. OE deorlice would be
more appropriate in both meanings and phonetic correspondence. Those people might have
used their linguistic instinct not always systematically but in more pragmatic ways.
ON skrr in the third line of the stanza does not seem to have its cognate in English. Be-
sides, it is a part of a kenning a high-mountain of hair = head, and the audience must know
the expression. Unfortunately, the English people have not appeared to borrow the kenning
phrase, but the word itself seems to be imported into Cumbrian and Scottish dialects, though
the meaning of cliff, ridge; a bare place on the side of a steep hill has been rather prevailing
(Jamieson: Scar, Skair, Scaur). We cannot tell whether the OE speakers could understand
the meaning of this kenning, but at any rate, the story is easy to follow.
The last line of the 24th stanza includes a difficult word ymblyt. Clark Hall defines its
meanings as circle, circuit, circumference, but Bosworth-Toller gives up its definition. In
my transliteration, it is based on its etymological correspondences and the meaning of the ON

435
ofljtt has been defined as grimmig, graus by Sijmons and Gehring, stygg by Heggstad,
very hideous by Cleasby-Vigfsson, and translated as widerwrtig by Kommentar. How-
ever, the content of this passage refers to rs attempt to hit the head of migarsormr with
his hammer, but he actually missed, and the serpent sank into the sea rather than floated in
death. I would rather take the word literally too shiny for its venomous gloss (eitrfaan, l.2) to
hit the very head of the ormr, and my OE transliteration may convey that meaning.
The 25th stanza is most difficult and full of cruces. ON glkn has been argued by the
scholars but its etymology is uncertain. We must content with its obscure meaning as a mon-
ster. Interestingly, ON hrein lava would have sounded like OE hran whale in the English
context where no volcanic activities had been found. Intriguing enough, but perhaps merely a
result of stylistic coincidence, the head of the migarsormr on the Fishing-Stone in Gosforth
appears to be rather whale-like, or at least it resembles his fellow monster fish on the panel.

Conclusion
This paper aims to experiment on how much the 21st century scholars could reconstruct the
situation in Gosforth when the OE speakers would see and hear the recital of a song about
Scandinavian pagan god r rowing out to sea in order to catch his rival, migarsormr.
When we see both OE and ON texts side by side, it is obvious that the two languages were in
so similar outlook. When we hear the sound, though only theoretically reconstructed as it may
be, the similarity must be more striking. As Gunnell puts it, if we could see the performance
of the recitalist who, undoubtedly, does not merely sing the poem but also displays the story
with his body movements, then we should have been able to recognize how much the words
could have been corroborated by human performance.
As is stated above, the mutual intelligibility does not work if there is one way of interpreta-
tion between the two languages, no matter how linguistically close they may be. The mutual
communication between two peoples living together with different background requires not
only dialect congruity but sympathetic human congruity, in which a certain aspect of religious
differences can be an obstacle. In the case of Gosforth parish church, the remnants of the fa-
mous cross, the hog-backs and the Fishing-Stone all visualize the collaboration of the two
religions, two peoples and two languages. If this paper can add something in our appreciation
of their collaboration, it also owes it to their efforts in producing their devotional monument
in their home parish.

Bibliography
sgeir Blndal Magnsson. 1995. slensk Orsifjabk. 3rd ed. Reykjavk: Orabk Hsklans.
Barnes, Michael P., 1993: Norse in the British Isles. In: Viking Revaluations. Ed. by A. Falkes & R.
Perkins. London: Viking Society for Northern Research. (Viking Society Centenary Symposium
1992.) Pp. 6584.
Bailey, Richard N. 1980: Viking Age Sculpturein Northern England. London: Collins. (Collins Ar-
chaeology.)
Bosworth, Joseph. Ed. and enlarged by T. Northcote Toller. 1898. An Anglo-Saxon Dictionary. Ox-
ford: OUP, 1898.
Bosworth, Joseph. An Anglo-Saxon Dictionary: Supplement. T. Northcote Toller. Revised and
Enlarged Addenda by Alistair Campbell. 1921. Oxford: OUP.
Campbell, Alister. 1959: Old English Grammar. Oxford: OUP.
Clark Hall, J. R. 1960. A Concise Anglo-Saxon Dictionary. 4th ed. with a Supplement by Herbert D.
Meritt. Cambridge: CUP.
Cleasby, Richard and Gudbrand Vigfusson, with a supplement by William A. Craigie. 1957. An Ice-
landic-English Dictionary. 2nd ed. Oxford: OUP.
Eddadigte II: Gudedigte. Ed. by Jn Helgason. 1965. Oslo, Kbenhavn, Stockholm: Dreyers, Munks-
gaard, Svenska Bokfrlaget.

436
Fellows-Jensen, Gillian. 1985: Scandinavian Settlement Names in the North-West. Copenhagen: C. A.
Reitzels.
Ferguson, Robert. [1873.] The Dialect of Cumberland with a Chapter on Its Place-Names. Carlisle:
Steel Brothers.
Gunnell, Terry. 1995: Eddic Poetry. In: A Companion to Old Norse-Icelandic Literature and Culture.
Ed. by Rory McTurk. Oxford: Blackwell. Pp. 82100.
Heggstad, Leiv, Finn Hdneb and Erik Simensen. 1977. Norrn Ordbok. 4. utgve av Gamalnorsk
Ordok ved Hgstad og Torp. Oslo: Det Norske Samlaget.
Hines, John. 1991: Scandinavian English: A Creole in Context. In: Language Contact in the British
Isles: Proceedings of the Eighth International Symposium on Language Contact in Europe, Doug-
las, Isle of Man, 1988. Ed. by P. Sture Ureland and George Broderick. Tbingen: Max Niemeyer.
Pp. 40327.
Ito, Tsukusu. 1993: Old Norse Idiomatic Phrases in Lay of Havelok the Dane. Geibun Kenkyu 64 (To-
kyo: Keio University) Pp. 1939.
Ito, Tsukusu. 2008. Considering the Differences of Scandinavian Tradition in Danelaw: Stone Sculp-
ture in Cumbria and Other Danelaw Areas. In: The Round Table 22. (Tokyo: Keio University) Pp.
4559.
Jamieson, John. 1882: Etymological Dictionary of the Scottish Language. 4 vols. Paisley: Alexander
Gardner.
KL = Kulturhistoriskt Lexikon.
Kommentar zu den Lieden der Edda. Ed. by Kalus von See & al. 1977. Vol.2. Gtterlieder. Heidel-
berg: C. Winter.
Die Lieder der Edda. 3 vols. Ed. by B. Sijmons and H. Gering. 18881927. Halle (Saale): Buchhand-
lung des Waisenhauses.
Noreen, Adolf, 1923: Altislndische und altnorwegische grammatik (laut- und flexionslehre) unter
bercksichtigung des urnordischen. 4. vollstndig umgearbeitete aufl. Halle (Saale). (Sammlung
kurzer grammatiken germanischer dialekte 4. Altnordische grammatik 1.)
Rollarson, David. 2003: Northumbria, 5001100: Creation and Destruction of a Kingdom. Cambridge:
CUP.
Smundar Edda mit einem Anhang. Ed. F. Detter and R: Heinzel. 1903. 2 vols. Leipzig: Georg Wi-
gand.
The Scottish National Dictionary. Ed. by William Grant and David D. Murison. 1974. 10 vols. Edin-
burgh: The Scottish National Dictionary Association.
Snorri Sturluson. Ed. by Anthony Falkes. 1988. Edda: Prologue and Gylfaginning. London: Viking
Society for Northern Research.
Townend, Matthew. 2000: Viking Age England as a Bilingual Society. In: Cultural in Contact: Scan-
dinavian Settlement in England in the Ninth and Tenth Centuries. Ed. by Dawn m. Hadley and
Julian D. Richards. Turnhout: Brepols.
Townend, Matthew. 2002: Language and History in Viking Age England: Linguistic Relations be-
tween Speakers of Old Norse and Old English. Turnhout: Brepols. (Studies in the Early Middle
Ages 6.)

437
 Aldeigjuborg of the sagas
in the light of archaeological data1
Tatjana N. Jackson, Institute of world history, Russian academy of sciences
Adolf Fririksson terms the method of using sagas and early historical writings in Icelandic
archaeology as literary analogy. He claims that by the application of literary analogy finds
are given function, age and meaning, and notes that the reliability of the conclusion of an
archaeological inquiry is dependent on the current views of the historicity of the sources it
uses (Adolf Fririksson 1994: 14).
In case of Russian studies, sagas are much less reliable as a historical source (Jackson
2005). Old Rus had never been in the focal point for saga authors sagas dealt with different
material and had other interests. Nuggets of information preserved in the sagas still need care-
ful examination and comparison with various source material. Among other things, literary
data have to be coordinated with methods and results of archaeological investigations.
Archaeology might prove the veracity of certain facts mentioned in the sagas. Thus, the issue
presented here puts Adolf Fririkssons premises, so to say, upside down: what will be dis-
cussed below might be termed as archaeological analogy.
Old Norse-Icelandic narrative sources have preserved twelve toponyms that are considered
by medieval authors, as well as by modern publishers and researchers, to have been the names
of Old Russian towns (Jackson 2003). One of them is Aldeigja / Aldeigjuborg. This name is,
as a rule, understood as a designation of Ladoga (Old Ladoga), a settlement in the lower
reaches of the Volkhov River, on the route from the Varangians to the Greeks described by
the Russian Primary Chronicle. Aldeigja / Aldeigjuborg is mentioned about forty times in
skaldic poetry and sagas, while it does not occur in runic inscriptions and geographical trea-
tises. The events described in the sagas and connected with Aldeigjuborg (Ladoga) can be
dated to the Viking age, i.e. up till the middle of the eleventh century, when Haraldr Sigurar-
son left Rus.

I
Going from Novgorod to Sweden, a medieval traveller would naturally sail along the Volkhov
River down to Old Ladoga, then into Lake Ladoga, and therefrom along the Neva River into
the Gulf of Finland. This is a manifest communication route, and that is why its details are
seldom reffered to in the sagas. Still, Aldeigjuborg (Old Ladoga) is sometimes mentioned as a
transitional station, a kind of gateway, on the water route from Scandinavia to Old Rus and
back. The travellers are said to make a halt there and to change their ships.

Um vrit snimma byrja eir fer sna Einarr ambarskelfir ok Klfr rnason ok hfu mikla
sveit manna ok it bezta mannval, er til var rndalgum. eir fru um vrit austr um Kjl til
Jamtalands, til Helsingjalands ok kmu fram Svj, ru ar til skipa, fru um sumarit
austr Garrki, kmu um haustit Aldiigjuborg. Geru eir sendimenn upp til Hlmgars
fund Jarizleifs konungs [] Vru eim gri seld til eirar ferar (Snorri Sturluson.
Heimskringla 2: 414415).

Early in the spring Einar Thambarskelfir and Klf rnason set out with a large company of men
picked from the best in all the Trondheim districts. They proceeded to Jamtaland in spring,
across the Keel, from there to Helsingjaland, and arrived in Sweden. There they procured ships

1
Supported by RFH, grant 07-01-00058.

438
 and in summer sailed east to Gartharki, arriving in fall at Aldeigjuborg. From there they sent
messengers to Hlmgarth and King Jarizleif [] They were given safe-conduct for the journey
(Hollander: 537, with my emendations).

Magns lfsson byrjai fer sna eptir jlin austan af Hlmgari ofan til Aldeigjuborgar. Taka
eir at ba skip sn, er sa leysti um vrit (Snorri Sturluson. Heimskringla 3: 3).

King Magns lfsson started on his journey from the east after Yule, first from Hlmgarth to
Aldeigjuborg. He and his men began to get their ships ready when ice broke up in spring (Hol-
lander: 538, with my emendations).

eir Kalfr dvauluz i Hlmgare ar til er lei iol. Foro ofan til Aldeigioborgar oc fluu ser
ar skipa; foro egar austan er isa leyste um varit (Orkneyinga saga: 57).

Kalf and his men stayed in Hlmgarth till the end of Yule. They went then down to
Aldeigjuborg and got themselves there ships; they sailed from the east as soon as ice broke up in
spring.

Enn at vri byrjai hann (Haraldr Sigurarson. T.J.) fer sna r Hlmgari ok fr um vrit til
Aldeigjuborgar, fekk sr ar skip ok sigldi austan um sumarit, snri fyrst til Svjar ok lagi
til Sigtna (Snorri Sturluson. Heimskringla 3: 91).

In the spring following he journeyed from Hlmgarth to Aldeigjuborg. There he got himself
ships and in summer sailed from the east, turning first to Sweden, and anchored in Sigtna (Hol-
lander: 538, with my emendations).

In her lecture at the Institute of material culture in April 1941, Elena Rydzevskaja, a Russian
scholar studying sagas as a source for Russian history, put forward an assumption proceeding
from the sagas that the vessels sailing in the Baltic and in the Volkhov River should have
been of different types, and that, correspondingly, there should have been craftspeople in
Ladoga occupied in ship repair and ship equipment. She also expressed hope that further ar-
chaeological excavations in Ladoga might bring to light some traces of local crafts, remains of
workshops, etc. (Rydzevskaja 1945). And in fact, as early as in 1958, at the horizon E1 which
dates to 870s890s, there was revealed a complex connected with iron and bronze working
that was thought to have been a smithy. A craftsman working there produced, among other
things, rivets, most likely, to repair northern ships coming to Ladoga (Davidan 1986). More-
over, fragments of ships are found in Ladoga excavations beginning with the earliest layers,
as well as iron boat rivets of the type known from excavations in Scandinavia (Ibidem). Pirjo
Uino is right in stressing (with reference to excavations of 1970s in the Varjazhskaja street in
Ladoga cf. Petrenko 1985) that local boat-building and the repairing of cargo vessels are
indicated by finds of boat and ship parts, which were secondarily used in the structures of the
houses and the wooden streets (Uino 1988: 217). Ship-building, or ship-repairing, was one of
the functions of a Scandinavian manufacturing complex of the mid-eighth century revealed in
197375 (Rjabinin 1980). Thus, we can see that predicting archaeological finds on the basis
of saga data can sometimes bring its fruit.

II
On the other hand, archaeological material is able to verify those saga stories that seem
unlikely at first sight. An example of this kind would be the description of Earl Eirkr Ha-
konarsons attack on Aldeigjuborg. Both Fagrskinna and Heimskringla refer in their narration

439
of this event to Bandadrapa composed by Eylfr daskld, the skald of Earl Eirkr, ca. 1010,
i.e. two centuries earlier than the two compendia, and Snorri even quotes the following stanza:

Oddhrar fr eya
(x hr at at) san
logfgandi (lgis)
land Valdamars brandi;
Aldeigju brauzt, gir
(oss numnask skil) gumna;
s var hildr me hlum
hr, komt austr Gara
(Den Norsk-Islandske Skjaldedigtning. B: 1912).

The author of Fagrskinna does not cite the corresponding strophe, but, like the skald, speaks
about the destruction of Aldeigjuborg (Hann braut Aldeigjuborg Fagrskinna: 165), while
Snorri adds to it the burning down of the entire town (braut og brenndi borgina alla (Snorri
Sturluson. Heimskringla 1: 338339). The latter supplement is thus likely to be Snorris own
invention, though in fact wooden towns were destroyed in the Early Middle Ages through
fires. According to the relative chronology in the saga of lfr Tryggvason in Heimskringla
this enterprise of Earl Eirkr took place in approximately 997. Still, in 1941 Elena Ry-
dzevskaja had to stress in her lecture the absence of any archaeological traces of fire in the
earthwork fortress area in Ladoga that could have been a result of Earl Eirkrs assault of 997
(Rydzevskaja 1945: 55). By the 1980s, however, archaeologists had accumulated certain data
that could back up Snorris narration. Excavations in the Varangian street (on the left bank of
the River Ladozhka) revealed the fact that all constructions of the second main layer (horizon
II), as well as many constructions of the third one (horizon III, both dated dendrochronologi-
cally within the tenth century), bear marked traces of fire destruction (Petrenko 1985: 91, 92,
115). The same data have been obtained in the earthwork fortress area in Ladoga, where a
badly preserved level XI, after 980, displays traces of destruction in the fire that could have
been the result of Earl Eirkrs attack on Ladoga (Machinskij, Machinskaja, Kuzmin 1986;
for the discussion of this new, based on dendrochronology, stratigraphic scale of Old Ladoga,
which includes eleven layers from mid-eighth through the tenth centuries, cf. Kuzmin 2000).

III
Finally, archaeological analogy turns out to be of use in toponymic studies, which might be
exemplified by the examination of the origin of the Russian town-name Ladoga () and
its connection with the Old Norse-Icelandic Aldeigja. The fact that Aldeigja is mentioned in
skaldic poetry (in the Bandadrpa by Eylfr daskld, as mentioned above) points at this
form as the original one for this name. The compound Aldeigjuborg that occurs in the sagas
(the earliest mention is in lfs saga Tryggvasonar by Oddr Snorrason, ca. 1190) was built
with the help of the geographical term borg (town, fortification) that served for the forma-
tion of town-names of Western Europe, and was not typical for the town-names of Old Rus.
The reason for this lies in the fact that the Scandinavians moved along the route from the
Varangians to the Greeks stage by stage. Thus, Ladoga, located at the gateway of this route,
was, according to archaeological materials, opened up by the Scandinavians as early as in the
middle of the eighth century, while at the remaining part of this route their archaeological
traces go back to the second half of the ninth century only. Those Scandinavians who settled
in Ladoga, and who are likely to have constituted there a relatively independent political or-
ganization (Lebedev 1975: 41), created, on the local basis (to be discussed further), the name

440
Aldeigja, and then transformed it into Aldeigjuborg, in accordance with the well familiar to
them toponymic pattern X-borg.
Scholars are unanimous in recognizing the genetic relation of the place-names Aldeigja and
Ladoga. However, their origin and correlation have been interpreted in different ways. The
town-name has been explained as having originated on the basis (i) of the name of Lake
Ladoga (Finnish *aaldokas, aallokas wavy < aalto wave) (Munch 1874: 260; Thomsen
1879: 84; Vasmer 1955: 448), (ii) as well as of the name of the river Ladoga modern Ladoz-
hka (Finnish *Alode-joki < alode, aloe low lands, and jok(k)i river) (Mikkola 1906: 1011;
Brim 1931: 222223; Rydzevskaja 1945: 6465; Rospond 1972: 53; Popov 1981: 5556, 90
91; Neroznak 1983: 101102; Schramm 1982), (iii) and even of the name of the river
Volkhov, or the Lower Volkhov (Finnish Olhava) (Schramm 1986: 369370. Having changed
his mind, Gottfried Schramm did not take into consideration an important toponymic regular-
ity which results in the fact that if a town grows at the mouth of a small river falling into a
bigger one it usually gets its name not from the main river, but from its tributary).
It may be considered practically proved by now that the first to arise was the river name,
then that of the town, and lastly, the name of the lake. The prevalent opinion today is that the
name comes from the Baltic-Finnish languages. Most likely, the original hydronym was the
Finnish *Alode-jogi (joki) Low river. According to A.I. Popov, the succession was as fol-
lows: 1) the Baltic-Finnish, or the Saami, original 2) the Russian transmission of this name
(my italics. T. J.), with further association with a trade center, namely the town of Ladoga,
3) the transition of the name from the town to the lake (Popov 1981: 5556, 9091). The Old
Norse-Icelandic place-name Aldeigja (Aldeigjuborg) is, as a rule, mentioned by scholars as a
parallel to the Old Russian name Ladoga (Ibidem; Neroznak 1983). However, as J.J. Mikkola
was quite right to show, the original combination of sounds could be only al, but not la, since
only the latter could have arisen from the former, but not vice versa (Mikkola 1906). Corre-
spondingly, we may assert that the likely development is from the Finnish river name *Alode-
jogi to the Scandinavian name Aldeigja (first for the river, and then for the settlement), and
only then (with a methathesis ald > lad) to the Old Russian Ladoga.
The origin of the Old Russian name Ladoga not directly from the substrate *Alode-jogi,
but via the intermediate Scandinavian Aldeigja, has to be explained. Not so long ago,
Gottfried Schramm could only put forward an assumption that, if the Slavic name had origi-
nated from the Scandinavian one, the Slavs had to have reached Ladoga some decades later
than the Scandinavians (Schramm 1986: 369). Today we have plausible arguments in support
of this supposition. As the latest analysis of archaeological materials from Ladoga has shown,
the first settlers in Ladoga were in fact the Scandinavians in 750s, while the first Slavs
reached this region not earlier than the 760s (Kuzmin 2000).

Bibliography
Adolf Fririksson, 1994: Sagas and Popular Antiquarianism in Icelandic Archaeology. Avebury.
Brim, V.A., 1931: Put iz varjag v greki: Izvestija Akademii nauk SSSR. Otdelenie obshchestvennyh
nauk. Leningrad. Vol. 2. Pp. 201247.
Davidan, O.I., 1986: Etnokulturnye kontakty Staroj Ladogi VIIIIX vekov: Arheologicheskij sbornik
Gosudarstvennogo Ermitazha. Vol. 27. Pp. 99105.
Den Norsk-Islandske Skjaldedigtning. Ved Finnur Jnsson. 1973. Kbenhavn. B: Rettet tekst. B. I:
8001200.
Fagrskinna Nregs konunga tal. Bjarni Einarsson gaf t. 1985. Reykjavk. (slenzk fornrit 29.)
Jackson, Tatjana N., 2003: The Image of Old Rus in Old Norse Literature (a place-name study): Mid-
delalderforum. Nr. 12. Pp. 2956.
Jackson, Tatjana, 2005: Russian History, Icelandic Sagas and Russian Historiography of the Eight-
eenth to Early Twentieth Centuries: Dialogues with Tradition. Studying the Nordic Saga Heritage.
Ed. by K. Zilmer. Tartu (Nordistica Tartuensia 14.) Pp. 1544.

441
Kuzmin, Sergej, 2000: Ladoga, le premier centre proto-urbain russe. Les Centres proto-urbains russes
entre Scandinavie, Byzance et Orient. d. par M. Kazanski, A. Nercessian et C. Zuckerman. Paris.
(Ralits Byzantines 7.) Pp. 123142.
Lebedev, G.S., 1975: Put iz varjag v greki: Vestnik LGU. Nr. 20. Pp. 3743.
Machinskij, D.A., Machinskaja A.D., Kuzmin, S.L., 1986: Rannie skandinavo-slavjanskie kontakty
po materialam Ladogi VIII vv.: Vsesojuznaja konferentsija po izucheniju Skandinavskih stran
i Finljandii. Moscow. Part 1. Pp. 164166.
Mikkola, J.J., 1906: Ladoga, Laatokka: Journal de la Socit Finno-ougrienne. B. XXIII. Pp. 112.
Munch, P.A., 1874: Om Nordboernes Forbindelser med Rusland og tilgrnsende Lande: Munch, P.A.
Samlede Afhandlinger. Christiania. B. II. S. 184274.
Neroznak, V.P., 1983: Nazvanija drevnerusskih gorodov. Moscow.
Orkneyinga saga. Udg. ved Sigurur Nordal. 191316. Kbenhavn. (Skrifter udgivet af Samfund til
udgivelse af gammel nordisk litteratur 40.)
Petrenko, V.P., 1985: Raskop na Varjazhskoj ulitse (postrojki i planirovka): Srednevekovaja Ladoga.
Novye arheologicheskie otkrytija i issledovanija. Leningrad. Pp. 81116.
Popov, A.I., 1981: Sledy vremen minuvshih. Iz istorii geograficheskih nazvanij Leningradskoj,
Pskovskoj i Novgorodskoj oblastej. Leningrad.
Rjabinin, E.A., 1980: Skandinavskij proizvodstvennyj kompleks VIII veka iz Staroj Ladogi:
Skandinavskij sbornik. Vol. XXV. Tallinn. Pp. 161177.
Rospond, S., 1972: Struktura i stratigrafija drevnerusskih toponimov: Vostochno-slavjanskaja
onomastika. Moscow.
Rydzevskaja, E.A., 1945: Svedenija o Staroj Ladoge v drevnesevernoj literature: Kratkie
soobshchenija o dokladah i polevyh issledovanijah Instituta istorii materialnoj kultury. Vol. 11.
Pp. 5165.
Schramm, Gottfried, 1982: Normannische Sttzpunkte in Nordwestrussland. Etappen einer Reichsbil-
dung im Spiegel von Namen: Beitrge zur Namenforschung. N.F. Bd. 17, 3. S. 273290.
Snorri Sturluson. Heimskringla 13. Bjarni Aalbjarnarson gaf t. 19411951. Reykjavk. (slenzk
fornrit 2628.)
Snorri Sturluson. Heimskringla: History of the Kings of Norway. Tr. by Lee M. Hollander. 1964. Aus-
tin.
Thomsen, V., 1879: Der Ursprung des russischen Staates. Gotha.
Uino, Pirio, 1988: On the History of Staraja Ladoga: Acta Archaeologica. Vol. 59. Kbenhavn, Pp.
205222.
Vasmer, Max, 1955: Russisches etymologisches Wrterbuch. Heidelberg. Bd. II.

442
 The Sea-Kings of Litla Sklda
Judith Jesch, Centre for the Study of the Viking Age, School of English Studies, University of
Nottingham, England
It is of interest, though not of immediately obvious significance, that the little treatise on po-
etic language known as Den lille Sklda (Faulkes 1998: xiv)1 makes reference on seven sepa-
rate occasions to skonungar sea-kings in connection with its summaries of kenning-types
for ships (not surprisingly), dwarves and giants (more surprisingly), battle, shields, armour,
striking weapons and the sea. In this, it contrasts with Skldskaparml in Snorris Edda which
includes skonungar among the determinants for sea- and ship-kennings, but not for the
broader range of referents (Faulkes 1998: 36, 74). The difference is illustrated in the way in
which the two texts summarise battle-kennings:

Orrostu m kalla namni nkkurs hreystis skarksamligs, glaum ea hljm, kenna vi herkli
ea vpn ea hlfar, ok v meirr, at skal kenna vi in ea skonunga, ef vill. [LS: 255]

Hvernig skal kenna orrostu? Sv at kalla ver vpna ea hlfa ea ins ea valkyrju ea
herkonunga ea gn ea glym. [Faulkes 1998: 66]

Its interest in sea-kings is not the only way in which it differs from Snorris work yet, with the
notable exception of Gurn Nordal (2001: passim), LS has received little scholarly attention.
The purpose of this paper is to bring out its characteristics as an independent analysis of
skaldic diction and possibly to speculate on its origins and function, and relationship to
Snorris work.

Litla Sklda
This text, here called Litla Sklda following Nordal, is a short compendium of poetic lan-
guage found in two manuscripts that also contain independent redactions (Nordal 2001:
2245) of Skldskaparml and related texts. The provenance and date of LS are uncertain,
though the earlier manuscript AM 748 Ib 4to is dated to the first quarter of the 14th century
and is therefore contemporaneous with the earliest manuscripts of Snorris Edda (Nordal
2001: 215). Nevertheless, LS is usually thought to postdate, indeed to derive from or be a
supplement to, Snorris Edda (Finnur Jnsson 1931: lix; Faulkes 1998: xlvii; Nordal 2001:
288). However, Faulkes elsewhere (1998: xiiixiv) admits the possibility that it might have
been a source for Snorri, and there are a number of reasons to think that it derives from the
kind of material used by Snorri, rather than from the Edda itself.
LS occupies only four-and-a-half pages in Finnurs edition and closely parallels Skldska-
parml in being an analytical account of poetic diction, though clearly it is much shorter than
Snorris work. Although Finnur Jnsson (1931: lviiiix) called it usystematisk and ikke
meget logisk, it seems to me to be organised into a reasonably coherent list of kenning-
referents, with the occasional interruption of other types of material (in square brackets be-
low):2

1. poetry, ships and drink

2. dwarves, giants and stones
3. [brief comment on kennings]

1
This is abbreviated LS and cited from the edition by Finnur Jnsson 1931: 2559.
2
For a rather different analysis of the structure of LS, see Nordal 2001: 2267.

443
 4. battle, and defensive and offensive weapons
5. blood and the sea
6. snakes
7. gold
8. arms (i.e. the human limb)
9. (drinking) horns
10. ice
11. dogs and other destructive things
12. night, snow and winter
13. wolves and carrion birds
14. men
15. women
16. body parts (hair, skull, ears, eyes, nose, mouth, breast, teeth, tongue, heart, hnetr)
17. the world and the heavens
18. the sun [and the moons companions Bil and Hjki]
19. [Grmnisml sts 4041]
20. kennings formed from parts of Ymirs body
21. [the story of the magic quern Grtti/Grtta]

Coherence is provided partly by the structure of each section and partly by links between top-
ics. For each referent, LS lists one or more base words and one or more appropriate determi-
nants. What is either a base word or a determinant for one referent can become the next refer-
ent to be listed, and this gives the text some cohesion, for example in the opening section:

Skldskapr er kallar skip dverga ok jtna ok ins ok fundr eira ok drykkr eira ok er rtt at
kenna sv, ef vill, bi skip ok drykk sem annars staar skldskap ok eigna eim, en skip m
kalla dra heitum ok fugla ok hesta ok kenna vi sj ok alt reii skips, en kalla hesta-heitum
einum, ef vi skonunga eru kendir. [LS: 255]

This opening statement allows for the incorporation of kennings into other kennings, but oth-
erwise the text shows no interest in embedded or extended kennings.
While the approach and structure of LS are reminiscent of Skldskaparml, the most im-
portant difference is that the kennings are explained but not exemplified. The only poetic quo-
tations in the whole text are two Eddic stanzas from Grmnisml, which lead into (and could
be said to exemplify) the kennings in the following sentence. Also, while Skldskaparml
lists both heiti and kennings, LS is almost exclusively about the latter. While it does through-
out recognise the role of heiti in providing variation in kenning-types (as exemplified in the
quotation above), there are only three examples of X heitir Y which are not followed by one or
more determinants:

Hnetr heitir fylvingar [] Sgr heitir sr, en simul stng [] [LS: 2589]

and all of these are more than a little obscure.

Also unlike Skldskaparml, there is very little retelling of Norse myth and legend, yet
such narratives are alluded to throughout.3 Thus, the section on kennings for poetry presup-
poses myths of the origin of poetry (as can be seen in the quotation above). inn is men-
tioned, often parallel with the skonungar, as a determinant in kennings for poetry (1), battle
(4), shields (4), armour (4), striking weapons (4), carrion birds (13) and men (14). The section

3
The mythological and legendary references more extensive than is suggested by Nordal 2001: 227, 312.

444
on stones (2) mentions Hamir and Srli, that on snakes (6) mentions the Migarsormr, and
that on gold (7) mentions King Fri, Fenja and Menja, King Kraki [sic], Fyrisvellir, Grani,
Sif, Mardll, giants, and Draupnir. The section on women (15) says that Regin heita go
heiin, bnd ok rgn, but does not go into any further detail, except to note that divine terms
can be used as base words in woman-kennings. The section on the cosmos (17) mentions the
four dwarves Norri, Suri, Austri and Vestri, and section (18) mentions the moons compan-
ions Bil and Hjki. The stories are assumed to be known and are therefore not related, with
the exception of a brief account of the magic ring Draupnir:

Hringr ht Draupnir, hina nundu hverja ntt draup af honum hringr jafnhfugr honum, v er
gull sveiti hans. [LS: 256]

The other exception is that the brief reference to Fri, Fenja and Menja in the section on
gold-kennings is picked up again at the very end of the text, in a short narrative about the
magic quern Grtti. I shall return to this passage later.
Gurn Nordal has argued (2001: 2259) that LS, together with a short account of the Fen-
rislfr and the redaction of Skldskaparml that follow it in its two manuscripts, represents a
new work on skaldic diction. While this may be true of this particular conglomeration of
texts in these two manuscripts, it still leaves open the possibility that the individual texts had
separate origins. Indeed Nordals comment (2001: 225) that LS was [] written with defer-
ence to Snorris Skldskaparml [and] associated with it at [an] early stage implies this pos-
sibility of separate origins. The overall structure of LS is identical neither to the redaction of
Skldskaparml which follows it in the manuscripts, nor to that in the Codex Regius (Clunies
Ross 1987: 90; Faulkes 1998: xlixl; Nordal 2001: 21621, 2247), and there are a number of
ways in which its analysis of poetic language differs from that of Skldskaparml (Finnur
Jnsson 1931: lix).
Nordal went on to analyse kennings for gold and for body-parts to demonstrate later devel-
opments in skaldic poetry. Thus, LS reveals not only the growing interest in and demand for
body-kennings in the late thirteenth century, but indeed the constantly developing ideas at
work in skaldic verse-making in the late thirteenth and fourteenth centuries, whereas Snorri
[] presents a conservative view of the practice of earlier poets (Nordal 2001: 239). How-
ever, she did not consider LS closely in her study of gold-kennings. Nordals book is primar-
ily about the poetry of the 13th and 14th centuries and LS fits into this context by virtue of its
preservation in two early 14th-century manuscripts. But a closer look at some of the kennings
shows that there is also overlap with the poetical practice of the 12th century, a time of great
innovation and transition in skaldic verse-making and analysis, yet also a century whose poets
are not so extensively treated in Snorris conservative view. The following examples arise
from my work on the 12th-century poetry of Rgnvaldr Kali Kolsson, jarl of Orkney (a poet
not cited by Snorri), and are illustrative rather than exhaustive.4

Feeders of ravens
Rgnvaldrs lausavsa 6 has a warrior-kenning brir bengagls feeder of the wound-gosling
[raven/eagle].5 Both parts of this belong to well-known types: the warrior as a feeder of one
of the beasts of battle (Meissner 1921: 283308) and a carrion bird as one of those beasts
(Meissner 1921: 11923). Both kenning-types are explained contiguously in LS:

4
All references to Rgnvaldrs lausavsur are from my forthcoming edition in Gade 2009, where further refer-
ences and more detail on the interpretations can be found.
5
Httalykill 37a (Finnur Jnsson 191215: IB 5056) has a warrior-kenning bvar hauka beiti-Nirir feeding-
Njrrs of the hawks of battle.

445
 Haukar ok hrafnar eru hrfuglar ok sv ernir. Rtt er ara fugla at nefna til ok eigna ni ea
kenna vi orrostu ea hr ea bl. Hvern karlmann m kalla feiti ea bri hrfugla [] [LS:
257]

Skldskaparml has a similar account of kennings for the raven and the eagle:

Alla ara fugla karlkenda m kenna vi bl ea hr ok er at nafn rn ea hrafn [] [Faul-

kes 1998: 90]

but the other part of the kenning is missing. The word brir does not occur anywhere in
Skldskaparml, nor does it specifically mention any agent-nouns meaning something like
feeder in relation to warrior-kennings. Moreover, according to Skldskaparml, bengagl
wound-gosling would not be a possible kenning for carrion bird, since the base word must
be a bird that is grammatically masculine, while gagl is neuter. Skldskaparml does cite a
verse (Faulkes 1998: 66, from orbjrn hornklofis Glymdrpa) which uses the kenning
benggl wound-goslings, but there it is used in the sense of arrows.
At the same time, kennings for carrion bird with gagl as the base-word are found in both
Httalykill 20a (hjaldrggl battle-goslings, Finnur Jnsson 191215: BI 496) and Httatal
62 (undgagl wound-gosling, Faulkes 2007: 27). Thus Rgnvaldrs verse accords with the
practice of Httatal and the theory of LS, rather than with the theory of Skldskaparml. This
pattern can be seen in other examples.

Women, gold and silver

Another of Rgnvaldrs lausavsur (17) has the woman-kenning hlanipt alindriptar head-
band-Nipt [norn] of forearm-snow [gold], i.e. the norn of the golden headband. Paul Bibire
(1988: 233) translated this kenning as broider-sister of the ell-drift and glossed it the
snow-drift of the forearm is silver. Woman-kennings are normally constructed with a word
or kenning for gold rather than silver as the determinant (Meissner 1921: 4134), as can be
seen in another of Rgnvaldrs lausavsur (4). Alindript forearm-snow is taken by all previ-
ous commentators to mean silver, presumably because snow-drifts suggest silver rather than
gold. Certainly Skldskaparml distinguishes clearly between red gold and white silver:

Gull er kallat kenningum eldr handar ea lis ea leggjar vat at er rautt, en silfr snr ea
svell ea hla vat at er hvtt. [Faulkes 1998: 61]

However, LS allows for the possibility of constructing gold-kennings with a base-word mean-
ing snow or ice:

enda m gull kenna til sns ok ss ok kenna til handar. [LS: 256]

There are similar kennings in both Httalykill and Httatal. Httalykill 4b has alnar dript
snow-drift of the fore-arm (Finnur Jnsson 1912-15: BI 489), which Meissner (1921: 224)
translates as Silber, but since the stanza is about Gunnarr Gjkason and the Niflung treasure,
gold is surely more appropriate. Httatal 43 has Grotta gladript joyful snow-drift of
Grotti (Faulkes 2007: 21). Meissner (1921: 224) translates this too as Silber, but the con-
struction with Grotti again makes a gold-kenning more probable (Faulkes 2007: 114).6
Rgnvaldrs woman-kenning must therefore be understood to include a kenning for gold,
as was traditional, but the gold-kenning he uses is not traditional. Again, this non-traditional
6
The parallels in LS and Rgnvaldrs poetry provide a more likely explanation for this kenning than Nordals
attempt (2001: 294) to link it with Freyjas tears.

446
kenning is consistent with the practice of Httatal and the theory of LS, rather than with the
theory of Skldskaparml.

Whales of the heath

Rgnvaldrs joint composition with Hallr rarinsson breimaga, Httalykill, twice has a
kenning for snake, serpent in which the base-word is a sea-creature and the determinant
some aspect of dry land (2b: heiar hvalr whale of the heath; 4a: urar lax salmon of the
stone-heap, Finnur Jnsson 191215: BI 4889). This kenning-type is explained in LS, which
cites both elements of heiar hvalr:

Orma er rtt at kalla fiska heitum ok hvala, ef eir eru kendir vi nkkut l, vi hraun, gras
ea grjt, gljfr ea heiar. [LS: 256]

Meissner (1921: 11213) lists a few examples of this type of kenning, all from later poetry.
Thus, heiar hvalr is found in Merlnussp, while frns lax and heiar lax can be found in
Harmsl and in a verse in Ragnars saga lobrkar, respectively. However, there is no equiva-
lent of this type of kenning in Skldskaparml, though its reverse, the kenning-type land of
whales [sea], is recorded there (Faulkes 1998: 63, 75).

Doors of battle
Httalykill 3a (Finnur Jnsson 191215: BI 488) uses both a simplex and a kenning in which
a shield is likened to a door: (gunnar) gtt door (of battle). Both the kenning-type and the
word gtt appear in LS:

Skjld m kalla [] vegg ea gar, blk ok brk, hur ok gtt, ili ok grind, hlea ok segl, tjld
ok refil, ok eigna jafnan orrostu ea ni ea skonungum. [LS: 255]

There is no equivalent of this kenning-type in Skldskaparml, nor does the word gtt appear
there in any other context. Gtt door-opening does appear in Httatal 89, though not in a
shield-kenning, but in a context where the poet requires a series of words ending in -tt (Faul-
kes 2007: 36). Meissner (1921: 167) has two other examples of this type of shield-kenning,
one from a lausavsa by Kormkr and one from the ugte vers of Njls saga.

Fris meal
The core of the Grotti myth is the magic hand-quern that grinds gold, so that its product can
be called Fris meal. Rgnvaldr uses the kenning Fra meldr for gold in his lausavsa
15. This kenning is explained in LS:

Gull er korn ea meldr Fra konungs, en verk ambtta hans tveggja, Fenju ok Menju [] [LS:
256]

While the basic kenning-type is widespread, the use of the word meldr is not. It does occur in
Httatal 43 (Fenju meldr, Faulkes 2007: 21). Skldskaparml has a different base-word,
(Fra) mjl, in its prose explanation of the kenning (Faulkes 1998: 51).7 Meldr does occur
twice in its quotation of Grottasngr (Faulkes 1998: 53, 57), though here in the meaning ac-
tion of grinding, rather than what is ground, flour.8 It is used in this way in a gold-kenning
7
Also in the Uppsala-Edda, which has a much shorter version of the story (Finnur Jnsson 1931: 136n, see also
120n.), discussed below.
8
An addition in another ms. of Skldskaparml (i.e. B that also contains LS) lists as gold-kennings both Fra

447
(Fenju meldr) in a stanza by Einarr Sklason, quoted immediately after Grottasngr in
Skldskaparml (Faulkes 1998: 57).
Rgnvaldrs kenning only makes sense if meldr is given the same sense as mjl, since part
of the point of the story is that Fri did not do any grinding. This meaning is also suggested
by the parallelism with korn in the LS explanation. The kennings of Httatal and Einarr
Sklason associate meldr with Fris slave Fenja, who did do the grinding, so that their
meaning is ambiguous and in them meldr could be either act of grinding or grain, corn,
meal, flour, though the former seems more likely, since the story makes clear that the grind-
ing is associated with the slaves while the end result, the gold, belongs to Fri. This distinc-
tion is implicit in the addition to ms B (quoted in n. 9) and explicit in LSs analysis of the
kenning-type.

Litla Sklda and Snorri

These examples demonstrate, firstly, that LS has some interesting correspondences with 12th-
century poetry, particularly that by Rgnvaldr, and, secondly, that many of these correspon-
dences also have parallels in Httatal. Snorris poem is generally acknowledged to be in-
debted to Httalykill (Faulkes 2007: xiixxi), yet the examples listed above are not paralleled
in Skldskaparml, indeed tend actively to contradict the theories of that text. It would seem
that Snorris own composition is grounded in the practice and the analysis of poetry from the
generation before him, in the 12th century. But once he had composed Httatal, generally
thought to have been before the rest of the Edda (Faulkes 2007: vii), he drew back from the
practice of his immediate predecessors and concentrated on establishing the conservative
canon that forms the basis of Skldskaparml. These are the hfuskld whose forn heiti he
wishes to codify in order to teach them to young poets (Faulkes 1998: 5). Although he cites
Einarr Sklason 35 times and Marks Skeggjason seven times, there is a much narrower
range of 12th-century poets cited than of those from earlier times (Nordal 2001: 778). The
aim of Skldskaparml is thus clearly normative rather than inclusive Snorri wishes to teach
only certain kinds of poetic language, on the whole that of before the 12th century. Then, as
Nordal has shown, the text of Skldskaparml becomes open for negotiation and rewriting in
its subsequent manuscript reception.
The correspondences outlined above suggest that LS is analogous to the earlier stage in
Snorris understanding of poetical language, when he was less censorious of certain poetic
developments of the 12th century. Thus LS is more likely to be derived from the kinds of ma-
terials he used, rather than having been written as a response to his analysis. This hypothesis
is supported by a closer analysis of the texts approach to gold-kennings.

Gold
Gurn Nordal has analysed the different manuscript versions of Skldskaparml to demon-
strate how the full account of all possible stories associated with gold-kennings found in the
Codex Regius is subjected to a gradual attrition and attenuation in subsequent mansucripts.
She argues that this came about because of a decreasing interest in the Danish narratives
which were the fashion in the 12th and early 13th centuries and an increasing interest in certain
Norwegian material (Nordal 2001: 31927). She does not, however, consider the evidence of
LS in this particular context.
The gold-narratives in Codex Regius include five derived from pagan myth and five in
which the gold originates [] in the world of myth, but the stories are transferred into the
domain of legends associated with Danish historical writing in the twelfth century (Nordal

mjl and Fenju meldr ok Menju (Finnur Jnsson 1931: 120n.; Nordal 2001: 328).

448
2001: 321). It is particularly these latter that are subject to omission or revision in the other
manuscripts, while the manuscripts that contain only Skldskaparml have a tendency to omit
the gold-narratives entirely (Nordal 2001: 3215). While LS does not have a full complement
of allusions to gold-narratives, seven of Nordals ten categories are represented in some form
(all quotations are from LS: 256):

1. The fire of gir: although gir is not mentioned by name, the idea of gold as the fire of
a body of water is exemplified in:

Gull skal kenna [] til sjar ok til vatna allra, kalla eld ok sl ok tungl ok stjrnu ok
kyndil ok kertil, dag ok leiptr, geisla ok blik ok alla birti []

2. The barr of Glasir: not recorded

3. The golden objects made by the dwarfs: haddr of Sif, Draupnir: haddr Sifjar [] Dra-
upnir
4. The tears of Freyja: tr Mardallar
5. The voice and words of giants: ml jtna
6. The compensation for the killing of Otr, and the myth of the Rhinegold, featuring
Sigurr Ffnisbani: for the Rhinegold, see no. 1 above, for Sigurr there is byrr
Grana and, indirectly, Orma jr er gull, rekkja eira ok gata.
7. Ragnarr lobrk [] : not recorded
8. The gold of Fri:

Gull er korn ea meldr Fra konungs, en verk ambtta hans tveggja, Fenju ok Menju

9. The legend of Hrlfr kraki and Ails: s Kraka konungs ok fr Fyrisvallar

10. The myth of Hlgi: not recorded

If Nordal is right in her analysis of the origins and fate of the interest in the different gold nar-
ratives, then this pattern suggests that LS reflects the earlier (12th and early 13th-century) stage
when the full range of narratives were of interest.

Fris meal again

An important aspect of Nordals argument is that certain legends underpinning gold imagery
are prominent in 12th-century culture and survive in Snorris treatise, but are not referred to
in later versions of the work and in Litla Sklda (2001: 31112). The examples listed above
show that in fact several (though not all) of these are referred to in LS. The fact that these are
referred to only briefly or allusively is in keeping with the nature of this text and the section
on gold-kennings is actually the longest in LS. The key legend is that of gold as Fris meal
which is not only alluded to in the section on gold-kennings, but is briefly narrated at the end
of the text (LS: 259):

Kvern heitir Grtti, er tti Fri konungr; hon ml hvetvetna at er hann vildi, gull ok fri.
Fenja ok Menja htu ambttir r, er mlu. tk Msingr skonungr Grttu [sic] ok lt mla
hvtasalt skip sn, ar til er au sukku Pttlandsfiri. ar er svelgr san, er sr fellr auga
Grttu. gnr sr, er hon gnr, ok var sjrinn saltr.

At this point the text seems more interested in the aetiological aspects of this narrative, its
explanation of the why the sea is salt, rather than its connection with legends of gold, and this
may reflect a particularly Orcadian take on the story (Jesch forthcoming). But what is more

449
important in this context is that LS is hereby shown to be closer to the versions of Skldska-
parml that are in manuscripts of the whole of Snorra Edda and particularly Codex Regius,
and quite unlike the versions (A and B) that are found in the same manuscripts as LS and
which omit almost all of the material on gold (Nordal 2001: 3226).
Here the case of Codex Uppsaliensis, which we are celebrating at this conference, is of in-
terest. Like LS, it has a very short version of the narrative, with no quotation of the poem
Grottasngr, yet its narrative is quite different (Finnur Jnsson 1931: 136n., corrected with
reference to Grape et al. 1977: 87):

her segir hvi gvll er kallat froa miol Gvll er kallat miol froa viat froi konungr keypti ambat-
tirnar fenio ok menio. ok a fanz kvernsteinn einn sva mikill. i danmorkv at engi feck dregit. En
sv nattura fylgi at allt miol at er vndir var malit var at gvllit. Ambattirnar fengv dregit stein-
inn. konvngr let r mala gvll vm hri. a gaf hann eim eigi meira svefn en kvea matti lio
eitt. Sian molo r her a hendr hanvm. Sa var havfingi fyrir er mysingi het spekingr mikill.

Faulkes (1998: xlii) calls this a summary of ch. 43 of Skldskaparml, but it lacks the linked
story of why the sea is salt that both LS and other versions of Skldskaparml have. This is
presumably to do with the fact that Codex Uppsaliensis rearranges the material to do with
gold very substantially and that subject is its focus at this point. LS, on the other hand, seems
to be more interested in stories to do with the sea, as its Grotti narrative follows on from top-
ics related to the natural world and concludes with Msingr and why the sea is salt. Which
brings us back to sea-kings.

Sea-kings

I highlighted above the particular interest in sea-kings shown in LS, where they appear as pos-
sible determinants for a wide range of kenning-types. Skldskaparml has a much narrower
range of kennings which sea-kings can determine, as noted above, but they are present else-
where in the text, as Msingr is mentioned in the Grotti-narrative, and the first of the ulur
begins with a list of 75 sea-kings, including Msingr (Faulkes 1998: 52, 10910). This list of
sea-kings appears in all of the manuscripts that have ulur, and heads the lists in all except B,
where the sea-kings are relegated to second place below kings (Clunies Ross 1987: 84).
Snorri himself explains what a sea-king is, in ch. 30 of Ynglinga saga (Bjarni Aalbjarnar-
son 1979: I 60):

ann tma herjuu konungar mjk Svaveldi, bi Danir ok Normenn. Vru margir sk-
onungar, eir er ru lii miklu ok ttu engi lnd. tti s einn me fullu mega heita skonungr,
er hann svaf aldri undir stkum si ok drakk aldri at arinshorni.

This comment comes immediately after a reference to the death of Hrlfr kraki, thus associat-
ing sea-kings with a particular historical period. In the context of the study of poetry, these
sea-kings can therefore also be associated with the poetry of an earlier period, the conserva-
tive poetry that Snorri was keen to codify in Skldskaparml.
Elena Gurevich (1992: 51) has identified the ulur as the first learned writing dealing with
the way the poetic language was constructed and has demonstrated in detail the process by
which sequences of former nomina propria, place-names and mythological names [were]
transferred from a number of unique beings and objects to those classes they belong to and
thus turned into common nouns (1992: 35). Gurevich stresses the dynamic process of
polysemy by which the ula is not so much a catalogue of real skaldic lexical stock as a gen-

450
erator of poetic synonyms (1992: 36) and she sees the ula as a learned activity which helped
poets to understand and compose poetry and which could thus undermine the mythological
origins of the names. Clearly, it is unlikely that every single name of a sea-king had a story or
legend behind it, as Sophus Bugge once thought (1875: 210). Yet the case of Msingr shows
that some certainly did.
It is fruitful to see the sea-kings of LS in the context of the sea-kings of the ulur. The two
go hand in hand: the kenning-templates of LS indicate that sea-kings are required in a variety
of kennings, while the ulur generate the actual names of sea-kings that can fill the slots. I
would like to suggest that these two forms, the list of kenning-types (i.e. LS), and the heiti
with which they can be filled (i.e. ulur), both represent early forms of analysis of skaldic
poetry, and therefore the kind of material which formed the basis of Skldskaparml. But LS
is not just a list of kenning-types: it demonstrates, very briefly, some of the ways in which this
analysis could be expanded, by the quotation of examples, by commenting on the kenning-
formation process, and by expanding some of the mythological names from the ulur with
stories about them, where these were known, as in the case of Msingr. This kind of expan-
sion is exactly what Snorri did at much greater length in Skldskaparml, and the dynamic
process continued in the different manuscript versions of this text, as demonstrated by Nordal.

Bibliography
Bibire, Paul, 1988: The poetry of Earl Rgnvaldrs court. In: St Magnus Cathedral and Orkneys
Twelfth-Century Renaissance. Ed. by Barbara E. Crawford. Aberdeen. Pp. 20840.
Bjarni Aalbjarnarson, ed., 1979: Snorri Sturluson. Heimskringla. slenzk fornrit, 268. Reykjavk.
Bugge, Sophus, 1875: Biskop Bjarne Kolbeinsson og Snorres Edda. Aarbger for nordisk
oldkyndighed og historie. Pp. 20946.
Clunies Ross, Margaret, 1987: Skldskaparml. Snorri Sturlusons ars poetica and medieval theories
of language. Odense.
Faulkes, Anthony, ed., 1998: Snorri Sturluson: Edda. Skldskaparml. London.
Faulkes, Anthony, ed., 2007: Snorri Sturluson: Edda. Httatal. London. (2nd ed. available at
http://vsnrweb-publications.org.uk/).
Gade, Kari Ellen, ed., 2009: Poetry from the Kings Sagas, 2: From c. 1035 to c. 1300. Turnhout.
Grape, Anders, et al., ed., 1977: Snorre Sturlassons Edda. Uppsala-handskriften DG 11, II.
Transkriberad text och palaeografisk kommentar. Stockholm.
Gurevich, Elena, 1992: ulur in Skldskaparml. An attempt at skaldic lexicology. Arkiv fr nordisk
filologi, 107. Pp. 3552.
Finnur Jnsson, ed., 1931: Edda Snorra Sturlusonar udgivet efter hndskrifterne. Kbenhavn.
Jesch, Judith, forthcoming: The Orcadian links of Snorra Edda. In: Snorra Edda i europeisk og
islandsk kultur. Ed. By Jon Gunnar Jrgensen. Reykholt.
Meissner, Rudolf, 1921: Die Kenningar der Skalden. BonnLeipzig.
Nordal, Gurn, 2001: Tools of Literacy. The role of skaldic verse in Icelandic textual culture of the
twelfth and thirteenth centuries. Toronto.

451
 Royal Women and the Frigerarsaga Episode
Jhanna Katrn Fririksdttir, Faculty of English, Oxford University, England

Introduction
The Frigerarsaga in Snorri Sturlusons lfs saga helga in Heimskringla has been de-
scribed by Lars Lnnroth as a narrative which centers upon some of the most fundamental
problems of medieval government and kingship (Lnnroth 1976:17). This episode, funda-
mental to Snorris subtle portrayal of power politics, relates a dispute between the Swedish
and Norwegian kings, lfr snski Eirksson and lfr digri Haraldsson, who are equally
ambitious but unlike in character.1 It involves several key participants: the two kings, the Ice-
lander Hjalti Skeggjason, operating as an international diplomat, the Swedish jarl Rgnvaldr,
and three aristocratic women: Ingibjrg Tryggvadttir (the sister of lfr Tryggvason),
Ingigerr lfsdttir and her half-sister strr lfsdttir. Previous scholarship has illumi-
nated the subtexts of Frigerarsaga in a political or ideological context; however, before the
final culmination of the episode, the dramatic confrontation between orgnr the lgmar and
King lfr snski at the Uppsala assembly, there is a long prelude with a relatively large in-
volvement of women which merits further attention. This part of the story is characterised by
delicate and covert efforts to settle the dispute, mainly conducted by Hjalti and Ingigerr, but
with significant contributions from Rgnvaldr jarl and especially his wife Ingibjrg, who
holds a personal grudge against lfr snski. The aftermath of the dispute is dramatic as well;
in some narrative traditions (i.e. versions of the Separate Saga of St lfr by Snorri as well as
the Legendary Saga), strr surreptitiously travels to Norway, meets King lfr and boldly
offers herself to him in marriage.
In this paper, I shall not seek to examine how the narratives ethically value the actions of
active female characters, nor is my goal to examine the historical veracity of the events;
rather, I will explore how the texts convey and interpret the relationship between gender and
power in this highest stratum of society. I will examine the considerable impact these three
aristocratic women have on politics, what role they play and what tools are available for them
to wield power. Sverre Bagge has argued that politics in Heimskringla are less governed by
concepts of ideological struggle, e.g. the source and scope of the kings power or the role of
the people in government, but more through the rational assessment of interests, e.g. the king-
doms or the kings own, and tracing how agents strive to protect and advance these interests
(Bagge 1991: 1089). This pragmatic and rational approach to gaining political support, fo-
cusing on power, not ethics, principles or doctrine, can be characterised as Realpolitik. This
equally applies to royal womens involvement in politics; their success in wielding power in
the Frigerarsaga is determined less by the womans legitimate right to act because of aris-
tocratic rank than the question of parallel or conflicting interests.

Ingibjrg Tryggvadttir
The Frigerarsaga episode in Heimskringla shows how delicate international negotiations
between royal courts could be. The kings quarrel is initiated by a dispute about which of the
two had the right to claim taxes in a region on the border of the two kingdoms that was previ-
ously held by Sweden. The disagreement quickly escalates into full-blown antipathy towards
lfr digri on lfr snskis part. The dispute affects the Swedes in Vestra-Gautland firstly
on a financial level because it makes trade with Norway difficult; secondly, it is a matter of

1
For clarity, they shall be referred to below as lfr snski and lfr digri.

452
international power politics. Ingibjrg Tryggvadttir, the wife of Rgnvaldr lfsson, the jarl
of Vestra-Gautland, is adamant that her husband should help lfr digri. Although it is a risk
to oppose his own king, it would weaken Rgnvaldrs position even more if he refused lfr
digris request for support as he would then be considered cowardly. Furthermore, Ingibjrg
has a personal motive in working towards the downfall of lfr snski, and she uses her in-
fluence to persuade her husband to form an alliance with lfr digri:

[Ingibjrg] gekk at me llu kappi at veita lfi konungi. Hon var aftakamar mikill um etta
ml. Helt ar til hvrt tveggja, at frndsemi var mikil me eim lfi konungi ok henni, ok at
annat, at henni mtti eigi fyrnask vi Svakonung at, er hann hafi verit at falli lfs Tryggva-
sonar, brur hennar, ok ttisk fyrir sk eiga tiltlu at ra fyrir Nregi. Var jarl af fort-
lum hennar mjk sninn til vinttu lfs konungs. (Heimskringla II 85)

Despite being married to a Swedish magnate, Ingibjrg has many reasons for supporting lfr
digri instead of her husbands king: the narrator emphasises a desire to uphold her honour (by
seeking revenge for her brother and preventing her husband from being accused of coward-
ice), family connections and finally an emotional investment in seeing her brothers enemy
fare badly. These reasons, emotional rather than strategic, seem to be the cause of her deter-
mination, expressed in the same urgent words as the servant-womans challenge to Hrafnkell
in Hrafnkels saga Freysgoa: Ltr grikonan ganga af kappi (Hrafnkels saga 127).
When the envoys of lfr digri, Bjrn stallari and Hjalti Skeggjason, arrive at Rgnvaldrs
court and address the dispute, Ingibjrg urges her husband to support lfr digri, regardless of
the risks involved:

Skjtt mun ek birta minn hug, at ek vil, jarl, at r leggi allan hug at stoa orsending lfs
konungs, sv at etta rendi komisk fram vi Svakonung, hverngan veg sem hann vill svara.
tt ar liggi vi reii Svakonungs ea ll eign vr ea rki, vil ek miklu heldr til ess htta
en hitt spyrisk, at leggisk undir hfu orsending lfs konungs fyrir hrzlu sakir fyrir
Svakonungi. Hefir til ess buri ok frndastyrk ok alla atfer at vera sv frjls hr
Svaveldi a mla ml itt, at er vel samir ok llum mun ykkja heyriligt, hvrt sem heyra
margir ea fir, rkir ea rkir, ok tt konungr sjlfr heyri . (Hkr II 90)

Ingibjrg points out the resources which Rgnvaldr has at his disposal: the support of their
kin and his official role as jarl, providing him with an arena in which to speak legitimately
and persuade the Swedes to support him, as well as the political weight of a magnate. Al-
though this monologue does not appear to be a hvt or incitement speech, involving accusa-
tions of effeminacy in the traditional sense, the husbands reply Ekki er at blint, hvers
eggjar implies that there is just such a subtext in her words (Hkr II 90). Rgnvaldr accedes
to Ingibjrgs request but insists on being in control of the plan; he wants to deliberate and
evolve a plan instead of taking rash measures. From the beginning, although he makes it look
as if his wife is compelling him to it, the jarl does not oppose the plot against lfr snski; he
is likely to gain from the kings loss. Ingibjrgs role in this matter might function as a literary
device, representing the voice of public (or narratorial) opinion, prompting Rgnvaldr to act
as well as give his reasons for doing so, but her grudge reminds the audience about lfr sn-
skis past dealings with the Norwegian people.
It is clear from lfr snskis speech at the Swedish assembly that he regards Ingibjrg to
bear considerable responsibility for what he claims is Rgnvaldrs treasonous support of lfr
digri: [lfr snski] segir, at allt slkt hlaut hann af eggjan Ingibjargar, konu sinnar (Hkr II
115). Ingibjrg is thus twice explicitly said to have incited (eggja) her husband to act; this
suggests that her appeal was considered a formal speech act since it had become public
knowledge. Although Rgnvaldr decided on the details of the plan, his wife is depicted as

453
having achieved her own separate agenda in the matter; their interests ultimately coincide. It
is not clear whether, by her participation in the strategising, the narrator regards Ingibjrg as
being within her rights, either as an aristocratic woman, i.e. being of royal descent and the
wife of a jarl, or simply as a woman, whose gender role it is to whet male kin to act. Consid-
ering that the narrator makes a substantial effort to give logical reasons and motivations for
Ingibjrgs decision to encourage her husband to undermine his king, I suggest that, whether
legitimate or not, her actions are viewed as justifiable; the question of legitimacy relies partly
on the appropriateness of whetting as a culturally-determined speech act.
Having ensured her husbands support, Ingibjrg consequently helps Hjalti to gain an au-
dience at the Swedish court by sending him with her token (jarteikn) as well as a bag of silver
which he is to give to the king as tax in order to win him over. By this, she shows a great deal
of shrewdness, identifying the financial interests that lie at the centre of the dispute and find-
ing a way to capitalise on them to her partys advantage. It is not clear whether the silver
comes out of her own funds or her husbands, but whichever the case, she seems to have the
right to distribute it. Ingibjrg also works through other unofficial channels, sending word to
the Swedish princess, Ingigerr, asking her to help Hjalti.2 This suggests that she has already
recognised the possibility of a marriage between Ingigerr and lfr digri and thus she suc-
ceeds in involving the princess as well as helping Hjalti get an ally at the Swedish court.
Ingibjrg remains a presence throughout the dispute and a pivotal member of the cast of
participants although she does not play a leading role after her initial involvement. She uses a
number of strategies in order to further her agenda: she persuades her husband to act by put-
ting forward rational arguments as well as prompting him in what is most likely a formal in-
citement speech, she provides economic resources to enable Hjalti to gain the kings favour
and she enlists Ingigerr as an ally.

Ingigerr lfsdttir
Following Ingibjrgs initiative, princess Ingigerr and Hjalti privately discuss the possibility
of reconciling her father and lfr digri on numerous occasions. Although she is not optimis-
tic that the king will be well-disposed towards the idea, she still makes an honest effort to
appeal to his better judgement, suggesting that he abandon his aim to conquer barren Norway,
reconcile with lfr digri and focus on regaining lands previously held in Russia. Hjalti had
previously tried to persuade the king of the same, but neither of them gains anything by their
attempts except to provoke lfr snskis anger. However, Hjalti does not let this discourage
him; he remains at the Swedish court and continues his talks with the princess.
Hjaltis next step is to test the waters carefully to see whether Ingigerr is open to marriage
with lfr; the princess blushes a little, considers the matter for a while and then replies
brtt ok stilliliga, slowly and calmly, and with what is most likely false modesty:

Ekki hefi ek hugfest svr fyrir mr um at, v at ek tla, at ek myna eigi urfa til at taka eira
svara, en ef lfr er sv at sr grr um alla hluti sem segir fr honum, mynda ek eigi
kunna skja minn mann annan veg, ef eigi er at, at r myni heldr hli gilt hafa marga
stai. (Hkr II 99100)

It must have occurred to Ingigerr at some point before this that through her advocacy she
could acquire a king as a husband; otherwise she would hardly have helped Hjalti or made
several attempts at persuading her father. After this suggestion of marriage, Bjrn stallari and

2
In the Legendary Saga, lfr snski consults Ingigerr and asks her opinion on Hjaltis character and when
they have agreed that he is vitr mar oc vl um sec, lfr decides to allow Hjalti to address the princess (lfs
saga hins helga: die Legendarische Saga 96).

454
Rgnvaldr become involved in the matter once again; Ingigerr states that her father is the
only one who can give her away in marriage, but otherwise, Rgnvaldr is the relative she
looks to for trustworthy advice. She thus hints to lfr digris envoys that she is open to cir-
cumventing her fathers authority if Rgnvaldr were to give her his support. Ingigerrs moti-
vation is clearly personal; her father does not seem to be making much of an effort to find her
a noble suitor and she probably does not want to run the risk of marrying beneath her rank. As
Jenny Jochens has pointed out, princesses were often married to lower-born men, as with
Ingibjrg Tryggvadttir, a princess who marries a jarl (Jochens 1986:16970). Thus,
Ingigerr is neither governed by a sense of duty nor obedience to her father; her loyalty is
only to herself and her ambition of making a suitably royal match.
After it has been established that Ingigerr is willing to use herself as a peace-offering in
the proposed reconciliation between the two Scandinavian royal families, against the will of
her father, lfr digris allies proceed to solve the matter at the Swedish assembly, using the
wise orgnr lgmar, the relative and foster-father of Rgnvaldr, as their mouthpiece.
orgnr demands that the king follow the will of the Swedes and make peace with lfr digri
under the threat of a revolt, and the dispute is settled with lfr snski grudgingly submitting
to these demands. However, nothing comes of the marriage between Ingigerr and lfr digri
because of her fathers refusal to bow to the Swedish assemblys coercion; instead, he calls
everything off and marries her to King Jarisleifr in Russia, a satisfactory outcome for
Ingigerr in terms of a suitable match (although perhaps not as good as marrying a future
saint). lfr digri ultimately marries her illegitimate half-sister strr; the narrator of Heim-
skringla presents this as a great victory for the king as it is Rgnvaldr who betroths strr to
lfr digri without the consent of her father, much to his humiliation and anger.
In Snorris carefully drawn portrayal of the Frigerarsaga, Ingigerr is one of the main
players. However, it complicates matters that her efforts to settle the dispute by becoming a
peace-weaver are not productive in the end; the princess does not have the efficacy to realise
her plans. She makes a failed attempt to persuade her father to settle with lfr digri, and then
tries with others to arrange to marry him with or without the kings consent, an agreement
which the king accepts unwillingly under duress, and has no intention of fulfilling. However,
Ingigerr does wield some degree of power, although clearly not legitimately, when, after her
betrothal, she succeeds in making her father promise to let her have one man of her choice to
go with her to Russia. After he has agreed, she announces that she wishes this person to be
Rgnvaldr jarl, enabling him to escape from Sweden unscathed despite his involvement in the
clandestine plot to marry strr to lfr digri, an act which lfr snski considers treason.
The king replies: Annan veg hefi ek hugat at launa Rgnvaldi jarli drttinsvikin, au er hann
fr til Nregs me dttur mna ok seldi hana ar til frillu eim inum digra manni ok eim, er
hann vissi vrn vin mestan (Hkr II 147) but nevertheless, he keeps his promise to his daugh-
ter. Thus, although unsuccessful in her efforts to marry lfr digri, Ingigerrs manoeuvring
ultimately manages to secure her own interest, driving the king to find her a suitable husband
and saves the skin of her ally.
With the exception of Heimskringla, all the historical sources which relate the events of the
Frigerarsaga (grip, Fagrskinna, the Legendary Saga) give the fathers irrational anger as
an explanation for the end of Ingigerr and lfr digris royal engagement. In the Legendary
Saga, lfr snski is initially portrayed as a sensible man and when Hjalti initially suggests a
peace-settlement with the union, the king decides that this is a good solution, consults the
princess, who also approves, and betrothes her to lfr digri. His only provision is that lfr
come to meet him, and show humility and willingness to reconcile. However, the abrupt end
of their engagement occurs after a hunting trip, when Ingigerr compares her father unfa-
vourably to lfr digri after he had boasted of his hunting prowess. The king angrily replies
that as a punishment, she will never have lfr, which, as Hans Schottmann points out, makes

455
much less sense psychologically than the Heimskringla version (Schottmann 1994:543). He
argues that Snorri adds characters not present in any of his sources (i.e. Rgnvaldr, Ingibjrg
and rgnr) and develops the rgnr-episode in order to rationalise the kings anger and
prepare the audience for his decision to break off the engagement. Otto von Friesen and most
recently Theodore Andersson have argued that Snorri made use of oral sources for the
orgnr-episode, which is not unlikely in terms of the additional rudimentary facts, but as
Schottmann compellingly argues, Snorris sophisticated and carefully-constructed narrative is
more convincing in literary terms than that of the other versions.3

strr lfsdttir
Nothing comes of lfr digris marriage to Ingigerr but the king still manages to form a un-
ion, albeit less prestigious, with a member of the Swedish royal family, the illegitimate
strr. The Legendary Saga includes an episode after Ingigerr has been packed off to Rus-
sia, in which strr travels to Norway with her foster-father Egill, presumably without per-
mission from her father, to meet lfr digri. The king has locked himself up in a loft, incon-
solable at the loss of Ingigerr. strr makes several attempts at persuading lfr to cheer up
and resume his duties as king, telling him that this is the express wish of her sister Ingigerr:
[hon] mllte, hrra, at er skilldu higgia af harme oc glia vini yra oc taka upp goa
sivniu, sem yr byriar. Gersc mikil briostar, sem kononge somer oc hans tign hfver
(Legendarische Saga 102). On her second visit, strr gives lfr a shirt embroidered
with gold, claiming that it is sent by Ingigerr along with her offers of eternal friendship. On
the third occasion, she proposes marriage:

En sva er mer boet af Ingigiri, at yr skilldim ver i hvrn sta vira um fram alla menn. En
firir a soc, at u ert sva harmfngenn, a er ess miri orf, yr at glia. o var igi su
hamingia konongs vars, at sa radahagr skilldi fram koma, sem tlar var, a mtte enn nokcora
bot a ui vinna, firir ui at igi man inmllt um vera, hvar oviring er miri, su er Olafr ko-
nongr gere yr i brigmlonom ea esse, at hann skal gi raa ea forsio firir hava firir vara
hond. En hlldr en igi faer u gli ina, a man ec at til lggia me umrom Ingigirar,
at fastna mik siolf yr utan hans vilia ne rada. Oc er btra, at biia goz ras oc goz konongs,
en iga ovirilegan mann, o at konongs namn bere. En o at at bere a, at hon se mestr
skarungr, a man at vitra manna or, at su er gofgazt, er ionar. (LS 1024)

King lfr is here depicted as neglecting his duties because of his heartbreak at the loss of
Ingigerr.4 strr tries to get him to follow her advice by reminding him of the proper behav-
iour of a king and appealing to his love for her sister; the gold-embroidered shirt, ostensibly a
gift from Ingigerr, appears to be a courtly love trope.5 strr then makes her final speech,
again claiming to be sent by Ingigerr and although she acknowledges his sorrow, she re-
minds the king that by marrying her without her fathers consent, lfr can avenge the hu-
miliation of having lost out on her sister. strr declares that she will give her own hand in
marriage, explaining that it is better to propose to a good husband and king than marry an
unworthy one, even if he is royal. Finally, strr highlights her virtues of obedience, which

3
For discussion on whether Snorri invents the orgnr-episode or makes use of oral sources, see Oscar Albert
Johnsen (1916), Otto von Friesen (1962:205270), Hans Schottmann (1994:53953), and Theodore M. Anders-
son (2008:538).
4
For discussion on lfrs lovesickness, see Anne Heinrichs (1999:3643).
5
The LS is dated to the first quarter of the thirteenth century and is preserved in a Norwegian manuscript from
the first half of the same century (Sverrir Tmasson 2006:4512); Brother Robert is said to have translated Tho-
mas Tristan in 1226. It is thus possible that the author (or editor/scribe) of the LS was familiar with romance
literature.

456
seem to make her an even more attractive match.6 After this speech, King lfr brightens up,
marries strr and resumes his royal duties.
Two themes are noteworthy in strrs discourse: firstly, the repeated evocation of
Ingigerr and her blessing, which perhaps reveals her aim to associate the two sisters in a
positive way in the kings mind, emotionally, as an object of his love, and politically, due to
the fact that both sisters have sided with their fathers opponents. Secondly, strrs proposal
to lfr seems highly subversive; as Jochens has argued, the idea that women should even
give their consent in marriage did not exist until the late twelfth or thirteenth century in Scan-
dinavia but is applied anachronistically by authors to earlier periods (Jochens 1986:16970).
Perhaps strrs bold proposal to lfr is influenced by romance: Rosemary Power suggests
that the unconventional idea of a woman initiating union, albeit sexual, with a man, is im-
ported into Old Norse literature from Marie de Frances Le Lai de Lanval, and the gift of the
gold-embroidered shirt further supports this suggestion (Power 1985:160).
This extraordinary account does not appear in Heimskringla, where it is Rgnvaldr who
betrothes the princess to lfr digri after she has given her consent, but the episode is interpo-
lated in some of the manuscripts of the Separate Saga of St lfr, an earlier work of Snorri.
One of them is Bergsbk (Perg. fol. nr. 1, Royal Library, Stockholm), a large and impressive
manuscript dated to ca. 1400, containing historical prose texts such as the sagas of the kings
lfr Tryggvason and lfr Haraldsson, as well as religious poetry and poems celebrating
kings (Lindblad 1963:12). In the Separate Saga, there are many interpolations, especially
concerning miracles and marvels, some of which do not appear in any other sources and have
been attributed to the manuscripts editor (Lindblad 8). This manuscript adds details which
give the scene a greater interest, e.g. noting that strr covers her face with a veil when she
visited the king for the first and second time, and it depicts her as even bolder than in the Leg-
endary Saga. strr confesses in this account that she travelled to Norway against the advice
of everyone else, and acknowledges her persistence in her attempts to win the king over. On
the third visit, she uncovers her face and proceeds with her final address, pointing out to the
king that although she is not as well-born as Ingigerr, he will be no worse off without her.
Her final argument manages to arouse the kings interest:

enn ott adr se gior mikil mvnr a mvn sv brat tignvzt sem er gengr nrst ok drotning er yfer
ollv landi med er. veit ek ath sva mvn ickia sem eingi kona mvni sva hafa til mannz melt sem
ek ok villda ek ok sva vi at ek tla ath ek eigia eim mvn til dra mannz at mla sem ek legg
meiri stvnd a enn adrar icki mer ok meiri vegr at mla diarflega til ess mannz er mer er
finleg giefa at en bida hins or stad er meir dregr til iafnnadar (Den store saga om Olav den hel-
lige, 76970).

strr here declares that it is likely that no woman has ever spoken thus to a man; further-
more, although her rank does not equal his, this will be insignificant after their marriage, since
the woman nearest to the king in status, i.e. his queen, is the noblest of all. She adds that she
knows this is unusually ambitious but she would rather boldly address the man who will be
her eternal good fortune than wait for a proposal from someone more equal in status.
The third version of this episode appears in AM 61 fol., dated to ca.1400, containing the
sagas of kings lfr Tryggvason and lfr digri, and in Tmasskinna (GKS 1008 fol.), dated
to the same period, containing Thmas saga, the story of the life of St Thomas Becket, as
well as the Separate saga of St lfr (Loth 1964:7). Tmasskinna thus has a distinct interest
in saints. In this version, strr is less feisty and more humble than in Bergsbk; neverthe-
less she appears as strikingly forceful in her last, urgent request to the king to agree to their
6
Anne Heinrichs argues that obedience is a Christian virtue, and furthermore, that strr represents Christian
values, while Ingigerr stands for secular and political values (Heinrichs 1985:458).

457
union in order to prevent warfare and the deaths of many Christian men. The narrator juxta-
poses strrs spakligt (wise) counsel and patient efforts to negotiate an agreement with what
she describes as the kings ra lyndi (stubbornness) in his refusal to accept a less advanta-
geous match than originally suggested (Den store saga 7701).
These versions of strrs excursion to Norway agree with her image elsewhere in me-
dieval sources as an assertive, intelligent, independent and eloquent woman, suggested by her
subversive behaviour towards her husband in ttars ttr svarta (especially in the Bergsbk
version) where she rewards the skald ttar for the mansngr he composed for her, and in
chapter 1 of Magnss saga ins ga in Heimskringla, where she persuades the Swedish as-
sembly to support her stepson Magns to the Norwegian throne. Snorris reasons for omitting
the episode in favour of involving Rgnvaldr jarl are unclear but perhaps he either found the
idea of strr proposing to the king outrageous and unlikely, or he decided to focus on the
struggle between lfr digri, Rgnvaldr and lfr snski. If, as Bagge has suggested, the
Heimskringla version of the episode is favourable towards lfr digri and aims at presenting
his marriage to strr as a victory over lfr snski, concealing the humiliation he suffers
when the engagement to Ingigerr is broken off, then reducing strrs role and increasing
lafr digris and Rgnvaldrs seems to be a part of that strategy (Bagge 1023).

Conclusion
In the Frigerarsagas many versions, royal and noble women are depicted as employing
various tools, verbal and financial, to wield power according to their own agendas; individual
authors arrange, develop and perhaps invent the details of this episode according to their own
narrative aims. Royal women are in the unusual position of having access to material re-
sources and, as wives and daughters, to members of the ruling class. Whether kings or noble-
men, they are, in their struggle for power, primarily concerned with furthering their own in-
terests; so too are the women, and these do not always match. By identifying these competing
interests and negotiating the subtle currents of power, women are able to cooperate with the
rulers, directly or through representatives, or to undermine them by surreptitiously working
with the opposing party. The difference between the sisters success in achieving their goal,
marriage to lfr digri, depends on how strongly they are under their fathers influence;
strrs freedom of movement in some versions enables her to remove herself from his au-
thority while Ingigerr is unable to circumvent his power over her. Whether they fail or suc-
ceed in carrying out their own agendas, royal women in every version of the Frigerarsaga
are active participants in international and domestic politics.

Bibliography
Andersson, Theodore M., 2008: The Oral Sources of lfs saga helga in Heimskringla. In: Saga-book
32. Pp. 538.
Bagge, Sverre, 1991: Society and Politics in Snorri Sturlusons Heimskringla. Berkeley.
Den store saga om Olav den hellige. Efter pergamenthndskrift i Kungliga biblioteket i Stockholm nr.
2 4to med varianter fra andre hndskrifter, vol. 2. Ed. by O. A. Johnsen and J. Helgason. 1941.
Oslo.
Friesen, Otto von, 1942: Fredsfrhandlingarna mellan Olov Sktkonung och Olav Haraldsson. In
[Svensk] Historisk tidskrift 62. Pp. 20570.
Heimskringla. Ed. by B. Aalbjarnarson. 194552. Reykjavk. (F vol. 23.)
Heinrichs, Anne, 1999: Wenn ein Knig liebeskrank wird: Der Fall der lfr Haraldsson. In Die Aktu-
alitt der Saga. Ed. by S. Anderson. Walter de Gruyter.
, 1985: Christliche berformung traditioneller Erzhlstoffe in der Legendarischen Olafssaga. In
Sixth International Saga Conference 28.7 2.8 1985: Workshop Papers. Copenhagen. Pp. 45167.
Hrafnkels saga Freysgoa. Ed. by J. Jhannesson. 1950. (F vol. 11.)

458
Jochens, Jenny M., 1986: Consent in Marriage: Old Norse Law, Life, and Literature. In Scandinavian
Studies 58.2. Pp. 14276.
Johnsen, Oscar A., 1916: Frigerar-saga: en kildekritisk underskelse. Kristiania.
Lindblad, Gustaf, 1963: Introduction. In Bergsbk. Perg. Fol. Nr. 1 in the Royal Library, Stockholm.
Copenhagen. (Early Icelandic Manuscripts in Facsimile, vol. 5.)
Lnnroth, Lars, 1976: Ideology and Structure in Heimskringla. In Parergon. Pp. 1629.
Loth, Agnete, 1964: Introduction. In Thomasskinna. Gl. Kgl. Saml. 1008 fol. in the Royal Library,
Copenhagen. Copenhagen. (Early Icelandic Manuscripts in Facsimile, vol. 6.)
lfs saga hins helga: die Legendarische Saga ber Olaf den Heiligen (MS Delagard. saml. nr.
811). Ed. and German trans. by A. Heinrichs et al. 1982. Heidelberg.
Power, Rosemary, 1985: Le Lai de Lanval and Helga ttr rissonar. In Bibliotheca Arnamagnana
38. Pp. 158161. (Opuscula 8.)
Schottmann, Hans, 1994: Frigerarsaga. In Studien zum Altgermanischen. Ed. by H. Uecker. Berlin.
Pp. 53953.
Sverrir Tmasson, 2006: Konungasgur. In slensk bkmenntasaga I. Ed. by V. lason, 2nd ed. Pp.
358401. Reykjavk.

459
 Den hviske Bsi. Herraus ok Bsa saga i genrernas grns-
land
Karl G. Johansson, Dept. for Linguistic and Scandinavian Studies, University of Oslo, Nor-
way
Herraus ok Bsa saga r inte den frsta saga vi skulle tnka p som hvisk. Den rknas tra-
ditionellt, alltsedan C.C. Rafn inkluderade den i sin samling frn 182930, till de s kallade
fornaldarsgur norrlanda. Primrt torde den kunna karakteriseras som en quest, allts en typ
av narrativ som r vanlig bde i versatta riddarasgur, i fornaldarsgur och i de inhemska
riddarasgur frn Island. En traditionell definition av en fornaldarsaga r att den ska utspela
sig i ett forntida Skandinavien och att innehllet ska framst som fantastiskt med hjltar,
hemska odjur och grna en kvinna som mlet fr hjltens handling. I innehllet finns ocks
mycket av detta i Herraus ok Bsa saga. Men det finns ocks mycket som ligger markant
utanfr definitionen och som snarare knyter verket bde till de versatta riddarasgur och de
inhemska riddarasgur. Herraus ok Bsa saga kommer drmed i all sin atypiskhet att fram-
st som en god representant fr utvecklingen av texter med ett fantastiskt innehll, men ocks
fr bruket av texterna under den lnga perioden frn 1200-talets introduktion av europeiska
verk till det sena 1400-talets tradering och kreativitet i islndska samlingshandskrifter, dr
bde inhemska traditioner och europeiska trender gr sig gllande. ven om vi i dag rent ge-
nerellt r tveksamma till en indelning i tre typer av sagor, framstr Herraus ok Bsa saga
srskilt som en hybrid mellan dem, och dessutom med drag som mer pminner om senare
traditioner fr det som karakteriseras som folksagor. Det har framfrts av Vsteinn lason
att sagan r en parodi p dessa genrer och att tankarna leds till den europeiska renssansen
(Vsteinn lason 1994). Sverrir Tmasson har dessutom pekat p mjlig kontakt med franska
fabliaux (1989:218ff. och 1996:66ff.). Det r mycket i sagans framstllning som knyter an till
framvxten av en helt ny form fr litteratur p kontinenten, representerad av frfattare som
tydligare markerar sin nrvaro i berttelsen, som t.ex. Cervantes och Rabelais. Vsteinn la-
son ppekar likheten med de tv europeiska giganterna. Han placerar sagan i grnslandet mel-
lan de tre genrerna:

The frame of reference of the saga is not only the fornaldarsaga, which is parodied, but also the
chivalric romance and probably also its Icelandic imitations, the Mrchensagas. (Vsteinn la-
son 1994:121)

Mnga av de iakttagelser som presenteras i det fljande knyter an till de synpunkter som
framfrts av de nmnda forskarna, ven nr jag inte hnvisar explicit till deras framstllning-
ar. De frgor jag vill diskutera gr det emellertid ndvndigt att ter presentera de grund-
lggande tendenserna i verket. Det som str i centrum hr blir sagans tydliga litterra karaktr;
det rr sig entydigt om en skriftlig produkt tillkommen i en milj dr skriftkulturen har ett fast
grepp, och dr den som sammanstllt verket explicit arbetar med litterra medel fr att struk-
turera sin berttelse. I utforskningen av de processer som rr litterariseringen av norrn kultur
och textualiseringen av denna kulturs narrativa traditioner ska Herraus saga ok Bsa drmed
placeras i ett sent skede.
Herraus ok Bsa saga finns i tv versioner. I det fljande r det den ldre versionen som
r i fokus. Denna freligger i tre manuskript frn senare delen av 1400-talet. De aktuella ma-
nuskripten r AM 586 4to, av utgivaren, Otto L. Jiriczek, betecknat som A och vanligtvis da-
terat till ca 14501500, AM 343 a 4to, Jiriczeks manuskript B, ven det daterat till sent p
1400-talet, och AM 577 4to, som av Jiriczek benmns D och dateras till 14501500. Ytterli-
gare ett tidigt manuskript, AM 510 4to, benmnt C av Jiriczek och daterat till ca 1550 ska

460
nmnas hr. Dessutom finns sagan i ett flertal manuskript frn 1600- och 1700-talet som inte
kommer att vara aktuella fr min diskussion. I min fljande framstllning kommer verket
Herraus ok Bsa saga att diskuteras nrmare, dock hela tiden med sagans funktion i sin di-
rekta kontext i de tre medeltida handskrifterna in mente. Centralt fr mitt resonemang blir
genomgende synen p det medeltida manuskriptet som kommunikativ handling. Handskrif-
ten AM 343 a 4to (B) innehller fljande texter:

orsteins saga bjarmagns

Samsons saga fagra
Egils saga einhenda ok smundar bersekjabana
Flres konungs saga
Vilhjlms saga sjs
Yngvars saga vifrla
Ketils saga hngs
Grms saga loinkinna
rvar-Odds saga
ns saga bogsveigis
Saulus saga ok Nikanors
Hlfdanar saga Eysteinssonar
Herraus ok Bsa saga
Vilmundar saga viutan
Meistara Perus saga

Som framgr av denna sammanstllning bestr manuskriptets samling av texter dels av det vi
skulle karakterisera som fornaldarsgur, som t.ex. de fyra Hrafnistu-sgur, dels av riddara-
sgur som torde ha tillkommit direkt p folksprket p Island. Meistara Perus saga faller del-
vis utanfr dessa bda kategorier och har traditionellt betraktats som ett exemplum. Till jmf-
relse kan innehllet i de tv andra 1400-talsmanuskripten stllas upp, frst AM 586 4to:

AM 586 4to (A)

Af rimr kumpnum
Af rimr jfum Danmrk
Af brytja ok bnda
Af meistara Pero ok hans leikum
Af Vilhjlmi bastari ok sonum hans
Fr ferum Roberts ok hans manna
Flres saga konungs ok sona hans
Herraus ok Bsa saga
Vilmundar saga viutan
Hlfdanar saga Eysteinssonar
Hrings saga ok Tryggva
rar saga hreu
Krka-Refs saga
smundar saga kappabana

Handskriften inleds med ett antal exempla och texter som omhandlar mnen som inte direkt
har anknytning till det nordiska. Det r frst med Herraus ok Bsa saga som handlingen
frlggs till Norden. De fljande texterna i samlingen har en inriktning p nordiska frhllan-
den, mjligen med undantag fr Vilmundar saga viutan som ju emellertid r knuten till Her-
raus ok Bsa saga genom slktskapet mellan Bsi och Vilmundr. Hr skulle man allts kun-

461
na argumentera att den som sammanstllt handskriften har stllt verk med en europeisk eller
mer exotisk geografisk placering mot verk som relaterar till en mer knd geografi.
Den tredje medeltida handskriften som innehller Herraus ok Bsa saga r annorlunda till
sammansttningen, samtidigt som mnga av texterna r gemensamma:

AM 577 4to (D)

Egils saga einhenda
Vilhjlms saga sjs
Herraus ok Bsa saga
Af bndasyni nokkrum kngsgari
orsteins saga bjarmagns
Vilmundar saga viutan
Flres konungs saga og sona hans

Som framgr r det allts genomgende samma textverk som samlas i de tre handskrifterna.
Intressant r t.ex. att Vilmundar saga viutan, vars huvudperson, Vilmundr, omnmns som
sonson till Bsi, inte bara frekommer i alla tre manuskripten, men i tv av dem r placerad
som en direkt fortsttning p Herraus ok Bsa saga. Denna logiska fljd fr texterna i tv av
de tre medeltida manuskripten ska emellertid inte vertolkas; en nrmare analys av de tre ma-
nuskriptens sammansttning och hur texterna i dem samspelar vore av nden fr att kunna dra
mer bestmda slutsatser om sammanstllarens intention. Redan i nulget kan det konstateras
att de som sammanstllt de tre manuskripten inte ndvndigtvis har gjort en skillnad mellan
fornaldarsgur norrlanda, t.ex. de fyra textverken om slkten frn Hrafnista i AM 343 a 4to
eller smundar saga kappabana i AM 586 4to, och de inhemska riddarasgur. Inte heller
tycks det ha varit dem frmmande att sammanstlla dessa tv typer av berttelser med det som
vi vanligtvis benmner exempla, vilket framgr av att Meistara Perus saga inkluderats i AM
343 4to och att Af meistara Pero ok hans leikum ingr i AM 586 to.
Det finns ingen omfattande diskussion om tillkomsten av Herraus ok Bsa saga. Den tra-
ditionella dateringen r till slutet av 1300-talet, en datering som frmst tycks grunda sig p att
det ldsta textvittnet freligger i en handskrift som tidigare har daterats till ca 1400. Jon Gun-
nar Jrgensen argumenterar fr att sagan har tillkommit under pverkan frn den tidiga itali-
enska renssansen. Han vill datera verket till ett gott stycke in i 1400-talet, lite skmtsamt
freslr han 1474 som ett lmpligt r. Drmed vill han ocks flytta fram dateringen av de ak-
tuella handskrifterna till helt i slutet av 1400-talet (Jrgensen 1997:103f.).
Ett drag i verket som ofta omtalas r sagans skildringar av erotiska scener. Dessa r vl i
dag inte srskilt uppsiktsvckande och r inte heller helt enastende i den norrna litteraturen.
Men de frknippas knappast med den hviska litteraturen, kanske med reservation fr Mttuls
saga. Fr tidigare generationer i forskningshistorien frn 1800-talet och framt har sagan
emellertid betraktats som bde frck och av mindre intresse fr studiet av sagalitteraturen. Det
finns anledning att ifrgastta bda dessa vrderingar. De erotiska scenerna pminner om
folksagornas omskrivningar fr sexuella relationer. Bsi och de kvinnor han med en modern
omskrivning sover med, i syfte att erhlla viktig information, omtalar den sexuella relatio-
nen i klassiska omskrivningar som hsten som leds till kllan eller hsten som ska stllas i
spiltan. Det hela utspelar sig utver detta i lsarens eller hrarens fantasi. Nr man har lst
sagan i lsgrupp ngra gnger har man ftt inblickar i hur texten fungerar och hur ven en
modern publik blir upprymd och lite rd om kinderna nr fantasin stts i rrelse. S har sagan
ju troligen ocks fungerat i en upplsningssituation i 1400-talets Island. Men det brande i
Herraus ok Bsa saga r nd det uppdrag som Herraur och Bsi ftt av Herraus far,
kungen i Gautland, och de tv drefter fljande utfrderna fr att ervra de tv kvinnor som
hjltarna slutligen gifter sig med. De erotiska scenerna fungerar hr som i folksagan med sina

462
tre episoder som stegvis hjer spnningen fram till dess att hjltarna kan genomfra sitt upp-
drag.
Den som komponerade sagan har tydligt arbetat inom en skriftlig, litterr tradition. Det blir
drmed intressant att nrmare diskutera hur beteckningen saga anvnds i verket, dels om ver-
ket i sig sjlv, dels i referenser till andra berttelser. Om vi brjar med den frsta typen av
omnmnanden r episoden dr Bsis fostermor Busla sgs ha erbjudit Bsi att lra honom
galdrar intressant:

Hn bau Bsa at kenna honum galdra, en Bsi sagizt eigi vilja, at at vri skrifat sgu hans,
at hann ynni nkkurn hlut me sleitun, ann sem honum skyldi me kallmennzku telja. (Jiriczek
1893:6 f.)

Hr lter allts den som sammanstllde sagan Bsi uttala sig om en skriven saga om honom
sjlv. Sammanstllaren av sagan leker allts med sin egen roll som berttare. Hnvisningen
till en tnkt, skriven saga om hjlten indikerar tydligt att sagans samtid hr frhller sig till en
tradition fr skrivna texter. Sammanstllaren upprtthller ocks sin roll som skrivande bert-
tare nr sagan avslutas1 med orden:

ok lkum vr hr n sgu Bgu-Bsa. (Jiriczek 1893:63)

Drmed blir referenser inom verket till andra berttelser med benmningen saga intressanta.
Nr sammanstllaren placerar Herraur och Bsi i tiden fr slaget p Brvellir sker det med
en hnvisning till en annan berttelse:

[] var settr tmi til bardagans Brvllum, er mestr hefir verit Norrlndum, sem segir
sgu Sigurar hrings, fur Ragnars lobrkar. (Jiriczek 1893:33)

Det r oskert vilket verk som avses med *Sigurar saga hrings, men det kan exempelvis
vara en kompilation av den typ som freligger i Heimskringla eller i en norrn version av
Skjldunga saga. En liknande referens ges i anslutning till omnmnandet av slaget till
*Haralds saga hilditanns, en berttelse som inte entydigt kan identifieras i det bevarade sa-
gamaterialet:

essi orrostu fell Haraldr konungr ok me honum fimtn konungar annars C, sem segir sgu
hans. (Jiriczek 1893:34)

Sammantaget indikerar bruket av benmningen saga hr att det fr sammanstllaren har hn-
visat till en skriven berttelse. Den skriftbaserade sagan har varit sjlvklar som referens och
sannolikt har ocks fenomenet saga fr samtiden varit textualiserat. En saga har drmed haft
en given inledning, givna strukturer och ett tydligt slut. Inledningen i sagan:

Hringr hefir konungr heitit, er r fyrir Eystra-Gautlandi [], (Jiriczek 1893:3)

blir drmed vntad; med liknande fraser inleds ju de flesta bevarade sagor, oavsett vilken gen-
re vi placerar dem i. Avslutningen, som citerats ovan och vars varianter diskuteras i det fl-
jande, kan ocks sgas motsvara vad vi skulle frvnta oss, och p mnga stt pminner den
dessutom om hur mnga av 1400-talets inhemska riddarasgur avslutas.
Hr blir det emellertid relevant att se nrmare p de olika versionerna av textverket. Inled-
ning och avslutning utformas nmligen markant olika i de fyra versionerna AD. I Jiriczeks

1
Se emellertid nedan fr en diskussion av de olika versionernas inledning respektive avslutning av sagan.

463
utgva av Herraus ok Bsa saga utelmnas en liten prolog som freligger i versionerna A, B
och D. Denna prolog terges endast i en fotnot, men r ytterst intressant fr den som vill fr-
st verkets placering i traditionen.

essi saga hefzt eigi af lokleysi eirri, er ktir menn skrkva sr til skemtanar ok gamans me
frligum setningum, heldr sannar hn sik sjlf me rttum ttartlum ok fornum orzk-
vium, er menn hafa iuliga af eim hlutum er essu fintri eru skrifair. (Jiriczek 1893:3)

Det r mrkligt att Jiriczek hr har valt att placera prologen i en fotnot. Drmed avgr han ju,
att den inte ursprungligen har utgjort en del av textverket. Med tanke p att de tre medeltida
handskrifterna faktiskt innehller prologen, torde det ur ett textkritiskt perspektiv vara svrt
att hvda denna stndpunkt.2 Med detta sagt blir det intressant att diskutera prologens roll, och
dess nrmaste paralleller i sagalitteraturen.
Hela tonen i prologen r skmtsam. Den inledande meningen slr an med sin negation av
lokleysa ndlst prat som glada mnniskor skrkva ljuger eller diktar upp med frli-
gum setningum okunniga meningar som bara r till skemtanar ok gamans nje och gldje.
Markeringen blir s verdriven att effekten trots att den explicit ska stlla den fljande tex-
ten i ett serist ljus blir, att vi inte ska uppfatta sagan som allvarlig. Drefter gr den som
sammanstllde prologen en dygd av att styrka trovrdigheten genom att hnvisa till ttart-
lum genealogier och fornum orzkvium gamla talestt som finns iuliga rikligt i detta
fintri. Benmningen fintri tycks hr f en dubbeltydighet; det kan antingen syfta p de
korta berttelserna till moralisk uppbyggelse, dit t.ex. Meistara Perus saga rknas, eller till de
ventyr som riddarna i riddarasgur gnar sig t. Tvetydigheten frstrker ytterligare den
parodiska tonen i prologen.
Det som gller fr inledningen av textverket gller ven fr avslutningen. Hr r det delvis
strre avvikelser mellan de olika textvittnena. Jag vljer drfr att citera Jiriczeks variantappa-
rat in extenso.

[ok lkum] sv essa sgu af Herrau konungi ok Bgu-Bsa, er ruddi sr til rkis ok konungr
var Bjarmalandi b. Nach Bgu-Bsa add. AC: ok signi [h (hn? hr?) C] seta Busla alla
[ add. C] sem hr [hafa til hltt A, til hafa hltt C] lesit ok skrifat. [A fgt noch hinzu: er hr
nkkut til fengit er gott at gjrt A-M-E-N.]. (Jiriczek 1893:63)

De varianter av avslutningen som framkommer hr leder frst och frmst tankarna till den typ
av avslutningar som terfinns i de inhemska riddarasgur som exempelvis i Vilmundar saga
viutan eller Vilhjlms saga sjs. Oavsett om avslutningen r ursprunglig i textverket (ngot
som Jiriczek tycks anse nr han placerar den i variantapparaten, och allts inte behandlar av-
slutningen p linje med den utelmnade prologen) eller om den tillfogats i en senare fas i tra-
deringen, placerar den Herraus ok Bsa saga tydligt i samtidens textkultur, men kanske ock-
s i dess upplsningskultur. Hr liksom i liknande epiloger i andra verk frn 1400-talets andra
hlft tilltalas ett kollektiv, de som til hafa hltt C] lesit ok skrifat, eller som i A ven hr
nkkut til fengit er gott at gjrt. Sammantaget tycks de hr epilogerna antyda en uppls-
ningssituation, dr upplsaren avslutar sin framstllning med en hlsning som omfattar dem
som lyssnat, den som sammanstllt texten och honom sjlv, den som lst. Vi skulle ocks
grna vilja veta mer om vad nkkut til fengit er gott at gjrt har hnsyftat till; r det andra
som kommenterat upplsningen eller kanske rentav deltagit i framstllningen t.ex. med tillrop
eller upptrdanden?

2
Sverrir Tmasson (1988:374) kommer till samma slutsats, att ingenting talar emot att se prologen som en del av
verket redan frn brjan.

464
 Om sagan explicit framstr som del av en skriftkultur, bekrftas detta intryck om man ser
nrmare p hur berttaren str fram i texten. Sagans struktur kan kort indelas i en inledande
del, dr de tv hjltarna presenteras och dr konflikten med kung Hringr leder fram till den
frsta frden fr att hmta ett gammsegg gamgg. Detta leder till att de tv hjltarna befriar
bjarmalandskungen Gomundrs syster, Hleir, frn det hedniska hov dr hon hlls fngen.
Hon frs till Gautland fr att bli Herraurs hustru, varefter de tv hjltarna snds till striden p
Brvellir. Nu kommer bjarmerna till Gautland. De ddar kung Hringr och fr med sig Hleir
hem till Bjarmaland. Nr Herraur och Bsi tervnder frn slaget fr de besked om vad som
skett och ger sig ut p en andra frd fr att terhmta Hleir och ta hmnd. Detta lyckas, var-
efter Bsi inleder en tredje quest fr att rva bort kungadottern Edda, ven detta en lyckosam
tur. Nr sagan betraktas med utgngspunkt i de tre frderna framstr berttaren sparsamt med
egna kommentarer eller lsangivelser, men han gr sig konsekvent nrvarande i samband med
strukturella grnser. I det inledande partiet som leder fram till den frsta frden hller bertta-
ren sig i bakgrunden, med undantag fr den ovan nmnda kommentaren om Bsi och Busla.
Det r frst efter att det inledande uppdraget r genomfrt att den frsta strukturella kommen-
taren frekommer. Det r nr de tv hjltarna terkommer frn Brvelleir:

hfu orit au umskipti Gautland, sem sar mun sagt vera, mean eir vru burtu.
(Jiriczek 1893:34)

Berttaren anger hrmed att han avser att bryta den kronologiska linjen i sin berttelse, och att
han tnker terkomma till denna del av sin framstllning. Direkt efter ovanstende kommentar
fortstter han:

N af v, at eigi m senn segja meir en eitt, verr n at at skra, sem fyrr hefir til borit
sgunni, ok er ar n til at taka fyst, []. (Jiriczek 1893:34)

Nu tervnder berttaren allts till den tid d Hleir hade frts bort av hovgyjan Kolfrosta.
Han berttar hur bjarmerna brjar leta efter henne, hur de fr reda p var hon befinner sig och
hur de besegrar kung Hringr. Nr berttaren tervnder till sina tv huvudpersoner sker det
med orden:

ar er n til mls at taka. (Jiriczek 1893:37)

Detta r en fras som terfinns i mnga av de norrna sagatexterna. Hnvisningen till olika
tidsskikt i berttelsen torde hr snarast indikera att det rr sig om en fras som helt hr hemma
i en skriftkultur.
Nr hjltarna utfrt sin hmnd och terhmtat Hleir, beskriver berttelsen bjarmernas fr-
beredelser fr att frflja dem, fr att drefter tervnda till sina huvudpersoner:

ok ltum n bazt, en vkjum sgunni aptr til eirra kumpna []. (Jiriczek 1893:49)

Bsi fr nu an i den tredje och sista frden fr att rva bort kungadottern Edda. Nr denna
uppgift r avslutad fr berttaren sina lsare eller hrare ter till bjarmerna dr Hrrekr och
Siggeir r klara med sina frberedelser:

at byrjazt n, sem eir brr hfu fullbit sitt li. (Jiriczek 1893:55)

Om vi vnder oss till beskrivningen av de tv kumpanerna finns det en intressant kontrast

mellan dem som leder tankarna bde till Egils saga Skalla-Grmssonar och ns saga bogsvei-
gis, i det att de beskrivs som varandras motsatser. Hans-Peter Naumann och Vsteinn lason

465
argumenterar fr att frfattaren har valt tv hjltar till sin historia fr att kunna kontrastera
dem och drmed framhva de hviska idealen (Naumann 1978:46; Vsteinn lason
1994:116). Framstllningen av de tv hjltarna ger dock enligt min mening inte std fr att
argumentera fr underliggande hviska ideal i sagan; snarare framstr den som en parodi ock-
s p dessa ideal. Herraur framstlls som en ljus och omtyckt person:

hann var mikill vexti ok frir snum, sterkr at afli ok vel at rttum binn, sv at fir menn
mttu vi hann jafnazt. Hann var vinsll af llum mnnum, en ekki hafi hann mikit strki af
fer snum. (Jiriczek 1893:4)

Bsi skildras dremot som mrk och hotfull:

hann var mikill vexti ok sterkr at afli, dkklitar ok ekki mjk frir, ok lkr mur sinni at skap-
lyndi ok skpun; ktr var hann ok keskimll ok rfylginn v, sem hann tk upp, ok eigi mjk
fyrirleitinn vi hvern, sem hann tti. (Jiriczek 1893:6)

Hr finner vi allts en motsvarighet i de tv brdraparen rlfr och Skalla-Grmr respektive

rlfr och Egill i Egils saga Skalla-Grmssonar, men ocks i brderna rir och n i ns
saga bogsveigis. Det r d intressant att notera att den senare sagan har diskuterats som i n-
gon mn byggd p den frra (Viar Hreinsson 1990). Vsteinn lason ppekar att Herraus
ok Bsa saga skiljer sig frn Egils saga Skalla-Grmssonar och ns saga bogsveigis, vilket
gr att den sistnmnda inte uppvisar samma entydigt nrvarande berttare i texten. Han kon-
staterar:

ns saga is certainly anti-royalist with a vengeance, and in contrast to Bsa saga there is no
trace of influence from the world of chivalry. An influence from the early Egils saga Skalla-
grimssonar, is clearly felt, and thus the literary and ideological preconditions of an authorial
presence are very different from those of Bsa saga. (Vsteinn lason 1994:128, fotnot 30)

Nr Herraus ok Bsa saga trots detta tycks framstlla sina tv hjltar p ett liknande stt
som de tv andra sagorna, kan detta tyda p att den som sammanstllde denna saga har haft
tminstone en av dem som frebild. I mitt fortsatta arbete med Herraus ok Bsa saga blir det
drfr ndvndigt att genomfra en parallell analys av skildringen av de fyra brdraparen; i
handskriftskulturen har de tre verken levt i en gemensam tradering dr grnser mellan genrer
knappast har varit avgrande. I det fljande begrnsar jag emellertid min lsning till exemplen
frn Herraus ok Bsa saga.
De tv kumpanernas agerande i frhllande till de tv kvinnorna Hleir och Edda kan
eventuellt ge en indikation om hur sagan frhller sig till den hviska litteratur som tminsto-
ne torde utgra dess bakgrund. Hr skildras hjltarnas agerande p ett stt som i mycket mot-
svarar beskrivningen av dem som terges ovan. Herraur erbjuder Hleir att flja med till
Gautland och gr klart att han inte tnker ska godknnande hos hennes slkt:

Eigi mun ek til eirra giptingar leita, segir Herraur, ok vil ek hr engan undandrtt hafa, v
ek ikkjumzt ngu r varboinn, ok skal leysa ik sem r. (Jiriczek 1893:32)

Hr finns det en liten, men signifikant skillnad i version D (representerad av AM 577 4to) av
sagan:

Eigi mun ek leita giptinga til frnda inna, en ngvan vil ek undandrtt essu mli, en eigi
skal nauoka ik til kaupa nkkra, v ek ikkjumzt r eigi varboinn, ok skal leysa ik sem
r. (Jiriczek 1893:32; min kursivering)

466
D-versionen understryker allts ytterligare det hviska exemplet som Herraur ger. Det r
ocks intressant att notera Hleirs svar:

Eigi veit ek ann mann, segir hn, at ek vil heldr eiga en ik, af eim, sem ek hefi st. (Jiriczek
1893:32; min kursivering)

I D-versionen svarar Hleir med mundi heldr kjsa fr heldr eiga vilket vl n tydligare
understryker att hon har mjlighet att vlja, ven i den situation hon befinner sig.
Bsi agerar emellertid annorlunda i sin frsta kontakt med Edda:

at er n sem takazt vill, segir Bsi, ok gjr n hvrt er vilt, at fara me mr viljug, ea geri
ek skyndibrlaup til n hr skginum. (Jiriczek 1893:53 f.)

Han beter sig inte p ett passande, hviskt stt, utan far i stllet med hot om att ta henne med
vld. Det skildras hur han och Herraur ddar eunucken som vaktar Edda, p ett stt som fr
tankarna till 1000 och en natt, och drefter setti Bsi konungsdttur handlegg sr (Jiriczek
1893:54). Hr mste vi frga oss om skillnaden mellan de tv huvudpersonerna r intentionell,
allts om den som sammanstllt berttelsen medvetet har stllt upp dem som kontraster, den
ljuse och vnlige Herraur som agerar hviskt i frhllande till Hleir stlls upp emot den
mrke och mindre vnslle Bsi och hans stt mot Edda. I det senare fallet rr det sig ju ex-
plicit om att kvinnan bortfrs med vld och under hot, medan det frsta vl ger ett gott exem-
pel till efterlevnad. Hr kan det vara vrt att ocks tervnda till Bsis tre erotiska eskapader;
en av dem leder ju explicit till att bondflickan fder en frilloson, Svai, som presenteras som
far till Vilmundr viutan, en hjlte som terkommer i den saga som fljer efter Herraus ok
Bsa saga i handskrifterna AM 343 a 4to och AM 856 4to, Vilmundar saga viutan. Bsi r
allts inte ngot gott exempel p en hvisk riddare och jag har svrt att hlla med Vsteinn
lason nr han konstaterar:

However, it is important to realize that Bsis vulgarity is superficial; he has in him the makings
of a nobleman. (Vsteinn lason 1994:117)

Snarare framstr Bsi fr mig som en uppkomling som fr framgng trots sin vulgra framto-
ning, inte p grund av att han har en underliggande hvisk kapacitet. Hr skiljer Bsi sig ock-
s markant frn Rabelais eller Cervantes hjltar; de representerar ju trots allt ett slags adelskap
ven om just riddargenren parodieras. Den som sammanstllde Herraus ok Bsa saga paro-
dierar drmed enligt min mening inte bara de aktuella genrerna men ocks den ideologi som
ligger bakom den hviska litteraturen.
De ovan presenterade iakttagelserna talar inte fr att placera Herraus ok Bsa saga bland
fornaldarsgur. Snarare r det mnga drag som pekar i riktning av de senare, inhemska ridda-
rasgur. Med tanke p den diskussion som frts under senare r om medeltidens genrer, och
fr de islndska sagornas vidkommande om hybrider, torde emellertid en placering av Her-
raus ok Bsa saga i den senare kategorin vara lika missledande. Kanske vill jag inte g s
lngt som till att avfra Herraus ok Bsa saga frn sagalitteraturen (se Jrgensen 1997:103),
men det r definitivt dags att sluta se den som en fornaldarsaga. Det r tid att vi lmnar de
lsta genrebegreppen och ser det enskilda verket som en kommunikativ handling. Ett nsta
steg blir d att fokusera det individuella textvittnet och dess kontext i den textbrare dr det
freligger; det torde framg av ovanstende kortfattade framstllning att de tre textvittnena,
A,B och D, dels framstr med olika formuleringar i fr vr frstelse viktiga sammanhang,
dels fungerar i samspel med andra texter i de manuskript dr de freligger. En analys av de
olika textvittnena i relation till kontexten i de manuskript dr de freligger tycks i dag vara en

467
mer relevant utmaning n att hlla fast det traditionella perspektivet p verket som en fast och
ofrnderlig enhet. I min framstllning hr har just verket varit i fokus, men i ett nsta steg
ska perspektivet flyttas s att handskriftskontexten kommer i centrum.
Bsi framstr inte som ngon hvisk hjlte i Herraus ok Bsa saga. Snarare r han att be-
trakta som en skmtsam kommentar till tidigare hjltar i den norrna litteraturen. Den fule,
men kanske nd exemplariske hjlten, i form av den islndske skalden Egill Skalla-Grmsson
eller den frn fornaldarsgur hmtade n, fr hr en annan beskrivning. Bsi uppfr sig inte
kurteist, men kommer sig trots detta fram i vrlden, frn att vara son till en bonde, blir han till
slut kung i Bjarmaland. Signifikativt r nog drmed sttet frfattaren lter honom framst nr
han vnder sig till bjarmena:

[] beiddi Bsi sr ar vitku ok taldi at til, at Edda tti land alt eftir fur sinn, er n var
orin eiginkona hans, ok segizt hann sv helzt mega bta landzmnnum ann mannskaa, sem
eir hfu af honum fengit, at vera konungr yfir eim ok styrkja me lgum ok rttarbtum,
ok me v at eir vru hfingjalausir, s eir ngvan sinn kost vnna, en taka hann sr til
konungs []. (Jiriczek 1883:61)

Hr r det uppkomlingen som gr sig bemrkt, den som slagit sig fram till sin position, fak-
tiskt genom att vlla den tidigare kungens dd. Man mste frga sig hur en dansk kungamakt
mot slutet av 1400-talet skulle ha uppfattat framstllningen av kungen som uppkomling, som
erbjuder sig att styra med lgum ok rttarbtum; med tanke p 1400-talets politiska historia
med flera tronpretendenter, kunde Bsi nrmast kunna uppfattas som det rent motsatta av en
hvisk kung, en usurpator av makten med std endast i r styrka och pgngsmod.

Litteratur
Jiriczek, Otto L. (utg.). 1893. Bsa-saga in zwei Fassungen nebst Proben aus den Bsa-Rmur. Strass-
burg.
Jrgensen, Jon Gunnar. 1997. Bsi ht annarr son eirra. I: Bergljt S. Kristjnsdttir & al. (red.),
Frejas psalter. En psalter i 40 afdelinger til brug for Jonna Louis-Jensen. Kbenhavn. S. 101106.
Naumann, Hans-Peter. 1978. Die Abenteursaga. Zu einer Sptform altislndischer Erzhlkunst.
Skandinavistik 8. S. 4155.
Sverrir Tmasson. 1988. Formlar slenskra sagnaritara mildum. Rannskn bkmenntahefar.
Reykjavk.
Sverrir Tmasson. 1989. Hugleiingar um horfna bkmenntagrein. Tmarit Mls og menningar 50:2.
S. 211226.
Sverrir Tmasson (utg.). 1996. Bsa saga og Herraus. Reykjavk.
Vsteinn lason. 1994. The marvellous North and authorial presence in the Icelandic fornaldarsaga.
Contexts of Pre-Novel Narrative. I: R. Eriksen (red.), The European Tradition. (Approaches to
Semiotics 114. Ed. T.A. Sebeok, R. Posner, A. Rey) Berlin, New York. S. 101134.
Viar Hreinsson. 1990. Hetjur og ffl r Hrafnistu. Tmarit Mls og menningar 51:2. S. 4152.

468
 Birners edition of the Frijfs saga ins frkna1
Vera Johanterwage, Ins titut fr Skandinavis tik, Johann Wolfgang Goethe-Universitt Frank-
furt/Main, Germany
In the 19th c. Frijfr the Brave was one of the best known heroes of old times, not least
thanks to Esaias Tegnrs romantic Frithiofs saga which was immensely popular not only in
Sweden, but also in Germany, Britain and the United States. In contrast to the wide and en-
thusiastic reception the Frijfs saga ins frkna received up to the beginning of the 20th c.,
little attention has been paid to the text in the last decades. In this paper I would like to show
that interesting questions remain to be discussed when it comes to the relationship between
the mss. and the early editions.
It is a commonplace that Tegnr became acquainted with the saga via Birners edition in
Nordiska Kmpa dater (1737). Already before Birners editio princeps the Frijfs saga
had been reproduced a number of times in Sweden. Despite the fact that Frijfrs adven-
tures mainly take place in Norway, it was first and foremost the Swedish historical interest of
the 17th c. that resulted in an intense production of copies. As far as I can see, Birners text
has not been compared thoroughly with the Swedish mss. (and that includes his exemplar!) or
with Tegnrs adaptation. In this context Birners Swedish and Latin translations of the saga
are of particular interest. Also the translations have been paid little attention (the only detailed
study I am aware of is an article by Hirvonen 1987). I have started studying Birners edition
just recently and would like to present a few aspects which I intend to analyse more closely in
the near future. Moreover I shall address questions about Birners Swedish rendering of the
Frijfs saga.

Manuscripts and editions of the Frijfs saga

To start with, some remarks on the surviving mss. and the editions of the saga are called for.
As is well known, two versions of the Frijfs saga ins frkna exist, furthermore a render-
ing in rmur. Up to the end of the 19th c. the longer version of the Frijfs saga was consid-
ered the original text. First in 1886 it was realized that the opposite assumption is correct
(Valdimar smundarson 1886:V). While Klbing and Calaminus ignored Valdimar s-
mundarsons edition and formli (Klbing 1887; Calaminus 1887), Falk confirmed his find-
ings and expressed his disbelief in Calaminus theories, among which especially the old idea
that the saga was based on an archaic Norwegian local tradition, was met with strong opposi-
tion (cf. Calaminus 1887:5162; Falk 1890:60). Nowadays it is undisputed that the Frijfs
saga does not contain any historical nucleus. Already the shorter version was written in Ice-
land and is considered a fornaldarsaga, even if its language and style might seem simple and
archaic compared to other texts of that genre (cf. Falk 1890:9598; Wenz 1914:LXXXII
CIX, CXXXIII).
The shorter version has come down to us in three mss., while the rmur are transmitted in
one ms. only (cf. Larsson 1893:XXVIXXXII). It seems that both the shorter version of the
saga and the rmur were used as a source for the longer version (cf. Wenz 1914:XXIXXVI).
The longer version of the saga has undoubtedly been more popular in the 17th and 18th cen-
turies, as the number of mss. indicate (for a description of the mss. see Larsson 1893:I
XXVI). The fact that the longer version is clearly influenced by the riddarasgur might have
contributed to the popularity. The differences between the two versions cannot be discussed in

1
I would like to thank Beatrice la Farge for helpful discussions and for generously lending me her private copy
of Birners Kmpa Dater.

469
detail here, but it is worth mentioning that a strategy lay behind the changes in the younger
version:

Andere Erweiterungen zeigen deutlich die Absicht des berarbeiters, die Einzelabschnitte, in
die die Saga zerfllt, mglichst in sich abzurunden und auszugestalten, die Hauptszenen durch
Erweiterung der Motive wirksamer durchzufhren, vor allem die Redeszenen auszufllen, den
Bericht durch Zufgung naheliegender Einzelbestimmungen zu vervollstndigen, schliesslich
den Charakter einzelner Personen schrfer hervortreten zu lassen. (Wenz 1914:XLVI; for a dis-
cussion of the differences between the versions see XLIIILXIII).

Of course it is not improbable at all that chance best explains why the longer version was cop-
ied extensively in Sweden: Ultimately, the Swedish mss. all derive from one Icelandic ms.
(Holm papp. 17 4) which ended up in Sweden somehow.

Swedish manuscripts of the Frijfs saga

Holm papp. 17 4 16401671 Icelandic
Holm papp. 56 fol. ca. 1685 copy of Holm papp. 17 4;
space for Swedish translation (not carried out)
Holm papp. 52 4 ca. 1685 Swedish translation
UppsUB R 704 4 ca. 1685 copy of Holm papp. 52 4 (containing corrections);
Swedish translation of the first chapter
Holm papp. 30 4 ca. 1685 copy of Holm papp. 56 fol.
Sfstaholmssaml (SRA) I Papp. 6 18th c. Swedish translation
Engestr B III, 1, 19 ca. 1820 excerpt; summarizing Swedish translation

The edition of Birners text is based on Holm papp. 56 fol. Apart from the first two leaves
which were written by Helgi lafsson, the entire codex was written by Arngrmur Jnsson
while he was working at Antikvitetskollegiet (16831691); in total the codex contains 13 sa-
gas (cf. Gdel 18971900:111). It was meant to offer Swedish translations alongside the Old
Norse texts, but the translation was not carried out and only the inner column has been filled.
The text of the Frijfs saga in Holm papp. 56 fol. follows its exemplar, Holm papp. 17 4,
closely, and yet has its own chapter division and adds chapter titles (cf. Larsson, 1893:VII
VIII).

Birners Nordiska Kmpa Dater

Before discussing Birners edition and Swedish translation of the Frijfs saga some gen-
eral remarks about the Nordiska Kmpa Dater need to be made. The work consists of 17 parts,
of which the first are a preface and a genealogy of the kings of Sweden. What follows are 15
Old Norse texts accompanied by a Swedish and a Latin translation: Norriges Upfindelse (i.e.
Fr Fornjti ok hans ttmnnum), Rimen om Karl och Grim (Rmur af Karl ok Grmi), Rolf
Krakes Saga (Hrlfs saga kraka ok kappa hans), Fridthiofs Saga (Frijfs saga ins frkna),
Alfs Saga (Hlfs saga ok Hlfsrekka), Romunds Saga (Hrmundar saga Gripssonar), Halfdan
Branas Fostersons Saga (Hlfdanar saga Brnufstra), Sorles Saga (Srla saga sterka),
Halfdan stenssons2 Saga (Hlfdanar saga Eysteinssonar), Samsons Saga (Samsons saga
fagra), Wolsunga Saga (Vlsunga saga), Ragnars Saga (Ragnars saga lobrkar), Ans Saga
(ns saga bogsveigis), Norna Gests och Helge Thoris Sons Saga (Nornagests ttr/Helga
ttr orissonar) and Thorstens Saga (orsteins ttr bjarmagns).

2
In the edition the umlaute are printed with the superscript letter e. In my quotations I make use of the modern
graphemes and . Also other special letter forms such as the tall s are not reproduced.

470
 As will be discussed below, Birner was mainly interested in the language of these texts.
Apart from the linguistic side of the matter, there are other details which easily explain the
fascination exerted by the Frijfs saga:

Neben der thematischen Verwandschaft [sic] mit den schon lange bekannten orsteins saga
Vkingssonar, Gautreks saga und Sturlaugs saga starfsama war fr die schwedischen Gelehrten
wahrscheinlich auch die sagenhafte Gestalt des Knigs Hring, der ber Svj geherrscht ha-
ben soll und aus einer Vielzahl anderer berlieferungen schon bekannt war, ausschlaggebend.
Weiterhin findet sich in der Frijfs saga frkna die berlieferung eines Baldurkultes, die aus
heutiger Sicht zwar anachronistisch und unrealistisch anmutet, zur damaligen Zeit aber neu und
unbekannt war und die Sicht auf die heidnische Vorzeit betrchtlich zu erweitern schien. (Busch
2004:149)

While Birners edition was to become one of the major sources about king Hringr and Baldr
eventually, it was not granted easy access to its envisaged audience at first: The publication of
the Kmpa Dater caused serious disturbance among the Swedish nobles, which resulted in
censorship. Birner had the Kmpa Dater printed without having shown the preface to his
superiors at Antikvitetsarkivet. The preface contained a quotation by the historian Caspar Sag-
ittarius about the shortcomings of new nobility, a statement which could potentially be seen as
a critical remark directed to the young Swedish noble dynasties (cf. Bothius 1924:480). The
distribution of the edition was brought to a halt, the quire in question was removed and re-
placed by leaves containing an unproblematic praise of old age (cf. Busch 2004:136139).
Not only did Birners edition cause discomposure, it was also criticized for being hardly sci-
entific (cf. Bothius 1924:481). Eric Julius Birner (16961750), kanslist and, since 1719,
translator at Antikvitetsarkivet, was a true Rudbeckian and initiated continuous discussions
with colleagues (cf. Bothius 1924:478480). Contemporary sources clearly indicate that
Birners quarrelsome disposition caused his opponents to launch rather severe attacks on his
old-fashioned scientific writings: Gustav Benzelstierna (16871746), censor librorum, once
stated (with regard to the introduction to Birners De ortographia) that he would rather ex-
press his belief in Birners theories despite their improbability than be forced to rea d the
introduction once more (cf. Bothius 1924:481). Ironically, while Benzelstierna and other
contemporaries aiming at a more critical way of dealing with Old Norse sources are almost
forgotten nowadays, Birner was to become one of the best known collectors of Old Norse
sagas thanks to the vivid reception the Kmpa Dater.

Birners edition of the Frijfs saga

It has been noted that Birners reproduction of the Old Norse text follows his exemplar
closely in all respects (cf. Larsson 1893:IX; Busch 2004:152). Generally speaking this obser-
vation is correct, yet there are some instances where Birner modified the text.
In Holm papp. 56 fol. proper nouns, a number of other specific terms and some statements
are underlined. In most cases Birner highlights those terms and statements by italicizing
them, which has been commented upon as follows: Die in Papp. fol. nr 56. unterstrichenen
Wrter und Wortpassagen entsprechen den kursiv gedruckten Stellen in BIRNERS Aus-
gabe. (Busch 2004:152) In some cases, though, Birner does not highlight exactly the same
terms, while in others, he adds accentuations.
With regard to proper nouns, be it personal or place names, Birner is more systematic
than the writer of the exemplar. He aims at italicizing every proper noun when introducing it.
Thus, Sygna fylki (BiFri:1), Framnesi (BiFri:2) and Biorn and Asmunur (BiFri:3) are
italicized despite no underlining in the exemplar (cf. Holm 56, 179r, 180r).

471
 In a number of cases Birner chooses to highlight single expressions rather than entire
phrases containing these expressions. Thus, enn ess vil ek biia yckur at i hafit langvini
a sem ek hefi hapt (Holm 56, 180r; s pecial letter forms in Holm 56 a re not reproduced) is
rendered as enn ess vil ek biia yckur, at it hafi langvini a sem ek hefe hapt (BiFri:3).
In the Latin translation the highlighted passage is even longer than in the exemplar: ideoque
vos monitos volo, illos quam diutis sime amicos habeatis, quos ego habui (BiFri:3). In this
particular case Birner might have felt the need to highlight the Old Norse term, since he does
not give a verbum pro verbo rendering in his Swedish text: at j hafwen dem lnge til wenner
(BiFri:3). By highlighting one word only in the Old Norse text, that expression is given im-
portance and is more likely to catch the readers eye. In many cases the words highlighted
independently of the exemplar can be considered key words in terms of expressing specific
Norse concepts, such as fostbrur (BiFri:3; cf. Holm 56, 180r) or at skia veitslu (BiFri:4;
cf. Holm 56, 180v).
On the whole, Birners edition of the Old Norse text is characterized by more highlighting
than the exemplar. While in the exemplar the focus is on phrases and passages which are of
importance with regard to the sagas plot, Birner gives special emphasis to particular Old
Norse terms. As we shall see when studying the Swedish translation, Birner is eager to
spread knowledge about Old Norse vocabulary and specific concepts expressed by it. To my
mind the technique of highlighting serves exactly the same purpose: Personal and place names
as well as key words are meant to attract the readers attention.

Birners Swedish translation of the Frijfs saga

In his preface Birner gives some insight into his way of translating. Ideally, a translation
captures the geist of the original language, thus many of the existing translations of Old Norse
expressions within dictionaries and saga editions are to be criticized:

[] Sagor ro s beskaffade / at deras / hlst Latinska / uttolkning / sllan uttrycker Gtska

sprkets rtta snille och must / warande den sama mera smakande af en alt fr fri och prlach-
tig utlggning / n ntt och granlaga fwersttning / samt sledes mindre tjenlig fr dem / som
sjlfwa sprket sig bemgtiga wilja. (BiFre:5)

Birner explicitly wants his audience to get to know the Old Norse language. Completely in
line with the theories developed by Rudbeck and other Swedish historians in the 17th c., he is
convinced that Swedish is particularly well-suited for translating Old Norse texts. The edition
and the preface (and later writings) are meant to prove this to be true:

Til slika bewis komer och srdeles detta / at Swenskan n i dag hafwer fram fr Danskan en
gonstickande strre liket / med s wl den gamla som nyare Islndskan / det jag med denna
Sagoflocks fwersttning har nogsamt allom daga lagt. (BiFre:4)

Due to the close relationship between Icelandic and Swedish the use of Icelandic vocabulary
is not considered problematic. Accordlingly, the use of old Swedish words is conceived of as
a possibility to establish a (natural) connection between the two languages. Birners Swedish
translations abound with archaisms and Icelandisms (for an overview see Hirvonen
1987:110121). As has been pointed out alrea dy (cf. Hirvonen 1987:107), it is clearly stated
in the preface that the use of archaisms is programmatic (cf. Hirvonen 1987:107):

[] warande doch frskrad / mig hafwa s ansat / ej allenast Swenska uttolkningen (skjnt
jag dri fornlskarom til tjenst / efter Stiernhielms och Verelii bermliga sedwana / ofta in-
mngt en hoper gamla / doch mustiga ord och krta talestt / som ej utan widlftigt ordaswef

472
 ltit sig uttydas; och skjnt de i frstone kunna fr hastiga lsare wara hinderliga/ doch / om de
sjlfwa meningen ngorlunda eftersinna / med frmodeligit gagn snart begripeliga) utan och
den Latinska uttolkningen drjmte s inrttat / at / ehuru jag dri merendels granneliga flgt
Gtiska hufwudsprket / ingen / utan den okunniga och afwundsjuka hopen / warder nekande /
at eho detta sprk ernar sig bemgtiga / han doch med detta arbetets hjlp / skjnt det ssom
andras / fwen granlagste mnniskors / ej r utan fel / skall kunna mycket redigare / n hrtils
r skjett i sit upst fortfara. (BiFre:56)

Birner aims at finding readers who might face some difficulties, but are willing to spend
time and energy in order to improve their command of Old Norse. His translations bear wit-
ness to this ideal: It is obvious that Birner was hoping that his abundant use of Icelandisms
would have an impact on the Swedish language and that words such as andas (die,
BiFri:4 and more often), rakna vid (regain consciousness, BiFri:29) or mannarn
(dangerous situation, BiFri:4) would become part of the Swedish lexicon. Based on dic-
tionary entries, Hirvonen discusses which Old Norse words used by Verelius and Birner
have found their way into modern Swedish (cf. Hirvonen 1987:137145). What I intend to
study more closely in the near future is the use of Icelandisms and archaisms in the literary
works derived from the Kmpa Dater.
As is expressed in the preface, Birner is well aware of the fact that the general audience
needs some help with Old Norse expressions. He offers this help by adding explanations
probably the most characteristic stylistic device made use of are word pairs, the first part of
which is an Old Norse word which is explained by the second (Swedish) word: Denna Inge-
borg war wen eller wacker tilandlete/och wetig til hug eller sinne / samt frmst eller
frnmst af Kongsbarnen. (BiFri:1, my accentuation). Usually the words are combined by
the conjunction eller, which is typical of interpreting/explanatory word pairs (cf. Hirvonen
1987:122). Hirvonen has analysed Birners use of such explanatory word pairs and has com-
pared it to Olof Verelius and Gustaf Bondes use of the same construction. The result is un-
ambiguous: While Bonde and Verelius hardly use explanatory word pairs, the construction
appears extremely frequently in some of Birners translations, among them the Fridthiofs
Saga (cf. Hirvonen 1987:122124).
Particularly striking is the fact that the explanatory word pairs are used to introduce an Old
Norse word. When the same word appears once more in the source text, only the Old Norse
word is used in the Swedish translation, cf. Kong Bele feck nu sot eller sjukdom []. Denna
sot mnde leda mig til bana (BiFri:3, m.a.). This technique clearly serves Birners didactic
purposes, thus the use of the construction should be considered more than just a manner.
From time to time explanations of customs are inserted in brackets: Kongarna sto p
deras faders hg ( efter sedwana / til at dr fwerwga stora rdslag / ssom under hans
nrwarelse) och helsade Fridthjofer dem wl (BiFri:5). Moreover, in some cases the Old
Norse expression is added in order to specify what exactly the translation is supposed to
mean: Fridthjofer sat nu wid taflspel (hnfwa tafl) (BiFri:7). The explanations are another
clear indication of the intended educational character of the edition.
There are additions, however, which illustrate that Birner was not only preoccupied with
the audiences knowledge gain, but also wanted his story to be coherent and comprehensible.
This becomes clear from the episode in which Helgi and Hlfdan send Hildingr to Frijfr in
order to ask him for support against King Hringr. Frijfr plays chess, when Hildingr ar-
rives. Hildingr asks for his help and Frijfrs reaction is described as follows:

Friiofur svarar honum a ungvo ok mllti til Biarnar er hann tefli vi, bil er ar nu fos
tbroir ok muntu ey brega vi, ok (BiFri 7)

473
The translation differs from the Old Norse text:

Fridthiofer swarade honom intet / och sade til Bjrn / den han tflade med : rum r dr nu
imellan brickorna/ fosterbroder/ och mnde du ej ndra det / men jag skall utwlja den rda /
och weta om han blir fri. (BiFri:7)

The deviating translation is easily explained: There is a line of text missing in the Old Norse
version, a line already missing in Holm 56 fol. (cf. 181v) and in Holm papp. 17 4, the Ice-
landic exemplar of Holm 56 fol. (cf. 357rv). Birner must have had access to a different ms.
(of the longer version) and translated the passage in accordance with the undeniably better
reading:

fridiofur suarar hanum nguo og mllti til biarnar er hann tefldi vid, bil er arna fostbroder
og muntu ei bregda ui, helldur mun eg setia ad hinni raudu tflunni og vita hurt henni er for-
dad (Larsson 1893:5).

Moreover, he offers some assistance to the readers trying to interpret Frijfrs ambiguous
statements about the chess figures. Frijfr does not only refer to Ingibjrg when talking
about the red chess piece (of course the original version in which the chess piece, the tafla, is
correctly referred to by the feminine pronoun underlines the ambiguity in a more sophisticated
way, but still Birners version works), he also goes on to declare war on Helgi and Hlfdan
by use of yet another chess metaphor: a mun ra at sitia fyrst at hnefanum []; d mnde
rdligast wara at draga frst Kongsbrickan til strids [] (BiFri:7; hnefi is the king). When
Hildingr is asked to explain what Frijfr might have meant, he gives the following explana-
tion: enn ar er hann liest sitia mundi at fogru taflinu, at mun koma til Yngibiargar systur
yckar, gitit hennar vel sva vist. The Swedish translation reads: men nr han ltsade sitja wid
fagert tafl (eller talte om rda spelbrickan) det lter komma an p eder syster Ingeborg / be-
waren hnne drfre wisserliga (BiFri:8). The explanation that the beautiful tafla (or rather
tafl in this case) is identical with the red chess piece is added by Birner. The technique is
similar to that of the explanatory word pairs already discussed, but here it is not only a ques-
tion of translating one Old Norse word. The word for chess piece has been introduced already,
thus we can conclude that Birner considered the passage to be particularly difficult (which it
undoubtedly is) and deliberately added information when translating it.
Hopefully I have been able to give some insight into the techniques used by Birner in or-
der to spread knowledge about Old Norse language and culture. This was not necessarily the
only aim, however. It seems to be generally agreed that the style of Birners translations is
rather inferior. According to Malm they belong to a period in which a mimetic approach is
applied:

[] Birners versttningar fljer liksom 1600-talets originalen s nra att mlsprket skruvas
och tvingas. Versifikatoriska formalia fljs noggrant, liksom ordstammar aktivt bibehlls, men
versttningen blir tvungen och den poetiska valren obetydlig. Detta hnger frsts samman
med bakomliggande avsikter att introducera (en uppfattning om) fornnordiskt sprk, kultur och
mentalitet i Sverige och en kraftig bild av originalens auktoritet. (Malm 1996:191).

Of course the observations are correct, but I am not entirely convinced that judgements of this
kind do justice to a work like Birners. I do not intend to enter the minefield of discussing
what can be considered poetic, yet I would like to suggest that Birner tried to create a piece
of art also in terms of style. At least one will have to admit that certain stylistic devices are
used throughout the text, thus suggesting that a stylistic ideal was followed.

474
 In cases where the Old Norse text contains word pairs bound together by ok, Birner often
uses word pairs with the conjunction och, for example when rendering var hun ar uppfdd
vel ok vanlega as wart hon dr wl och granneliga uppfdd (BiFri:2) or me ofsa ok ojaf-
nai as med hgfrd och obillighet (BiFri:7). To my mind this clearly indicates that the
overall frequent use of word pairs is also an expression of Birners interest in reflecting the
style of the Old Norse source in his translation.
The same can be said of the attempt to imitate ACI-constructions, e.g. De swarade / sig
icke wilja lra det i unga hren as a rendering of eir kvaust ecki vilia lra at a unga alldri
(BiFri:6). One might not regard this construction as particularly elegant or aesthetically
appealing, but I doubt that Birner chose this expression only to give a close translation of his
source. On the contrary, it seems likely to me that he congratulated himself for finding ways
of making his Swedish sound learned and archaic which in his view certainly were positive
attributes. And if Birner felt enthusiasm for constructions induced by Latin and Old Norse, at
least some of his contemporaries might have shared that sensation. In order to be able to dis-
cuss these questions on solid ground, it will be helpful to analyse whether or not the literary
works based on Birners text adapt his stylistic devices.
To come to an end, I would like to discuss one stylistic trait which is abundantly used
throughout Birners translation: There are innumerable instances of present participles. In-
terestingly, Birner does not follow his exemplar in this respect (as mentioned above, the
Frijfs saga is not strongly influenced by the courtly or the florid style), but renders finite
verbs as participles: yckir eim a vnlegt um sina fer vs. tyckande de d wl om sin frd
(BiFri:14, m.a.); logu undir sig vs. lggande under sig (BiFri:31, m.a.); Ok geck Fri-
ofur inn ok kva visu vs. och geck Fridthjofer in / kwdande sdan wisa (BiFri:39, m.a.).
Birner does not even hesitate to make frequent use of the present participle of vera, a highly
marked and artificial construction.3 So far I have not found a single example of this construc-
tion reflecting the source text. Typically, eir brir foru suur til Jadars ok funu Hring
kong i Soknarsundi, vi hafi Hringur kongur mest reyst, at [] is rendered as Brderna
foro nu sder til Jader/ och funno Kong Ring i Sognasundet/ warande Kong Ring drfwer
mst wreder at [] (BiFri:8, m.a.). The construction also appears in the preface: Warande
frst til wetandes / at uphofsmann til detta Wrks tryckande / har warit []; Warande alts
wr tids rtta Atle / som de gamla Yfwerboars wetenskaps himel / kraftige arlar brande
[]. (BiFre:2, m.a.). Since the preface is written independently of a source, one may as-
sume that Birner is rather convinced of the elegance of the participle construction.
Sometimes the sense of the text is modified due to the use of a participle form. When
Frijfr has been asked to go to the Orkneys to gather Helgis and Hlfdans tribute there,
his men encourage him to make peace with Helgi:

eir spuru Friiof hvort hann villdi ecki fara til Helga kongs ok sttast vi hann, ok biia af
ser reii Balldurs, hann mllti, at mun ek heitstreingia, at ek mun ei Helga Kong friar biia;
eptir at geck hann a Ellia ok helldu eir ut eptir firinum Sogni. (BiFri:12, m.a.)

The Swedish translation narrates this detail as follows:

Dessa sporde Fridthjof til / om icke han wille fara til Kong Helge och frlikas med honom /
samt afbedja sig Balldurs wrede ? Han swarade: det lftet skall jag gjra / at ej bidja fred af
Kong Helge ; och efter detta geck han p skepet Ellida/ hllande widare ut efter fjrden.
(BiFri:12, m.a.)

3
The use of this construction in Old Norse is discussed in detail in my forthcoming dissertation (Die Barlaams
ok Josaphats saga eine hfische Legende am norwegischen Knigshof. Heidelberg. (Skandinavistische Ar-
beiten)).

475
In the translation it is Frijfr alone who sails out along the fjord. Of course it is hard to tell
whether or not Birner chose the participle construction for merely stylistic reasons. In order
to give a well-founded judgement one would have to gather all the participle forms in all his
translations, something I intend to do in the near future. At the moment it is nothing but a pos-
sible (yet in my view appealing) interpretation, when I suggest that in the quoted passage the
participle was chosen both for stylistic and for narrative reasons: Frijfr appears as an even
more active protagonist than in the Old Norse saga, and it is underlined that he takes his deci-
sion not to make peace with his opponents wholeheartedly, thus turning his action into an
assertive statement.

Conclusion
This paper was mainly intented to illustrate that Birners Nordiska Kmpa Dater are well
worth being studied more thoroughly than has been the case so far. I hope to have shown that
Birner had a keen interest in educating his fellow Swedes and to spread knowledge about the
Old Norse language and the Scandinavian Middle Ages. Undoubtedly, Birner was not a
child of his times, but a true Rudbeckian who could have published a century earlier and go
unnoticed. The impact of the Kmpa Dater on later writers, however, is reason enough to ana-
lyse his use of the Old Norse sources.

Bibliography
Valdimar smundarson (utg.), 1886: Fornaldarsgur Norrlanda. Bd. 2. Reykjavk.
Birner, E. J. (utg.), 1737: Nordiska Kmpa Dater, i en sagoflock samlade om forna Kongar och Hjl-
tar. Stockholm. (with separate pagination for each text).
BiFre = Fretal til Lsaren. In: Birner: Nordiska Kmpa Dater. Pp. 136 (separate pagination).
BiFri = Sagann af Friiofe Frkna. In: Birner:Nordiska Kmpa Dater. Pp. 143 (separate pagina-
tion).
Bothius, B., 1924: Birner (Birner), Eric Julius. In: Svenskt biografiskt lexikon. Bd. 4. Stockholm.
Pp. 477487.
Busch, Kay, 2004: Grossmachtstatus & Sagainterpretation. Die schwedischen Vorzeitsagaeditionen
des 17. und 18. Jahrhunderts. Vol. I: Beschreibung. Inauguraldissertation Universitt Erlangen
2002. http://www.opus.ub.uni-erlangen.de/opus/volltexte/2004/51/ [urn:nbn:de:bvb:29-opus-511,
12.04.09]
Calaminus, Wilhelm, 1887: Zur Kritik und Erklrung der altnordischen Frithjofssage. Diss. Jena.
Falk, Hjalmar, 1890: Om Frijfs saga. In: ANF 6. Pp. 60102.
Gdel, Vilhelm, 18971900: Katalog fver Kongliga bibliotekets fornislndska och fornnorska hand-
skrifter. 4 bd. Stockholm. (Kongliga Bibliotekets handlingar 1922.)
Hirvonen, Ilkka, 1987: Fornlskarom til tjenst om sprket i E.J. Birners Nordiska Kmpadater.
En jmfrelse med Olof Verelius sagaversttningar. In: Diverse. Vnskrift till Bertel Fortelius.
3.8.1987. Utg. av Dido Bjrkblad & al. bo. (Filologiska Institutionen vid bo Universitet, Nor-
disk Filologi. Meddelanden 6.) Pp. 105161.
Klbing, Eugen, 1878: Ueber die verschiedenen bearbeitungen der Frijfssaga. In: Beitrge zur
vergleichenden Geschichte der romantischen Poesie und Prosa des Mittelalters. Breslau. Pp. 207
217.
Larsson, Ludvig (utg.), 1893: Sagan och rimorna om Friifr hinn frkni. Kbenhavn. (Samfund til
Udgivelse af Gammel Nordisk Litteratur 22.)
Larsson, Ludvig (Hg.), 1901: Frijfs saga ins frkna. Halle/S. (Altnordische Saga-Bibliothek 9.)
Malm, Mats, 1996: Minervas pple. Om diktsyn, tolkning och bildsprk inom nordisk gticism.
Stockholm/Stehag.
Wenz, Gustav (Hg.), 1914: Die Frijfssaga. In ihrer berlieferung untersucht und der ltesten Fas-
sung kritisch hg. Halle/S.

476
Where Old West and Old East Norse literature meet. A project
outline
Regina Jucknies, Institut fr Skandinavistik / Fennistik, Universitt zu Kln, Germany

Project aims
Old East Norse text have been playing a minor role as a subject of international research, es-
pecially when compared to the lievely scientific work that has been and is being done on Old
West Norse topics. Fortunately, Old Swedish and Old Danish manuscripts and texts have
come into focus in recent times, and some of the questions raised deal with comparative as-
pects and / or with the transmission of these texts.
In the project I am just starting to do research on, my aim is to put together the results of
the different (historical and current) research fields. I am going to try to give a survey on how
texts were materially transferred as well as in what ways the texts were changed or adapted
when imported to the North-Eastern and -Western vernacular. This, I hope, will contribute to
a broader perspective on the mechanisms of text transmission in the Nordic Middle Ages. I
have not yet come very far, but I will at least try to outline my project task.
Firstly (and for this conference paper), I want to focus on texts that were transferred from
North-West to the North-East. There are two examples I am going to refer to, but there are
more.
Secondly, I am going to concentrate on texts coming from the South that were translated to
both Old West and Old East Norse like Barlaam saga or Lucidarius, and I will try to find out
which factors played a role in the different reception and adaptation of the texts.
As a third step, I would like to take a closer look at the relations between texts dealing with
similar content but emerging in different forms in the various areas. Why, for example, would
the Knittel verse be so popular in Sweden and almost not taken into account in the West when
treating historical issues? Why did Icelanders and Norwegians prefer the genre of the saga,
while Swedes and Danes wrote rimkrnikor?
In some cases, the answer could be trivial, but still I think it to be worthwhile to to com-
pare the ways of literary dealing with similar contents in different contexts. I have not yet
decided on on the all text genres I am going to take into account, but I think I will take into
account literary genres as well as non-fictional texts. And I think I will not be able to avoid
dealing with religious texts, although this is a very broad task, as the Christian culture is of
such a great importance to the transmission of written knowledge.

Ways of transmission
On a basic level, the traditional way of communicating texts in the Middle Ages are well
known: they were copied by the hands of scribes. But beyond this fact, questions arise, not
only due to the destruction of text witnesses through the ages by a variety of factors. Fortu-
nately, we do have information on how scribal work was organized in some of the scriptoria
in Scandinavian cloisters (cf. the Vadstena project), but research in this field has not yet come
to an end and is far from that.
Besides the inter-cloister transmission of texts, trade and private travels are other means of
literary transport, although in a much minor range. Like cloisters and episcopal sites, towns
and market places served as platforms for the exchange and distribution of information and
knowledge, but the written word played no decisive role. Much more important were the po-
litical centres like the royal courts and, in Iceland, the sites of the hfingjar as well as the

477
regular thing assemblies. Part of my project will be the tracing of these aspects of material
transmission, that is the the transport and the distribution of manuscripts.

Northeastern and Northwestern texts. A comparison

The following table is to give a first idea of the texts in question. I am fully aware that this
table does not take into account the time of emergence of the texts that varies highly, and that
some of the texts are not to be compared directly, partly because their sources are different.
Of course, the time of origin of the texts as well as the time of writing will influence both con-
tent and form. It might be interesting to see from a broader point of view what kind of content
was transferred and why and which form the writers choose for their adaptations. The specific
historical background is to be taken into consideration in all cases. The texts mentioned in the
table are only examples and far from complete.

Table 1. Old West and Old East Norse parallel content.

Type of literature Iceland Norway Sweden Denmark
Heilagramannasgur x Barlaams saga ok Js- Fornsvenska legenda-Hellige kvinder
aphats riet
Family sagas x Gutasaga
Translated ridda-vens saga Herr Ivan Ivan lveridder
rasgur Flres saga ok Blan- Flores och Blanzeflor Flores og
kiflr Blanseflor
Hertig Fredrik af Hertug Fredrik af
Normandi Normandie
Karlamagnssaga Karl Magnus Karl Magnus
krnike
idreks saga af Bern Didrikssagan
Sagas of antiquity Alexanders saga Konung Alexander
Trojumanna saga Historia Trojana

Annals / Chronicles Annales Reseniani Erikskrnikan Roskildekrniken

Annales vetustissimi Engelbrektskrnikan Jyske krnike
Hyersannll Karlskrnikan
Annales regii Prosaiska krmikan
Sklholtsannll Lilla rimkrnikan
Annlabrot fr Skl- Sturekrnikan
holti Gutakrnikan
Lgmannsannll Vidhemsprstens an-
Gottsklksannll teckningar
Flateyjarannll
Aarbger rbkur Sjaellandske Aarbog
Skaanske Aarbog
Yngre Lundeaarbog
Rydaarbogen

National Laws Grgs Gulaingslg Gutalagen Jyske Lov

Jrnsa Frostaingslg Vstgtalagen Skaanske Lov
Eisivaingslg stgtalagen Valdemars
Borgaringslg Upplandslagen sjllandske Lov
Magnus lagabtirs Dalalagen Eriks sjllandske
landslov Sdermannalagen Lov
Vstmannalagen
Hlsingelagen
Tionhradslagen
Magnus Erikssons

478
 landslag
Town Laws Magnus lagabtirs Magnus Erikssons Slesvigs bylov
landslov stadslag Flensborgs bylov
Bjrkartten Ribe bylov
Sderkpingsrtten Roskilde
Kobenhavns by-
lov
Birke bylov

Medicine AM 194 8vo Harpestreng

AM 87 8vo

Frstenspiegel Konungs skuggsj Konunga styrelsen

Religious writings
prayers Marubnir
Instructive literature Vira slar ok lka- Trtan mellan sjl och
mans kropp
Sjlinnar trst Sjlens trst

The following second table lists some of the texts that were transmitted both to Old West and
to Old East Norse. I have not indicated where the texts came from.

Table 2. Texts transmitted to both Old West and Old East Norse.
Iceland Norway Sweden Denmark
Elucidarius Lucidarium Lucidarius
Barlaams saga ok Josaphats Sagan av Barlaam och Josaphat
Olafs saga helga Historia Sancti Olai
vens saga Herra Ivan Ivan lveridder
Stjrn Pentateukparafrasen
Karlamagnssaga Karl Magnus Karl Magnus krnike
ireks saga af Bern Didrikssagan
Duggals leisla Visio Tnugdali (Visio Tnugdali)

Transmission patterns?
What would be the benefit of a combining investigation of all these very different texts that
came into existence under different circumstances in different times and places and that were
written for different audiences? In my opinion, it could help to improve our knowledge about
mechanisms that determined the ways of adaptation and transfer of these texts.
I wonder if it will be possible to discern certain patterns of transmission. Is there a general
way of transmissin for a certain kind of text genre or does transmission sometimes happen in
incidental ways? What individual ways of transmission are there? What aspects of transmis-
sion are crucial for the adaptation of texts?
Probably the ways of transferring texts are not unique compared to the ways common in
the rest of Europe, but the adaptation itself of the texts certainly is.

Two examples: Ivan and Olaf

I will now shortly mention two texts that went from Norway eastwards. The way this hap-
pened was very different. The first text, Yvain ou le chevalier au lion by Chretien de Troyes
was translated from Old French to Old Icelandic and to Old Swedish. An Old Danish version
is extant, which is a translation from the Swedish. Hrra Ivan is part of the so-called Eu-
femiavisor; three verse novels that were produced on demand of Queen Eufemia, the German

479
wife of king Hkon Magnusson of Norway. Stefanie Wrth (now Gropper) discussed in an
article (Wrth 2002:13), why Queen Eufemia would have Yvain translated into Old Swedish
from the French, not from Old Icelandic: the West Norse saga form must have been quite un-
familiar to her as a German, and she used the opportunity of her daughter Ingeborgs engage-
ment and wedding with duke Erik from Sweden to put the text into Old Swedish knittel. So,
in this particular case, it was the personal background of the patron that determined the form
and not the milieu in which the text was converted.
Like the Eufemia-texts, lafs saga helga was transferred from West to the East. But this
time, it was not a German princess having the text converted in Norway, but a Swedish king,
Karl Knutsson Bonde, who became king of Norway in 1449 and went there for the coronation
ceremony. From Trondheim, he took a manuscript of lafs saga with him to Sweden and had
the text translated into Old Swedish knittel. Rolf Pipping says about this adaptation: Dikten
r dessvrre srdeles trkig och talanglst skriven i schablonmssig stil. (Pipping 1943:89).
About the Old Swedish Historia sancti Olai, there has not been done much research (apart
from Carlquist). Probably, Klemmings commentary in his edition (Klemming 188182:522)
as well as Pippings harsh judgement prevented scholars from dealing with this text. From my
perspective, the literary quality of the text is irrelevant, as I am more interested in the fact that
someone thought it to be necessary to produce an abridged and versified version of lafs
saga. Of course, the transmission of the St:Olaf legends are a vast area of research, and for a
more thorough investigation of the transmission there is more to it than this one example. For-
tunately, there has been quite a number of scholars who have done research on these legends
(though without special emphasis on the East Norse transmission), and so I will be able to
benefit from their findings (cf. Helgonet 1997, Lidn 1999:5565 and Mortensen/Mundal
2003:353395).

Summary
The task of my project is the combined view of transmission of texts happening during the
Nordic middle Ages both into Old East and Old West Norse. It is both questions of material
transfer of manuscripts and the transformation of content that I am interested in. The range of
texts is wide as I hope to be able to find patterns of transmission that are of more general im-
portance. Unfortunately, I have not yet been able to present more than a draft of my research
plans yet, accompanied by two very small examples. Due to the great number of texts in-
volved, I will not be able to investigate all text genres myself, but I think that this will at least
give a picture of the work that is yet to be done in this field. As I still stand at the beginning of
my research on these topics, I will be glad for any comment or critical view on my ideas and
thoughts.

Bibliography
Carlquist, Jonas, 1995: Den helige Olafs historia. Stockholm. (MINS 41)
Helgonet i Nidaros. Olavskult och kristnade i Norden. utg. av Lars Rumar. 1997. Stockholm. (Skrifter
utgivna av Riksarkivet 3)
Lidn, Anne, 1999: Olav den helige i medeltida bildkonst. Legendmotiv och attribut. Stockholm.
Mortensen, Lars Boje and Else Mundal: Erkebispesetet i Nidaros arnestad og verkstad for olavslitte-
raturen. In: Ecclesia Nidrosiensis 11531537. Skelys p Nidaroskirkens og Nidarosprovinsens hi-
storie. utg. av Steinar Imsen. 2003. Trondheim. (Senter for middelalderstudier, NTNU 15) Pp. 353
395
Pipping, Rolf: Den fornsvenska litteraturen. In: Litteraturhistoria. A. Danmark, Finland och Sverige.
utg. av Sigurdur Nordal. 1943. Stockholm Oslo Copenhagen. (Nordisk kultur 8A) Pp. 64128
Svenska medeltids dikter och rim, utg. av Gustav E. Klemming. 188182. Stockholm. (SFSS 25)

480
Vadstenaklostret som text- och handskriftsproducerande milj produktion, tradition och reception,
project leader Lars Erik Edlund,
http://www8.umu.se/littnord/Vadstena/index_vadstena.htm (last visit 30/04/09)
Wrth, Stefanie, 2000: Eufemia: Deutsche Auftraggeberin schwedischer Literatur am norwegischen
Hof. In: Akten zur 13. Arbeitstagung der deutschsprachigen Skandinavistik in Oslo, hg. v. Fritz
Paul. Frankfurt. Pp. 269281
(http://homepages.uni-tuebingen.de/stefanie.wuerth/schriften/eufemia_wuerth_2000.pdf, last
visit 16/04/09)

481
Boleslaw The Brave, his father Mieszko, and Harald Fairhair
progenitors of royal dynasties described in texts bearing a po-
litical message: Chronica Polonorum and Haraldar saga
harfagra
Anna Kaiper, Polish Academy of Science
Chronica Polonorum written by Gallus Anonymus and Snorri Sturlusons Haraldar saga
hrfagra have something in common. Both of them concern the history of the rulers, and not
only present important events in the history of their kingdoms, but also convey a political
message. Though the construction of the two texts is quite different, there are many similari-
ties. Descriptions of the progenitors of the dynasties; the first, the strongest and the most im-
portant persons in the history of their countries, are constructed in the same way and use the
same means to create their image. One can not only find it in the description of their appear-
ance, but also in facts from their private life and their way of ruling.
The times of the reign of Mieszko and Boleslaw The Brave in Poland and Harald Fairhair
in Norway were the times of change, the turning points in the history of both countries. The
spouses of the rulers were portrayed as the initiators and intercessors of these changes in both
texts. Harald, inspired by Gyda, decided to unite Norway, to rule it as a single, independent
king and to establish legal order. Dobrawa, the Czech princess, requested of her husband, Mi-
eszko, the adoption of Christianity and, through the act, the introduction of his kingdom to the
Christian world.
Both Boleslaw and Harald greatly expanded the territories they inherited. They waged
many wars and proved themselves to be great warriors and charismatic leaders commanding
devoted armies. They were not only able to expand their power but also to unify conquered
lands into homogenous kingdoms. At the end of their long reigns both rulers attempted to
prevent future fighting for domination among their heirs by leaving dispositions in their last
wills. Their plans failed. Both texts describe the reigns of the progenitors of the two royal
dynasties as the golden ages for their countries, with the domestic situation deteriorating sig-
nificantly after their death. Their legitimate successors were unable to maintain political au-
thority and lost the power passed on to them. Boleslaws and Haralds reigns so profoundly
differed from those of their predecessors and successors that it was the reason, as the texts
explain, why they were so popular and beloved by their subjects during their life and after
their death.

482
 Heads and tales: Mmir, Vlsi, and the pursuit of prophecy
Merrill Kaplan
The prophetic head of Mmir is frequently assigned a Celtic geneology, and the many
wisdom-spouting heads of medieval Celtic literature make some such connection seem likely.
However, there may be another useful parallel in Vlsa ttr. The manner of inns
acquisition, preparation, and use of Mmirs severed head as described in Ynglinga saga re-
sembles the housewifes treatment of the horse phallus in Flateyjarbks Vlsa ttr. Both
head and phallus are leftovers from an unrelated slaying or death: Mmir is beheaded by the
dissatisfied Vanir after the first war, and the horse dies of apparently natural causes and is
relieved of its member while being butchered. Both body parts are salvaged and smeared with
herbs (urtur in Ynglinga saga and laukar ok nnur grs in Vlsa ttr) to prevent them from
decaying (funa or rotna). Both inn and the hsfreyja speak magic over them: he kva ar
yfir galdra and she pronounces formlar and kver vsu. The effects are comparable: the head,
as the object of the verb magna, is increased in some magical way while the phallus grows
and is strengthened (vex [] ok styrknar). Both are re-animated to a degree that, while
limited, restores to each an essential function: Mmirs head can speak, and Vlsi can stand on
its own. inn creates an object through which he can acquire information. To what end the
housewife creates Vlsi is not entirely clear; the ceremony is interrupted, and the reader never
gets to see its intended results. However, the etymological links among Vlsi, vlr (staff), and
vlva (seeress, lit. staff-wielder) suggest that a Vlsi-wielder might have access to prophecy.
This paper argues that we should see the use of both Mmirs head and Vlsi as
representations of seir made to very different literary ends in their respective texts. In
Ynglinga saga, Snorri is at pains to set inn in a specific light. In Vlsa ttr, Jn rarson
reveals the ugliest underpinnings of seir in a tale about the Christian monopoly on revealed
knowledge.

483
 Sweden of the Sagas
Kri Gslason, Creative Writing and Literary Studies, Queensland University of Technology,
Australia
In this paper, I discuss the representation of Sweden and Swedes in the slendingasgur, with
an emphasis on identifying patterns across the works, both in terms of narrative structure and
content. The aim in doing so is to shed light on modes of representing non-Icelanders in the
slendingasgur, as well as on medieval Icelandic conceptions of Sweden as a distinct region
within Scandinavia. I also aim here to add to a longer-term project that examines the place of
foreign visitors to Iceland in the saga corpus more generally.
As the scope of this paper is limited to Swedish characters, I am cautious about drawing
broad conclusions about their representation observations given here will need to be framed
by a wider study, and one that reads for the characterisation of Swedes in the context both of
other genres of saga literature and representations of characters from other regions beside
Sweden. However, it is clear that some similarities exist in saga episodes involving Swedish
characters: in four of the slendingasgur, Swedes are given roles as intruders or outsiders
who threaten the community of the saga and whose deaths bring about a change in the for-
tunes of their killers.

A Norwegian in Sweden
At the beginning of Svarfdla saga, the audience meets orsteinn orgnrsson, the unpromis-
ing younger brother of rlfr. orsteinn is abrasive, very big, and loath to leave the comfort
of the hall. He is so lazy and still that a pile of ashes gathers around him. People trip over his
feet. While his older brother is bringing honour and wealth to the family, orsteinn attracts
only the indignation of his father, to the point where orgnr will not hear orsteinn referred
to as his son. Eventually, after rlfr has had a turn at tripping over orsteinn, the brothers
agree to go trading together. rlfr accepts orsteinns proviso that orsteinn should always
have the last say in any disagreements between them, and, much to their mothers surprise,
there is a call made for the running of a bath: orsteinn is to be cleaned up.
orsteinn, though, is not particularly well-suited to trading. He is more concerned with
adding to his bravery than his wealth, and, in pursuit of a fame that can only be won through
the hardest of fights, he sets himself up against Ljtr. In the lead-up to the encounter, or-
steinn declares his pleasure in the fact that by autumn he will either be dead or have killed
Ljtr. He also tells rlfr that during the summer he should go trading that he does not
want any harm to come to his older brother. rlfr rejects the instruction, joins the expedi-
tion against Ljtr, and is killed.
This is the sequence of events that, by chapter six of the saga, has brought orsteinn to
Sweden and to the hall of Earl Herraur. There, orsteinn breaks local custom by entering
fully-armed. He declares who he is and asks for the use of the hall for a memorial service for
rlfr. Herraur consents, and through his hospitality establishes a basis for orsteinns later
assistance to the hall. During the winter, Herraur and his followers become quiet and down-
cast. They are being hounded by Moldi, a half-berserk Viking who wants to marry the earls
daughter, Ingibjrg. orsteinn insults Moldi and accepts a challenge to a duel, the hall cheers
up, and, with the help of Herraur he knows about Moldis sorcery and how to beat it or-
steinn kills the berserk. In doing so, he avenges rlfr (Moldi is Ljtrs brother) and rids the
hall of a grave threat. He marries Ingibjrg and, after another winter in Sweden, returns to
Norway.

484
 The plot elements of this opening episode might be listed in this way: 1) a difficult young
man, lazy and taciturn, is introduced; 2) he leaves home and, because of his bravery and
strength, is more successful than anyone had expected; 3) because of his unrestrained personal
ambition, he takes on the strongest opponent; 4) the subsequent fight results in a win but also
personal loss; 5) he then travels to a hall (in Sweden) that is threatened by a hostile intruder;
6) he fights the intruder and is rewarded by his host. As we will see later, the series of events
bears a number of similarities to other episodes in the slendingasgur that have a significant
involvement by Swedish characters.

References to Sweden
Svarfdla sagas extended setting of action in Sweden is unusual in the context of the slend-
ingasgur, which make relatively few direct references to Sweden or to those of Swedish de-
scent, by which I mean those specifically referred to as snskr, Svar, or as being from
Svj. In this, I am following the sagas in their differentiation of Svj from other re-
gions, such as Gautland and Jamtaland, that form provinces within Sweden today. The forma-
tion of a unified Sweden is felt to have come relatively late, a fact that the saga authors ap-
pears to recognise. In Brennu-Njls saga, for instance, Atli, the son of Arnvir jarl, is de-
scribed as being from East Gautland, and Gunnar and Kolskeggr travel specifically to
Smlnd; in Harar saga og Hlmverja, Hrr is off to Gautland to meet Haraldr jarl; and in
the opening of Droplaugarsona saga, we find that Ketill is riding east not to Sweden but to
Jamtaland.
The identification austmar, a term that is used to refer to Norwegian traders and some-
times as a nickname, is much more common. Austmenn, too, are more likely to be given sig-
nificant roles in the events of the sagas: they become part of the feud narrative, take sides, and
have an impact on the Icelandic characters. This is the case in, for instance, Fljtsdla saga,
in which the Austmenn have a prominent role throughout. The kaupdrengr rn in Hnsa-
ris saga is another example.
By contrast, of the references to Sweden that do occur in the slendingasgur, many are
minor in terms of plot, character development, or theme. In Bjarnar saga Htdlakappa, we
learn merely that Bjrn and Auunn travel to Sweden before moving on to Denmark; in
Brennu-Njls saga that, by the time of his encounter with Hrtr, Atli has been outlawed in
both Sweden and Norway, and that orkell hkr travels through Sweden on his way to and
from his fights with the fantastic beings located still further east. Droplaugarsona saga men-
tions the return of Grmr and Ormar from their raiding in Sweden; in Egils saga Skalla-
grmssonar, Arinbjrn cites the poet Bragis conduct before the Swedish king Bjrn as an
example for Egill to follow; in Fstbrra saga, ormr Kolbrnarskld rejects the oppor-
tunity of travelling to Sweden; and in Eyrbyggja saga, Bjrn travels to Sweden after he has
been outlawed. While these references to Sweden may be significant to the saga characters
and the audience of the saga, for the modern reader there is little indication of their narrative
importance.

Icelanders in Sweden
More detailed narratives set in Sweden can be found in Gunnlaugs saga ormstungu and Hall-
frear saga vandrasklds. In both, we find the poets enjoying the kind of welcome and
hospitality that we often see granted to Icelanders during their travels in Norway. That is,
these two sagas observe the conventional presentation of Icelanders abroad: straight away,
they are recognised as important to the court and given opportunities to address it. Their skill
is clear, and their leaving the court is seen as a loss. Flamanna saga, too, places a reasonably
detailed event in Sweden, orgils fight with the sorcerer Randvir. As with orsteinns fight

485
with Moldi in Svarfdla saga, orgils is in a position to prevent a marriage between the
hosts daughter, Sigrr, and a suitor of whom the host is afraid. orgils volunteers to fight on
the farmers behalf, kills Randvir, then two other Vikings, Snkollr and Snbjrn, and the
summer after sails to Iceland.
In both form and content, the events in Flamanna saga bear a number of similarities to
those in Svarfdla saga that I describe above. During the journey in Sweden, the traveller
arrives at a hall/farm that is beset by an intruder who threatens violence and who is capable of
sorcery. In both instances, the intruder wants to marry the daughter of the owner of the
hall/farm. The traveller offers to fight the intruder on the owners behalf, and, despite the su-
pernatural abilities of his opponent, wins the duel and relieves the hall/farm of its troubles.
The traveller is rewarded with gifts, and shortly afterwards leaves.

A Swedish ghost in Iceland

A match with Svarfdla saga is also found in Grettis saga smundarsonar, when Grettir
takes on the ghost of Glmr, described by the saga as a Swede who has recently come to Ice-
land. Grettir, like orsteinn, is slow to show his potential ekki brgrr, mean hann var
barnsaldri (36); he was not accomplished while he was in his childhood years. He is also
quickly identified as a quarrelsome young man with whom it will be difficult to deal, and as
immensely strong. Like his father, Grettir does not like farm work and seems ill-suited to rou-
tine life. And, as in Svarfdla saga, father and son appear to dislike and provoke one another.
The characters destined to take on the two sagas Swedish intruders are alike.
An ironic comment by the author, margir bu hann vel fara en fir aftur koma (974)
many wished Grettir well on his travels abroad but few wished him to return comes at the
beginning of an adult career of excessive personal ambition and troublemaking. A pattern of
behaviour soon emerges, one of taciturn laziness, disruptiveness, rudeness, followed by bursts
of intense activity. The last of these behaviours means that Grettir can be of use to those
around him, as is the case when he defeats Glmr as well as during a number of other inci-
dents, such as his efforts during his voyage to Norway, his strength in acquiring fire for his
shipmates, his fight with the berserk Snkollr, and his fight with a troll woman in Sandhaugar
and her male friend.
And yet, like many of the ambitious characters in the slendingasgur, Grettir is seldom
able to restrain himself when it would be best for the community or indeed himself if he did.
As a result, rarinn hinn spaki worries about Grettirs suitability for important matters, a
concern which his foster-son Bari confirms when Grettir confronts him near reyjargnpr:

Grettir svarar: Bleyask ykki mr , Bari, sagi Grettir, ef orir eigi at berjask vi
mik. Kalla at sem vill, segir Bari, en rum sta vilda ek, at kmir fram jaf-
nai num en vi mik; er at eigi lkligt, v at n gengr r hfi offors itt. Gretti tti illar
spr hans, ok efar n fyrir sr, hvrt hann skyldi ra til einhvers eira, ok snisk honum at
forsjligt, er eir vru sex, en hann einn. (106)

Grettir answers: You seem cowardly to me, Bari, said Grettir, if you dare not fight with
me. You may call it what you like, says Bari, but I would prefer your overbearing nature
to be fulfilled in other places than here with me. It is not unlikely, because your insolence now
exceeds all moderation. Grettir thought ill of his prophecy, and hesitates now about whether he
ought to attack one of them, but this seemed unwise as they were six in number and he was
alone.

At the heart of this confrontation is Grettirs socially disruptive desire to constantly challenge
his own strength as well as that of others, a clear similarity with orsteinn in Svarfdla saga.

486
Many, like Bari, recognise the compulsion and attempt to avoid becoming victims of it. Oth-
ers seek to restrain Grettir, as is the case prior to the fight with Glmr Grettir, like his coun-
terpart in Svarfdla saga, is only encouraged by the warnings he receives.
A key difference between Grettis saga and Svarfdla saga lies in the extent of Glmrs, or
the intruders, back story. Grettis saga is careful to explain how it is that a poor farmer from
Forsludalr ends up with the services of an abrasive, heathen Swede. Such details help to
ground the incidents to come in the everyday world of a farming community, and in this way
serve to highlight the strangeness of foreign characters and the events they bring on. We read
that Skafti the Lawspeaker sees Glmr as a solution to the haunting in the valley, and that,
during the Assembly, Glmr agrees to work for rhallr so long as he is allowed to do as he
pleases. In the following winter, he joins the household. Immediately, there is tension: Glmr
is rude, faithless, and unpopular. rhallrs wife is more troubled by Glmr than others are, a
fact that may be explained later by the ghosts harassment of her and rhallrs daughter: as
in Svarfdla saga and Flamanna saga, the threat of violence appears to have a sexual ele-
ment. Another feature in common with Svarfdla saga is the concentration of the threat dur-
ing the Yule festivities.
The solution to the problem seems at first to rest with Grettirs strength and courage, and
indeed with his excessive ambition. Famously, he is able to rid the farm of Glmr, but in
Glmrs final moments encounters a terrifying vision of his own future Glmrs stare con-
firms that Grettir will always be on the edge of Icelandic society, an outlaw. Clearly, at this
point the saga is making a much more complex use of the intruder figure than we see in
Svarfdla saga, as the episode has moved beyond its function of confirming Grettirs bravery
and strength towards a questioning or at least broader thematising of it his desire to test
himself against the ghost has damaged him, and, despite the gratitude of the household, the
social benefit may have come at too great a cost to the individual and the people with whom
he will deal in the future. Grettir himself declares that the fight with Glmr has worsened his
temperament, and that he now has more trouble restraining himself than he did before. In fact,
excessive ambition has not solved the problem of Glmr; rather, it has established Grettir as a
problem figure for the community, by no means as terrifying as Glmr but certainly as diffi-
cult to control.

The Swedish berserks

In Eyrbyggja saga, we again see foreign characters threatening the stability of the community.
This time there are two, the Swedish berserks Leiknir and Halli, whom Vermundr has brought
across from the Norwegian court of Earl Hkon. There, Vermundr was warned against taking
berserks to Iceland, Earl Hkon pointing out that there could be little use for such types in a
small farming community. Vermundr, though, is keen to gain ground in his relations with his
brother, Styrr, and thinks the berserks will help. Before leaving Norway, Earl Hkon warns
that the berserks must be treated with respect. Just as in Grettis saga when rhallr agrees to
employ Glmr, Eyrbyggja saga creates a strong sense of mismatch and foreboding rhallr
is a relatively minor social figure who has to fetch his own horses, while Vermundr is a
farmer ill-equipped to deal with fighters like these.
In Iceland, it is soon apparent that Vermundr is out of his depth. Halli wishes to marry, and
Vermundr, unsure of how to respond, is forced to consult Styrr. Styrr agrees to take the ber-
serks off Vermundrs hands, and is then faced with the same issue of marriage that has been
troubling Vermundr: Halli expresses a wish to marry Styrrs daughter, sds. Thus, in a fa-
miliar turn in episodes of this kind, the threat posed by the berserks comes to be focussed on
the hosts daughter. And, as is the case in Svarfdla saga, the threat to the daughter will be
taken away by her future husband. Snorri goi suggests that Styrr set the berserks a number of

487
difficult tasks, upon completion of which Halli expects to be given sds in marriage. Ex-
hausted by the tasks, the berserks are weak enough to be attacked by Styrr. Rather like Grettir
and Glmr, the berserks exert their strength to the point where they become vulnerable.
In contrast, Snorri goi is never a victim of his strengths: his self-possession sets him apart
from other characters and gives him a noticeable advantage in his dealings with them. He
does not act rashly or hastily, and tends not to be physically violent himself, preferring to
achieve his aims through others acts of violence (see, for example, chs. 4144). He is also a
relatively silent character. The combined effect of these characteristics is that, for those
around him, his precise position tends to be unclear and his actions difficult to predict. One
early presumption about him (that he is poor) is seen to fail in an amusing way when Snorri
outwits Brkr in order to make a profitable purchase of the land at Helgafell.
It appears that Snorri gois characteristics also combine to make him unpopular, for at the
close of the saga we are told that Snorri goi became more popular later in life (p. 180).
Snorri is by no means a noble hero, and in this respect is contrasted to his more popular rival,
Arnkell. And though he seems to share in the far-sightedness of characters like Njll orgeirs-
son in Brennu-Njls saga, it would be difficult to claim Snorri goi as one of the men of good
will featured in other slendingasgur. His handling of the Swedish berserks, then, is as con-
sistent an expression and development of his personality as is Grettirs fight with Glmr.
While Grettir openly expresses a desire to fight Glmr, and appears to have no ulterior motive
other than to beat a stronger opponent and thereby achieve greater fame and honour, Snorri is
secretive about the full nature of his involvement, and benefits not so much from the act itself
as its consequences: he has helped Styrr, sds is free of Halli, and the region has been ridden
of its dangerous new arrivals. While Grettir leaves Forsludalr in a worse position than in
which he arrived, Snorri goi has worked matters out well in advance and profits as a regional
authority figure.

Patterns
In Svarfdla saga, Eyrbyggja saga, and Grettis saga, hostile Swedish intruders threaten the
safety and stability of a local community. A central character then enters, is told of the prob-
lem, and is asked to help or offers to help. In the lead-up to the confrontation, one of the non-
fighting characters gives a warning or offers advice in dealing with the threat. A fight occurs
and the intruder is killed. The killer is offered a reward, and in both Svarfdla saga and Eyr-
byggja saga that is the reward of marriage to the woman whom the intruder had sought out.
An important point of differentiation is the way in which the confrontations are ultimately
brought to a close. As, in these sagas, the intruder is being opposed by a major character, the
confrontation is in keeping with that characters development as well as function in advancing
both plot and theme. Thus, Snorri gois characterisation as an expert tactician is confirmed
by an ability to outwit the berserks he produces a social benefit to the region and he
strengthens his own position. In contrast, Grettir is not tactical but instead seeks to take Glmr
on in terms that Glmr himself understands well: a one-to-one a fight to the death. His mode
of conflict is more likely to attract honour and popularity than is Snorri gois, but it is also
more dangerous and, in the end, more destructive. While Snorri goi and the berserks fight
without ever meeting, Grettir comes so close to his enemy that he is bound by his influence
for the rest of his life.

Implications
As I mentioned in my introduction, it is rather early in this project to speak conclusively about
the implications of similarities that we find in the structure and content of episodes involving
Swedes and Sweden. Certainly, we are left with as many questions as answers, and the next

488
stage in a study of this kind lies both with the representation of other non-Icelandic characters
in the slendingasgur and further afield. The kings sagas, for instance, offer a far more de-
tailed and substantial representation of Swedish culture and society than we find in sagas
based predominantly in Iceland.
Whether there is a connection between the ethical outlook implicit in these sagas and the
Swedish identity of the characters who threaten the community is also far from clear. The fact
that so few Swedish characters appear in the slendingasgur undercuts their representative
value, especially as characters like Glmr and Halli are probably more readily identifiable as
berserks, sorcerers, or ghosts than as Swedes Sweden may well be a flag of convenience, so
to speak, and stand mainly for the idea that the characters are from far away. Swedes, we
know, tended to travel towards the east rather than the west, and Sweden was late in adopting
Christianity. A geographical and historical context of that kind may well support the literary
use of Sweden as a berserk- and sorcerer-producing region.
The two episodes set in Iceland appear to thematise ambition and restraint, and to share in
a broader ethical concern, previously identified in a number of other sagas, about the balance
of personal ambition and community well-being. Snorri goi, whose ambition is framed by
tactical awareness and a strong sense of regional power, seems best able to judge a course of
action that will both end the threat to the community and further his own desires. Grettir, on
the other hand, driven by excessive personal ambition, goes too far in taking on Glmr. The
Swedes, it appears, are to be handled with greater care than that.

Bibliography
Bjarnar saga Htdlakappa. 1938. Ed. by Sigurur Nordal & Guni Jnsson. (slenzk fornrit 3.) Reyk-
javk.
Droplaugarsona saga. 1950. Ed. by Jn Jhannesson. Reykjavk. (slenzk fornrit. 11.)
Egils saga Skalla-Grmssonar. 1933. Ed. By Sigurur Nordal. Reykjavk. (slenzk fornrit 2.)
Eyrbyggja saga. 1935. Ed. by Einar l. Sveinsson. 1935. Reykjavk. (slenzk fornrit 4.)
Fljtsdla saga. 1950. Ed. by Jn Jhannesson Reykjavk. (slenzk fornrit 11.)
Flamanna saga. 1991. Ed. by Bjarni Vilhjlmsson & rhallur Vilmundarson. Reykjavk. (slenzk
fornrit 13.)
Fstbrra saga. 1943. Ed. by Bjrn K. rlfsson & Guni Jnsson. Reykjavk. (slenzk fornrit 6.)
Gurn Kvaran. 1994. Nfn Austmanna slendingasgum. Sagnaing helga Jnasi Kristjnssyni
sjtugum 10. aprl 1994. Reykjavk. Pp. 269276.
Gunnlaugs saga ormstungu. 1938. Ed. by Guni Jnsson & Sigurur Nordal. Reykjavk. (slenzk forn-
rit 3.)
Grettis saga smundarsonar. 1936. Ed. by Guni Jnsson. Reykjavk. (slenzk fornrit 7.)
Hallfrear saga vandrasklds. 1939. Ed. by Einar l. Sveinsson. (slenzk fornrit 8.)
Harar saga og Hlmverja. 1991. Ed. by Bjarni Vilhjlmsson & rhallur Vilmundarson. Reykjavk.
(slenzk fornrit 13.)
Harris, Joseph. 1976. The Masterbuilder Tale in Snorris Edda and Two Sagas. (ANF 91.) pp. 66101.
Hawes, Janice. 2008. The Monstrosity of Heroism: Grettir smundarson as Outsider. (SS 80.1.) Pp.
1949.
Hieatt, Constance B. 1989. Hrtrs Voyage to Norway and the Structure of Njla. Sagas of the Ice-
landers: A Book of Essays. Ed. by John Tucker. New York. Pp. 27279.
Hollander, Lee M. 1959. The Structure of Eyrbyggja saga. (JEGP 58.) Pp. 222227.
Hnsa-ris saga. 1939. Ed. by Sigurur Nordal & Guni Jnsson. Reykjavk. (slenzk fornrit 3.)
Jones, Gwyn. 1972. Kings, Beasts and Heroes. London.
McCreish, Bernadine. 197879. Structural Patterns in Eyrbyggja saga and Other Sagas of the Conver-
sion. (Mediaeval Scandinavia 11.) Pp. 271280.
McTurk, Rory. 1986. Approaches to the Structure of Eyrbyggja saga. Sagnaskemmtun: Studies in
Honour of Hermann Plsson. Ed. by Rudi Simek, Jnas Kristjnsson, & Hans Bekker-Nielsen. Vi-
enna. Pp. 223237.
Motz, Lotte. 1973. Withdrawal and Return: A Ritual Pattern in the Grettis saga. (ANF 88.) Pp. 91
110.

489
Sawyer, P. H. 1982. Kings and Vikings: Scandinavia and Europe AD7001100. London.
Schach, Paul. 1977. Some Observations on the Generation Gap Theme in the Icelandic Sagas. The
Epic in Medieval Society: Aesthetic and Moral Values. Ed. by Harald Scholler. Tbingen. Pp. 361
81.
Svarfdla saga. 1956. Ed. by Jnas Kristjnsson. Reykjavk. (slenzk fornrit 9).
Vatnsdla saga. 1939. Ed. by Einar l. Sveinsson. Reykjavk. (slenzk fornrit 8.)
Whitaker, Ian. 1977. Some Anthropological Perspectives on Gretla: A Response to Motz. (ANF 92.)
Pp. 14554.

490
 Sweden and the Swedes in English language surveys of the
Viking period
John Kennedy, School of Information Studies, Charles Sturt University, Australia

Introduction
The Vikings are a part of the mental furniture of the average modern persons mind, even as
far from their original homelands and centres of operations as Australia. The major football
team in Australias capital city, Canberra, is known as the Canberra Raiders, and a particu-
larly ferocious and mean-looking horn-helmeted Viking forms a major element in the clubs
official insignia. A search for Viking in the online version of the business telephone direc-
tory for Australias largest city, Sydney, yields over forty entries. A few of the names, such as
Viking Sauna and Husqvarna Viking Sales & Service, probably indicate a wish to draw atten-
tion to the Scandinavian associations of the companys products. Most, however, seem simply
to reflect an association of Vikings with bold, vigorous masculine action and with the sea (or
even just water). Thus one encounters Viking Car Spares & Accessories, Viking Cargo Logis-
tics, Viking Diesel & Electrics, Viking Freight, Viking Marine, Viking Pools, Viking Pumps,
Viking Scaffolding, Viking Sheet Metal & Engineering, Viking Tool & Die, and Viking Win-
dow Cleaning.
Many factors contribute to popular perceptions of the Vikings. The role of book length
surveys of the Viking period intended for the general reader in creating these perceptions (and
amending them where it seems necessary) is difficult to calculate, and may be less than the
authors of such surveys clearly hope, but this does not seem to have deterred authors and pub-
lishers of such books in the last fifty years. The purpose of this paper is to look at an aspect of
these books, their treatment of Viking Age Sweden and the Swedes, with a view to discover-
ing the picture of these they present to non-specialist readers interested enough in the Viking
phenomenon to read a book about it. It will endeavour to detect major elements in the treat-
ment, and changes that have occurred over the years.
The books examined for this purpose are listed in the bibliography. They are limited to
English language books that treat the Viking phenomenon comprehensively, and that are not
restricted to specific aspects of it. For this reason they exclude books devoted solely to Viking
activity in limited parts of their world, such as the British Isles or Russia; books on groups
within Norse society, such as women, kings, or slaves; and studies focused on topics such as
Viking Age religion, ships, or warfare. Books on restricted time spans within the Viking pe-
riod are also excluded, as are those in which the Vikings are but one focus among several,
such as the Time-Life publication Fury of the Northmen (Amsterdam, c.1988) that also in-
cludes treatment of Pre-Columbian Amerindian civilisations and Heian Japan. Books intended
specifically for children and young people are also excluded from this examination. Were
they included, the listing would be far longer than it is. The Vikings are a very popular subject
in juvenile literature, presumably in part because of a hope that boys, notoriously of concern
to educators as being generally more reluctant readers than girls, will enjoy reading about
people who led adventurous lives, fought a lot, and were rather uncouth in their behaviour.
Excluding revised editions and reissues from the count, there are fifty-seven books listed.
With only one exception, the authors, or their publishers, seem to have envisaged that some at
least of the readership would be non-specialists. Even in the case of Peter Sawyers The Age
of the Vikings, a work which set out to challenge conventional scholarly wisdom, the dust
jacket to the second edition quotes a reviewer from The Economist who states that Sawyer
opens up the work of Scandinavian experts [] which would, without his book, remain inac-

491
cessible not only to the general reader but also to most scholars. The exception is a recent
publication, The Viking World, edited by Stefan Brink and Neil Price. In the preface the edi-
tors state that unlike the general syntheses which appear with some regularity the new work
is intended for an academic audience (2008: xix). The observations to be made in this paper
do not generally apply to this work of 717 pages, by some seventy different scholars.
The books listed are nevertheless very diverse. Many are large-format, lavishly illustrated
works; a few, such as the book by Richards and the 1999 Haywood volume, are in a concise
pocket book format. Several are by distinguished historians, archaeologists and literary
scholars in fields relevant to the Vikings, while others are by authors who have also written
on very different topics. Dust jackets indicate that Allan has produced books on Americans in
1920s Paris and on Edgar Allan Poe, while Clements has published Confucius: A Biography.
With only four exceptions, the works of du Chaillu (1889), Mawer (1913), Kendrick, and Ol-
rik (both 1930), the books listed all date from 1960 or later. Before 1960 English language
publishers apparently saw little market need for Viking books aimed at the general reader.
Since then the stream of new publications has not run dry: excluding revisions and reissues
seven books appeared in the 1960s, twelve in the 1970s, six in the 1980s, eleven in the 1990s,
and seventeen have appeared to date in the 2000s.

Terminology
Eric Christiansen (2002: 1) has suggested that referring to all Scandinavians of the centuries
around 1000 as Vikings is a little like using the term cowboy to designate all citizens of the
United States. The books dealt with here frequently observe that there is a case for employing
Viking only to refer to those who went on raids, but that the practice of using it far more
generally for Viking Age Scandinavians of all ages, most callings and both genders is very
well established (e.g., Sawyer 1982: 1). Almost all fall in with the general practice. Among
the books listed, only that of Oxenstierna does not contain Viking or Vikings in its title.
Kirsten Wolf, uniquely, makes it her practice to use Viking only of those who engaged in
banditry, raiding, trading, or military or political action, and to employ Scandinavians for
those who remained at home (2004: 3). Her wide-ranging book is however entitled Daily Life
of the Vikings. Viking clearly is seen by publishers as a selling point.
Many of the books listed note that in the Viking Age (usually seen as extending from the
late eighth century to the second half of the eleventh century) the boundaries of Sweden were
not those of the modern nation of that name. Konstams Historical Atlas of the Viking Age
(2004: 8) shows the salmon pink of Viking Denmark extending well into what is now south-
ern Sweden, the yellow of Viking Norway covering most of the territory of the present Nor-
wegian nation, but the green of Viking Sweden confined largely to the southeast of present
day Sweden, although it does extend to pockets of territory in modern Finland. More rarely
consideration is given to the more basic issues of whether there was in a modern sense a Swe-
den in the Viking Age at all, and whether we can safely refer to Swedes rather than Svear.
Gotar, Gotlanders, etc. They [the Vikings] came from the Norse lands that would one day be
known as Denmark, Norway, and Sweden, but at the start of the Viking Age they did not
think of themselves as Danes, Norwegians, or Swedes, observes Allan (2002: 10). Most au-
thors do use Sweden Swedish and Swedes freely, a mode of description which may in-
deed be questionable but which could be matched easily enough in the many modern accounts
of the early Middle Ages in what are now England, Ireland, and France.

Sweden
One message emerges sharply and repeatedly from reading what Viking books for the general
reader relate about Sweden. It is that Sweden is the part of Viking Scandinavia about which

492
we know least, the part that was the slowest to achieve unity, and the slowest to embrace the
Christianity that tended to be associated with unification, integration into wider European
culture, and the advent of recorded history. Writers display no doubt that Viking Age Sweden
was part of the Viking homelands, and they freely use artefacts discovered in Sweden to illus-
trate and enhance their general observations about the Viking Age. In fact, Swedish artefacts,
notably the runic inscriptions far more prevalent there than elsewhere and the picture stones
from Gotland (an island not all writers regard as Swedish territory in the Viking Age) provide
some of the more impressive illustrations that adorn many of the books. But there are frequent
admissions that for much of the period we do not know what was happening in the territory
now know as Sweden. It was, according to Colleen Batey and her colleagues, always the
wildest and most remote of the Scandinavian lands (1994: 206).
This situation provides something of a contrast with that of the centuries immediately pre-
ceding the Viking Age, a point some writers make. Kendrick suggests that pre-Viking devel-
opments give Sweden title to claim herself the oldest state in Europe (1930: 70); Brndsted
describes pre-Viking Sweden as the most advanced of the Scandinavian countries (1965:
22); and Magnus Magnusson claims that Sweden had a Golden Age [] before she had a
Viking Age (1973: 1314). But few books stress this aspect, and the few which attempt to
outline the political or social developments of specific peoples in Scandinavian during the
Viking period find themselves with limited material relating to the Swedes. Brndsted (in the
Skov translation) provides eleven pages on the ninth century Danes, eight on the Norwegians,
and five and a half on the Swedes. His figures for the tenth century are nine each for the
Danes and Norwegian, and two and a half for the Swedes, and in his chapter on the eleventh
century the Danes and the Norwegians receive ten pages, the Norwegians considered alone
eight, and the Swedes only two. The decades after Brndsted wrote did not produce a marked
improvement in scholarly knowledge, judging from Christiansens consideration of the three
peoples in his 2002 book: four pages are devoted to the Danes, nine to the Norwegians, and
only one to the Swedes (117132). A moderately diligent reader of surveys of the Viking Age
is unlikely to miss the point that where the Viking Age is concerned it is the territories of pre-
sent-day Sweden that are darkest for us today.
What evidence is available is of course put to use. What can be learnt from the Vita An-
skarii is frequently paraphrased or summarised, along with Adam of Bremens rather lurid
description of the pagan temple at Uppsala. Archaeological evidence from the Birka cermeter-
ies, and more recently from the town itself, frequently receives considerable attention. The
artefacts unearthed in or near the town suggest links with much of the known world, and on
the basis of them a colourful picture of a flourishing cosmopolitan town in frequently drawn.
Even writers whose style is generally magisterial allow themselves a rhetorical flourish when
describing Birka:

This meadow-like area, now dotted with birch and juniper, was once a thriving market, thronged
in the the summer months by traders from the whole of Scandinavia, from the realms of the
Carolingians, from Russia, from Arab states of Asia and from Constantinople and other parts of
the Eastern Empire. [] Here was all the noise of workshops, animals and creaking carts,
curses in twenty languages and prayers to a dozen gods. (Foote and Wilson 1970: 207)

There is of course some danger in believing that artefacts from far away lands were necessar-
ily brought to Birka by inhabitants of those lands, or even by people who had themselves
travelled to the locations where the items were manufactured.

The Swedes abroad

493
The Scandinavians of the Viking centuries have captured the modern imagination because
they did not remain at home. If Scandinavian history in the years 8001100 had been a story
of a people who mainly looked inward, gradually unifying and converting to Christianity and
fighting the occasional war among themselves or with their immediate neighbours, there al-
most certainly would be few books relating to them intended for the general reader in the
English-speaking world. It is not surprising, therefore, that what attention the Swedes receive
in the books under consideration largely relates to their adventures outside Scandinavia.
In book after book on the Vikings the point is made that the Swedes of the Viking Age di-
rected their activities east, across the Baltic and into the lands that are now Russia and
Ukraine, and along the great rivers, so that extensive contact was made with Islamic realms,
particularly those in Central Asia, and with Constantinople. A contrast is drawn with the
Norwegians and Danes, who are said to have gone west to the British Isles, the lands on the
western European seaboard and in the western Mediterranean, and across the North Sea to
Iceland, Greenland, and beyond. Some writers do take care to point out that the division was
not absolute, and that the early eleventh century in particular saw significant Swedish partici-
pation in the Danish-led attacks on England, possibly as a result of a drying up of opportuni-
ties in the east (e.g., Sawyer 1971: 56). But the motif of an east-west, Swedes-others division
of Viking activity emerges sharply. Michael Kirkby comments self-referentially on the phe-
nomenon, when after two chapters entitled Early Raids and Colonization he begins one
entitled The Eastward Movement by observing

The comparted nature of the Viking movement is indicated by the fact that it has been possible
for the last two chapters to be written with scarcely a mention of Sweden or eastern Europe and
that during the course of the present one hardly any reference will be necessary to any point on
the map west of longitude 15. (1977: 104)

Only slightly less ubiquitous is the idea that the nature of the activity of the Swedish Viking
in the east was different from that of their fellow-Scandinavians in the west. It is widely sug-
gested that whereas the latter engaged in raiding and plunder, and later in conquest and the
establishment of farms, the former were primarily traders. The picture is sometimes modified
in various ways: it is observed that trading did also take place in the west; that Vikings in the
east needed to fight at times, and were not averse to taking opportunities for raiding and plun-
der when they arose; and that circumstances in the Finnish and Slavic lands, which lacked
monasteries and significant towns to pillage and organised states to blackmail, help explain
the differences.
Nevertheless, the idea of the Swedes in the east as primarily traders has been persistent. In
1913 Mawer wrote that The Swedes were for the most part interested only in Eastern Europe,
and there by way of trade rather than battle [] (9). Poertner in 1975 observed that the
Swedes in western Russia [] were mostly traders interested in the turnover of goods, rather
than pirates and brigands killing for booty (50); Wilson in 1989 proclaimed that trade is
practically the whole basis of the Viking activities in the East (101); and as recently as 2002
Konstam could write:

The Viking presence in Russia was completely different from that in the rest of Europe. They
came as traders, then governed small townships, which in turn developed into city-states. Power
was the by-product of trade. (170)

The quotation from Konstam serves as a reminder that even those who view the Swedes in the
east as primarily traders acknowledge that they may also have had a role as leaders and state
builders. This, of course, is at the core of the so-called Normanist controversy, which now has

494
an acrimonious history of over 250 years. Perhaps it is not surprising that books that often
seem intended to celebrate the Viking achievement by indicating the breadth of their geo-
graphical reach and displaying the artefacts of Viking culture in strikingly artistic photographs
should incline, often implicitly, to a Normanist view that argues Swedes did have a significant
role to play in the formation of the early Russian state. But while Kendrick could write in
1930 that The Swedes were the folk who achieved the mightiest and most remarkable tri-
umph of viking history, namely the creation of an independent Swedish-Russian state (910)
more recent writers virtually without exception present a more moderate view. A very com-
mon claim is that the debate has cooled, with the abandonment on both sides of more extreme
claims, a process said to be aided by a political environment allowing less ideologically moti-
vated access to increased archaeological evidence.
Archaeological finds, and in particular the discovery of vast quantities of Islamic dirhems
in the soil of what is now Sweden, leave no doubt that Viking Age Sweden was influenced by
its contacts with the east (From Viking to Crusader, 1992: 78), but there is little discussion of
the social effects of such influence in the books under consideration, probably for want of
firm evidence. The coin hoards, with the coins carefully cleaned so as to draw attention to the
Kufic script, provide some striking photographs, and it is often stated that the inflow of silver
which the coins indicate had a role in stimulating the economy of the entire Scandinavian re-
gion. There are some suggestions that Orthodox Christianity, stemming from Kiev and ulti-
mately from Constantinople, may have had some influence in Sweden (e.g., Roesdahl 1991:
166), but none of the books seriously considers the possibility that Sweden could have been
drawn permanently into an Orthodox Christian orbit. Arbman and his colleagues, who do
raise the issue, quickly dismiss the possibility:

Byzantium was too far away to exert serious religious influence. The Roman Church prevailed
in Scandinavia because the bishopric of Hamburg was close by. (1974: 140).

Changing perspectives
In Christiansens view The map, and the story of Nordic immigration to the east is changing
too fast for a survey to catch (2002: 217). However, the books considered here do not gener-
ally give an impression of such rapid development in regard to our understanding of Sweden
and the Swedes abroad. Examined in chronological order they do indicate that more compre-
hensive archaeological evidence from within Sweden has become available over the years,
notable in the case of Birka, and that the years since the end of the Soviet Union in 1991 have
brought more archaeological evidence from what were once its territories, along with a
greater freedom to examine and study it. Holman detects in scholarship a desire to move be-
yond the simplistic image of the Vikings as destructive raiders in the West and constructive
traders in the East (2005: 12). But an attempt to highlight how Viking Age surveys consid-
ered here have changed over time (apart from becoming more sophisticated in terms of physi-
cal book production) would probably focus not on Sweden and the expeditions east but on
such factors as enhanced attention to the role of the Smi in Viking Age Scandinavia, and on
more serious attempts to explore as a cultural phenomenon the modern fascination with the
Vikings.

Role of the Swedish experience

Some surveys of the Viking Age have little to say about the Swedes abroad: Viking Empires
by Forte, Oram, and Pedersen (2002) and The World of the Vikings by Hall (2007) are impor-
tant recent examples. Much more commonly the expeditions to the east are given a chapter of

495
their own. But it is commonly just one chapter out of half a dozen or more recounting Viking
exploits abroad, with far more attention devoted to the expeditions west, to Britain and Ire-
land, the Frankish realms, Iceland, Greenland, and Vinland. This of course reflects the fact
that the perspective of the writers is frequently a British one, even if very few would be com-
fortable with Magnus Magnussons observation that At 6 p.m. on Saturday 14 October 1066
the world changed. The Viking Age was effectively over (1980: 313). It seems fair comment
that in Viking books for the English-speaking general reader the activities of the Swedish Vi-
kings receive less attention than their role in European history deserves.
Consideration of the expeditions east is included in the books partly because it would be
difficult to argue that an overview of the Viking Age was complete without them. One might
just possibly work on the premise that Vikings were people who went raiding in longships,
and that this excluded most of the eastern activity, but such a view would also exclude much
activity in Western Europe that most scholars would want to consider Viking, not to mention
the fascinating voyages of exploration and settlement across the North Atlantic. However, the
eastern expeditions have another important function in the surveys. They bring to the Viking
story, which many of the books seem written to celebrate, an element of the exotic east, plac-
ing into the Viking story the misty beginnings of Russia and glimpses of Byzantium and Cen-
tral Asia, of Kiev and Novgorod, Constantinople, Bukhara, Tashkent, and Samarqand. Thanks
to them, the Viking world can be shown to stretch from the shores of Newfoundland almost to
the borders of China. The opportunity to include with so much else some illustrations of ex-
otic places and of artefacts from far off lands is understandably one that appeals to authors
and their publishers, even if their attentions are focused far more on the greyer lands of west-
ern Europe.

Bibliography
Allan, Tony, 2002: Vikings: The Battle at the End of Time. London: Baird.
Arbman, Holger, 1961: The Vikings. Trans. by A. Binns. London: Thames & Hudson.
Arnold, Martin, 2006: The Vikings: Culture and Conquest. London: Hambledon Continuum.
Arnold, Martin, 2007: The Vikings: Wolves of War. Lanham, Maryland: Rowman & Littlefield.
Baker, Alan, 2004: The Viking. Hoboken, New Jersey: Wiley.
Batey, Colleen, Clarke, Helen, Page, Raymond, and Price, Neil S., 1994: Cultural Atlas of the Viking
World. Ed. by J. Graham-Campbell. Oxford: Andromeda.
Birkebk, Frank and Barren, Charles, 1975: Sea Wolves: The Viking Era. Gothenburg: Nordbok.
Brent, Peter, 1975: The Viking Saga. London: Weidenfeld & Nicolson.
Brndsted, Johannes, 1960: The Vikings. Trans. by E. Bannister-Good. Harmondsworth: Penguin. Re-
translated 1965 by K. Skov. Harmondsworth: Penguin.
Chartrand, Ren, Durham, Keith, Harrison, Mark and Heath, Ian, 2006: The Vikings: Voyagers of
Discovery and Plunder. Botley, Oxford: Osprey.
Christiansen, Eric, 2002: The Norsemen in the Viking Age. Oxford: Blackwell.
Clements, Jonathan M., 2005: A Brief History of the Vikings: The Last Pagans or the First Modern
Europeans? New York: Carroll and Graf.
Clements, Jonathon M., 2007: Vikings. New York: Sterling.
Cohat, Yves, 1992: The Vikings: Lords of the Seas. Trans. by R. Daniel. London: Thames & Hudson.
Donovan, Frank R. and the editors of Horizon Magazine, 1965: The Vikings. London: Cassell.
Du Chaillu, Paul B. 1889: The Viking Age: The Early History, Manners, and Customs of the Ances-
tors of the English-Speaking Nations. Illustrated from the Antiquities Discovered in Mounds,
Cairns, and Bogs, as well as from the Ancient Sagas and Eddas. 2 vols. London: Murray.
Foote, Peter and Wilson, David M., 1970: The Viking Achievement. London: Sidgwick & Jackson.
Reissued 1980 with Supplement. London: Sidgwick & Jackson.
Forte, Angelo, Oram, Richard, and Pedersen, Frederik, 2005: Viking Empires. Cambridge: CUP.
From Viking to Crusader: The Scandinavians and Europe 8001200. Ed. by E. Roesdahl and D. M.
Wilson. 1992. New York: Rizzoli.

496
Graham-Campbell, James, 1980: The Viking World. London: Lincoln/Weidenfeld & Nicolson. 2nd
edn. 1989. London: Lincoln.
Graham-Campbell, James and Kidd, Dafydd, 1980: The Vikings. London: British Museum.
Hall, Richard, 2007: The World of the Vikings. New York: Thames & Hudson.
Haywood, John, 1995: The Penguin Historical Atlas of the Vikings. Harmondsworth: Penguin.
Haywood, John, 1999: The Vikings. Stroud, Gloucestershire: Sutton.
Haywood, John, 2000: Encyclopaedia of the Viking Age. London: Thames & Hudson.
Healey, Tim, 1996: Life in the Viking Age. London: Readers Digest.
Holman, Katherine, 2003: Historical Dictionary of the Vikings. Lanham, Maryland: Scarecrow.
Jones, Gwyn, 1968: A History of the Vikings. Oxford: OUP. Rev. edn. 1984. Oxford: OUP.
Kendrick, Thomas D., 1930: A History of the Vikings. New York: Charles Scribners.
Kirkby, Michael H., 1977: The Vikings. Oxford: Phaidon.
Klindt-Jensen, Ole, 1970: The World of the Vikings. Trans. by C. Gibbs and G. Unwin. London: Allen
& Unwin.
Konstam, Angus, 2004: Historical Atlas of the Viking World. London: Mercury.
La Fay, Howard, 1972: The Vikings. Washington: National Geographic.
Logan, F Donald, 1983: The Vikings in History. London: Hutchinson. 2nd edn. 1991. London:
Routledge. 3rd edn. 2005. London: Routledge.
Madsen, O. 1976: The Vikings. Trans. by D. Macrae. Geneva: Minerva.
Magnusson, Magnus, 1973: Viking Expansion Westwards. London: Bodley Head.
Magnusson, Magnus, 1980: Vikings! London: Bodley Head/British Broadcasting Commission. Reis-
sued 2003 as The Vikings. Stroud: Tempus.
Mawer, Allen, 1913: The Vikings. Cambridge: CUP.
Olrik, Axel, 1930: Viking Civilization. Completed by H. Ellekilde. Trans. by J.W. Harrtman and H.
Astrup-Larsen. New York: American-Scandinavian Foundation.
Oxenstierna, Eric, 1966: The Norsemen. Trans. and ed. by C. Hunter. London: Studio Vista.
The Oxford Illustrated History of the Viking Age. Ed. by P.H. Sawyer, 1997. Oxford: OUP.
Pendlesonn, K.R.G., 1980: The Vikings. New York: Mayflower.
Poertner, Rudolf, 1975: The Vikings: Rise and Fall of the Norse Sea Kings. Transl and adapted by S.
Wilkins. London: St James Press.
Richards, Julian D., 2005: The Vikings: A Very Short Introduction. Oxford: OUP.
Roesdahl, Else, 1991: The Vikings. Trans. by S.M. Margeson and K. Williams. London: Allen Lane.
Sawyer, Peter H., 1962: The Age of the Vikings. London: Arnold. 2nd edn. 1971. London: Arnold.
Sawyer, Peter H., 1982: Kings and Vikings: Scandinavia and Europe AD 7001100. London:
Methuen.
Simpson, Jacqueline, 1967: Everyday Life in the Viking Age. London: Batsford. Reissued 1980 as
The Viking World. London: Batsford.
Time-Life Books, 1993: Vikings: Raiders from the North. Alexandria, Virginia: Time-Life.
Time-Life Books, 1998: What Life was Like when Longships Sailed: Vikings AD 8001100. Alexan-
dria, Virginia: Time-Life.
The Viking. Ed. by B. Almgren et al. 1974 Gothenburg: Tre Tryckare/ n.p.: Crescent.
The Viking World. Ed. by S. Brink in collaboration with N. Price. 2008. London: Routledge.
Vikings: The North Atlantic Saga. Ed. by W.W. Fitzhugh and E.I. Ward. 2000. Washington: Smith-
sonian Institute in association with the National Museum of Natural History.
Wernick, Robert and The Editors of Time-Life Books, 1979: The Vikings. Amsterdam: Time-Life.
Wilson, David M., 1971: The Vikings and their Origins. London: Thames & Hudson. 2nd edn. 1989.
London: Thames & Hudson.
Wolf, Kirsten, 2004: Daily Life of the Vikings. Westport, Connecticut: Greenwood.
Wooding, Jonathan, 1997: The Vikings. Sydney: Lansdowne.

497
 Celtic and Continental handicraft traditions;
Template use on Gotlandic Picture Stones analysed by 3D-
scanning
Laila Kitzler hfeldt, Dept. of Archaeology and Classical studies,
Stockholm University, Sweden
The Gotland Picture-Stones belong to the most spectacular and informative artefacts of the
Iron Age and Viking Age in Sweden. This treasure of images is generally considered to illus-
trate myths and cultural phenomena in the Icelandic sagas. In this study, a number of Viking
Age Gotland picture stones have been documented and analysed by optical 3D-scanning, with
regard to template use. The results will be discussed in the context of handicraft traditions.

Geometrical construction Celtic Ornament

It has earlier been noticed that the interlace ornament on the Gotland picture stones might
show a Celtic influence (Lindqvist 1941). Continental and Insular influences on Scandinavian
art styles have been extensively discussed in earlier research (e.g. berg 1941, 1948; Nerman
1935, 1975; Holmqvist 1952, 1977; Eshleman 1983 and others), and will not be further dis-
cussed. Dr Uaininn OMeadhra has pointed out to me, though, that Celtic interlace is applied
on objects according to Geometrical principles, demanding a mathematical approach where
intersection points are marked with a tool and the ornament lines adapted to these points. This
method implies a certain level of intellectual insight into the principles of the ornament (cf
OMeadhra 1987). When this art was performed to its full extent, it was an expression not
only for aesthetics but also expressed the striving for recreation of divine principles of har-
mony and Geometry, the actual base for Creation and a quest for spiritual truth and aesthetical
perfection (Brown 2003:297).
The Loughcrew Slips is a find category that might have been a means for composition of
art motifs according to Geometrical principles. Points have been marked by using compasses
in order to show the principles for construction rather than the complete ornament. These
finds have been dated to Early Iron Age and have been interpreted as remains of the trial
pieces of a skilled craftsman. Probably, they have been used as a form of pattern book for a
metal handcrafter or sculptor (OMeadhra 1987:128, fig. 88).
In some degree, the interlace patterns well known from the picture stones also appear on
motif-pieces of bone (e.g. OMeadhra 1979, Pl.24. No 58). The distribution of motif-pieces
indicates that they are of Irish origin and finds in other places suggest a connection with the
Insular area. The aim with motif-pieces is to illustrate a design for complicated ornament in
concrete form, in stead of working it out mathematically. In contrast to the Loughcrew slips,
there are no grids or similar aids to achieve intricate patterns (OMeadhra 1987:173f). Motif
pieces are found in Scandinavia not earlier than c.1000 AD, in urban find contexts and even
then rarely and of a simple design. As far as I know, none have been found on Gotland so far.
The difference between the Loughcrew slips and the later motif-pieces is that the first men-
tioned illustrate principles of construction, whereas the latter show complete motifs.
In manuscripts, complicated patterns may be constructed by grid points and lines, con-
nected by circle segments drawn by compasses (Henry 1965; Bain 1951; Blidmo 1976:67f
and literature cited therein). Another characteristic feature is the transition of motifs in panels,
which can be inserted as units into larger compositions.
Both geometrical principles and motif panels have been applied in a most sophisticated
manner in the manuscript The Lindisfarne Gospel (Brown 2003). Actually, there are some

498
similarities between the Gotland picture stones and this manuscript. However, only the more
simple variants of entrelac have been applied on picture stones. Several variants of entrelac
appear within the same section of ornament, perhaps an analogy to the variation of knots
found on e.g. the picture stones Ardre VIII and Alskog Tjngvide I (Fig. 2; Kitzler hfeldt
2009).
The similarities between picture stones and manuscripts reappear in insular stone sculpture,
for example on Isle of Man and Lindisfarne. The recurrent variation in the entrelac is found
on stone crosses on Isle of Man, e.g. The Thor Cross from Bride (Kermode1907: Pl. 97a). On
the same cross we also find the dog-like creature with hunched back and rolled up tail, very
similar to the dog (?) on the picture stones Alskog Tjngvide I and Ardre VIII (Fig. 2). A
stone cross from Ballaugh has an interesting shape, reminding of the so called mushroom-
shaped picture stones (Kermode 1907:154, Pl.77A). On The Sigurd Stone in Halton, we find
figural scenes in panels, e.g. a smithy and smithing tools (Kermode 1907:173, Fig.55). On
Ardre VIII several scenes seem piled vertically on top of each other, as on a pillar. The panels
with intertwined animal bodies have a certain likeness to manuscript art (Brown 2003 Pl.20).
Finally, the name-stones from Lindisfarne are worth considering as parallels to the Gotland
picture stones. The name-stones are dated to the 7th and 8th centuries AD and are decorated
with borders of entrelac. Furthermore, they have bilingual inscriptions in English and Latin
written with Roman letters, runes and half-uncial scripts. The last mentioned script has evi-
dent parallels to The Lindisfarne Gospels (Brown 2003:227 Fig.91). Usually, reproductions
illustrate the richly sculptured Insular stone crosses with deep cut plastic relief but these more
plain name-stones have more in common with the Gotland picture stones.

Templates and pattern books Continental tradition

In Roman handicraft too, Geometrical drawing techniques were used. These had roots in Ar-
chitecture and were applied on floor mosaics. The mosaic patterns were primarily constructed
by compasses and ruler according to Classical ideals, but also by templates (Arrhenius 1971;
Blidmo 1976:56 and literature cited therein). However, the extensive use of art on every kind
of object promoted the use of templates within Roman art (Wilson 1999). The Roman heri-
tage lives on through the Late Antique period and into Merovingian and Carolingian art. In
this context, the widespread use of templates and pattern books can be understood as an inte-
grated part of Continental handicraft tradition.
Among the Anglo Saxon stone crosses, where the entrelac of the high-qualitative
crosses has been constructed by a square grid, we also see instances of copying and template
use (Adcock 1978:33). According to Richard Bailey, templates have been used so that figures
could be mirrored, divided into sections, adapted in size or combined in various ways. The
same human figure could be rendered with different attributes, which completely altered the
meaning of the image. The same template appearing on different sites has been interpreted as
a sign of a central workshop or an itinerant sculptor. In each case Bailey means that monu-
ments sharing the same templates are contemporary within one generation (Bailey 1978:184f;
1980:240253). On rune stones too, templates might have been used for some parts
(OMeadhra 1987; Herschend 1998:105f).
The consequence of use of pattern books is that iconographical motifs as well as formal
elements can be transmitted from place to place and wander through studios and generations
(Scheller 1963:3). Only a few have been preserved. There are two known finds from 3rd C AD
Egypt but then there is a gap until 10th C AD, after which point they become more usual
(Scheller 1995). The aim was to transmit motifs to those who had no access to the original.
Therefore, it is more likely that an itinerant craftsman owned a pattern book than a sedentary
one (Scheller 1995:41f). Collections of models do not presume intimate cooperation between

499
handcrafters. On the contrary, these models can be used freely. The compiler of the original
has no control of how the ornaments come to use later. As a consequence, there may be large
time gaps which of course effect the dating of the later objects. For example, a few elements
of Ringerike style are to be found in Anglo Saxon manuscripts. Signe Horn Fuglesang means
that this is more likely a result of motifs being combined in a pattern book than an instance of
collaboration between Scandinavian handcrafters and English illuminators (Fuglesang
1978:212). Another example is a pattern book dated to 10th C AD, which include Late An-
tique and Carolingian art as well as Insular entrelac and a Merovingian animal motif (Scheller
1963:49). This means that if carvers of picture stones have had access to pattern books, it
might prove extremely difficult to find the nearest source of inspiration.

3D-scanning of Gotland Picture Stones

The interpretation of the picture stones as pictorial counterparts to Icelandic sagas is most
often dependent of the identification of animals and attributes. 3D-scanning may help to re-
veal details in the carvings which make these identifications more obvious or change the
meaning totally. The point of departure for present interpretations is the paintings made in the
1930s or even earlier (Lindqvist 1941-42). But the carved lines of the motives are very tiny
and earlier interpretations can be very much debated. 3D-scanning is useful for the reading
and interpretation of inscriptions and ornament on weathered stone surfaces and the visual
impression of the 3D-image is not disturbed by colours. We can study the picture-stone as a
new find, without prejudices about what we are going to see. Thus scholars can become more
independent of tradition. Another application concerns dating, namely the inner relative chro-
nology of motifs and inscriptions. Which was applied first? Is the runic inscription secondary
to the pictures or is it integrated into the composition? Finally, local groups of handicraft tra-
ditions can be identified.
This study includes five picture stones of Sune Lindqvists group D; Alskog Tjngvide I,
Alskog K, Ardre VIII, Garde Bota and Nr Smiss I. Sune Lindqvist dated these stones to the
latter part of the 8th C AD, possibly into the 9th C AD (Lindqvist 1941:120f; Lamm 2006).
Equipment used is a high resolution, non-touch optical 3D-scanner; ATOS II (GOM). The
resolution used is 0,27mm between the measuring points. This study has been published
elsewhere (Kitzler hfeldt 2009), therefore only the main results will be discussed and one
example only is illustrated.
Since Lindqvists chronology leaves a lacuna for the 10th C AD, several scholars have tried
to revise it. Eshleman means that the picture stones of groups C and D are dependent of the
Carolingian renaissance in 790830 AD, and therefore cannot be earlier than that (Eshleman
1983). Lisbet Imer balanced the arguments for dating according to motifs, border ornament
and runic inscriptions (Imer 2004). In contrast to the opinion of Nerman (1947) and Lindqvist
(1942), Imer does not attach any great weight on the border ornament as a dating criteria. The
only chronological stratification in the border ornaments is that they seem to increase in com-
plexity (Imer 2004:100). There might be a problem of secondary runic inscriptions, because
the runological datings are younger than indicated by the border ornaments (Sanness Johnsen
1968:80). This objection is valid, as is shown by the fact that on some rune stones, the runes
clearly have been adapted to the ornament (e.g. Alskog Tjngvide). Understandably though,
Imer still chooses to attach greater weight to the runological dating since there is no precise
dating available of the separate style elements of the border ornament (Imer 2004:86, 94).
According to Imer, Alskog Tjngvide I, Alskog K and Ardre VIII can be dated to 10th C AD.
Garda Bote is dated to 9th C AD and Nr Smiss I to c.750900 AD (Imer 2004:104f, Tabel
1718).

500
 Fig. 1. Horses on Alskog
Tjngvide and Ardre VIII. The
outer contours of the main body
coincide, but details vary.

Results make it evident that templates in scale 1:1 frequently have been used, but with vary-
ing skill or care. This cannot always be seen only by studying the painted motifs by the naked
eye, but appears when the carving lines are studied in the 3D-images and the outer contours of
separate motifs are superimposed. In this case, the criterion for motifs being applied by a tem-
plate is that the outer contours and starting and crossing points of extremities coincide (Fig.
1). On the other hand, details inside the figure may vary. The stone surfaces are sometimes
bulging and the template might have moved while drawing the contour, which has to be ac-
counted for. In addition, weathering and wearing have contributed to the figures obscure ap-
pearance. With these disadvantages in mind, the figures still fit together to a surprisingly high
degree.

Fig. 2. Template relations on Alskog Tjngvide and Ardre VIII. Modified after Lindqvist 194142.

501
The picture stones Alskog Tjngvide and Ardre VIII have several templates in common (Fig.
2). Details in the carvings reveal that the templates have been used by two individuals, where
the one is more careful than the other. It can be noticed that there is variation in attribute the
objects in the riders hand are different. In my view, Ardre VIII shows a static relation to the
motif. This can be seen on the horse, which does not really fit into the available space. The
hinter legs have simply been cut off and made shorter (Kitzler hfeldt 2009:141, Bild 15).
The horse probably would have been better adapted to the available space, had it been drawn
by free hand. The knots in the border ornaments have been applied by templates too, not by
construction with grid points. The templates have sometimes been superimposed in spite of
lack of space, sometimes figures have been squeezed in instead of decreasing the size of the
desired motifs. Another example of this is the border ornament on Nr Smiss I. On the same
stone, the fighting opponents have been produced by mirroring the same warrior (Kitzler h-
feldt 2009:146f, Bild 3435).
Another example of template use is the procession on the picture stone from Garda Bote.
At a first glance, there is no striking resemblance between the figures but a closer study re-
veals that they fit together at easily identifiable points. The single human figure in the upper
part of the stone has been made with the same template, and has been adorned with a drinking
horn as an attribute (Kitzler hfeldt 2009:145, Bild 3033).
To conclude, templates have been saved and used on several stones, in order to repeat hu-
man figures in a procession (Garde Bote), to mirror figures (Nr Smiss I, Garde Bote) and to
apply knots and interlace (Alskog Tjngvide, Ardre VIII, Nr Smiss). The figures have been
adorned in various ways and it must be said that the templates have been used in a creative
manner. There are some similarities to Insular manuscript art and stone sculpture but the
mathematical principles of Celtic entrelac have not been applied. The use of templates for
border ornaments may indicate an attempt to apply Insular (Celtic) ornament without master-
ing the underlying principles of construction. The templates are more dumb and associates
to a Continental tradition of full-scale templates and pattern books.

Templates as signs of workshops

Template use may be a sign of the existence of a picture stone workshop, in the sense of a
group of carvers sharing a set of motifs and tools. Whether an instance of template use should
be interpreted as a sign of workshop also depends on how the templates have been used. If the
templates have been produced in a tolerably resistant material, this may tell us that the pro-
ducers had expectations to have use for them. Which indeed they had, regarding for example
the above mentioned Ardre VIII and Alskog Tjngvide.
The subject matters of itinerant or sedentary craftsmen, their social status and the existence
of guilds have been extensively discussed in Archaeology (cf OMeadhra 1987:169 and litera-
ture cited therein). Templates show that design is an activity separated from production. In
other contexts, this separation may indicate increasing specialization, professionalization and
possibly mass production (cf Blidmo 1976). In the Viking Age, some organisational devel-
opment appears to take place on Gotland and in Southern mainland Sweden, possibly caused
by political circumstances. The singular and unique artifact types produced in the 8th and 9th C
AD seem to be replaced by standardized products in the 10th C AD (Thunmark-Nyln
1995:118f; Callmer 1995:65f).
Do these changing circumstances have relevance for picture stones? The picture stones are
local, immobile and they are monuments neither utility goods nor jewelry. Picture stones
have signs of co-operation between carvers, but I doubt that a very great demand would be the
reason for this. What is actually a great demand for picture stones? They cannot be com-

502
pared to e. g. jewelry production, where some types appear in a large number but in few vari-
ants only (Arrhenius 1973; Blidmo 1976:12). The Gotland picture stones are locally confined
to the island Gotland, a fact that limits the scope for large organizations.
Anders Carlsson has suggested that monumental art may maintain archaic feature to a
higher degree than small finds (Carlsson 1983). As seen above, this might also be a conse-
quence of using pattern books ornament may gain a long lifetime. Imers dating of some of
the picture stones to the 10th C AD has the consequence that motif-pieces might be brought
into consideration when searching for models. In the 10th C AD, pattern books too might have
been more widely used than before, depending on how we interpret the find situation. Thus, if
picture stones have been decorated by means of templates in the 10th C AD, this is in accor-
dance with a contemporary trend in society.
I suggest that the carvers were attracted by the Celtic entrelac, but in order to apply it they
rather used full scale templates than geometrical principles of construction. The picture stones
have elements similar to traits in manuscripts and grave monuments from Lindisfarne and on
stone crosses in Northumbria and Isle of Man. Nevertheless, the connection is not necessarily
particularly tight. The templates may well have been designed by copying from a pattern col-
lection brought to Gotland. It would have been neat and nice to regard a geometrically de-
signed interlace as a characteristic of direct Western Insular influence and the use of templates
as a pure Southern and Continental influence. Regretfully, it is not that simple. Pattern collec-
tion may wander freely and loosen the personal connections. The compositions become di-
vided into smaller units and later combined again into new constellations.

Summary
I this study, Gotland picture stones dated to c. 7501000 AD have been analysed with an opti-
cal 3D-scanner with regard to template use. Results show that templates in scale 1:1 have
been used to apply figural motifs and border ornament. The use of templates for knots and
entrelac on the Gotland picture stones may indicate an attempt to apply Insular (Celtic) orna-
ment without mastering the underlying principles. The templates are more dumb and rather
associate to a Continental tradition of full-scale templates and pattern books. Template use
might indicate the presence of a picture stone workshop in the sense of a group of stone ma-
sons sharing the same kit of motifs and tools. This tells us something about the social circum-
stances in which picture stones with Saga-like motifs were produced. Another consequence of
template use is that ornament may have a long life time, which complicates dating. For inter-
pretation of the motifs on the picture stones in a Saga-context, it has to be remembered that
the motifs may seem clear enough in earlier publications, but attributes are difficult to identify
on weathered picture stones and earlier interpretations can be debated. Varying attributes may
completely change the meaning of the legendary sequences.

Acknowledgements
I wish to thank Dr Uaininn OMeadhra for valuable comments. This study is a part of my 2-
year project 3D-scannning of the Gotland Picture Stones; Workshops, Iconography and Dat-
ing financed by The Anna Ahlstrm & Ellen Terserus Foundation. The equipment has been
financed by The Swedish Research Council.

Bibliography
Adcock, George. 1978: The theory of interlace and interlace types in Anglian sculpture. In:
Anglo-Saxon and Viking Age Sculpture and its Context: Papers from the Collingwood
Symposium. Ed. by J. T. Lang & W. G. Collingwood. Oxford. (BAR 49) Pp.3345.

503
Arrhenius, Birgit. 1971. Granatschmuck und Gemmen aus nordischem Funde des frhen Mit-
telalters. (Studies in North-European archaeology. Series B). Stockholm.
Arrhenius, Birgit. 1973. East Scandinavian Style I A Review. Medieval Archaeology, Vol.
17. Pp. 2642.
Bailey, Richard. 1978. The Chronology of Viking Age Sculpture in Northumbria. In: Anglo-
Saxon and Viking Age Sculpture and its Context: Papers from the Collingwood Sympo-
sium. Ed. by J. T. Lang & W. G. Collingwood. Oxford. (BAR 49) Pp. 173204.
Bailey, Richard. 1980. Viking age sculpture in Northern England. London.
Bain, George. 1951. Celtic Art. The Methods of Construction. Glasgow.
Blidmo, Roger. 1976. Smyckeproduktion inom yngre jrnlderns metallkonst. En studie i
produktionsbetingelser och ritteknik. D-uppsats i arkeologi. Stockholm.
Brown, Michelle. P. 2003. The Lindisfarne Gospels. Society, Spirituality and the Scribe.
London.
Callmer, Johan. 1995. Hantverksproduktion, samhllsfrndringar och bebyggelse. Iakttagel-
ser frn stra Sydskandinavien ca. 6001100 e.Kr. In: Produksjon og samfunn. Om erverv,
specialisering og bosetning i Norden i 1. rtusen e.Kr. Ed by H. G. Resi. Oslo. (Varia 30)
Pp. 3972.
Carlsson, Anders. 1983. Djurhuvudformade spnnen och gotlndsk vikingatid. (Studies in
Archaeology 5). Stockholm.
Eshleman, Lori. E. 1983. The Monumental Stones of Gotland: A Study in Style and Motif.
Vol. 12. Minnesota.
Fuglesang, Signe Horn. 1978. Stylistic Groups in Late Viking Art. In: Anglo-Saxon and Vi-
king Age Sculpture and its Context: Papers from the Collingwood Symposium. Ed. by J. T.
Lang & W. G. Collingwood. Oxford. (BAR 49) Pp. 205216.
Henry, Francis. 1965. Irish Art in the Early Christian period (to 800). Part 1. London.
Herschend, Frands. 1998. UbiR, YbiR, ybir r det U485 Ofeg pir? Fornvnnen 93:2. Pp.
97111.
Holmqvist, Wilhelm. 1952. De ldsta gotlndska bildstenarna och deras motivkrets. Fornvn-
nen 1952. Pp.120.
Holmqvist, Wilhelm. 1977. Vr tidiga konst. Stockholm.
Imer, Lisbet. 2004. Gotlandske billedsten dateringen af Lindqvists gruppe C og D. Aarbger
for nordisk oldkyndighed og historie 2001. Pp. 47111.
Kermode, Philip M. C. 1907 (Reprint 1994). Manx Crosses. With an introduction by David
M. Wilson. Angus.
Kitzler hfeldt, Laila. 2009. Keltiskt eller kontinentalt? Om mallanvndning p Gotlands
bildstenar. In: Spaden och pennan. En minnesskrift till Erik B Lundberg och Bengt G S-
derberg. Ed. by C. Hedenstierna-Jonson, T. Svensson & E. Lundberg. Visby. Pp. 131154.
Lamm, Jan Peder. 2006. Tjngvide. Reallexikon der Germanischen Altertumskunde. Band 31.
Berlin.
Lindqvist, Sune. 194142. Gotlands Bildsteine. (Kungl. Vitterhets-, historie- och antikvitets-
akademien ; Monografier 28). Stockholm.
Nerman, Birger. 1935. Die Vlkerwanderungszeit Gotlands. (Kungl. Vitterhets-, historie- och
antikvitetsakademien. Monografier 21). Stockholm.
Nerman, Birger. 1947. Arkeologisk datering av vendeltidens nordiska runinskrifter. 6. Bild-
sten frn Tjngvide, Alskogs sn, Gotland. Fornvnnen 1947. Pp.133136.
Nerman, Birger. 1975. Die Vendelzeit Gotlands 1:1 Text. (Kungl. Vitterhets-, historie- och
antikvitetsakademien ; Monografier 55). Stockholm.
Nyln, Erik & Lamm, Jan Peder. 2003. Bildstenar. Stockholm.

504
OMeadhra, Uaininn. 1979. Motif-Pieces from Ireland. Early Christian, Viking and Roman-
esque Art. 1. Catalogue. (Theses and Papers in North-European Archaeology 7). Stock-
holm.
OMeadhra, Uaininn. 1987. Motif-Pieces from Ireland. Early Christian, Viking and Roman-
esque Art. 2. A Discussion. (Theses and Papers in North-European Archaeology 17).
Stockholm.
Sanness-Johnsen, Ingrid. 1968. Stuttruner i vikingatidens indskrifter. Oslo.
Scheller, Robert. W. H. S. 1963. A Survey of Model Books. Haarlem.
Scheller, Robert. W. H. S. 1995. Exemplum. Model-Book Drawings and the Practice of Artis-
tic Transmission in the Middle Ages (ca. 900 ca. 1450). Amsterdam.
Thunmark-Nyln, Lena. 1995. Spnnekonstruktion och spnneproduktion. In: Produksjon og
samfunn. Om erverv, specialisering og bosetning i Norden i 1. rtusen e.Kr. Ed. by H. G.
Resi. Oslo. (Varia 30) Pp. 117122.
Wilson, Eva.1999. Roman designs. (British Museum pattern books). London.
berg, Nils. 1941. Keltiska och orientaliska stilinflytelser i vikingatidens nordiska konst.
(Kungl. Vitterhets-, historie- och antikvitetsakademiens handlingar, 46:4). Stockholm.
berg, Nils. 1948. Vendeltida frbindelser med fastlandsgermanska och insulra kretsar.
Fornvnnen 1948. Pp. 111122.

505
 Eksotiske vesener i islandsk ridderlitteratur fremstilling og
funksjon
Karoline Kjesrud, Institutt for lingvistiske og nordiske studier, Universitetet i Oslo, Norway
I en del av ridderlitteraturen nedskrevet p Island i det 14. og 15. rhundret, reiser gjerne
vestlige helter mot sr og st i sken etter eventyr. Heltene er ofte vestlige, men kan ogs
komme fra for eksempel Syria (Adonias i Adonias saga) eller Afrika (Blvus i Viktors saga
ok Blvus, opptrer riktignok sammen med den franske Viktor). Felles for heltene er at de
mter en rekke utfordringer de m overvinne for underbygge sine helteposisjoner.
Litteraturen er i frste rekke underholdende. Den rike manuskriptproduksjonen antyder at
sagaene har vrt svrt populre i senmiddelalderen p Island.1 Med en s stor produksjon av
underholdende ridderlitteratur i denne perioden, kan man tenke seg ulike forml liggende til
grunn. Det underholdende, kanskje noe eventyraktige, ved sagaene, kan ha utgjort en
overordnet funksjon. Hver enkelt saga/tekstverk br derimot forsts som selvstendige enheter
som realiseres gjennom tekstvitnene i bevarte manuskripter.
Mtet med eksotiske dyr og fantastiske skapninger (heretter: eksotiske vesener) utgjr i
flere av tekstverkene en stor del av de utfordringer aktrene gr gjennom. Det kan vre mter
med lver, draker, kameler og elefanter, eller det kan vre mter med skapninger som har
menneskehoder og dyrekropper, store hoder eller ett ye. Noen skapninger har egenskaper i
skifte skikkelse, for eksempel berserker som kan forvandle seg til lve eller drake. De var alle
skapninger som tilhrte en fjern verden. Litterre beskrivelser av en rekke skapninger som
lter fantasiaktig i vre rer var nedskrevet i autoritre verk/encyklopediske verk i
middelalderen, og var antakeligvis en del av en virkelighetsoppfatning. Kjennskapen til og
derav benyttelsen av de eksotiske vesenene i ridderlitteraturen varierer. Variasjonene som
kommer til syne i det islandske manuskriptet fra sent 1400-tall, Holm perg 7 fol, er i fokus for
denne studien.2
De eksotiske vesenene som er fremstilt i Holm perg 7 fol gir flere fortolkningsmuligheter.
Funksjonene de har hatt, kan forsts bde ut fra et internt og et eksternt perspektiv. I et internt
perspektiv fungerer vesenene som elementer i sagaens struktur og handling. Det eksterne
perspektivet representerer en bevissthet som ligger til grunn for produksjonen. For f
innblikk i det eksterne perspektivet, m de interne funksjonene frst gjre redes for. Frst og
fremst har de funksjon som handlingsmarkrer internt i sagaens forlp. Der ett eller flere
eksotiske vesener opptrer der skjer handling. Det ser ikke ut til at noen eksotiske vesener er

1
Manuskriptproduksjonen p Island i det 14. og 15. rhundret var svrt rik. Fra denne perioden markerer det seg
srlig stor nedskrivning av ridderlitteratur, dvs. oversatte riddarasgur og selvstendige islandske riddarasgur
(fornsgur surlanda). Flere av de oversatte riddarasgur ser ut til ha blitt oversatt p 1200-tallet. I denne
perioden ble ogs en del islandske sagaer, hvilke i stor grad skiller seg fra kongesagaer og islendingesagaer,
produsert: fornaldarsgur norrlanda. Disse tre gruppene av sagaer oppviser en rekke likhetstrekk som gjr at
man kan betrakte de forskjellige sagaene i et slags kontinuum. Det strre forskningsprosjektet Translation,
Transmission and Transformation. Old Norse Romantic Fiction and Scandinavian Vernacular Literacy 1200
1500, ved Universitetet i Oslo, diskuterer blant annet disse sjangerdefinisjonene. Som deltaker i dette prosjektet
arbeider jeg i min avhandling frst og fremst med de sene islandske riddersagaene. I min avhandling
argumenterer jeg for at det, til tross for likhet i form i ridderlitteraturen, finnes store interne forskjeller mellom
ulike tekstverks funksjoner. Jeg mener derfor at hvert tekstverk br forsts som en selvstendig enhet, realisert
med egne funksjoner som tekstvitne i manuskripter og derav ogs som del av et strre manuskripts helhetlige
funksjon.
2
Holm perg 7 fol er et islandsk manuskript fra siste halvdel av 1400-tallet. (Sanders 2000) Manuskriptet er
skrevet av fire ulike hender, og inneholder flgende tekstvitner: Rmundar saga keisarasonar, Elss saga ok
Rsamundar, Sigurar saga turnara, Bevers saga, Konrs saga keisarasonar, Hektors saga, Gibbons saga,
Viktors saga ok Blvus, Sigurar saga fts, Partalopa saga, Adoniass saga i denne rekkeflgen.

506
satt inn i sagaene tilfeldig, eller uten et visst forml. De eksotiske vesenene har ofte, i tillegg
til vre handlingsmarkrer, ogs funksjoner som sosiale markrer innad i sagaene. Slik
representerer de et mte med en verden langt borte, det ville og usiviliserte, og skaper en
kontrast til helten og heltens omgivelser.3
Internt i sagaenes strukturelle oppbygging kan de eksotiske vesenene ogs fungere som
leseanvisninger for mottakeren. Ved at de eksotiske vesenene fremstilles som lrde og
moralske referanser, skaper de fortolkningsrammer for leseren.
Bakenfor de funksjoner fremstillingen av eksotiske vesener har internt i sagaene finnes en
ekstern bevissthet. De eksotiske vesenene blir ogs markrer for et eksternt senderperspektiv.
Handlingsmarkrene markerer en bevissthet om strukturell oppbygging av en historie, sosiale
markrer kan forsts som en skildring av en geografisk forestilling. Som lrde referanser
fremstilles de eksotiske vesenene med beskrivelser som kan knyttes til en encyklopedisk
tradisjon, en tradisjon som bygger p antikke forestillinger bde hos grekere og romere,4 og
frst og fremst er gjort kjent p Island ved hjelp av Isidor av Sevilla (500600-tall) og hans
Etymologia.5 De skapningene som for oss virker eksotiske og virkelighetsfjerne, forklarte
Isidor under samme kategorier som mennesker og dyr, og de gir inntrykk av ha vrt en del
av en virkelighetsforstelse. Referanser til Isidor finnes i manuskriptet AM 194 8vo, fra 1387.
I dette manuskriptet finnes et avsnitt om rise-raser, som er besynderlig likt Isidors kapittel om
underlige skapninger, portenta (Klund 1908:3436).
I tillegg til at eksotiske vesener har vrt tilknyttet en encyklopedisk tradisjon, har ogs en
rekke eksotiske vesener tatt del i en rik tradisjon av verker som beskriver vesenenes
allegoriske fortolkninger. Bestiarier representerer et hydepunkt i fremstillinger av dyrs
allegoriske funksjoner. Her ble samlet illustrasjoner og beskrivelser av dyr ut fra deres
moralske kvaliteter, samt hvordan de kan forsts allegorisk, i religise og moralske
forestillinger. Bestiariene kunne blant annet vre nyttig for moralsk og religis veiledning
(Hassig 2000:xi, Hermannsson 1938:5). Bde eksotiske dyr og fantastiske skapninger er
fremstilt i den islandske Physiologus, i manuskripter fra omkring 1200.6 Visse eksotiske
vesener som blir fremstilt i sagaene kan knyttes til allegoriske fortolkningsmuligheter,
liknende for eksempel de som er fremstilt i Physiologus, og fungerer sledes som moralske
markrer.
I Holm perg 7 fol er det et spenn fra, p den ene siden, handlingsmarkerende skapninger
til, p den andre siden, skapninger med tydelig tilknytning til encyklopediske verker og med
moralske/allegoriske fortolkningsmuligheter. I det flgende vil jeg underske hvordan de
ulike fremstillingene kan pvirke lesningen av manuskriptets tekstvitner og forstelsen av
tekstvitnenes individuelle funksjon. Jeg vil derfor gi eksempler p hvordan fremstillingen av
eksotiske vesener fungerer som ulike markrer, frst internt, og deretter vil jeg lfte
fortolkningen til et eksternt perspektiv.

Handlingsmarkrer
3
Flere har arbeidet med problemstillinger om eksotiske vesener i norrn litteratur. Katja Schulz har skrevet om
kjemper i eddadikt og sagalitteretur i: Riesen: Von Wissenschtern und Wildnisbewohnern in Edda und Saga
(2004), Annette Lassen har gitt bokomtale av denne i Collegium Medievale 2005. Masteravhandlinger fra
Universitetet i Oslo av Nordian Nifl Heim og Silvia Rasheva i 2005 tar for seg henholdsvis varulver og alver i
norrn litteratur.
4
John Block Friedman skriver i The Monstrous Races in Medieval Art and Thought om middelalderens
frekvente fremstilling av eksotiske vesener som tydelig bygger p antikke forestillinger om hvilke vesener som
finnes i sten. Forestillinger om disse ble introdusert i vesten av blant annet Ktesias (gresk 400300-tall f.Kr.),
Megasthenes (gresk 300-tallet f.Kr.) og Plinius den eldre (romer dd i 79 A.D.) (Friedman 2000: 525).
5
[Isidore]
6
Det er to fragmentariske manuskripter fra omkring 1200 som utgjr den islandske Physiologus (Hermannsson
1938: 7).

507
Alle eksotiske vesener heltene mter fungerer som handlingsmarkrer. Ofte markerer
vesenene handling mellom helt og eksotisk vesen. De kan ogs vre handlingsmarkrer for
drive handling videre. Eksempel p dette finnes i Gibbons saga der Gibbon ser en lve og en
hjasse (et dyr med lange rer) i lek i en lysning.

Sia eir fram vr skoginvm hvar hleypr ein leona suo fogr ok frid at alldri litv eir vnna dyr
Litlo sidar sa eir fram koma einn hiasa med storum eyrum ok klifr vpp a leonit suo sem med
nockurum leik en eftir lidna stvnd huerfa av bdi aftr j morkina af eirra avgsyn.7

De s at en lve, s fager og gild, at de aldri hadde sett vakrere dyr, kom lpende ut av skogen.
Litt senere s de enn hjasse med store rer komme frem, og klatrer opp p lven, som en slags
lek. Og etter en liten stund har gtt, forsvant de begge tilbake inn i skogen, ut av deres yesyn.

Gibbon flger etter dyrene, inn i skogen. Dette frer ham imidlertid til en annen lysning, der
annen handling foregr og lven og hjassen er borte.
I dette eksempelet anser jeg dyrene for bare fungere som handlingsmarkrer.

Sosiale markrer
Andre eksotiske vesener fungerer i tillegg som interne sosiale markrer. I frste rekke blir
vesenene fremstilt i motiver der sagaenes hovedpersoner mter en ukjent verden. Som sosiale
markrer representerer vesenene det ukjente fra en eksotisk verden. Eksotiske vesener med
truende atferd, srlig grunnet strrelse, fremtredende utseende og overnaturlige egenskaper
finnes i beskrivelser av berserker, jotner, riser, troll og blmenn. Slike vesener blir ofte
beskrevet om hverandre. Disse vesenene fungerer tydelig som interne sosiale markrer.
Beskrivelser av dem omhandler ofte strrelse, farge og stygghet og som sosiale markrer
skaper de dermed distinksjoner mellom en kjent verden og en eksotisk verden langt borte, i
tillegg til en distinksjon mellom det kultiverte og det ikke-kultiverte. Draker er ogs et vesen
som stadig opptrer i utfordrende motiver for helten, og som skaper en kontrast til heltens egne
dyr, for eksempel hesten.
Det finnes ogs eksotiske vesener som ikke er truende, men som bidrar til heltens vekst og
fremgang. Dette gjelder i frste hnd dverger, og i enkelte sagaer alver. Dvergene opptrer ofte
som heltenes hjelpere forut for de utfordringer de skal mte. De bidrar med gjenstander som
innehar eksepsjonelle gjerne overnaturlige egenskaper, eller rd. Slike bidrag fr helten
enten mot betaling, eller som takk for en tjeneste helten har gjort dem.
Bde dverger, draker eller berserker med ulike ytre egenskaper fungerer som interne
sosiale markrer da de markerer en distinksjon mellom helten og den verden helten mter, og
opprettholder slik et spenningsforhold mellom de ulike aktrene. Skildringen av de eksotiske
vesenenes annerledeshet er i frste rekke med p definere heltens tilhrighet, og deres
kvaliteter fungerer derfor dels som kontrast til heltens egenskaper, dels som de hinder han m
overvinne og dels som tilskudd for styrke heltens uovervinnelighet.
I Viktors saga ok Blvus blir flgende beskrivelse gitt av berserker som Viktor og Blvus
utfordrer til duell:

[] eir eru blaer berserker ok suo miklar hamhleypur at eir bregdazt j ymissa kuikinnda liki.
eru ymiszt j iordu edur /aa/. spyia eir eitri j bardogum ok eingi jarn bita a. er at ok eingra
menskra manna at eiga vid a j orrustum.8

7
[Gibbons saga]: 80.
8
[Viktors saga ok Blvus]: 2731.

508
 [] de er svarte berserkere og s kraftfulle hamskiftere, at de kan ikle seg ulike skapningers
utseende, enten p jorden eller over. De spyr eter i kamper og ingen jern biter p dem. Ingen
menneskelig mann kan slss mot dem i kamp.

Dette er et eksempel p hvordan vesenene fungerer som interne sosiale markrer. Berserkene
uttrykker en klar distinksjon til heltene ved deres eksotiske egenskaper som gir
uovervinnelige krefter. Mtet med berserkene virker styrkende p heltene, da de overvinner
dem til tross for deres fryktinngytende fremferd.

Lrde referanser
De eksotiske vesenene kan ha tilhrighet i et strre europeisk encyklopedisk perspektiv og
utgjr slik lrde referanser som skaper nye fortolkningsrammer. De lrde referansene
markeres p flere mter. De realiseres gjennom beskrivelser av ytre egenskaper som
sammenfaller med for eksempel Isidors beskrivelser, slik som i Ectors saga:

[] hann bijta einngi iarnn nema au se miog uaunduth. e(dur) duerga smijde. uij hann deyfir
eggiar ij hueriu uopni. er hans yfirlith lijkari trllzligum glyrnnum enn manligum asionum. hann
hefir hrossa ftur ath hniam nedan galldra madur og seidskratti.9

[] intet jern biter p ham, foruten de som er kunstferdig smidde eller dvergesmidde, fordi han
kan dyve eggingen i hvert vpen. Hans skikkelse er likere trollslig utseende enn menneskelig.
Han har hesteben fra knrne og ned, (han er) trollmann og seidmann.

Dette motsvarer en beskrivelse av Ippopedes i AM 194 8vo, en rise-rase fra Sithia som har
hesteben (Klund 1908:35).
En annen type av referanser til encyklopedisk litteratur uttrykkes ved at distinksjoner
mellom beslektede dyr fremkommer i samme tekstvitne, som forskjellen mellom kamel og
hyrela i Ectors saga. Hyrela er et forelpig ukjent dyr for meg, men man kan ane dets likhet
til kamelen ut fra forholdene den blir beskrevet i. Flere av Ectors riddere mter truende
eksotiske vesener ridende p kameler.10 Ector selv mter en berserk ridende p et annet stort
dyr. Denne berserken er nemlig s stor at ingen hest kan bre ham, knapt en kamel
berserken har derfor temmet seg en hyrela.11 I Ectors saga ser det ut til at kamelen er utvalgt
som de eksotiske vesenenes fremkomstmiddel, blant annet med hensikt si noe om
berserkenes strrelse. Kamelen er sterkere enn en hest. Den temmede hyrelaen fungerer som
en forsterkning av utfordringene som sagaens hovedperson mter. Berserken som rir dette
dyret er mye strre enn alle berserker som er presentert tidligere i sagaen. Ector selv fr ren
av overvinne denne.12
Det finnes ogs beskrivelser av eksotiske vesener som nrmest kan betraktes som
etterlikninger av lrde referanser. Det forekommer fremstillinger av vesener med detaljerte
beskrivelser av hyde eller ytre egenskaper som kan minne om for eksempel Isidors
beskrivelser, men som tilsynelatende er egenkomponerte (eller av annen ukjent
inspirasjonskilde).

9
[Ectors saga]: 99.
10
For eksempel [Ectors saga]: 94.
11
[Ectors saga]: 153.
12
Det er en lakune i begynnelsen av Ectors saga i Holm perg 7 fol. I andre tekstvitner blir Ector gitt en dromedar
da han blir sltt til ridder. Dromedaren er nrt beslektet med kamelen, men det blir beskrevet blant annet hos
Isidor at dromedaren er raskere enn kamelen. Kan hende blir Ector gitt nettopp en dromedar for gi ham kt
hastighet i forhold til de eksotiske vesenene som senere dukker opp i sagaen. I disse tekstvitnene av Ectors saga
er distinksjonen av beslektede dyr mellom dromedar, kamel og hyrela. [Ectors saga]: 86.

509
 [] og er hann er buinn ser hann hvar fram kemur madur ef mann skal kalla og rijdr einum
vlfallda. s/aa/ uar bdi har og digur. hofudit uijlict sem hraunklettur. augun huijt sem
hiegeitlar. arf ar ecki sogu ur ath gera. ath einga ottiz hann leidiligri skepnu hafa seth. hann
hafdi kylfu ser ij hendi og fleina ij. gengr nu /aa/ mot honum og heilsar honum.13

[] og da han var utstyrt ser han der hvor det kommer frem en mann, om mann kan kalles, som
rir en kamel. Han var bde hy og tykk. Hodet likner en bergknall, ynene hvite som flint. Det
trengs ikke lages en saga om det at ingen syntes ha sett flere skapning. Han hadde en
klubbe i hnden og to spyd. Han gr n mot ham og hilser ham.

Moralske referanser
I middelalderens religise bevissthet spiller dyrene en vesentlig rolle som part av
skaperverket, samtidig fungerer de i en symbolsk verden en allegorisk fortolkningssfre.
Motiver der dyr og skapninger fungerer som moralske referanser, bygger ndvendigvis til en
viss grad ogs p kjennskap til en strre litterr tradisjon. I denne sammenheng er som
tidligere nevnt Physiologus og den bestiariske tradisjonen i sentrum, hvor dyr og skapninger
gis en allegorisk tilhrighet som gjenspeiles i den kristne tro. Det finnes likevel grader av
moralske referanser. Noen er direkte, det vil si at de fremstilte eksotiske vesenene direkte
henspeiler til allegoriske fortolkninger, i for eksempel Physiologus. Andre dyr og skapninger
kan fremst som indirekte moralske referanser. Disse er tydelige markrer av for eksempel
ondskap og bygger p en kristen fortolkningssfre, men kan ikke direkte knyttes til definerte
allegoriske fortolkningsrammer.
Eksempler p sistnevnte, indirekte moralske referanser, finnes for eksempel i Gibbons
saga:

en a tialldino hia honum sa eir standa einn varg mikinn ok illilegann sem fiandan kolsuartan at
lit.14

og p teppet ved siden av ham, s de at en stor ulv stod. Forferdelig som djevelen og kullsvart
se p.

Det er f direkte moralske referanser i Holm perg 7 fol, men flgende eksempel fra Konrs
saga keisarasonar viser hvordan et motiv delvis kan gjre bruk av allegori for understreke
handlingen.

Ha<n> gat at lta huar drek ogurligar ok suo mikll at hann ottsz eck kuknd hafa sied
jafnmikt essh/aa/tar. hann hafdi mikla ung. Konradr s/aa/ at hann hafdi vndir sier dyr d
oarga at stod vp. hann ottizt at hellzt af kenna at dreknn mundi flogit hafa med dyrit uiat
hann hafdi festar hremsur j bgum dyrsns. enn hann hafdi halanum vafit wm mitt dyrit ok uilld
fliuga medur at edur draga upp j fiallt. []15

Han kom til se hvor en fryktelig drake og s stor at han syntes ikke ha sett en like stor
skapning av det slaget. Den hadde store vinger. Konradr s at han hadde det uredde dyret
(lven) som stod opp under seg. Han syntes det s ut til at draken hadde flyet med dyret, fordi
han hadde satt klrne sine i dyrets bog, og han hadde viklet halen sin rundt dyrets midje og ville
fly med det, eller dra det opp i fjellet.

13
[Ectors saga]: 111.
14
[Gibbons saga]: 87.
15
[Konrs saga keisarasonar]: 6667.

510
I den islandske Physiologus str det skrevet i begynnelsen av beskrivelsen av panteren (lven)
at draken er dette dyrets uvenn. Senere str det at Jesus Kristus er den sanne panter
(Hermannsson: 19). Det er ikke ndvendig forst dyrene som allegorier for Kristus og Satan
i dette motivet, men dyrenes allegoriske fortolkningsmuligheter fungerer som en
leseanvisning for understreke at lven er et godt dyr i denne sammenheng. Konrar redder
lven ut av drakens klr, og lven blir deretter Konrars flgesvenn. Spenningen mellom de
to allegoriske dyrene lve og drake sammenfaller med spenningen i Physiologus.

Eksternt perspektiv
Tilsynelatende liknende vesener kan ha ulike funksjoner i de forskjellige tekstvitnene, for
eksempel lven i Gibbons saga og lven i Konras saga keisarasonar. Sorteringen mellom
handlingsmarkrer, sosiale markrer, lrde og moralske referanser kan lftes til et eksternt
perspektiv. Der lven og hjassen i Gibbons saga fungerer som handlingsmarkrer, fungerer
de ogs i et eksternt perspektiv, hvor de viser en bevissthet i oppbyggingen av en historie.
Strukturen veksler mellom handling og pause, de eksotiske vesenene blir et slags
pauseinnslag som forflytter handlingen videre.
Berserkene i Viktors saga ok Blvus har spesielle ytre egenskaper. I tillegg til formidle
den sosiale avstanden mellom helten og motstanderen, gis berserkene egenskaper som til dels
kan bunne i en faktisk virkelighetforstelse. Vesener som kan skifte skikkelser kunne finnes i
en ukjent verden. Berserkene i Viktors saga ok Blvus gir uttrykk for en bevissthet om en
eksotisk verden, uten henspeile direkte til noen lrde tradisjoner.
De lrde referansene som er eksempelgitt i Ectors saga kan vre med p uttrykke
individuelle kunnskapsforhold og kunnskap om encyklopedi i et senderperspektiv. En del av
vesenene (berserkene) fremstilt i Ectors saga fungerer som markrer p et visst
forholdningssett til en strre encyklopedisk tradisjon som har vrt kjent p Island. Jeg har her
forholdt meg til Isidor av Sevilla og den nrliggende gjengivelsen i AM 194 8vo.
De moralske referansene vil, som flge av den tid litteraturen ble nedskrevet i, vre farget
av en kristen fortolkningsramme. Det finnes mange indirekte moralske referanser i sagaene.
De er med p underbygge den kristne forstelseshorisonten som preget samfunnet og tiden
det ble skrevet i. De direkte moralske referansene viser ytterligere bruk av religise, moralske
kilder for gi allegoriske fortolkningsmuligheter, slik som spenningsforholdet mellom lven
og draken i Konrs saga keisarasonar.
Ut fra dette kan man f et inntrykk av at de ulike tekstvitnene i Holm perg 7 fol representer
ulike typer sagaer, og derav ulike funksjoner. Noen oppviser en bevisst benyttelse av lrde og
moralske referanser i en ellers underholdende form, mens andre er rent underholdende. De
lrde referansene for eksempel i Ectors saga kan forsts som eksterne sosiale markrer som
kan plassere Ectors saga i en fortolkningsramme for en lrd type sagaer.

Oppsummering
I Ectors saga kommer det tydelig frem at de eksotiske vesenene de syv ridderne mter p sine
ferder tar del i det felles utgangspunktet som blir lagt da Ector og ridderne hans bestemmer
seg for dra ut verden hver for seg, for styrke sine ridderegenskaper. De vil vre
uovervinnelige. Dette representerer noe grunnleggende felles for motivene vesenene opptrer i
i alle tekstvitnene i Holm perg 7 fol. De fremstilles i motiver der heltene bekjemper
utfordringer, utfordringer som ofte fremstiller den gode i bekjempelsen av det onde. Som
interne sosiale markrer har eksotiske vesener ofte en funksjon i motivet de fremtrer i at de
gir personlig vekst til hovedpersonen ved bekjempe det truende eksotiske og det ikke-
kultiverte.

511
 Manuskriptet som helhet fremstiller en rekke eksotiske vesener. Alle disse eksotiske
vesenene kan, som jeg har gitt noen eksempler p ovenfor, forsts som markrer til ulike
perspektiver og fortolkningsniver. Vesenenes funksjoner skiller seg merkbart mellom de
ulike tekstvitnene. De fungerer som strukturelle markrer internt i sagaens forlp. De
eksotiske vesenene som opptrer internt i sagaene gir samtidig referanser til de eksterne sosiale
forholdene rundt tekstvitnenes komposisjon i et senderperspektiv, for eksempel om
avsenderen er, eller vil fremst som lrd. Bde markrer for lrde og moralske referanser kan
knyttes til lrde, encyklopediske verk og vil vre eksterne sosiale markrer for en lrd type
sagaer som forholder seg til en strre encyklopedisk og allegorisk fortolkningsramme. Der de
eksotiske vesenene utelukkende fungerer som handlingsmarkrer og interne sosiale markrer
skaper det antydninger til at tekstvitnet er en rent underholdende saga.
Her har jeg gitt utvalgte eksempler av en tendens som synes vre tilstede i hele
manuskriptet. Lesningen av de eksotiske vesenenes ulike funksjoner kan knyttes til de enkelte
tekstvitners funksjon som type saga. I min phd-avhandling vil jeg utfre flere element- og
motivanalyser i manuskriptets tekstvitner for danne en mer kompleks argumentasjon for en
slik typeinndeling.

Bibliografi
[Ectors saga]: Late Medieval Icelandic Romances, ed. Agnethe Loth, 1962. Editiones Arnamagnan,
Series B, vol.20. Kbenhavn.
Friedman, John Block, 2000: The Monstrous Races in Medieval Art and Thought. Syracuse University
Press. New York.
[Gibbons saga]: ed. R.I. Page, 1960. Editiones Arnamagnan, Series B, vol.2. Kbenhavn.
Hassig, Debra, 2000: Introduction. In: The Mark of the Beast. Ed. D. Hassig. Routledge, New York
and London.
Hermannsson, Halldr, 1938: The Icelandic Physiologus. In: Islandica vol. 27. Ithaca, New York.
[Isidore]: Isidore of Sevilles Etymologies. Vol 1&2. Trans.: Priscilla Throop. 2005. MedievalMS,
Vermont.
Klund, Kristian 1908: Alfri slenzk. Islandsk encykklopdisk litteratur I. Cod. MBR. AM. 194,
8vo. S. L. Mllers Bogtrykkeri, Kbenhavn.
[Konrs saga keisarasonar]: ed. Otto J. Zitzelsberger. 1987. American University Studies/01. Vol. 63.
Lassen, Annette 2005: Riesen: Von Wissenschtern und Wildnisbewohnern in Edda und Saga.
Bokanmeldelse. I Collegium Medievale vol.18. Oslo.
Nifl Heim, Nordian, 2005: The Norse Werewolf. Masteroppgave ved Universitetet i Oslo.
Rasheva, Silvia Bisserova, 2005: Elves in medieval Scandinavian imagination. Masteroppgave ved
Universitetet i Oslo.
Sanders, Christopher, 2000: Tales of Knights : perg. Fol. nr 7 in the Royal Library, Stockholm. I
Manuscripta Nordica. Ed. Peter Springborg. C.A. Reitzel, Kbenhavn.
Schulz, Katja, 2004: Riesen: Von Wissenschtern und Wildnisbewohnern in Edda und Saga. Universi-
ttsverlag Winter, Heidelberg.
[Viktors saga ok Blvus]: Late Medieval Icelandic Romances, ed. Agnethe Loth, Editiones Arna-
magnan, Series B, vol.20. Kbenhavn, 1962.

512
 When small words make a big difference
On adaptation and transmission of texts in Late Medieval
manuscripts
Elise Kleivane, Department of Linguistics and Scandinavian Studies, University of Oslo,
Norway
The Eirks saga vfrla (Esv) is preserved in five medieval manuscripts and several Post-
Medieval manuscripts.1 My concern here will be the medieval manuscripts of which two pre-
serve the entire saga without lacunae, whereas the three others are fragmentary in varying
degrees. The saga text, as far as it is preserved in each of these five manuscripts, show much
variation when compared to each other. In fact, where all five can be compared, only se-
quences of up to 13 words are identical not considering orthographic and palaeographic
variation. However in spite of the differences between the individual text witnesses, there is
general consensus that they are all Esv.
Eirks saga vfrla is a short saga in Old Norse probably composed in Iceland sometime
around 1300 (Jensen 1983: XIV). It is about Eirkr, the son of rndr, king of rndheimr.
One Christmas Eve Eirkr strengi heit to travel the world in search of the place called
dinsakr. His first stop is Denmark, where he is joined by his namesake Eirkr, son of the
Danish king. They travel south with their men to Miklagar, where they help the Gara-
konungr against some of his enemies, and are well received. One day Eirkr starts a conversa-
tion with the Garakonung. This conversation is structured as questions from Eirkr and an-
swers from the king, and they focus on God, the Trinity, the world and universe, and finally
the way to dinsakr. Much of this conversation is based on the Elucidarius, either on a Latin
version or a translation to Old Norse, and the first book of the Imago mundi.2 After the stay in
Miklagar, the two Eirkar travel east following the kings instructions. They travel through
India until they reach a river they recognise as the Phson. On the only bridge across the river
lies a terrible dragon. On the other side of the river they see a beautiful land. Together with
one man, the Norwegian Eirkr enters the dragons mouth, sword drawn, and the Danish
Eirkr and the rest of the men turn home believing Eirkr is dead. But Eirkr and his compan-
ion emerge on the other side of the river. The two explore the land, and the description of
what they see is similar to descriptions of holy places in other stories. They discover a tower
hanging unsupported in the air, and in it they find food, drink and soft beds, and they realise
that they have reached their goal. After falling asleep, Eirkr has a conversation with an angel
in which the angel answers Eirkrs questions e.g. on angels and he explains about the place
they have found. This conversation also draws on Elucidarius. Eirkr is given permission to
return home, since his telling about what he has experienced will, according to the angel, ease
the Northerners forthcoming conversion. The angel also says that he will come and get Eirkr
after a certain time. On their way back, they stop in Miklagar, telling the Garakonungr eve-

1
Jensen (1985: 501) counts ca. 60 manuscripts.
2
Elucidarius is a theological work written in Latin (probably) by Honorius Augustodunensis (1075/801156) in
the beginning of the 12th century. It was translated to several vernacular languages, including Old Norse. The
oldest preserved Old Norse manuscript containing Elucidarius-text is AM 674 a 4to dated to ca. 11501200.
(This and all further dating of Old Norse manuscripts are from Ordbog over det norrne prosasprog. Registre.)
Imago mundi is a scholarly work also attributed to Honorius and written in Latin. There is no known Old Norse
translation of this work, but it has been used as source in more than Esv. Some passages in Esv are very close to
Elucidarius, so close that Elucidarius must have been a direct source either written (Jensen 1983: xx) or through
memoria (cf. Carruthers 2005). Simek 1984 argues the source was a Latin Elucidarius-text. The use of Imago
mundi in Esv is not as direct as with Elucidarius (see e.g. Springborg 1985: 204ff).

513
rything, and then continue to rndheimr. And after the period designated by the angel, Eirkr
disappears during morning prayers.
In this paper I will address two questions. One is why different text witnesses of the same
saga display so much variation when compared, and the other is how these variants function
within each text witness. I will address the latter question first, by presenting variants from
the text witnesses of Esv.
When describing and discussing the material, I use the threefold distinction, due to Wendt
(2006), of the notion of text: 1) A manuscript or text carrier (textbrare) is a specific in-
stance of a text as a document, i.e. as a material object (such as a copy of a newspaper), in-
cluding the material object itself. 2) A text witness is the immaterial text that is manifested in
individual text carriers. One text witness can be manifested in more than one text carrier, but
this is extremely rare in manuscript culture. My use of the term text witness is narrower than it
is often used in philological writings, since I reserve this term for the immaterial text, and
define the manifestation of the text witness as (part of) the material text carrier. The five text
witnesses of Esv that have been manifested in medieval manuscripts are all different text wit-
nesses. 3) The text work is an aspect based on the reception of text witnesses. When there is
agreement that text witness X, Y and Z all are the same, despite variation, they are text wit-
nesses of the same text work. This aspect is not to be confused with original or archetype
etc. I am referring to the text work when using the title Esv. When referring to manuscripts I
use the catalogue signatures, and when referring to text witnesses I use a combination of letter
and number, based on Jensen (1983). (See Table 1 and 2.)
Elsewhere I have demonstrated how the material context the different text witnesses are
manifested in can give a text work like Esv very different rooms for interpretation.3 Here I
will focus on the variation on the level of text witnesses.

Variants
Before going any further, it is necessary to define what I mean by variants and changes. As
undergraduates we learn that the medieval scribes copying Old Norse texts made changes to
the text as they were copying, and that some changes are simply mistakes and there are sev-
eral categories of mistakes defined by the (probable) reason for the scribes making the mis-
takes or they are changes made on purpose. And changes made on purpose were acceptable
since copyright or authors right did not exist in Medieval Europe. But unless we deal with
textual criticism and wish to distinguish between significant and insignificant variation, the
variants rarely get more attention than filling the critical apparatus in a decent critical edition.
We know they are there, but seldom make use of them more than for confronting isolated
variant readings.
Variants as I define the term for the purpose of this paper, are first of all restricted to varia-
tion within the semantic field, so palaeographic and orthographic variation, including dialectal
variation, is not considered. There are two main types of variation: a) changes that are the
result of mistakes and that are recognisable as mistakes, such as misspellings, and b) variation
that is not recognisable unless comparing different text witnesses.
The first type, i.e. the mistakes, are involuntarily made, and render a more or less meaning-
less piece of text (by which I mean a piece of a physically manifested text witness) that any
reader will recognise as a mistake and probably try to correct when reproducing the text
witness either orally when reading or in writing in the process of copying.
The second type, which this paper will focus on, can be result of changes made either vol-
untarily or involuntarily. This second type of variation will of course include results of what

3
Kleivane 2009, and forthcoming PhD-thesis.

514
on an earlier stage of the transmission might have occurred as the result of a correction of a
mistake, but is no longer visible as either mistake or variation within the text witness. The
second type thus contains a great diversity of variation, from exchanging a single word with a
different word to adding or omitting longer passages.
Studying these variants we can learn more about how medieval scribes worked when copy-
ing texts, what they changed and what they did not change. This also opens for studying what
consequences the changes made have had on the individual text witnesses, that is to say how
the conditions governing the reception of the text witness is affected. Since few instances of
mother-daughter-manuscripts have been preserved (and none of these are among my mate-
rial), and since variants are not visible when reading the text witnesses individually, the
method for undertaking such examinations will have to be compare and trace. Comparing
in order to identify the variation and tracing in an attempt to identify tendencies within indi-
vidual text witnesses.

Variants in Eirks saga vfrla

Jensen (1983) has demonstrated how the text witnesses of Esv that are preserved in medieval
manuscripts must be said to belong to two different redactions, called A and B.4 Of a total of
five text witnesses, three belong to the A- and two to the B-redaction. All manuscripts are
dated to the 14th and 15th centuries.5 Only one text witness in each of the two redactions is
preserved without lacuna. Of the two in the A-redaction that are not preserved completely, the
preserved text does not overlap, and so it has not been possible to compare the same piece of
text in more than two text witnesses within one redaction. (See Table 1 and 2 for the amount
of text preserved in the respective manuscript.) Since orthographic or palaeographic variation
is not under consideration, I will normalize the examples given in this paper.

Table 1. Number of words in the text witnesses in the A-redaction. Distributed according to the lacu-
nae in a1 and b3. Not counting Jn rarsons epilogue in GKS 1005 fol.
Manuscript GKS 1005 fol (A3) AM 720 a VIII 4to (a1) AM 557 4to (b3)
371 354 -
204 - -
993 - 981
1102 - -
263 - 264
Total words 2933 354 1245

Table 2. Number of words in the text witnesses in the B-redaction. Distributed according to the lacuna
in B2.
Manuscript AM 657 c 4to (B1) GKS 2845 4to (B2)
341 326
1401 -
1139 1076
Total words 2881 1402

As most of the variants in this material are insignificant variants in text critical terms, it is
irrelevant which of two variants is the changed and which is original, since in most instances
both may have been the result of a change. I do not evaluate the variants according to any
4
Jensen (1983) delimits two more redactions when also considering text witnesses preserved in Post-Medieval
manuscripts.
5
The part of GKS 1005 fol which Esv belongs to is dated to ca. 13871394. Manuscript AM 720 a VIII 4to is
dated to ca. 14001450; AM 557 4to to ca. 14201450; AM 657 c 4to to ca. 13401390 and GKS 2845 4to, to
ca. 1450.

515
stemma, but only how they work within each text witness. The examples here are only a very
small selection of the variants found when comparing these five text witnesses, but will serve
to give an impression of what kind of variation there is, and how they affect each text witness.
Comparing B1 and B2, there is a tendency in B1 to formulate conditioned or argumenta-
tive clauses where the choice in corresponding clauses in B2 are descriptive clauses, like in
Example 1.

Example 1. Conditioned vs. descriptive

B1 (10r1314) B2 (59v1214)
Ok er hann var ess vss at eir vru normenn Ok hann var ess vss at eir vru normenn ok
ok tluu va at kanna heiminn, helt hann mikils virar ok tluu at fara va um heiminn.
vel i llum hlutum. Konungr helt vel llum hlutum.

In B1 the er [] strengthens the relation between the information the king obtains and the
result that he treated the Northerners well. The corresponding clauses in B2 simply states two
things: the information the king obtains and the fact that he treated them well. In Example 1
there is also evidence of other variation. First there is an instance of amplification in B2 (ok
mikils virar), compared to B1, which adds to the level of precision with a detail that is not in
B1. Second, two variant expressions are used: in B1 Eirkr and his crew aims at va at kanna
heiminn (widely get to know the world), where they in B2 will fara va um heiminn
(travel widely in the world). The expressions have more or less the same meaning, but the
emphasis in B1 is knowledge whereas in B2 it is more the adventure of travelling.
Another tendency in B1, as opposed to the other text witnesses, is to avoid using introduc-
tory phrases of the kind that refers to the following having been told. (See Example 2.) In Ex-
ample 2a) and 2c) B2 uses at er/var sagt, and A3 and a1 from the A-redaction have the same
or similar phrases in the corresponding place. The exception is 2c) in a1 where the phrase is
similar to B1. The sentence in Example 2b) is omitted in B2, but a comparison to A3 and a1
strengthens the claim that introductory phrases referring to something having been told, is
avoided in B1.

Example 2. Introductory phrases in B1, B2, A3 and a1 (b3 has lacuna here).
a) B1 (10r34) B2 (59v24) A3 (4va69) a1 (1r271v1)
Ok einn jlaaptan at var sagt at einn jlaap- ess er getit eitt jlakveld, ess er getit eitt jlak-
strengi hann heit at tan strengi hann heit at strengdi Eirkr ess heit veld, at Eirkr strengi
fara sur heim unz fara sur heim unz hann at fara um allan heim at ess heit at fara um allan
hann fyndi dinsakr fyndi dinsakr at leita ef hann fyndi sta heim <at> leita ef hann
heiinna manna lzku. heiinna manna lzku. ann er heinir menn kalla fyndi sta ann er
dinsakr en kristnir heinir menn klluu
menn Jr lifandi manna dinsakr en kristnir
ea Paradsum. menn kalla Jr lifandi
manna ea Parads.

b) B1 (10r17) B2 A3 (4va2930) a1 (1v2325)

Ok hfst fyrst --- Sv er sagt at ar af Sv er sagt at ar hefist
normanna smi gerist fyrst normanna ok fyrst normanna smi
Grikkjakonungs rki. smi t Miklagari. Miklagari.

c) B1 (10r1718) B2 (59v1718) A3 (4va3132) a1 (1v2526)

Einn dag spuri Eirkr at er sagt einn dag, at at var sagt at einn dag at var einn dag at Eirkr

6
This no is written at the end of a line, and is an example of a mistake a variant of the first type mentioned
above, in this case a dittography. The <at> in the first passage quoted from a1 is similarly an example of a mis-

516
 rnski konung: Hefir Eirkr rndarson mlti til spuri Eirkr norrni spuri konung hvrt no6
nkkurr grt himin ea konungs: Geri nkkurr konung hverr grt hefi nkkurr geri heiminn7
jr? himin ok jr? himin ea jr. ok jr.

Example 2 also illustrates variation in tempus. In A3 the verb kalla is used once, in 3.p.pl.,
present tense: er heinir menn kalla, and in the following clause A3 uses a paratactic con-
struction: en kristnir menn Jr lifandi manna. In a1 the verb is first given in 3.p.pl., past
tense: er heinir menn klluu, and in the following clause the verb is repeated, but now in
present tense: en kristnir menn kalla []. None of these three verbs are abbreviated. The so
called historical present which is used for describing the to modern minds sometimes il-
logical switching between present and past tense, can not be used for explaining this varia-
tion, because this is referring to what Eirkr is promising to do, and not describing an action.
Rather, the result is an underlining in a1, from the point of view of the scribe, that people used
to be heathen, but are now Christian. This is an example of the extent to which individual
scribes made independent decisions regarding the choice between present and past tense. To
illustrate this further, the following (Example 3) is an example of the variation of the use of
the historical present in A3 and a1.

Example 3. Switch between present and past tense.

A3 (4va1418) a1 (1v912)
Ok um vrit eptir rst Eirkr danski fer me Um vrit eptir rst Eirkr danski til ferar me
nafna snum. Vru eir ok tlf saman. Sigldu n ok nafna snum. Vru eir ok tlf saman. Sigla n
tla til Miklagars, ok ar koma eir ann tma [] ok tla til Miklagars, ok ar koma eir.
sem Grikkjakonungr bau t her moti vkingum er ann tma hafi Grikkja konungr boat her ti
gengu mjk rki hans. moti vkingum er gengu mjk rki hans.8

The variation in tense concerns the verb sigla, which in A3 is 3.p.pl. past tense sigldu,
whereas in a1 it is 3.p.pl. present tense sigla. The preceding verbs are past tense in both, rst
and vru, and the following verb is present in both, koma.9 The two text witnesses display two
different choices as to when the audience will be brought closer to the narrative by the use of
present tense: In A3 it happens when they arrive in Miklagar and in a1 when they sail from
Denmark.
The final example I will give here is one where seemingly small variations cause larger
and important differences between the text witnesses. In A3, b3 and B1, the Greek king
equals dinsakr, Parads and Jr lifandi manna (a1 and B2 have lacunae in this part of the
conversation). In Eirkrs dream in the tower (see Example 4), the angel can enlighten Eirkr
further. In A3 the angel states clearly that Jr lifandi manna is the place he has found, and
that Paradise is somewhere else close by. However, when the angel explains that they were to
show Eirkr Jr lifandi manna by some image, it seems what Eirkr has found is merely an
image. This is probably what causes Ashurst (2006: 78) to argue that Eirkr never really ar-

take, and corrected here. In this case it is an omission of a word necessary for making sense of the sentence, and
it is easy to correct a medieval reader might even not have noticed the word missing. I have chosen to add at
here, but ok could also have been used.
7
Heiminn and jr have more or less the same meaning since himinn ok jr is a very common expression it is
possible that heimenn should be regarded as a mistake and that readers have read himin for heiminn. But it
clearly reads heimenn in the manuscript.
8
The part of a1 not quoted in Example 3 is a description of the route they sail: vestr fyrir lnd t til Nrfasunda
ok vestr Grikklands haf (a1: 1v910). There is no corresponding passage in any of the other text witnesses.
9
What follows after koma eir, shows a variation that concerns more than choice of tempus. In A3 the is an
adverbial clause explaining at which time the Eirkar came to Miklagar. The corresponding passage in a1 is an
independent sentence, describing the situation in Miklagar.

517
rives at any paradisiacal land, that it is all a vision. Ashurst refers to the Fornaldarsgur
Norrlanda (188589), which prints the A3.10
In B1 and B2 it seems that Jr lifandi manna is another name for the earthly Paradise, and
that the place Eirkr have found is something else, a place used as an image or simile (mynd
ok lking) of Paradise. But it seems this place still is something more than a vision.

Example 4. What is the place that Eirkr has found?

B1 (12v25) B2 (60v1861r1)) A3 (5rb3542)
En sj star er r snist inn En sj star er r snist inn En s star er sr hr, er sem
fegrsti, er at vira vi Parads fegrsti, er sem eyimrk at vira eyimrk til at jafna vi
sem eydimrk, er bygu inir vi Parads, er bygu inir fyr- Paradsum.
fyrstu menn, ok n byggja andir stu menn ok n byggja andir yfir- En skamt hean er s star, ok
yfirfera ok spmanna. fera ok spmanna. fellr aan s er stt. angat
En s star er skamt hean, ok En s star er skamt hean, ok skulu engir lfs koma, ok skulu ar
aan fellr s er r frut yfir. aan fellr sj. byggja andir rttltra manna.
En r r komut hingat, En r kemr angat bau En s star er hefir hittan heitir
bau Gu oss at vkva ok Gu oss at vkva ok blomga Jr lifandi manna.
blomga enna sta til at sna enna sta til at sna r Jr En r komt hingat bau Gu
r Jr lifandi manna me lifandi manna me nkkurri mynd oss at vakta enna sta ok sna
nkkurri mynd ok lking, at gera ok lking, ok gera r veizlu ok r Jr lifandi manna me
r veizlu ok launa r sv launa r sv erfii itt. nkkurri mynd, og gera r veizlu
erfii itt. ok launa r erfii itt.

However, the most interesting variation in this passage is in my opinion between B1 and B2,
in the beginning of the last paragraph starting with En r. The variation I am referring to is
the difference in tempus (B1 past and B2 present tense) and the choice of adverb. B1 says
But before you (pl.) came here, God instructed us [] whereas B2 says But before you
(sg.) will come there, God instructed us []. Jensen discusses the present tense used in B2
(1983: LIII) and concludes that it is meaningless. But she does not mention that B2 has an-
gat following, which none of the other text witnesses have. The question is what angat
(there) refers to, and in my opinion it is clearly the previously mentioned s star which is
not far away that the river flows from, and that it Paradise already introduced in the begin-
ning of the passage quoted in Example 4. This is the earthly Paradise which Adam and Eve
lived in, and where the patriarchs and prophets of the Old Testament now live. So, in B2, the
angel says that before Eirkr gets there, God wanted to show him an image of that place as a
reward for his effort. This could be a hint that Eirkr is seen as an Old Testament prophet
for the Nordic countries, which has not yet received the Gospel, and may suggest that he will
await Judgement Day in the earthly Paradise. However, later the angel says in both B1 and B2
that God has chosen this place for Eirkrs bones to be kept until Judgement Day, the refer-
ence being to the place he has already found. In the corresponding passage A3 (b3 has par-
tially a lacuna) does not say as clearly what place it is his bones and his soul shall be kept.11
Hence not just the text work Esv, but each individual text witness are in their own way
commenting on the huge and important medieval discussion of where the heavenly and the
earthly Paradise were situated, whether or not they were populated now, prior to Judgement
Day, and by whom.12 Especially B1 and B2 offer a specific solution for Eirkr after he is bod-
ily removed from this world by the angel. Such a removal instead of ordinary death also hap-
pened to the Old Testament prophets Enoch and Elijah (2. Kings 2:11 and Heb. 11:5). They

10
Fornaldar Sgur Nordrlanda (182930) prints the B1 with some variants from A3 and B2.
11
B1 12v1922; B2 61r1821; A3 5va68 and b3 44r13. a1 has lacuna.
12
On different medieval views concerning this, see e.g. Ashurst 2006.

518
are also mentioned in Elucidarius as having been translated (The Old Norse Elucidarius
1992: 8687).

Why so much variation?

The question of why the comparison of these five text witnesses reveals so many variations is
not a question particularly related to the Esv, but rather to the transmission of Late Medieval
vernacular texts in general. Disregarding obvious mistakes, what I called variants of the first
type, why did the medieval scribes make so many deliberate changes when copying the Old
Norse texts? So many in fact, that the process of copying often looks more like reproduction.
It is not a sufficient explanation to claim that this is simply a result of the manuscript culture.
That is certainly part of the explanation, but if it were sufficient, one would expect Latin texts
to have been subjected to the same kind and amount of changes during the copying process as
Old Norse texts. Latin texts are, certainly, also subject to such changes, but not to the same
extent. On the other hand one can not claim that this is exclusive for the transmission of Old
Norse texts, since other medieval vernacular texts display the same kind and amount of varia-
tion.13
I argue that the process of copying texts written in Latin was regulated by a different set of
norms than the copying of vernacular texts, and that this is the result of different views of the
fixity of written texts and of how one should treat written texts.14 Several elements jointly
contribute to the shaping of such sets of norms. For Latin texts, the quality of the language is
one element (Machan 1991). People living in the Medieval period were aware that Latin dis-
played much less variation regionally and over time than did vernacular languages. As a re-
sult, higher expectations of fixedness and stability were connected to Latin. Also, since Latin
was no longer anybodys first language (perhaps with a few individual exceptions), most
scribes had less skill for improvising and improving on Latin texts than they would have with
vernacular texts. Another element is that the auctores had chosen Latin for their writings,
thereby further adding to Latin as a language chosen for texts imbued with important and last-
ing knowledge.15 Finally, there is the question of genre. When composing a text, there is the
question of what form would be the most opportune and suitable the classical rhetorics de-
mand for aptum. This involves choices regarding style, content, and composition, in addition
to the text being adjusted to the situation (kairos) and the competence of the receivers (doxa).
Educated medieval people were trained in this, and would choose the most suitable language
for the text they were composing; texts in some genres would be more suitable in a vernacular
language than in Latin.
The qualities connected to the Latin versus vernacular languages in the Medieval period
seem to be the main reason why scribes could take greater liberties in the copying of vernacu-
lar texts. But why would they take such liberties? In asking this question, one is asking for the
intention behind the production of manuscripts and the selection of text witnesses to be mani-
fested in them. Except for the occasional meta-commentary, our sources to these intentions
are the manuscripts themselves, the material variation between them, and textual variation
between text witnesses of the same text works. The answers these sources present are that
changes are made for esthetical reasons; for making opinions clearer; to adjust to new con-

13
See e.g. for medieval German texts Hver 1995, for medieval Spanish texts Dagenais 1994, for medieval Eng-
lish texts Machan 1991 and Jacobs 1992, and for medieval Anglo Norman/French texts Busby 2002.
14
By norms I do not mean rules than one has to follow, but rather a competence in what is expected and of the
conventions relating to the activity in question. Clearly, the same person can be competent in and use more than
one set of norms.
15
On auctores and auctoritas see Minnis 1988, Machan 1991 and Carruthers 2005.

519
texts and uses; and, when a text witness appears confused or damaged, to try to make sense
out of it.

Literature
Ashurst, David, 2006: Imagining Paradise. In The Fantastic in Old Norse/Icelandic Literature. Sagas
and the British Isles. Preprint papers of The 13th International Saga Conference, Durham and York,
6th12th August 2006: 1, 7180. Eds. J. McKinnell et. al. Durham University.
Busby, Keith, 2002: Codex and context: reading old French verse narrative in manuscript, 2 vols. Ro-
dopi, Amsterdam.
Carruthers, Mary J., 2005: The Book of Memory. A Study of Memory in Medieval Culture. Cam-
bridge University Press.
Dagenais, John, 1994: The Ethics of Reading in Manuscript Culture. Glossing the Libro de buen amor.
Princeton University Press, Princeton.
Fornaldar Sgur Nordrlanda eptir gmlum handritum, 182930, 3 vols. Ed. C.C. Rafn. Kaupmanna-
hfn.
Fornaldarsgur Norrlanda, 18851889, 3 vols. Ed. V. smundarson. Reykjavk.
Hver, Werner, 1995: Zum Stand der Methodenreflexion im Bereich der altgermanistischen Editionen.
In Altgermanistische Editionswissenschaft: 126137. Ed.: T. Bein. P. Lang, Frankfurt am Main.
Jacobs, Nicolas, 1992: Regression to the Commonplace in Some Vernacular Textual Traditions. In
Crux and Controversy in the Middle English Textual Criticism: 6170. Brewer, Woodbridge, Suf-
folk.
Jensen, Helle, 1983: Eirks saga vfrla, utg. Editiones Arnamagnan Series B, vol. 29. C.A.
Reitzlers forlag, Kbenhavn.
Jensen, Helle, 1985: Eirks saga vfrla: Appendiks 3. In Workshop papers I. The Sixth International
Saga Conference 28.7.2.8. 1985. Arranged by Det arnamagnanske Institut, Kbenhavns
Universitet: 499512.
Kleivane, Elise, 2009: Sagaene om Oddr og Eirkr. Ei teksthistorisk tilnrming til to fornaldersagaer.
In: Fornaldarsagaerne. Myter og virkelighed: 2747. Eds. A. Ney, . Jakobsson and A. Lassen.
Museum Tusculanums Forlag, Kbenhavns Universitet.
Kleivane, Elise, forthcoming: Reproduksjon av norrne tekstar i seinmellomalderen. Variasjon i Eirks
saga vfrla. PhD-thesis. Universitetet i Oslo.
Machan, Tim William, 1991: Editing, Orality, and Late Middle English Texts. In Vox intexta: orality
and textuality in the Middle Ages: 229245. Ed.: A.N., Doane & C.B. Pasternack. University of
Wisconsin Press.
Minnis, Alastair J., 1988: Medieval theory of authorship: sholastic literary attitudes in the later Middle
Ages. Middle Ages series. University of Pennsylvania Press, Philadelphia.
Ordbog over det norrne prosasprog, 1989: Registre. Udgivet af Den Arnamagnanske kommission.
Kbenhavn.
Simek, Rudolf, 1984: Die Quellen der Eirks saga vfrla. In Skandinavistik 14/2: 109114.
Springborg, Peter, 1985: Weltbild mit Lwe. Die Imago mundi von Honorius Augustodunensis in der
Altwestnordischen Textberlieferung. In Cultura Classica et Cultura Germanica Settentrionale:
167219. Ed. P. Janni et. al. Macerata.
The Old Norse Elucidarius. Original Text and English Translation, 1992: Ed. E.S. Firchow. Camden
House, Columbia.
Wendt, Bo-A., 2006: En text r en text r en text? Om en terminologisk tredelning av textbegreppet.
Arkiv fr nordisk filologi, vol. 121: 253274.

520
 Rune stones and Saga
Lydia Klos, University of Kiel, Germany

Introduction
Rune stones and saga is the title for a whole session at the 14th International Saga Confer-
ence in Uppsala. In this paper I would like to ask if both sources have anything to do with
each other. Are there any rune stones mentioned in saga texts? What do both sources have in
common and what is the difference between them?
Living in Scandinavia today means that you cant avoid to meet one or another rune stone
in your life, because they are still visible in the landscape and a lot of them are even preserved
in museums or churches. Because of their monumentality and position in the landscape rune
stones have already been an object for scientific investigations in the early stages of research.
As early as in the 16th century Johannes Bureus came out with his Runaknslones lrespn. I
would argue that the reason for this early and ongoing interest in rune stones is the fact, that,
in contrast to smaller personal items or archaeological remains under the surface, rune stones
were and still are present in the landscape.

The stones
Although it is still difficult to date rune stones properly we know that the earliest stones from
Vetteland and Einang were raised in the 4th century. The earliest stones were mostly found in
Norway and Sweden and a little later also raised in Denmark, where the erection cumulated
between 900 and 1000, while the Swedish rune stone production had its climax between the
years 1000 and 1100. Later rune stones went out of fashion and stones with runic or Latin
inscriptions were laid directly on the graves of the deceased at the Christian cemetery. Today
around 3000 rune stones are known from Scandinavia, 2500 alone from Sweden.

The literature
The Old Norse literature is my last introductionary field. Dating it is even more complicated
than dating rune stones, but I hope that you will agree when I postulate that most literature
was written down between the 12th and 14th century, in some cases going back to an older
oral tradition. This means that they have to be seen as secondary sources for the Viking Age
and the period when rune stones were erected. In some cases it is possible that the topic of the
story goes back to an older tradition which was spread orally, but we have to take into account
that this tradition was altered during the centuries. It is also possible that the collectors and
authors of the written texts changed or shortened passages for several reasons. Therefore we
have to be very careful when we take the written sources as an evidence for the Viking Age.
In contrast to continental medieval documents the Old Norse Sagas give a much more per-
sonal view of the characters everyday life. We can for example read in detail about the fur-
nishing of the long house or the bright coloured clothes the hero wears. Therefore it can be
seen quite often in modern Viking Age research that the Old Norse literature is taken as a
main source for the Viking Age everyday life.1 There can be much truth in the texts, but at the
same time it is a big risk to transfer medieval texts back into the Viking Age without a careful
source criticism.2 Therefore it is important to state that the literary sources dont reflect the
1
Look for example at Price 2002 or Nordberg 2003.
2
Meulengracht Srensen states: [] ist es allgemein gebter Brauch geworden, die erzhlenden Texte des
Mittelalters als im groen und ganzen unbrauchbar abzulehnen. [Meulengracht Srensen 1990:58].

521
times they are dealing with, but the perception of the writer: the writer is of the opinion that in
Viking Age Iceland the hall building was furnished with a bright tapestry, but he has not seen
a Viking Age hall with a tapestry. He has maybe heard of it, but not seen. Old Norse sources
are no direct copies of the past, but a reflection.

Rune stones in literature

As those texts claim to portray the Viking Age it seems natural to look for rune stones, be-
cause compared to long houses, ships and weaponry rune stones are fragments from the mate-
rial culture and strongly linked to the Viking Age.
Unfortunately my research about rune stones in Old Norse texts ended faster than ex-
pected: There are no rune stones in Old Norse literature! Not a single saga uses the word rune
stone, no skaldic, no eddic stanza mentions it.3 And here it becomes obvious that Old Norse
literature cant be the only source in discussing Viking Age Scandinavia: If we would only
use literature as our source we have to conclude that rune stones did not exist during the Vi-
king Age. But reality paints another picture: From northern Germany to southern Norrland: In
Scandinavia, and especially Sweden, rune stones are present and we can be sure that there
were many more stones, which are lost today. Every traveller who made his way through Up-
pland in the late Viking Age has seen at least one rune stone, clearly visible in the landscape.
But also in Denmark and Norway rune stones were raised at prominent places: Those visiting
the kings at Jelling would surely pass the brightly coloured stone between the mounds.
How is it to explain that rune stones are totally unknown to Old Norse literature, while
they are visible to the naked eye for everybody? After my unsuccessful search for rune stones
in literary sources I would like to present some possible explanations for the fact that they are
obviously absent in literature.

The wrong word?

The first explanation could be that I was searching for the wrong word. No runic inscription4
designates the rune stone as a rune stone. The single characters are named as rnar r5or as
stafiR runa6. Even the inscription which emphasizes the long life of the stone from Nbbele in
Smland (Sm 16) differs between stone and runic symbol. The inscription is:

Inscription Old Norse English

A rostein * auk * eilifR * aki : A Hrsteinn ok Eilfr, ki ok A Hrsteinn and Eilfr (and) ki
auk * hakun : reisu * eiR * sue- Hkon reistu eir sveinar eptir and Hkon, these lads raised the
inaR * iftiR sin * faur * kubl sinn fur kuml kennilikt remarkable monument in memory
keni*likt * of their father

B In memory of Kali/Kalli the

B ftiR * kala * tauan : y : mun B eptir Kala/Kalla dauan. v deceased. So the good man will be
* ko [] -m kitit * uera * me mun g[s manns u]m getit vera, commemorated while the stone and
* sin * lifiR * auk * stafiR * run mean steinn lifir ok stafir rna. the rune-staves live.

3
In contrast runes in general are used frequently in all genres of Old Norse literature, look for example at Dill-
mann 1995.
4
All runic inscriptions are quoted after Samnordisk Runtextdatabas, version 2.0
5
Inscription from the rune stone in Rk, stergtland (g 136). The complete inscription is much too long to be
cited here. It starts with the runes: aft uamu stonta runaR aR > Old Norse: Eptir Vm/Vm standa
rnar r. > Translation: In memory of Vmr/Vmr stand these runes.
6
Sm 16, Inscription on side B: y : mun * ko [] -m kitit * uera * me * sin * lifiR * auk * stafiR *
run > Old Norse: v mun g[s manns u]m getit vera, mean steinn lifir ok stafir rna. Translation: So the
good man will be commemorated while the stone and the rune-staves live.

522
Hrsteinn and his brothers do not call the rune stone a rune stone, but a steinn with stafir
rna.
Even if we take a look at the carvers formula we can see that the carvers name stands to-
gether with the verb rista = carving, for example U 893 ubiR risti,7 and sometimes the carv-
ers name stands together with the phrase risti rnar8. Not at any time the carving of a whole
rune stone is mentioned. Was it because of the fact that carver, erector and the person who
prepared the stone were different persons and the rune master wasnt allowed to claim that he
made the whole monument? Some people commissioned a stone and they are named as erec-
tors of the stone, in some cases a carver cuts the runes into the stone, but no one makes a rune
stone. Or was is because this monument wasnt simply called rune stone?
If we assume that the word rune stone was simply unknown and therefore not mentioned in
literature, which word do we have to expect?
Only one type of stone is erected in literature: the bautasteinn. This type of stone is nor-
mally defined as an upright standing stone without inscription or ornamentation and is to be
found in combination with graves since the Bronze Age [Burenhult 1999]. The German philo-
logist Heiko Uecker claims that people used the word bautasteinn for heathen rune stones
[] fr die heidnischen Runensteine das Wort bautasteinar zu verwenden. [Uecker
1966:112]. This argument is quite weak, because most Swedish rune stones are clearly Chris-
tian and very few clearly heathen. Only the earliest rune stones written in the elder futhark
lack the Christian sign and formula, but why should literature only mention the old stones and
not the younger ones?
But lets take a closer look at the bautasteinn in literature:
The first king who is cremated in Ynglingasaga is Vanlandi. About his funeral we can read
in Ynglingasaga chapter 16:

Old Norse English

Svar tku lk hans, ok var hann brendr vi er The svear took his body and he was cremated at that
Skta heitir. ar vru settir bautasteinar hans. place, which is known as Skuti (= rock spur) today.
And there were his bautasteinar raised.

According to lfs saga Tryggvasonar bautasteinar were also raised at the grave of Egill ull-
serkr, and likewise right on the graves of king Dmarr and gvaldr [Pesch 1996:112]. In gen-
eral the Ynglingasaga states that the graves of the brunna-ld were provided with bau-
tasteinar [Uecker 1966:111].
As the only saga of the Icelanders the Egilssaga names a bautasteinn in describing the bur-
ial of rlfr in chapter 22. The text says:

Old Norse English

Bjuggu eir um lk rlfs eftir sivenju, sv sem ttt To Thorolfs body they gave all the customary hon-
var at ba um lk gfugra manna, settu eftir hann ours paid at the burial of a man of wealth and renown,
bautasteina. and set over him a bautasteinn.

The burial of rlfr takes place in Norway and due to the fact that bautasteinar are only
mentioned in this saga of the Icelanders, referring to Norwegian custom, and in the Ynglin-
gasaga, referring to Swedish and Norwegian custom, we can assume that the custom of rais-
7
Old Norse: pir risti. Translation: pir carved.
8
For example on the stone from Kllbo/Uppland (U 1042) the carvers formula is: kuli risti run > Old Norse:
Kjli(?) risti rnar. Translation: Kjli carved the runes.

523
ing bautasteinar simply refers to Norwegian or Swedish custom. Archaeological investiga-
tions on Iceland showed in conformity with the literal account that upright standing stones did
not occur on Icelandic Viking Age graves, so this may indeed be a mainland custom.
The last source for the word bautasteinn is the well known Havaml stanza 72, which says:

Old Norse English

Sonr er betri, 73. (72) A son is better,
tt s s though late he be born,
of alinn and his father
eftir genginn guma; to death have fared;
sjaldan bautarsteinar seldom bautarsteinar
standa brautu nr, stand by the road
nema reisi nir at ni. save when kinsman honours his kin.

The statement of the stanza that the named bautasteinn is raised close to a road contradicts the
observation that bautarsteinar both in literature and in the archaeological data were raised in
connection to graves. By looking at the context of the stanza the whole meaning deals with
memory and death and it would also be possible that the bautasteinn should be raised as a
memorial for the deceased close to the road so that everybody can see it, but at the same
time close to the graves of the deceased which became apparent and thus not worth men-
tioning.
However we interpret the Havaml stanza we cannot be sure, that the named bautasteinn
should be a rune stone, because no inscription is mentioned in the poem. As easy as naming
the bautasteinn in the poem it would have been possible to name a memorial inscription or a
rune stone, but it was not. The poem says that people were commemorated by stones and
not with words or runes. We cannot translate the word bautasteinn as a rune stone by implica-
tion, because the bautasteinn as an upright standing stone without inscription exists in the
archaeological account and not at any time it is mentioned in literature, that the named bau-
tasteinn is provided with an inscription. Different types of memorial stones were found from
the Viking Age, but only one type, the simple bautasteinn is mentioned in the texts.
So lets look for other evidence for rune stones in literature:
Snorris description of the kings funerals is very stereotypically [Pesch 1996:149]. His
used vocabulary gives no room for special features. There is only one funeral in Snorris
Heimskringla which differs from the other stereotypical descriptions: The funeral of king
Gulaugr in Ynglingasaga 26. In a stanza, which Snorri assigns to the scald Eyvindr
skldaspillir, the kings grave is described in the following manner:

Old Norse English

ar er fjlkunt This is widely known
um fylkis hrr the stonegrave of the leader
steini merkt, marked with stones
Straumeyjarnes. in Straumeyjarnes

The king was buried in a hrr, which is a special type of grave, where the grave mound is
build up out of smaller stones. This kind of grave is mentioned in at least two runic inscrip-
tions (S 47 & S 176) and well known in the archaeological account. Besides this special
type of grave construction the mound is steini merkt, marked with even more stones. To flag a
mound which is build up out of stones you need even more and bigger stones to mark it visi-
ble in the landscape. May these marking stones refer to a rune stone? Or is it a repetition of
the smaller stones, where the mound was build up with? The stanza needs the s in this verse

524
for the alliteration, the s bears the stave in this line together with Straumeyjarnes in the
last line. So how are these stones to be understood? As the stones from the mound? As upright
standing stones to mark the mound (bautasteinar)? As a purely metrical instrument to keep
the rhythm? Or do we have a rune stone standing at Straumeyjarnes? Unfortunately we cant
decide.
In the same way uncertain is my last example from literature: The Hldskvia, probably
one of the oldest eddic poems [Marold 1998:22], describes in stanza 10 the land which Hl
wants to archieve:

Old Norse English

Hrs at it mra, Famous forest,
er Myrkvi heita, folk call the Myrkvi,
grf ina helgu, at the holy grave
er stendr Goju, of the Gothic highway,
stein ann inn fagra, that famed boulder
er stendr stum Danpar, on the Banks of Danp,
halfar herborgir, half the war-gear,
r er Heirekr tti, that Heidrek owned,
lnd ok la land and people
ok ljsa bauga. and pretty rings.

At the holy grave and close to the main road through the land of the Goths a beautiful stone is
standing. Due to the fact that most stones are normally grey and far from being beautiful, this
stone may be a special one: Maybe coloured, with ornaments or maybe an inscription? At
least this stone marks the grave of an important ancestor, and by naming the grave a claim for
both land and ownership is combined. Moreover the grave and the stone are positioned close
to a main road, which reminds us of the Havaml stanza, where the bautasteinn for the dead
ancestors should be raised close to the road. All this indicates a special place in the landscape
which would be strongly suitable for a rune stone. Another argument would be that the poem
presumably derives from Sweden and the action mostly takes place there9.
But, unfortunately, again no inscription is mentioned in the text.

An unknown word?
After all these examples for special stones close to graves and roads in the Old Norse texts
and the observation that a stone with inscription is never named in literature, although we
know about over 3000 rune stones which are still preserved from the Scandinavian Iron Age,
a possible conclusion would be that the word rune stone was simply unknown when the Old
Norse sources were written down. This would be the easiest explanation, but unfortunately it
is wrong.
In a letter dated to the 19th of March 1287 from the hands of king Magnus Laduls (1275
1290) a rune stone is named for defining the juridical borders in Vstergtland. The text is:

Latin English
Hos terminos inter premissa pascua statuimus, videli- As borders between the named grazing land we decide
cet. de vinnbro. jn jadhurin. jnde in Runustenen. jnde the following: from Vinnbro crossing Jadhurin [a
in mdhalstenen. jnde in Hallin own widh odhens river11], crossing the rune stone, crossing the Mdhal-
kyldu.10 stone and upside Hallin with Odens spring.

9
Uppsala is for example named as the seat of the king.
10
Cited after Jungner, Svrdstrm 1970:150
11
Jadhurin is probably an older term for the rivers Hornborgan or Slafsn.

525
Together with a bridge, a river, another stone and a holy spring the rune stone marks the bor-
ders between the medieval village of Hkantorp and the monastery of Gudhem in Vstergt-
land. The named rune stone may be Vg 90 from Torestorp. The document shows that the
word runustenen was known in 13th century Sweden and rune stones were already in medie-
val times used as landmarks for juridical purpose. They were observed, mentioned and left on
its place. Therefore we have to conclude, that the stones under the name rune stone were
already known in the times when the Old Norse texts were collected and written down.

Results
The results of my paper can be summed up to the following:
1) Rune stones were important elements of the Viking Age memorial culture
2) Rune stones existed in the Viking Age, where the action of most of the Old Norse texts
takes place. They also existed in medieval times, where most of the Old Norse texts were
written down. And they still exist today
3) The term runusten = rune stone was already known in the 13th century
4) Both runes and stones occur in literature, but no rune stone is mentioned. People threw
stones, they hit their head on bigger stones, but never meet a rune stone. Runes are carved in
wood or horn. Far away, in Norway and Sweden, they raised upright standing stones at the
graves of their forefathers, but no runes are mentioned.
To explain this obvious discrepancy, my interpretation is the following:
Rune stones were known in Sweden, Norway and Denmark, where they were still visible
in the landscape, but not on Iceland, where the texts were written down. The collectors didnt
know what was meant when someone raised a stone for his ancestor or king.
This conclusion may lead to a much bigger result: The Old Norse texts reflect the life on
Iceland in a retrospective manner, they describe Viking Age Iceland out of their medieval way
of thinking, when already a lot of elementary knowledge about the Viking Age was forgotten.
In how far can we trust the Old Norse texts in explaining Viking Age Norway, Denmark
and Sweden? Do these sources provide any useful information for research about the Viking
Age Scandinavian mainland?
Apart from any source criticism it is obvious that rune stones and saga are two very differ-
ent types of sources for the Viking Age. They were made at different times, by the hands of
different people and with different intentions. Rune stones were erected in the Viking Age,
saga texts were written down in the medieval period. Rune stones were erected in Norway,
Sweden and Denmark, while sagas were written down mostly on Iceland. Male scholars wrote
on the vellum, while men and women from different social classes erected stones. They
erected stones to praise and honour other men and women, while the sagas were written down
to preserve a special kind of history for a special purpose. But there is one thing, both have in
common: They were written down to be read by future generations.
We have to thank all the erectors who raised their stones in the hope that they would stand
mean verld vakir (G 343), as long as the world is awake. Otherwise a lot of men and
women were forgotten today, people of the Viking Age. To understand the Viking Age prop-
erly we need those relics and cannot rely only on the written sources. In combination of will-
ful and unwitting relics the ideal picture, painted by the written sources, is widened. By study-
ing rune stones a living society can be reconstructed. Historical data may give all relevant
information about when and where and who. But to understand history properly, subjective
sources from all strata of society paint a vivid picture of the past. Maybe rune stones were too
personal, too archaic to be named in sagas. But they provide us with information, which oth-

526
erwise would have been forgotten: All the names, men and women, their family relationships,
journeys and their beliefs.
Therefore I would like to thank Jarlabanke and Odendisa and thank you for reading!

Bibliography
Bjarni Aalbjarnarson 1941: lafs saga Tryggvasonar. In: Heimskringla I. Reykjavk. (=slenzk forn-
rit; 26)
Bjarni Aalbjarnarson, 1941: Ynglinga saga: Heimskringla I. Reykjavk. (=slenzk fornrit; 26)
Burenhult, Gran, 1999: Bautastenar och domarringar. In: Arkeologi i Norden 2, Stockholm, Pp. 254
257
Dillmann, Francois-Xavier, 1995: Runorna i den fornislndska litteraturen. En versikt. In: Scripta
Islandica. Islndska sllskapets rsbok Pp. 1328
Jungner, Hugo / Elisabeth Svrdstrm 1970: Vstergtlands Runinskrifter. Stockholm. (= Sveriges
Runinskrifter Band 5.)
Kuhn, Hans / Gustav Neckel: Edda: die Lieder des Codex Regius nebst verwandten Denkmlern. 4.,
umgearb. Auflage. Heidelberg 1962. (=Germanische Bibliothek: Reihe 4, Texte)
Marold, Edith, 1998: Die Augen des Herrschers. In: Meier, Dirk (Red.) Beretning fra syttende
tvrfaglige vikingesymposium. Kiel, Pp. 730
Meulengracht-Srensen, Preben, 1990: Der Runen-Stein von Rk und Snorri Sturluson-oder Wie
aussagekrftig sind unsere Quellen zur Religionsgeschichte der Wikingerzeit? In: Ahlbck, Tore
(Red.) Old Norse and Finnish religions and cultic place-names. Based on Papers read at the Sym-
posium on Encounters between Religions in Old Nordic Times and on Cultic Place-Names held at
bo, Finland, on the 19th21st of August 1987. Stockholm (Scripta Instituti Donneriani Aboensis
XIII.)
Nordberg, Andreas, 2003: Krigarna i Odins sal. Ddsfrestllningar och krigarkult i fornnordisk reli-
gion. Stockholm
Pesch, Alexandra, 1996: Brunald, haugsld, kirkjuld. Untersuchungen zu den archologisch ber-
prfbaren Aussagen in der Heimskringla des Snorri Sturluson. Frankfurt am Main (Texte und Un-
tersuchungen zur Germanistik und Skandinavistik, Band 35)
Price, Neil, 2002: The Viking Way. Religion and War in Late Iron Age Scandinavia. Uppsala (Aun
31)
Sigurur Nordal 1979: Egils saga Skalla-Grmssonar. Reykjavk. (=slenzk fornrit; 2)
Uecker, Heiko, 1966: Die altnordischen Bestattungssitten in der literarischen berlieferung., Mnchen

527
 Sverris saga in Uppsala De la Gardie 3
James E. Knirk, Runic Archives, Museum of Cultural History, University of Oslo, Norway
The text of Sverris saga is preserved in five complete or fairly complete medieval vellum
manuscripts, twenty vellum fragments coming from eight otherwise basically lost medieval
manuscripts, six paper copies of varying length of lost vellum fragments with text-critical
value, one somewhat abbreviated Swedish translation from the seventeenth century of a now
lost vellum, one Danish epitome from the sixteenth century based on a lost medieval manu-
script, and a multitude of post-reformation paper copies of preserved manuscripts.
There is no modern critical edition of Sverris saga, and therefore the Fornmanna sgur edi-
tion of 1834 is still useful, particularly concerning readings from the vellum fragments in the
Arnamagnan Collection in Copenhagen. The four most complete medieval manuscripts con-
taining the text (AM 327 4to, Eirspennill, Flateyjarbk, Sklholtsbk yngsta) have each been
published in separate critical editions, and the earliest parchment fragment has also appeared
(Jensen 1979). AM 327 4to contains only Sverris saga and is considered to be the codex opti-
mus, in spite of three lacunae near the conclusion.
The slenzk fornrit edition (hereafter F), vol. 30 edited by orleifur Hauksson, appeared in
2007 with the text from AM 327 4to normalized and with a number of eclectic corrections
taken from readings in other manuscripts. The choice of corrections was based mainly on the
genealogical stemma of the major manuscripts.
In the F volume selected readings from the fifth vellum, Sth. perg. 8 fol., are presented.
The manuscript preserves 50% of the text, but can be supplemented by a mid-seventeenth-
century paper copy, AM 304 4to, made when the original was more complete; together they
preserve 70% of the text. In addition, selected readings are included from the fragmentary and
now lost Gullinskinna version (only chs. 116, now preserved in paper copies).
A genealogical stemma for the four major vellum manuscripts was established by Gustav
Indreb in his diplomatic edition of AM 327 4to (1920:xxxili). His results were challenged
by Lrus H. Blndal in 1982, who proposed a separate stemma for the placement of Flat-
eyjarbk in particular in the first half of the text (chs. 1100), but Hallvard Magery (1990)
refuted Lrus Blndals suggestion. orleifur Haukssons attempt to place the Sth. perg. 8 fol.
recension in the stemma led to the proposal of a slightly different and somewhat more refined
stemma (2007:xliliii, particularly xlvi).
There are some weaknesses in the F edition. Unfortunately, not even selected readings
were included from several of the minor text witnesses (in particular the vellum fragments),
and one recension is only commented on in passing, namely Uppsala De la Gardie 3. orleifur
Hauksson mentions Jn Rgmanns seventeenth-century Swedish translation twice, relating
that it was based on De la Gardie 3, a fourteenth-century manuscript destroyed in the fire in
1702, and stating simply (2007:xlvii, lxxviii):

Srleshttir ingunni bera me sr a etta glataa handrit hefur veri skylt B-handritum
[i.e., all manuscripts other than A = AM 327 4to], en erfitt er a afmarka a nnar ar sem te-
xtinn er talsvert styttur. [] Sverris saga er allmiki stytt ingu Jns og v erfitt a kvea
um stu hins glataa handrits ttarskrnni, a ru leyti en v a msir srleshttir eru
sameiginlegir B-handritunum.

In the present paper, the version of Sverris saga in the lost Uppsala manuscript will be ex-
amined using all sources available, and an attempt will be made to place that recension in the
genealogical stemma. In the following presentation, the sigla from the F edition are em-
ployed (cf. 2007:2): A = AM 327 4to, E = Eirspennill, F = Flateyjarbk, Sk = Sklholtsbk

528
yngsta, 8 = Sth. perg. 8 fol., 304 = AM 304 4to, 42 = AM 42 fol. The siglum *U is introduced
for the lost manuscript Uppsala De la Gardie 3. In addition, the siglum X is introduced for
vellum fragments in AM 325 X 4to. All textual references are to the page and line in the F
edition, designated Sv. In order to facilitate comparison of the different versions, the standard
chapter numbers are provided. In cases when readings in normalized orthography from manu-
scripts other than A were not included in the F edition, they are cited, normalized, from the
diplomatic edition of the manuscript or from the manuscript itself.
Whether Indrebs stemma or that of orleifur Hauksson is to be accepted, is not the issue
here, and for the genealogical placement of *U, this question turns out to be of little impor-
tance.

Uppsala De la Gardie 3
Count Magnus Gabriel De la Gardie (162286) possessed in the 1660s a vellum manuscript
containing various kings sagas (Johnsen & Jn Helgason 1941:1077). It was among those
donated by De la Gardie to Uppsala University in 1669 where it is termed in the letter of do-
nation Histori Regum, alias Konunga Sagurne. The manuscript, Uppsala De la Gardie 3,
was on loan from the university library to Olof Rudbeck and perished in the conflagration in
Uppsala in 1702. There are, however, various sources for information on its contents.
In the 1660s the Icelander Jn Rgmann made a translation of the manuscript into Swed-
ish, and this translation was published at Visingsborg in 1670 as Norlandz Chrnika och
Beskriffning (hereafter called R, for Rgmann; Sverris saga concludes the translation). John-
sen and Jn Helgason examined the separate lfs saga helga version in *U (1941:107888),
a manuscript which they date to the fourteenth century, and concluded that the text was not
interpolated and should have been used in their edition, but that this fact was realized at too
late a stage; besides, a reconstruction would have been a great and probably unrewarding task.
In their characterization of the translation of lfs saga helga in R, they relate, among other
things, that the language is frequently tainted by Icelandic expression and that skaldic verse is
as a rule dropped; other than that, the translation does not seem to be abbreviated. They men-
tion that the version preserved in *U of the following kings sagas up to King Sverrir, an in-
terpolated reworking of Heimskringla III, is closest to the Eirspennill recension, and state that
the text of Sverris saga that follows is forkortet, srlig mot slutningen.
The text in the second part of *U, covering the Norwegian kings from 1030 until 1177
(Heimskringla III), was examined in detail by Jonna Louis-Jensen (1977:3440). In this part,
*U was confirmed to be a sister text of Eirspennill.
Another source for the contents of *U, often containing direct quotations of short phrases
or longer contexts, is Olof Vereliuss Old Scandinavian dictionary published in Uppsala in
1691 as Index lingv veteris Scytho-Scandic (hereafter termed V). Here one finds some 190
citations of words or text from Sverris saga identified as deriving from *U. The dictionary
quotations complement the Swedish translation: phrases omitted by Rgmann frequently ap-
pear in the dictionary, and at times translations turn out to be fairly free reformulations.
There are, of course, great difficulties connected with attempting to employ a seventeenth-
century abbreviated Swedish translation as the main witness for an Old Norse text, even if one
has a number of dictionary supplements in Old Norse. Various methodological problems will
be illustrated below.

Rgmanns translation of Sverris saga

Rgmanns translation of Sverris saga in Norlandz Chrnika och Beskriffning covers pp.
41152[3]. The chapters are numbered consecutively, and the saga Om Konung Swerre
begins with ch. 451 and ends with ch. 596.

529
 The text of lfs saga helga in *U was not abbreviated, and the same is basically true con-
cerning the text from Heimskringla III. The text is apparently not shortened to any degree at
the beginning of Sverris saga either, and the Swedish retelling follows the Old Norse text
closely. Although there is a general simplification and reduction throughout the text, orleifur
Haukssons statements concerning the considerably/greatly abbreviated form of Rgmanns
translation of Sverris saga is an oversimplification and an exaggeration. There are indeed
some sweeping shortenings of long battle presentations and also of long or repetitive speeches
in the first half of the saga, but the impression of an abbreviated text applies in the main to the
last half or third of the saga concerning chiefly the conflict with the Eyjarskeggjar in 1194,
with the Baglar from 1196 onward, and particularly with the farmers in Oslo in 1200. Com-
paring the length of the translation with the length of the printed text in the F edition gives an
indication of the relative degree of shortening. For chs. 1100 (excluding a lacuna in *U), one
page of translation corresponds to 2.1 pages of the edition, whereas for chs. 10175 (where
*U broke off, see below), one page corresponds to 2.9 pages of the edition. Comparing only
chs. 15075, one page of the translation corresponds to 3.5 pages of the edition; thus in com-
parison with the first hundred chapters, there is a reduction of 40% in the translation. It seems
reasonable then that Rgmann either tired of his task, or ran out of time as work progressed,
and gradually was forced to or decided to abbreviate more and more. The first half to two-
thirds of the text, however, provides a reasonably good source for an attempt to reconstruct
the text of Sverris saga in *U.
Fifteen parenthetical comments, probably as a rule written by Rgmann, are included in
the translation, generally in petit roman type. The parentheses usually contain the explanation
of a word, e.g., R425.67/ch.18 een Einstig/ (een smal Wg/ som intet kan gs mehr n aff
een Man/) and R468.4/ch.78 Gullwnden (Sceptrum Regale,), but there is also one cross-
reference (R436.3233/ch.30) Hecklingar/ (vide Cap. 484. infra/). In connection with the
lacuna in *U encompassing Sv10.2218.7/chs.610, a parenthetical statement (R415.2023)
explains that the following text is taken from Den Danska Per Clauons Version; there is no
comment, however, when translation from *U again begins at R416.26. Likewise the fact that
the end of the text, from Sv272.16 on, was missing in *U and was taken from Peder Claussn
Friiss translation is related parenthetically.
In addition to the use of individual Icelandic words such as Einstig mentioned above and
the Icelandic spelling especially of several personal names and place-names, Old Norse direct
quotations are three times included in the text: (1) R438.3738/ch.33 Fall er farar Heill; (2)
R448.2930/ch.47 Ty ecki kiemr ufrigum [!=feigum] i hel/ ock ecki ma frigum forda i
Flotta er fall werst; (3) 449.58/ch.47 lla [!=tla] eck mier ena Mru / Mune fagru
Jorunni / Huegi er fundr med freygium / Ferr Magnus ok Suerri. (Poetic stanzas, as in the
third quotation, are otherwise as a rule dropped, although they are usually introduced or men-
tioned, e.g., R446.15/ch.44 T bleff een Wijsa qwden.)
There are abundant typographical errors in the text. Of more interest are the mistakes
clearly made by the translator. Abbreviations in *U were sometimes apparently misunder-
stood, e.g., when R428.16 mentions twelve Brder (Sv36.16/ch.21 Bndr tlf). Likewise
reference to a person only by title could be misunderstood. Thus in the description of Sverrir
rowing between his ships in a small boat and giving instructions during the battle at Fimreiti,
one finds (R475.3) K. Magnus rodde och s emellan Skepen; the mistake was caused by the
reading Sv141.23/ch.91 Konungr referring to Sverrir.
Further interesting readings from the text will be discussed below.

Vereliuss entries from Sverris saga in his Index

530
Olof Verelius (161882) compiled the first printed Old Norse dictionary. His Index lingv
veteris Scytho-Scandic, published posthumously (1691) by Olof Rudbeck, was actually an
Old Scandinavian dictionary based on Old Swedish provincial laws and a multitude of Old
Icelandic and Old Norwegian texts (laws, sagas of various kinds including the entire Orms-
bk, religious literature, learned texts, Eddic and skaldic poetry, etc.).
The dictionary entries consist of headwords, translations into Swedish and Latin (or expla-
nations, sometimes with additional cultural comments or etymologies), a reference with page
or chapter specification to the text where the word was registered, and often citations present-
ing the keyword in context and accompanied by a Swedish and/or Latin translation of the
phrase or sentence. A sample entry will illustrate the dictionary. (Gothic typescript is the
standard and is here represented by roman type, whereas roman type is here underlined; page
numbers in V are included, with column, to compensate for the somewhat haphazard alpha-
betization in the dictionary.)

Sibyrt, & sidbyrt, Skepzbord/ Latus navis Ab Sida Latus & Bord latus. Kong. pag. 223. Kong.
S. pag. 223. Ef eir hefdu sibyrt, Si latus navis, lateri hostilis navis, applicuissent. ibid. Han
lagdi skip sibyrt vid skip eirra. Han lade om bord med Fienden. [] (V219a)

It is not easy to determine the exact number of citations from Sverris saga in V. Two have not
yet been identified as definitely coming from this text (see below). In the sample entry just
quoted, there are two individual citations for one headword. On the other hand, Sald
(V214a), with the cited phrase [] romborat. Salld [], corresponding to Sv154.8/ch.99,
and also the entry Romboral [!] salld (V210b) appear, both refering to the same citation.
Since there are two headwords, they are nonetheless included as two separate citations in the
total of 192 registrations in V from Sverris saga. On the other hand, the entry Birkbenar
(V35b), simply from Kong. S. with no page specified, could be from either Heimskringla
III or Sverris saga, and is not included in the 192 registrations.
There are at least seven mistakes in the page numbers for citations, making it at times dif-
ficult to identify the quotations. An example is Flemta (V73b) with the citation Han fle-
mptar yfr [!=ydr] af rikeno corresponding to Sv192.1011/ch.126 and with pag. 288 incor-
rectly for 238. The last quotation from Sverris saga, Hus (V128b), comes from pag.
262 (corresponding to Sv272.1516/ch.175), and in a parenthetical note starting in the very
next line in Rgmanns translation (R521.35522.4), it is related that the manuscript breaks
off there. Such mistakes with page numbers necessitated examination of all citations from
Kong. S. in case any of the others might actually represent text from Sverris saga. This,
however, did not prove to be the case. In general, entries with bare references to somewhere
before Sverris saga began, but that could perhaps have come from Sverris saga, were as a rule
words found in other dictionaries with references to Heimskringla III, Morkinskinna or the
Hulda-Hrokkinskinna compilation, thus from the kings saga texts in part two of *U. Only
one such headword was identified as referring to Sverris saga: Blistra, cited from pag.
176, which corresponds, based on the citation Atti han a af [!=at] blistra i spor honom, to
Sv38.1213/ch.23.
A minor problem with the dictionary is that headwords do not always correspond with the
keywords in the citations, e.g., the headword Byrleidi gott (V047b) with the citation Byr
rann (Sv201.1/ch.134). Typographical mistakes also occur, e.g., Eidstafr (V55a) with the
citation Ganga lidstaf [!] (Sv96.67/ch.60). The headword Flur (V74b) with the citation
Flur oc hueti illustrates, when compared with Rgmanns translation, that Verelius could
rephrase the original; R482.31 has Hwete/ Honing/ Floor och Klde, corresponding exactly
to Sv159.1011/ch.104.

531
 Several incorrect translations by Verelius occur, e.g., Gryla (V100b) defined as Gry/
gryning/ Diluculum (from pag. 187; this word comes from the preface, thus establishing
the first page of Sverris saga in *U). In reconstructing the text of *U, however, translation
mistakes by Verelius as such are not of interest.
There remain two unidentified entries attributed by Verelius to Sverris saga. arnast
(V261b), clearly for arfnask (defined as quasi Tarfuas, Carere, amittere, Fattas) and with
the corrupt and unidentified quotation Hu arnar ssar huegra at? Hwar till tarfwades det
hgre, could not be found in any version of the saga where it is supposed to occur (some-
where between Sv91.23/ch.58 and Sv99.17/ch.62). Harfengiliga (V109b) for harfengiliga
does not occur in any other recension of Sverris saga in the description of the battle at Fim-
reiti, although the bare reference implies that it occurred between Sv141.11/ch.91 and
Sv146.10/ch.94.
Three problematic entries have been identified with some probability. Harri (V109b)
on pag. 204, corresponding to somewhere between Sv63.19/ch.38 and Sv73.7/ch.47, may
represent a singular reading for Sv64.18/ch.39 ofsi. The bastard word Mandyrdarmenn, He-
roes [], Hieltar [] (V168a), with a bare reference to pag. 220 and thus supposedly
occurring somewhere between Sv125.11/ch.81 and Sv134.13/ch.87, probably corresponds to
Sv128.21/ch.83 margir lendir menn AFSk 8; the variants margir mikilshttar menn E333.25
and margir gtir menn X5vb1314 might allow one to assume that the original reading in
*U was margir drarmenn, or perhaps rather that it included this reading, as indicated by the
doublet in R471.89 mnge frnehme och Lndamn. The unknown Old Norse word
Traka (V258a), defined as Transenna, Stacket hakelwrke, is said to have been on pag.
228, which would correspond to somewhere between Sv159.10/ch.104 and Sv164.20/ch.107.
The probable identification is with Sv269.30/ch.172 flaka (nom. flaki wicker-work shield),
mentioned in connection with Sverrirs attacks during the siege of the fortification in Tns-
berg. This solution is attractive, in spite of the fact that both the spelling and the page refer-
ence would be incorrect (R520.28 Flaackar would have been on pag. 261 or 262).
Verelius has taken excerpts from the entire text but was clearly more interested in the vo-
cabulary of certain sections. There are a total of 192 citations from the 76 pages of text, thus
about 2.5 per page. Several pages, however, have not been excerpted, whereas others have
more than their share of quotations. The largest numbers are from pag. 223 and 224 concern-
ing the battle at Fimreiti in chs. 9196 where twelve and fifteen quotations were registered
respectively, and pag. 241 concerning Sverrirs conflict with the Baglar and Bishop Nikols
in chs. 13134 with twelve citations. The proportion of quotations where text corresponding
to Vereliuss headwords or citations is not found in some form or other in the translation,
around one-third, gives an indication of the amount of abbreviation. The sample entry quoted
above for sbyrt illustrates the relationship. The first quotation is translated by Rgmann
(475.2122) om the hade laget Sijda wid Sijda (corresponding to Sv142.2324/ch.91),
whereas the second quotation is not represented in Rgmanns translation since the text is
abbreviated (at R475.29, Sv143.1415/ch.92).
At times the translation in R can be misleading as to the exact wording in *U, and V can be
used to correct this. For instance, R422.12 Tijdender corresponds to Sv27.9/ch.16 tendi
A, but the synonymous headword Nymli (V183b, with a bare reference to pag. 193)
illustrates that the reading in *U corresponded to the variant nymli EFSk. A singular and
interesting reading from *U is the headword Tuisynt (V260a) with the citation Tuisynt
otti huart fyrr myndi hrodit it mikla skipit corresponding to Sv143.1112/ch.92 at engi
mar ttisk vita hvrt fyrr [] mundi it mikla skipit vera hroit; this word does not occur in
any other version, nor is it reflected in Rgmanns translation, which has simply R475.2930
at man icke gierna kunde see hwilken ther skulle winna.

532
 The genealogical affiliation of *U
In the following, readings from *U are compared first with manuscript readings presented by
orleifur Hauksson in his discussion of the stemma (2007:xliliii) and then with the variant
readings provided in footnotes to his F edition. Frequently the readings from R cannot be
used to place *U conclusively with respect to the other manuscripts, since there are simply too
many possible sources of error. Such instances are passed over below without further notice.
Of the three shared errors that establish the B-class of manuscripts (2007:xliii), two are
found in *U: R418.1/ch.12 Rijkes Hradt and R467.35/ch.77 Strindar Fiord. R also has
een Grd/ som heeter Steen (R446.24/ch.44), agreeing with A and E (binn Steini A,
binn Stein E) against Beinstein Sk and Bensteini F. orleifur Hauksson suggests here that
this mistake was in all the B-manuscripts, but was corrected in E; it would then have to have
been corrected twice, in E and *U, or in a common original for the two. In connection with a
mistake shared by F and 8, about Eilfr in ch. 86 with the by-name ra E (A, Arason F, Era-
son 8), *U agrees with orra Sk (R471.29 Orra), a reading that could be just as correct as
Es. The common mistake found in F and 8 in ch. 164, Svartabum for Starabum, also
occurs, perhaps independently, in *U (R516.20 Swarta-bod; A has a lacuna here). Concern-
ing the other cited variants, R appears a few times to be closer to Sk, e.g., in ch. 1 (2007:xlvii)
som vhr ett Jrn/ som hefftigt blses th i Smediesian (R412.1718). The examples of
doublets (2007:lilii) are discussed in the conclusion below.
A comparison of the textual variants in notes to the F edition shows clearly that E, Sk and
*U constitute a group of closely related manuscripts. The following are two striking ex-
amples. In connection with orsteinn kgars surrender of the fortress near Niarss and the
Kuflungars subsequent taxation of the townsmen, only ESk*U, not AF 8, relate that the
townsmen paid the tax (at Sv167.8/ch.108): en bjarmenn greiddu heldr gjaldit (R485.16
hwilcken Borgarena vthgofwo). The conversation between Sverrir and lfr jarlsmgr in
Sv179.22180.5/ch.118 is found in AF 8, but not in ESk nor in *U (at R490.15).
Readings also point in other directions, but these instances may be only individual anom-
alies. In connection with Sv24.8/ch.14 Vazfell E (correct only here, for present-day Vassfjell
in South Trndelag), *U apparently had the same minor misspelling (R420.12 Wasa Fil)
as Vazafell A; the other text witnesses have Vaafell Sk (and 42), Nafnafell 304, F. A shared
cardinal mistake even separates E and Sk from *U, namely Sv127.26/ch.82 (cf. Sv135.3/
ch.88) Haugastrandar ESk against Hugastrandar AF 8 X5va23 *U (R470.32 Hugastrand),
but both readings are mistakes for present-day Hagastranda in Sogn.
The majority of readings point toward a closer connection between *U and Sk than be-
tween *U and E. For example, *U and Sk have the same reading of the by-name Sv69.27
/ch.45 Skuu- A, namely Skrddu- (R447.3 Skryddu), and here each manuscript otherwise
has its own reading (Skru- F, Skru- E); by-names are, however, not entirely reliable as
evidence, i.a. since they may be remembered from elsewhere and introduced by a copyist. The
special reading corresponding to Sv52.7/ch.32 en myndi ykkja vnligast um eira r
found otherwise only in Sk, namely en horfisk , is shared with *U (R438.1 n hwad
gonnen tinligit war). Likewise, the headword efur (V262a), with the cited phrase
Hrfa efr af valnum (corresponding to Sv146.21/ch.95 hrvadaun af valinum), affiliates
*U and Sk124.7 against all other recensions, including E344.35.
Other readings, a smaller number, point in the opposite direction and group *U with E
against Sk. Some appear convincing, however, in particular in ch. 16 concerning the trans-
porting of ships over land either a distance of Sv29.12 fim rastir AF 304/fim mlur Sk, or over
ei mikit E (yfir missing); *U too has the large isthmus (R423.2829 fwer et stoort Eed).
E, Sk and *U form a group of manuscripts with related readings. Either all three are simply
sister manuscripts, deriving from the same original, or two of them are sister manuscripts, and

533
their original is a sister to the third. The impression arrived at from all the examined variants
is that Sk and *U are sister manuscripts, and their original is a sister manuscript to E.
There might be one clear error shared by R and Sk that would underpin this suggestion for
the genealogical placement of *U. The mistake concerns Magns Erlingssons man Sigurr
Nikolsson mentioned in chs. 14, 34, 36 and 37. He is in ch. 34 concerning 1179 (at Sv56.2
3) in Sk not only confused but combined with Sigurr af Saltnesi, Sverrirs man mentioned in
ch. 14 and whose death in 1178 was related in ch. 28. Sk42.2324 reads kom til Sigurr af
Saltnesi Nikols<son> ok Jn af Randabergi, and *U has (R440.1718) kom Sigurd Niku-
lasson aff Saltns til honom/ tillijka med Jon aff Randaberg. Since, however, there is an ad-
ditional mistake in Sk (-son is missing, and thus the reading implies three men, including Nik-
ols and Jn, both af Randabergi), this shared mistake may not be entirely conclusive.

Conclusion
If *U is indeed a sister manuscript of Sk (and their original a sister manuscript of E), it can be
used to identify several readings that most likely were in their common original, particularly
mistakes, deletions, and innovations in Sk. Regardless of its exact placement in the genealogi-
cal stemma, readings from *U can be used to check some of orleifur Haukssons choices of
specific corrections to A. In addition, support for readings in A can be found in cases where
*U as the only B-group manuscript contains the same reading as A. These uses of *U are il-
lustrated in the following.
orleifur Hauksson (2007:lilii), following Lrus Blndal, considers the use of double ex-
pressions in Sverris saga, especially in A, as a good explanation for much synonymous varia-
tion among manuscripts: tautologies could be simplified to single expressions and choices
could vary. For example, the assumed original doublet Sv35.1011/ch.20 handfesti ok vpna-
tak A was simplified to handfesti E and vpnatak FSk. The same tautology as in A is found in
the B-class manuscript *U (R427.25 Handslag och Wapnetack), confirming its existence in
both classes and thus as far back as the original *X. The other double expression listed there
from the same chapter is also in *U, although with a slight mistranslation: R427.29 brdt aff
hg Miskund (Sv35.14/ch.20 bra ok hleita miskunn). *U even contains additional tau-
tologies that most likely were in the original of Sverris saga. The variants in Sv31.9/ch.28
troa FSk 304 and ganga AE can be explained as simplifications of the tautology in R424.37
g och trda. Perhaps even the variants in ch. 16 mentioned above, where a length of five
rests/miles alternated with a topographic statement, over a large isthmus, could be ex-
plained as resulting from the simplification of two-fold information: five rests/miles over a
large isthmus.
*U can contribute to the understanding of the genealogical relationships between the other
manuscripts of Sverris saga. It can be used to correct corrupt readings in Sk, to confirm some
readings from A as more original than other readings, and to demonstrate additional doublets
probably in the original text. The lost Uppsala manuscript is thus an important witness con-
cerning the text of Sverris saga.

Bibliography, abbreviations, sigla

A = AM 327 4to = Sverris saga etter Cod. AM 327 4to. Ed. by G. Indreb. Kristiania, 1920.
E = Eirspennill AM 47 fol. Nregs konunga sogur: Magns gi Hkon gamli. Ed by Finnur
Jnsson. Kristiania, 1916.
F = Flateyjarbk (Gl. kgl. sml. 1005 fol.) = Flateyjarbok. En samling af norske Konge-sagaer. [Ed. by
Gubrandur Vigfsson & C. R. Unger.] 3 vols. Christiania, 186068.
Friis, Peder Claussn, 1633: Snorre Sturlesns Norske Kongers Chronica. Trans. by P. Cl. Friis. Ed.
by O. Worm. Kjbenhafn. [Sverris saga is taken from Mattis Strssons translation: Norske Kongers
Krnicke oc bedrifft, Kibenhaffn, 1594; cf. Den norske krnike, ed. by M. Srlie, Oslo, 1962.]

534
F = slenzk fornrit.
Indreb, Gustav, 1920: Innleiding in the edition of A. Pp. vlxxviii.
Jensen, Helle, 1979: Fragmenter af et kongesagahndskrift fra det 13. rhundrede. In: Opuscula, vol.
6. Kbenhavn. (Bibliotheca Arnamagnana 33.) Pp. 2473.
Johnsen, Oscar Albert & Jn Helgason, eds. 1941: Saga lfs konungs hins helga/Den store saga om
Olav den hellige. Oslo.
Lrus H. Blndal, 1982: Um uppruna Sverrissgu. Reykjavk, 1982. (Stofnun rna Magnssonar
slandi, Rit 21.)
Louis-Jensen, Jonna, 1977: Kongesagastudier: Kompilationen Hulda-Hrokkinskinna. Kbenhavn.
(Bibliotheca Arnamagnana 32.)
Magery, Hallvard, 1990: Om skyldskapen mellom tekstene i fremre delen av Sverris saga (kap. 1
100). In: Maal og Minne 1990. Pp. 15369.
Norlandz Chrnika och Beschriffning. Ed. by O. Rudbeck. [Trans. by Jn Rgmann.] Wijsingzborg
[Visingsborg], 1670.
R = Rgmanns translation of *U, see Norlandz Chrnika och Beschriffning.
Sk = Sklholtsbk yngsta (AM 81a fol.) = Det Arnamagnanske Haandskrift 81a Fol. (Sklholtsbk
yngsta). Ed. by A. Kjr and L. Holm-Olsen. Kristiania/Oslo, 191086.
Sv = Sverris saga. Ed. by orleifur Hauksson. Reykjavk, 2007. (Islenzk fornrit 30.)
orleifur Hauksson, 2007: Formli (pp. vxc), and edition. In: Sv.
*U = Uppsala De la Gardie 3.
V = Verelius, Olof, 1691: Index lingv veteris Scytho-Scandic sive Gothic. Ed. by O. Rudbeck.
Vpsal.
X = AM 325 X 4to (plus AM 325 VIII 3a 4to; readings from the manuscript).
8 = Sth. perg. 8 fol. (readings from the manuscript or from Sv).
42 = AM 42 fol. (one of sgeir Jnssons copies of the Gullinskinna fragment of Sverris saga).
304 = AM 304 4to (readings from the manuscript or from Sv).
= dropped or missing.

535
 When was the Battle of Helge?
Annette Kruhffer, Farum, Denmark
The power balance between the three Scandinavian countries was challenged during the reign
of Cnut the Great. Alliances between countries were based upon personal relations between
powerful people. When King Swein Forkbeard of Denmark died, the firm coalition that se-
cured peace in Scandinavia during the years 9991014 crumbled. His successor on the Danish
throne was, however, capable of maintaining good relations to the Swedish neighbour. This
was partly due to the fact that his sister, the Danish princess Estrid, was a sister to the Swed-
ish king as well. That is, her mother Sigrid was at the time of her marriage to king Swein a
widow of King Erik of Sweden by whom she had given birth to king Oluf of Sweden.
Since the Battle of Svold in 999, Norway was governed by the brothers Erik and Swein,
sons of Earl Haakon. Erik had married a daughter of King Swein of Denmark, while Sweins
wife was a sister to King Oluf of Sweden. The Norwegian earls ruled Norway in concert with
their families in law, and all of Scandinavia was dominated by strong mutual interests and
united with ties of kinship.
After the death of King Swein, Cnut travelled to England to recapture the country, and he
was assisted by Earl Erik. At this time at the latest, Earl Eriks son Haakon was made a co-
ruler in Norway with his uncle Swein. Olav Haraldson a descendant of King Harald Haarfa-
gre of Norway fought on English side, but afterwards he went to Norway and claimed the
throne. The political stability of Scandinavia was shaken as Haakon and Swein were driven
out of Norway.
In the beginning of his reign, King Olav Haraldson was rather isolated in a Nordic context.
After some years he married a Swedish princess and when her brother, King Anund Jakob
acceded to the throne, the possibility of reconciliation between Norway and Sweden appeared
as a reaction towards the strong Danish-English royal power.
The growing tension between the two groupings resulted by the end of the 1020s in
Cnuts sovereignty in Norway, while Sweden remained an independent state. But it is not
quite clear how things came to that point. The only military battle we know of from that pe-
riod is the Battle of Helgeaa. Even it has been investigated by many historians (e.g. Steenstrup
1972, Moberg 1941, Lawson 1993 and Lund 1994), the events before, during and after the
battle are still dimly elucidated. This is due to the difference and frequent contradiction be-
tween the sources.
The enmity between the Scandinavian kings is firstly expressed in a request of submission
from Cnut to Olav. Olavs rejection and subsequent alliance with King Anund Jakob resulted
in a Swedish-Norwegian raid on Denmark. Cnut met the aggression with a Danish-English
army and haunted the allied troops to Helgeaa. Here, the two armies met in a battle ending up
with a suspended victory to Cnut. Anund Jakob was capable of sailing home but Olav was
forced to leave his ships in Sweden and travel home by land. Cnut headed for the coronation
of Conrad II in Rome. On his way home to England he made a detour to Denmark, but Norse
sources do not mention his errand. Only a letter, written home by Cnut, indicates some sort of
conflict that had to be settled. The next year Cnut left England with the army and sat sails for
Norway. Arriving there, he was elected king by the people while Olav escaped to Russia,
eventually returned and fell in the battle of Stiklestad in 1030.
This very brief version covers the general perception of the events and their chronological
order, build on a variety of sources among which are the above-mentioned letter written by
Cnut (Whitelock 1979:476). This letter has been crucial for the dating of events since Cnut
from Rome seems to be addressing the battle of Helgeaa. A lot of questions rise from this
fact, of which just three are to be mentioned here: Why did Cnut choose to go to Rome and

536
thereby prioritising his European contacts when at the same time he was threatened by well
equipped and hostile neighbours? Why did he make the decision to travel back to England via
Denmark? And, considering the fact that the three armies split up and went each to their des-
tination after the fight, who won the battle of Helgeaa?
I intend to challenge the interpretation of that specific passage in Cnuts letter which is
used to date the battle of Helgeaa and, in doing so, provide a new explanation of the order of
the event; an explanation which more logically explains Cnuts route home from the corona-
tion, and which contributes to an understanding of time and outcome of the Battle.

Snorres chronology and other sources

The most detailed account remaining of the events is to be found in the Norse sources and the
far most coherent of them is Snorre Sturlusons saga on Olav Haraldson. This saga is part of
Heimskringla, which Snorre finished in the second quarter of thirteenth century 200 years
after the events described. The saga constitutes a crucial part of the entire work, and Snorre
wants to tell the story of king Olav who lost life and land in the final battle of Stiklestad but
nevertheless ended up as Rex Perpetuus Norvegiae.
Despite the long time past, Snorre and his contemporaries had a detailed remembrance of
Olav. As a great-grandson of Harald Haarfagre who united all of Norway, people took interest
in documenting and preserving memory of his deeds from his very youth. Only a little more
than a year passed after his death before the first stories of his sanctity were told, and he was
soon to be worshipped as a saint. The tradition emerging in his cult is therefore likely to be
close to contemporary knowledge at least in its origin. The memory of Olavs political life
was vivid already when he was first mentioned a saint. As a consequence, the freedom of the
historians to arrange the history for their own purposes was somewhat restrained.
Snorre gives a detailed account of Olavs life from the moment he set foot on Norwegian
soil, and he arranged the episodes in a strictly chronological order. Even the inserted secon-
dary stories are fit in very neatly: time and geography of the outset or the end of the story hits
the main epic at the proper place. It is vital for Snorre to date every incident. To mark the
passing of time Snorre usually uses the names of seasons or holidays. When spring came or
in summertime are expressions found frequently during the narrative. Snorre gives a special
remark on the 3rd, 7th, 10th and 15th year of Olavs reign and these explicit years are in ac-
cordance with the epic time of the story.
Snorre recounts the continuation of bad relations between Sweden and Norway in the be-
ginning of the reign of Anund Jakob. Olav keeps armed forces ready at the border between the
two countries in case of need. Snorre describes how the two kings came to an agreement, and
contributing to this is Cnuts request of submission by Olav. This made Olav contact Anund
Jakob during the autumn. The two brothers in law reached an understanding, and when Cnut
sent letters of friendship to Anund Jakob they were received nok s koldt (Sturluson
1979:344), Snorre reports. This incident happened in the winter of 1025/26 according to
Snorres chronology.
Olav and Anund Jakob had a meeting in Elv in 1026. The outcome was not seen until the
following winter. Olav gathered his army and when spring broke, he sailed south from Trond-
heim where he had spent the 13th winter of his reign.
After a few stops, he sat sails for Denmark and started to plunder and harass on Sjaelland.
King Anund, too, gathered an army and attacked Skaane. When Cnut heard about the raid on
Denmark he also gathered his army and sailed, via The Limfjord, against the aggressors. His
two adversaries escaped but were caught up at Helgeaa. With fairly even forces on their
hands, the outcome was a dubious victory in Cnuts favour. Olav and Anund Jacob were ca-
pable of leaving the battlefield with ships and men, but Cnut was too strong for them to fight.

537
Anund Jacob sailed home but Olav did not dare to sail his ships through Oeresund through
Cnuts blockade. He therefore left his fleet in a Swedish harbour and travelled by land to
Norway.
Snorre gives a fairly accurate dating of the battle of Helgeaa. Ahead of the battle, he states
an indication of time: This was the year where Olav went to Nidaros to spend the winter. And
it was the 13th winter of his reign, Snorre writes (Sturluson 1979:354). Later in the saga he
states the day of Olavs death as Wednesday July 29th (Sturluson 1979:440). This combination
of weekday and date points to the year of 1030 (Einarsdttir 1964). When Snorre marks the
time of Olavs departure from Norway in the saga, he says:

Da hadde kong Olav vrt konge i femten r nr vi tager med den vinteren da han og Svein jarl
var i landet begge to, og s denne som vi n holder p at fortelle om. Det var alt over jul da han
forlot skipene sine og gikk opp i landet, som vi n fortalte. (Sturluson 1979: 404)

Counting the years, from this time to the Battle of Stiklestad makes 1 year or two winters
in Snorres chronology. According to the text, Olav leaves Norway right after Christmas at
New Year 1028/1029. He must have ruled 13 winters by New Year 1026/27, and we can
therefore conclude that Snorre dates the Battle of Helgeaa to late summer or early fall 1027.
Nevertheless Snorre does tell us, that Cnut stayed in Denmark during the winter 1025/26.
He wrote Anund Jacob in order to make an alliance between the two countries and at the same
time, he installed Earl Ulf in power along with Hardeknud. We only learn about this journey
to Denmark from Snorre and usually the historians ignore this information. However, there
are reasons to consider its validity anyway.
We consider the position of Ulf as ruler of Denmark on behalf of Cnut as a fact, but we
dont know when he took up this position. In Fagrskinna, considered a few years older than
Heimskringla, the only date of the battle is hidden within the information about Earl Ulf gov-
erning Denmark at this time (Finlay 2004). According to the Anglo-Saxon chronicle, Cnut
appointed Earl Thorkell as his substitute in Denmark in 1023 and left his son with him. That
is about the last we hear of Thorkell, and he is therefore believed to have died shortly after.
He must have been followed by Ulf but we have no information about this.
The Anglo-Saxon Chronicle dates the battle of Helgeaa to 1025:

In this year King Cnut went with ships to Denmark to the Holme at the Holy River, and there
came against him Ulf and Eilaf and a very great army, both a land force and a naval force, from
Sweden. And there very many men on King Cnuts side were destroyed, both Danish and Eng-
lish men, and the Swedes had control of the field. (Whitelock 1979:100)

No other sources support this date, and the Battle of Helgeaa is far outside the geographical
and political environment of the Anglo-Saxon clergy. Consequently, Lawson suggests (Law-
son 1993:96) that the wrong date origins from the fact that next entrance of the Chronicle is
1028. The Battle could therefore easily be recorded a year or two on the wrong side without
the author himself having any opportunity to discover his mistake.
But the argument is pointing in the other direction as well. Since no recordings seem to
have been made simultaneously during these years, it is possible that Cnut actually travelled
to Denmark that winter as claimed by the Norse sources. In that case it would be an under-
standable mistake to connect the purpose of the trip to the battle of Helgeaa not least be-
cause Cnut actually had received a rejection of his proposal to the Swedish king at this point.
Another unexplained discrepancy may be worth noting. Snorre writes:

538
 Da det lei p vren, gjorde kong Knut seg ferdig med hren til reise vest til England; han satte
Horda-Knut, snnen sin, igjen I Danmark, og satte Ulv jarl, snn til Torgils Sprakalegg, hos
ham. (Sturluson:350)

This has puzzled historians, since Snorre seems to date this event to 1026, while the Anglo-
Saxon Chronicle for the year 1023 reports about Cnut leaving his son in Denmark with Thor-
kell to rule the country. Immediately after Cnuts return to England in 1023, he is recorded as
present at the translation of Archbishop Aelfheah along with Queen Emma and their son
Hardecnut. The son to be left in Denmark in 1023 may therefore have been Harold whose
mother was Aelfgiva of Northampton, Cnuts first wife.
On the other hand, all sources agree that Hardecnut was the royal son present at the time
for the battle of Helgeaa. The coins left us may point in the same direction. Hardecnut became
king of Denmark a few years later, and it is therefore not surprising to find coins minted in his
name in Denmark. However, it is worth considering that C.J. Becker suggested a date of 1030
at the latest and possibly some few years earlier for a group of coins with a lindworm on the
adverse minted in Lund in Hardecnuts name (Becker 1988:123).
We may therefore conclude that the younger brother replaced the older during the period
10231026. The trip to Denmark in 1025/26 could have been due to Thorkells death. Cnut
may have wanted to appoint the new person in charge as well as to plan the succession of
royal power in England and Denmark. Since Harold had strong relations to the English nobil-
ity through his mother, Hardecnut may have been left with Denmark.
Fagrskinna and Snorre seem to point to the fact that Cnut was in England when he heard
about Olavs attack on Denmark. They both specify that Cnut went with a big fleet of ships
through the Limfjord to pursue the two Scandinavian kings round the eastern part of Skaane
and Smaaland, and they both quote the memorial drpa composed for Cnut by the scald Sigh-
vatr. However, the poem itself does not indicate that Cnut was in England while getting the
news:

Knutr was under the heavens []

Who heard from the east of peace, princes son
Piercing-eyed, of the Danes.
From the west the wood, wake shining, glided,
Bearing the adversary
Of Adalrdr from there. (Finlay 2004:147)

The sources lack information about Cnuts trip to Rome, though Fagrskinna mentions it but
dates it wrongly to the year 1031, same as the Anglo-Saxon chronicle. This event was
evidently not important in the collective memory of the Norwegians and Icelanders of the 13th
century and they had no reasons to research on Cnuts specific point of departure for the trip
leading to Helgeaa. Not aware of the trip to Rome and taking into account that the fleet came
through the Limfjord, they assumed that Cnut came from England.

Regno et vita privare

Snorres account of the events is basically supported by other Norse sources, but the chronol-
ogy in Snorres Heimskringla is not accepted by historians. This is solely because of a single
sentence in a letter which Cnut wrote to the English people while in Rome.
Independent sources confirm that Cnut participated in the crowning of Emperor Conrad II
in Rome. The Anglo-Saxon Chronicle dates like Florence of Worcester (and like Fagrskinna)
the ceremony to 1031 but from continental sources there is no doubt about the date, Easter
Sunday March 26th 1027. Cnut had a prominent role at the ceremony as he and the king of

539
Burgundy, Rudolph III, escorted the newly crowned emperor back to the celebration dinner
after the solemn ritual in the church.
This is undoubtedly the event described in Cnuts letter to the English people. The letter is
not dated but Cnut tells about his travel to Rome where he has met the pope, the emperor
Conrad and many other princes. A single passage of that letter has been decisive for histori-
ans:

Ego itaque uobis notum fieri uolo, quod eadem uia qua exiui regrediens, Danemarciam eo,
pacem et firmum pactum omnium Danorum consilio cum eis gentibus et populis compositurus,
qui nos et regno et uita priuare, si eis possibile esset, uolebant, sed non poterant, Deo scilicet
uirtutem eorum destruente, qui nos sua benigna pietate in regno et honore conseruet omniumque
inimicorum ostrorum potentiam et fortitudinem deinceps dissipet et adnichilet. (Diplomatarium
Norvegicum)

This manifest declaration of somebody who wished to bereave Cnut of power as well as of
life has since Johannes Steenstrup been linked closely to the battle of Helgeaa, since this is the
only known military encounter involving Cnut during this period. The sentence accentuated
above is understood as a reference to a specific and violent meeting, but the words are rather
vague. We therefore have to interpret the meaning.
Regno et vita privare is not a commonly used clich. However, a similar expression is to
be found in the Book of Ester, chapter 16. In the modern version of the Bible, the Book of
Ester contains only 10 chapters. A version directly translated around A.D. 400 from the Greek
Septuagint, called The Vulgate, presented 16 chapters and was recognised by the church as
the official Latin version. Only following the Reformation and Luthers aversion of Esters
Book the last six chapters were deleted.
The Book of Ester tells us about the Persian king, Ahasverus, who made Ester queen. Ester
was Jewish. Her guardian, Mordechai, prevented an assault on the king. But Haman, who had
a trusted position at Ahasverus court, was offended by Mordechai who refused to kneel for
him. He therefore sought the life of all Jews, and particularly of Mordechai, to revenge him-
self. Ester revealed Hamans intentions to the king and Haman was hanged for his crime.
Chapter 16 quotes a copy of a letter written by Ahasverus about the Jews and about the
plot planned by Haman against the Jews and the kings wife Ester:

qui in tantum arroganti tumorem sublatus est, ut regno privare nos niteretur et spiritu.

The king had not been subject to an attack in person, and neither had he participated in some
sort of conflict. The expression is used to describe a situation where a person, because of his
greed for power, abuses the power delegated to him from the king and thereby disregards an
obligation. Since the Vulgate was recognized by Cnuts contemporaries, the term regno pri-
vare nos niteretur et spiritu was well-known in the context mentioned. It may therefore have
made sense to use the expression about an attack on Denmark in the kings absence. In Cnuts
letter to the English people, the expression is followed by the sentence:

Si eis possible esset, uolebant, sed non poterant, Deo sciliset uirtutem eorum destruente.

This declaration does not necessarily refer to the Battle of Helgeaa but could point to the har-
assment of the enemy which had not (yet) defeated the Danish defence.
The expression Vita et regno privare is also found in an English charter, S 406, which is
dated to 930. This document has been kept in the archive of Worcester. Athelstan donates
land to Worcester Minster and refers to a trophy gained by king Anlaf of the Normans

540
 tropheum ex Anolafo rege Norannorum. qui me vita et regno privare disponit possim armis
superando adipisci (Barrow 1996)

The document is not genuine but was written in the 1090s with the purpose of documenting
the right of the monks to their possessions.
Cnuts letter from Rome has survived only in a Latin version in Florence of Worcesters
chronicle. M. K. Lawson expresses some reluctance in accepting the letter word by word.

It could have been written by its bearer, given by Florence as Abbott Lyfing of Tavistock, who
eventually became bishop of Worcester, for this would further explain why Florence had a copy.
(Lawson 1993:64)

And in the beginning of the letter Cnut claims to be king of Norway as well as of part of Swe-
den. This prediction of a future victory may be a verbal underlining of the claim on Norway
as Peter Sawyer remarks. (Sawyer 1994: 19) It may also be a later adjustment of the wordings
of the original letter. We should therefore be prepared to accept the same for certain phrases
within the letter.

Rome
Cnuts letter to the English people has made the historians insist on dating the battle of Hel-
geaa to the time before the coronation of Conrad II. The above mentioned arguments raise the
question whether this dating is valid against other sources solely because of the letter itself. If
the text is merely a reaction to information about the attack of Olav and Anund Jakob then we
have to suppose that Cnut knew of the harassments of the Swedish and Norwegian troops. We
cannot document the access of such knowledge in Rome, but Snorre tells about possible ways
in which information could reach King Cnut: Erling and his sons left Norway to go to Cnut in
England in the early spring of 1027. Olav watches them when passing Jaeren on his way
south, according to Snorre. And ahead of that, Stein Skaftason had left Norway according to
Snorre:

Tidlig p vren drog han vestover til England og s til kong Knut den mektige og ble hos ham
lenge og var velsett der. (Sturluson 1979:360)

These two contacts between Norway and England document departures from Norway before
Olav sat sails for Denmark but after it may have been known that Olav sent out a call for the
levy.
No one has explained why Olav chose that particular moment supposedly in 1026 to at-
tack Denmark, and historians has just accepted the fact that Cnut, according to the general
assumption, decided to travel abroad in a time of crisis. If Olav attacked Denmark after Cnut
left for Rome it makes sense. It also explains Cnuts own description of the route he chose on
his way home:

And therefore I wish to make known to you, that, returning by the same way that I went, I am
going to Denmark, to conclude with the counsel of all Danes peace and a firm treaty with those
nations and peoples who wished, if it had been possible for them, to deprive us both of the
kingdom and of life, but could not, since God indeed destroyed their strength. (Lund 1994:38)

We know very little of his actual choice of route. But we do know that the encomiast, who
wrote the appraisal of Queen Emma, saw Cnut in the monastery of St. Omer when he was on

541
his pilgrimage to Rome. (Campbell 1949:37) Knytlingesaga underlines this fact too (gidius
1977:33). Cnut may have started the trip in Flanders and passed Burgundy on his way south.
He is generally thought to have departed Ribe to reach Flanders, and from there taking off
for Rome. That would have been a detour unless he sailed to Flanders and left his ships there.
We dont know whether Cnut originally departed from England or Denmark, but if he sailed
to Flanders from any of the two places he may have had to fetch his ships there on his way
home. His letter tells us therefore nothing specific about where he started his travels.
A remark of William of Malmesbury points in either direction as well. William explains
that Lyfing was bishop of Crediton after having been abbot in Tavistock.

[] (he) had attached himself to his retinue on his journey to Rome. Even when Cnuts busi-
ness in Rome was concluded and the king had hurried off to Denmark by land, Lyfing sailed to
England carrying the kings letters and to execute his command. Before Cnut himself arrived in
England, Lyfing had wisely and skilfully completed all the tasks laid upon him by the king
[] (William:134)

This statement confirms that Lyfing had been in Denmark with Cnut, but it may not have
been immediately before the trip to Rome. However, William stresses the fact that Cnut hur-
ried off to Denmark, while the abbot was sent home to England with certain tasks. This may
indicate that Cnut received information which urged him to go to Denmark as quick as possi-
ble.
As mentioned earlier Fagrskinna has quite another order of the events. Cnuts trip to Rome
is dated 1031 and thereby after the battle of Helgeaa, the ousting of Olav, his return and death
in the battle of Stiklestad. It is not until this time that Ulf jarl deceived Cnut who let him kill
in return. And straight after that, Cnut went west to England and never came back to Denmark
but he did go to Rome, says Fagrskinna (Finlay, 2004:164):

King Knutr set out by sea from England on a pilgrimage to the Holy See

Fagrskinna documents the story by some verses from Sighvatr, who is rather more modest:

Desire to journey
staff in hand,
came to the king
who thought of war;
Dear to the emperor,
close to Peter,
the road to Rome quickly
the ruler shortened (Finlay, 2004: 165)

Sighvatr suggests that the king had a future war in mind. Another of Sighvatrs verse found in
Heimskringla lets the king express a similar intention. Sighvatr came from Rouen and headed
for Norway. For some reason he looked up Cnut to gain permission to go home. Sighvatr
finds himself outside a locked door behind which Cnut is planning a military action. Sighvatr
consider himself a faithful follower of Olav but realises that the Norwegian king are to face an
immense power. Snorre quotes Sighvatr with the words:

Da Sigvat merket at kong Knut rustet seg til hrferdmot kong Olav, og han fikk vite hvor stor
styrke kong Knut hadde, da kvad Sigvat:

Den dsle Knut og Hkon

542
 vil prve velte Olav,
alt har han som fr ute;
for kongens dd jeg engstes.
M verneren leve, enda
ei Knut og jarlene vil det.
Godt bliver et mte p fjellet
frst nr du slapp fra det. (Sturluson: 376)

Olav had definitely not travelled out in the preceding 13 years, but we know for sure that
Cnut did. Snorre doesnt reflect on this information. He does not mention anything about
Cnuts journey to Rome and generally, Snorre has a rather uncomplicated view on where to
find Cnut: When not in Denmark, he stays in England. It is therefore not striking in the saga
that Sighvatr goes to England to find Cnut but looking at a map, it seems peculiar that a trav-
eller should go to England in order to reach Norway from Rouen. Naturally, one would follow
the continental coastline up along Jutland and from there to cross north to Norway. Even
though the ships could manage the open sea, there were advantages by staying close to land.
A rational mind as Snorres may have known about Sighvatr visiting Cnut. In Kntsdrpa
the scald tells about a meeting with Cnut before the battle and it is therefore necessary for
Snorre to find a way for the king and the scald to meet. Logically Snorre lets Sighvatr look up
Cnut in England. If we are not bound to presume that Cnut is in England, the two Scandinavi-
ans may have crossed each other somewhere on their way the king hurrying home to meet
the enemy and the scald afraid of the war to come.

Conclusion
Contrary to what has hitherto been believed, the letter from Rome is not an authority to define
the Terminus Ante Quem as it is usually assumed. The decisive sentence in the letter can not
for certain be said to address an already fought military battle, and related passages in two
other documents seem to describe situations characterised by a mere threat.
The Norse sources on the other hand are very certain concerning the dating of the battle to
late summer or fall in 1027, i.e. after Cnuts trip to Rome. This chronology would explain
why Cnut went back to England via Denmark. If he heard about the attack while in Rome he
would of course feel an urge to go to Denmark. Lyfing was given a letter to the English peo-
ple to explain the change of plans. He has had other tasks to do as well, and they may have
included orders for the army to go to Denmark and meet Cnut at a suitable place. According
to Snorre, two fractions of the Danish-English army actually met in Limfjorden. That would
be the natural way to enter Danish waters no matter if you came from England or Flanders.
Such an explanation also eliminates the fact that Cnut should have left Denmark for a corona-
tion in Rome after a battle, that ended almost unsettled and with half of his enemies still vig-
orous.
Accepting the year 1027 for the battle of Helgeaa means, that Cnut hastened to Denmark
when he heard about the threat. Olav was forced to leave his ships after the battle, and he left
the battlefield as the undisputed looser. Cnut went to England in the winter 1027/28, but left
the army guarding Oeresund and thereby prevented Olav from regaining the ships. Upon real-
izing, that he had lost not only the ships but also the confidence of the Norse chieftains, Olav
left for Russia in the winter. Next spring Cnut was back in Denmark and set out for Norway,
where he was hailed as the victor of Helgeaa and king of Norway.

Bibliography

543
Barrow, Julia: The Chronology of Forgery Production at Worcester from c. 1000 to the early twelfth
century; In Saint Wulfstan and his world, (1996) ed. by J. Barrow and N.P.Brooks
Becker, C. J.: Danske Mnter som historisk kildemateriale i 1000-tallet; In: Festskrift til Olaf Olsen
paa 60-aarsdagen, 1988
Campbell, Alistair (ed.): Encomium Emmae Reginae, London 1949
Diplomatarium Norvegicum
Einarsdttir, lafia: Studier i Kronologisk Metode i tidlig islandsk historieskrivning. CWK Gleerup
1964
Finlay, Alison (ed.): Fagrskinna, A catalogue of the kings of Norway, Brill Leiden Boston, 2004
Florence of Worcesters Krnike; (ed) by E. Albrectsen, Odense 1982
Knytlingesaga; translated by Jens Peter gidius, Kbenhavn 1977
Lawson, M. K.: Cnut, London 1993
Lund, Niels: Cnuts Danish kingdom; In: The Reign of Cnut, ed. Alexander Rumble, London 1994, p.
2742
Moberg, Ove: Knut den stores motstndere i slaget vid Helge, Scandia 51, p. 717
Moberg, Ove: Olav Haraldson, Knut den store och Sverige. Lund 1941
Sawyer, Peter: Cnuts Scandinavian Empire; In: (ed.) Alexander Rumble: The Reign of Cnut, London
1994, p. 1023.
Steenstrup, Johannes: Normannerne, Vol. 3. Reprint 1972, Vol.3, p. 350 ff
Sturluson, Snorre: Norges kongesagaer, ed. Anne Holtsmark og Didrik Arup Seip, 1979
The Vulgate, Latin Bible, e.g. www.newadvent.org/bible
Whitelock, Dorothy (ed.): English Historical Documents c. 5001042, 1979
William of Malmesbury: The deeds of the bishops of England; (ed.) D. Preest, 2002
Wolfram, Herwig: Conrad II 9901039, Pennsylvania 2006

544
 Frithjof and Rde Orm: Two Swedish Viking impersonations
Hans Kuhn, Australian National University
It took me a while to get approval for talking about such a modern topic. Old Norse scholars,
like Classicists, are comfortable with the limited number of texts they deal with and, with
honourable exceptions, a little reluctant to extend the range of their field of study.
Swedes, unlike Norwegians and Icelanders, do not figure large in the sagas, despite their
Viking raids and settlements, especially in the East. Yet the Swedes were eager, over the cen-
turies, to claim these texts, or at least the activities and attitudes they reflect, as part of their
heritage, as Anna Wallette recently showed in her examination of the treatment of the Viking
age in popular histories of Sweden from the 17th to the 20th centuries (Wallette 2004). They
also inspired writers, artists and musicians, especially with the coming of Romanticism, when
the Dark Ages were no longer seen as dark but as unspoiled by the march of civilisation and
foreign influences. The founding of Gtiska frbundet in 1811 marked its arrival in Sweden,
and in the first number of its periodical Iduna, Geijer published the poems Vikingen and
Odalbonden, where the adventurous, roaming Viking was posited as a second national arche-
type next to the steady and tradition-bound free farmer (cp. Kuhn 1983). More than 20 years
ago, I sketched the treatment of Hervarar saga by the Romantic writers Grundtvig, Hertz and
Ling (Kuhn 1986). But while those works never reached beyond their countries of origin,
the two works discussed here, Tegnrs Frithiofs saga, completed 1825, and Bengtssons
Rde Orm, published 194145, found an international audience and echo. Tegnr was, like his
Danish colleague Oehlenschlger, whom he was to crown King of Nordic poets in Lund
Cathedral in 1829, an eclectic, and Oehlenschlgers Helge of 1814 may well have inspired
him as well as a desire to do better than his countryman Ling in his epic poem Asarne and his
tragedies. He knew the subject matter of the mythic-heroic Frijfs sags ins frkna since his
boyhood, when he read Birners Nordiska kmpadater of 1737, the first Swedish collection
of translated sagas. His verse treatment in 24 cantos is classically inspired, but unlike Homer,
he uses a different metre in every canto. It is very much a story on the folktale pattern where
the protagonist, a loner on the move, ends up with the princess, here called Ingeborg, and the
kingdom, here located in Ringerike in Norway; his enemies are Ingeborgs two unworthy and
treacherous brothers who do their best to get him out of the way, including by weather magic.
It also follows the folktale pattern by including, when the hero already seems to have reached
his aim, a reversal that sends him on another round of adventures. The fact that Ingeborg is
married off to a much older suitor turns it into a triangle drama with echoes of Tristan and
Isolde.
One could see why this story should be particularly appealing to a 19th c. audience. The
love between Frithiof and Ingeborg, who grow up as foster children, unifies the rambling tale;
the villainy of Helge and Halfdan is balanced by the support of Frithiofs friend Bjrn and the
magnaminity of old King Ring, Ingeborgs husband. So, Frithiofs saga combines the ele-
ments to make it effective as an adventure story, a love story and a morally uplifting tale: the
good and patient prevail in the end, and they can expect happiness ever after. And while the
fairly short saga, which is studded with with 35 (rather mediocre) lausavsur, only gives a
fairly rough outline, Tegnr expands both the descriptive, the emotional and the reflective
elements. He is at pains to show Frithiof as a dashing and daring, but also, in contrast to
Bjrn, as a moral Viking. In Canto XV, Vikingabalk, he draws up rules for a Vikings proper
behaviour; the following cantos, where, after reflection, Frithiof abstains from taking advan-
tage of King Ring, were the first that Tegnr wrote. A religious dimension is added by
Frithiof accidentally setting fire to Balders temple, his meeting-place with Ingeborg in hap-
pier days; after that, he feels guilty, an outlaw, and is at a loss how to be reconciled with the

545
god. In a vision at his fathers grave, he is shown the new temple he has to build, and when he
first visits it, the high priest, who, in a long sermon, even anticipates the coming of Christian-
ity, shows him that he also has to seek reconciliation with humans, in his case Ingeborgs sur-
viving brother Halfdan. All of this is Tegnrs addition, and in passing, as later on in Canto I,
the reader gets a refresher course in Northern mythology and skaldic diction; this may have
been part of the works attraction. In some respects, Tegnr renders the story more archaic
than it is in the saga. There, King Ring dies a natural death, while Tegnr lets him end his life
with his spear to escape a straw death, and then he devotes a whole canto to Rings arrival in
sgarur, again a pure fantasy. In a commentary, written 14 years later, he said that he had
tried to give his hero ngot individuelt nordiskt, det lefnadsfriska, trotsiga, fvermodiga, som
hr eller tminstone fordom hrde till nationallynnet (Tegnr 1872, p. 281). Tegnrs
Frithiof shares with his author an open, enterprising and spontaneous disposition but also
brooding and bouts of melancholy. Tegnr could not be aware of the likely oriental origin of
the story (Gould 1923; Mundt 1991), and the heros name, implying a compulsive fighter
(Peace-Thief), did not prevent him from turning Frithiof into an exemplary peacekeeper in
the end, the matured young man of a Bildungsroman. In literary style, being visually concrete,
sublime, and concise in expression, Schiller was the example Tegnr followed.
He thought Frithiofs saga untranslatable; yet the year after the original was published in its
entirety, three different German translations appeared, and many others were to follow. There
were four English translations in the 1830s, one with the subtitle A Scandinavian legend of
royal love, and Andrew Wawn has written about the cult of Stalwart Frith-thjof in Victorian
Britain (Wawn 1994). There were three Danish and at least one French translation. Selma
Lagerlf wrote a libretto for the organist Elfrida Andre, but the resulting opera was only per-
fomed privately, in 1899 (Wallette 2004, p. 265). By far the most influential musical setting
was Tolf snger ur Frithiofs Saga by the Finnish clarinet virtuoso Bernhard Crusell; they
spread to Denmark and Germany, and in 19th c. Danish songbooks I have often seen these
melodies used as timbres for other songs, which shows how well-known they were. In 1828,
Hedda Wrangel, P.C. Boman and Karoline Ridderstolpe published music for other parts of
Frithiofs saga, and in the 1830s, the German composers Joseph Panny and Friedrich Silcher
wrote music for parts of it, followed in 1866 by Max Bruch (for Crusell and Bruch, see Nor-
denfors 2008). Visual artists, too, responded to the work in great numbers, starting with Hugo
Hamilton in 1828; in Adolph Tidemans Frithiofs avsked frn Ingeborg from 1836, he is a
very Romantic-looking hero saying goodbye to a young lady dressed in Empire fashion. The
best-known illustrations were August Malmstrms, published in book form in 1868 and re-
printed many times both in Germany and Sweden, last in Niloes facsimile edition of 1984.
Purely in size, none was a match for the 26 meter high statue by Max Unger the Emperor
Wilhelm II, a great fan of the Scandinavian North, had erected in Sognefjord in 1913 (Grimm
1997; for other artists, see Mjberg 1967f.).
There are a number of features that Tegnr and Frans G. Bengtsson (18941954), the au-
thor of Rde Orm, share. Lund was their spiritual and, for long periods, also their actual
home, and in that university environment scholarship, wit and the art of repartee were refined.
They both knew how to express themselves concisely, elegantly and in memorable form, and
they both could be provocative and paradoxical. They both grew up in the countryside;
Bengtssons farmer was a steward on the country estate Rssjholm in Skne, and Frans
worked there with cattle after he had been diagnosed with a serious kidney complaint by mili-
tary doctors. When he went back to Lund in 1920, he was seen in cafs rather than lecture
rooms; he had some contact with Fredrik Bk and his students but never took any course in
history, although he turned into a gentleman historian of distinction as shown by this two-
volume biography of King Charles XII, also the subject of one of Tegnrs most memorable
poems. He started as a poet but made his mark first as a brilliant essayist where he could sug-

546
gestively evoke figures or incidents from the past that charmed him and scourge aspects of the
present that irritated him such as psychology and social realism and political commitment in
imaginative literature. He had a fabulous memory and could recite Icelandic sagas practically
by heart (Holmberg 1963, p. 204); no wonder, since they had the features he valued in narra-
tives, action and dialogue, not interior monologues or stream of consciousness (Bengtsson
1951). He claimed that the idea for Rde Orm was born from thinking about the evocative
bynames of medieval Scandinavian leaders, concentrating on Harald Bluetooth of Denmark,
and his contemporary Almansur in Southern Spain (Bengtsson 1948); to connect the two, he
needed an adventurous Viking, a young redhead from Kullen in Skne that he named Rde
Orm. So this was not the re-telling of an existing saga but the birth of a new one fed by what
the sagas told about Vikings (cp. Bengtsson 1931); the one expanded historical element used
is King Ethelred the Unready and the Battle of Maldon (see Ploman 1993). Eric Linklater,
whom Bengtsson translated, may well have been an inspiration.
Orm shares with Frithiof his fairy-tale hero character whose story begins when he leaves
home; he is not embedded in a genealogical tree as the protagonists of the family sagas are.
And again, his is a rags-to-riches story; his adventures in the West, which fill the first book,
land him for a while on the lowest rung of the social ladder, as a galley slave, from where he
rises, via service as a guard at Almansurs court, through bravery, daring and resourcefulness,
ending up at the Danish court in Jutland and getting his princess, Harald Bluetooths daughter
Ylva. Like Frithiof, he has a somewhat naive but trusty friend, here called Toke. But unlike
Frithiof, he is not a moral Viking; he measures happiness in material terms (Egill Skalla-
grimsson may have provided a model), and in Book 2, when he has settled as a family man in
the Ginge area, he sets out to the East to bring back the treasure of the Bulgars that his
brother had hidden in the River Dnjepr. His encouragement of the Christian missionary Willi-
bald is also partly motivated by economic considerations. Orm has matured, but not so much
morally as into what one reviewer called en myndig betnksam storfretagare, a self-
assured, wary big businessman (Tykesson 1948), and he lacks the tragic heroism that can be
glimpsed in skaldic verses and in some of the family sagas.
Bengtsson, writing during the Second World War, did not glorify war; his description of
fighting and killing was a great deal more realistic than Tegnrs. But he also recognised that
for the individuals involved, fighting was not just a test of strength and character but part of
an exciting adventure. It must have been a reaction to the high seriousness of much fiction in
those years that he spun a tale that was full of drastic incidents and humorous dialogue; by
calling it a rvarroman, he indicated that his main aim was to entertain. Escapism was one
of the negative characterisation the work received, and it was also claimed that it exaggerated
the grotesque elements that can be found in saga literature, and that the sober factuality of
action and dialogue was undermined, more strongly in the second book, by a spirit of parody
that betrayed the spex tradition that was especially strong in Lund; the illustrations by Gunnar
Brusewitz reinforced a humorous view. But it did not take long before the book, as Frithiofs
saga in earlier times, became a much-loved school text, a position evidenced by the publica-
tion of a 117-page Studiehandledning (Larsson 1979).
No wonder the high priests of literature gave it only grudging recognition (Lgn i Lund
was Victor Svanbergs characterisation according to Gran Hgg 1996), while the public,
both in Sweden and abroad, took it to heart. The publisher risked printing only 4000 copies
when when it first came out in 1941; by the mid-1960s, around 330000 copies had been sold,
and Bengtssons collected works were re-issued for the 100th anniversary of his birth in 1994.
German and English translations of Part I appeared already in 1943, both parts in 1951 in
Germany and 1954 in England; The Long Ships sold about 42,000 copies within one year. It
was probably on the back of their success that the two volumes of Karl XIIs levnad, the only
other work of Bengtssons that could be called a novel, were translated, 1957 in German,

547
1960 in English. According to Wikipedia, Rde Orm has been translated into 22 languages.
Yet the most powerful medium for spreading knowledge about the book was one that did not
exist in Tegnrs time: the film. It is true that Richard Widmark treated Part I freely and rather
selectively in his The Long Ships of 1964, making the bell of Santiago that Orm (here called
Rolfe) and his companions steal from the Moorish court the central motif of the story and
ending with Vikings, sent by King Harald Bluetooth to recover his abducted daughter, defeat-
ing the Moors. But it can be assumed that the film significantly contributed to the success of
the book on which it was based.
Both Frithiofs saga and Rde Orm differed from the sagas by fleshing out the descriptions
a modern reader has come to expect in narrative fiction. Both works reacted against the ways
Old Norse literature and mythology had been used by contemporary writers. Tegnr, though
formally a member of Gtiska frbundet, thought it absurd to pretend that the world had not
moved on since the early Middle Ages; he reacted to writers such as Ling as Oehlenschlger,
the poetic voice of his nation as Tegnr was of his, reacted to Grundtvig in the years of the
latters asarus. But Tegnr lived in an age when it was thought to be one of literatures func-
tion to extend and improve the mind of readers; idealism was part of the cultural scene.
Bengtsson felt under no such obligation; he was a sceptical observer of human nature and
made no claim for Orm to be the embodiment of a national ancestor; if anything, he could be
seen as a typical skning. He was content to tell a story graphically. He was not the only
writer of his generation who re-created the Sweden of the Viking period in fiction. Jan
Friidegrd wrote his Viking-time novels Trgudars land (1940), Gryningsfolket (1944), Of-
ferrk (1949) during the same years; but Bengtssons Viking was, like Tegnrs in his time,
the only one to reach an international audience.

Bibliography
Bengtsson, Frans G., 1931: Silverskldarna, Stockholm. His views of Viking activities are formulated
on pp. 231241 of the 1951 edition.
Bengtsson, Frans G., 1948: Hur Rde Orm blev till. In: Modersmlslrarnas Frenings rsskrift 1948.
Pp. 2028 = Bengtsson, Folk som sjng, Stockholm 1955. Pp. 1121.
Bengtsson, Frans G., 1951: Egil Skallagrimsson p svenska. In: Silverskldarna. Stockholm. Pp. 188
201.
Gould, Ch. N., 1923: The Frijfssaga an oriental tale. In: Scandinavian Studies 7. Pp. 219250.
Grimm, Birgit, 1997: In: Wahlverwandtschaft: Skandinavien und Deutschland 1800 bis 1914. Ed.
Bengt Henningsen. Berlin/Stockholm/Oslo. Pp 108f.
Holmberg, Olle, 1963: Skratt och allvar i svensk litteratur. Stockholm.
Hgg, Gran, 1996: Den svenska litteraturhistorien. Stockholm. P. 491.
Kuhn, Hans, 1983: The Farmer and the Viking: Forms of Romantic Nationalism in 19th c. Scandina-
via. In: Romantic Nationalism in Europe. Ed. by Ch. Eade. Canberra. Pp. 81100.
Kuhn, Hans, 1986: Die Geschichte vom blutrnstigen Schwert: Bearbeitungen der Hervarar saga in
der skandinavischen Romantik. In: Festschrift Oskar Bandle. Basel/Frankfurt. Pp. 223232.
Larsson, Gunnel, 1979: Rde Orm och vikingarna. Stockholm.
Mjberg, Jran, 1967f.: Drmmen om sagatiden. Stockholm. Bengtssons use of his sources and mod-
els are described in bd. II, pp. 333343.
Mundt, Marina, 1991: Zur Adaption orientalischer Erzhlstoffe im hohen Norden. Der Fall der
Frijfs saga. In: Deutsch-nordische Begegnungen, Odense 1991. Ed. by K. Braunmller. Pp.
164176.
Nordenfors, Ola, 2008: Adaptationer av Esaias Tegnrs Frithiofs saga. In Grnser i nordisk littera-
tur. Ed. by Cl. Zilliacus. bo. Bd. I, pp. 197208.
Ploman, Lennart, 1993: Frans G. Bengtsson, Rde Orm och vikingatiden. Ed. by Lennart Ploman.
Lund. The most relevant contributions in this popular publication are by Lars-Arne Norborg (Rde
Orm och den historiska verkligheten, pp. 6982) and Jran Mjberg (Frans G. Bengtsson och den
islndska sagan, pp. 88105).
Tegnr, Esaias, 1872: Samlade Skrifter, Bd. 4, Stockholm.

548
Tykesson, Elisabeth, 1946: In: Bonniers Litterra Magasin 15, p. 61.
Wallette, Anna, 2004: Sagans svenskar. Synen p vikingatiden och de islndska sagorna under 300 r.
Malm.
Wawn, Andrew, 1994: In: Northern Antiquity. Hisarlik Press. Pp. 211254.

549
Bik hugstra verung hressfrs jfurs heyra, hv olak vs:
Sigvatrs Austrfararvsur between praise poetry and lausavsur
Ann-Drte Kyas, University of Kiel, Germany
These first words of Sigvatrs Austrfararvsur show a mixture of elements from both praise
poetry and lausavsur. Sigvatr uses the bid for a hearing (bik hugstra verung hressfrs
jfurs heyra I ask the mighty-hearted retinue of the energetic ruler to hear), which is a
common formula in praise poetry, to introduce his poem. However, he does not praise the
princes deeds, as one might expect, but he speaks about himself (hv olak vs how I en-
dured hardship), which is a common feature of lausavsur.
In this paper I want to demonstrate by means of Austrfararvsur that the distinction be-
tween praise poetry and lausavsur is sometimes not clear-cut, because some elements are
found in both genres and sometimes praise poems focus not only on the prince but also on its
poet. I will thus compare expressions of Austrfararvsur with those elements which occur in
praise poetry and lausavsur, and I will examine the function of the poet in both genres.
Finally, I want to answer the question about how to classify Austrfararvsur, that means, if
these verses are a sequence of lausavsur or if they can be regarded as a praise poem in which
the poet focusses on his voyage to Sweden. This question is intended to be a starting point for
a discussion of the classification of skaldic stanzas and whether a third category is needed
besides praise poetry and lausavsur.

550