Feinman&Marcus Archaic States
Feinman&Marcus Archaic States
Stote~
(ontents
y Cj ~.6
List of Illustrations
ix
List of Tables I xi
Preface I Xlll
Gary M. Feinman and]oyce Marcus
Introduction I 3
]oyce Marcus and Gary M. Feinman
2
The Ground Plans of Archaic States I 15
Kent V. Flannery
3
The Peaks and Valleys of Ancient States:
An Extension of the Dynamic Model I 59
]oyce Marcus
4
Scale and Social Organization:
Perspectives on the Archaic State I 95
Gary M. Feinman
viii I CONTENTS
5
Beyond Centralization:
Steps Toward a Theory of Egalitarian Behavior
in Archaic States I 135
Richard E. Blanton
6
Uruk States in Southwestern Iran I 173
Henry T. Wright
Illustrations
7
Order, Legitimacy, and Wealth in Ancient Egypt and Mesopotamia I 199
John Baines and Norman Yoffee
8
Sociocultural Complexity Without the State:
The Indus Civilization I 261
Gregory L. Possehl
2.L
2.2.
2.3.
9
Inka Strategies ofIncorporation and Governance I 293
Craig Morris
10
Warfare and Status Rivalry:
Lowland Maya and Polynesian Comparisons I 311
David Webster
References I 353
Index I 421
2.4.
2.5.
2.6.
2.7.
2.8.
2.9.
2.10.
2.1L
2.12.
2.13.
2.14.
2.15.
2.16.
2.17.
2.18.
2.19.
2.20.
2.2l.
2.22.
3.l.
3.2.
3.3.
3.4.
x I ILLUSTRATIONS
3.5.
3.6.
3.7.
3.8.
3.9.
3.10.
3.11.
3.12.
3.13.
3.14.
4.1.
4.2.
4.3.
4.4.
4.5.
4.6.
4.7.
4.8.
6.1.
6.2.
6.3.
6.4.
6.5.
7.1.
7.2.
8.1.
8.2.
8.3.
8.4.
8.5.
8.6.
9.1.
9.2.
10.1.
10.2.
10.3.
10.4.
10.5.
Tables
n e of the most challenging problems that faces contemporary archaeology concerns the operation and diversity of ancient states. Because archaeologists, historians, and other social scientists have tended to focus
on the rise and decline of early polities, questions concerning how archaic
states worked (or broke down) have been given less attention. It can be argued, however, that it is difficult to understand a state's rise or fall if we do
not comprehend what it looked like at its peak. Furthermore, can we evaluate general explanations of state origins if we do not apprehend the full
range of variation in how states operated?
We proposed this session to the School of American Research in the fall
of 1989. At that time, there had not been a major comparative seminar on
the operation and structure of ancient states for more than 15 years. That
gap in the recent literature on archaic states remains as significant now as it
was when the meeting was held in November 1992. As the cycle from conference papers to finished chapters nears completion, we believe that the articles in this book collectively take significant strides in addressing questions
of state diversity and operation, as well as in helping define key avenues for
research and discussion in the decades ahead.
The participants in this session were selected to provide both broad
geographic coverage and a range of theoretical viewpoints. We wanted to
include diverse scholars, including several who work principally with documents, others who mostly examine the archaeological record, and a third
group who studies both. In regard to archaic states, there is relatively little
xiv I PREFACE
that we all would agree on other than the deep interest that we share in the
topic. It was this passion and the wealth of experiences we brought to the
seminar table that kept our discussions going far into the night.
As editors, we wish to thank all the participants for their thoughtful presentations and papers. We also are grateful to them for responding quickly
and with good humor to Gary's bibliographic queries as we readied the
book for press. George L. Cowgill and an anonymous reviewer provided
valuable reviews of the entire volume. It is a better book as a result of their
suggestions. Linda M. Nicholas, Sandra Austin, Lane Fargher, and Steve
Germain assisted in numerous ways as we edited and prepared the manuscripts for publication. We also appreciate the great care and comfort generously provided by Doug Schwartz and the entire School of American
Research staff in Santa Fe, as well as the production and editorial support
given us by Joan O'Donnell and the staff of the SAR Press.
We dedicate this volume to Robert McCormick Adams, a pioneer in the
comparative analysis of Old and New World states.
Anhoi( Stotes
1
Introdu(1:ion
IDYU MUCUS AND GUY M. H1HMAH
uring the past 25 years, ancient states have been the subject of numerous
volumes and conferences around the world. Two dominant themes of
this work have been the "origins" and "collapses" of such states.'
Believing that many other themes-including the operation and behavior of ancient states through time-had not been exhaustively treated, we
assembled ten scholars at the School of American Research between October 30 and November 6, 1992. One of our goals was to produce a volume
that could be used in conjunction with previously published books on the
origins and collapses of ancient states. We hoped to do this by providing new
comparative studies of early states in the Old and New Worlds, emphasizing similarities and differences. We also hoped to shed new light on a number of theoretical issues. Included among those issues are the operation and
structure of ancient states, their scale and territorial extent, the nature of
state-level political and economic institutions, the diversity of ancient states,
and the value of their systematic comparison. These topics are ones we share
with other scholars, such as Southall (1991: 77), who recently wrote, "I am
now more interested in how states work, what basic similarities in their
working (processes) can be found (a) at different eras of world history, (b) at
different stages of their development, (c) in different climates and ecologies,
and (d) in different historical streams."
Given that most of the seminar participants were archaeologists, we decided to focus on archaic states, those that arose early in the history of their
particular world region and were characterized by class-endogamous social
strata with royal families, major and minor nobles, and commoners. Most of
the states we consider were first-, second-, or third-generation states in their
region. We have not included more recent states with democratic institutions, presidents, parliaments, and global market economies.
Among the themes discussed in this volume are the scale, size, and organization of ancient states (Gary M. Feinman); an evaluation of state typologies and corporate political economy (Richard E. Blanton); a comparative
study of the ground plans of buildings in archaic states, aimed at helping the
archaeologist determine if a state is in fact present and whether certain institutions and groups of personnel have emerged (Kent V. Flannery); the
strategies ofstate expansion, the incorporation of autonomous polities in the
state, and their subsequent governance (Joyce Marcus, Craig Morris, Henry
T. Wright); evidence for warfare and status rivalry in chiefdoms and states
(David L. Webster); evidence for order, legitimacy, and wealth in ancient
states (John Baines and Norman Yoffee); and the political and social complexity of states (Gregory L. Possehl). World regions represented include the
Near East, India and Pakistan, Egypt, Mesoamerica, and the Andes.
Another theme discussed by various participants at the seminar was the
relationship between "state" and "civilization." The former was seen as a
political or governmental unit, while the latter was seen as the cultural
sphere associated with that political or governmental unit. In this volume,
"the state" is not considered to be isomorphic with "civilization" in the Indus region (Possehl), the Near East (Yoffee and Baines), or the Maya region
(Marcus). Even for Egypt, it is only during the politically more centralized
periods called "kingdoms" that there was a reasonably close areal fit between
"state" and "civilization."
Such historical differences were a central theme in seminar discussions.
We pondered why in some regions several states arose that shared a single
ethos of civilization, while in other areas the boundaries of a state and its
civilization were roughly coterminous. Although we could not explain this
historical variation, all seminar participants agreed that this diversity would
be an interesting question to examine in the future.
TERMINOLOGY
One of the first topics discussed at the seminar was, "What is an archaic
state?" Although definitions varied among participants, many believed that
in contrast to modem nation-states, archaic states were societies with (minimally) two class-endogamous strata (a professional ruling class and a commoner class) and a government that was both highly centralized and internally
specialized. Ancient states were regarded as having more power than the
rank societies that preceded them, particularly in the areas of waging war,
INTRODUCTION I 5
INTRODUCTION I 7
while denying its citizens the use of personal, individual force; and (7) the
establishment of governmental laws and the ability to enforce them. Although not all participants concurred with each of these criteria, a degree of
support generally was found for each.
Participants also agreed that ancient states were dynamic political systems
that changed throughout their specific historical trajectories. Each state is,
therefore, different in some respects from other states, but early states as a
group do display similarities. In this volume some authors tend to emphasize the similarities among states, while others emphasize both similarities
and differences.
SOME POINTS OF AGREEMENT
Although the seminar participants chose to disagree on a number of topics,
many felt that the use of two terms-"segmentary states" and "citystates"-was problematic. The first is an oxymoron, resulting from the illadvised phrase originally applied to a nonstate society. The second is a term
coined for the Greek polis and indiscriminately applied to dozens of other
societies that are not similar.
Let us consider "segmentary states" first. The concept can be traced back
to African Political Systems, a volume edited by Fortes and Evans-Pritchard
(1940). Fortes and Evans-Pritchard contrasted noncentralized lineage groups
(acephalous societies) with centralized hierarchical governments (unitary
states). They noted that stateless societies displayed many structurally equivalent segments, and "if one segment defeats another it does not attempt to
establish political dominance over it; in the absence of an administrative machinery there is, in fact, no means by which it could do so" (Fortes and
Evans-Pritchard 1940: 14).
The African ethnologist Aidan Southall (1956) was unable to fit the Alur
into Fortes and Evans-Pritchard's categories of"unitary state" or "acephalous
society," so he created an intermediate category he called a "segmentary
state." Southall thought that this concept would be useful in describing the
interaction between the Alur and their political economy; eventually other
scholars (such as Burton Stein [1977] in his historical analysis of the Chola
"empire" of medieval south India and Richard Fox [1977] in his account of
the Rajput states of north India) adopted the segmentary state concept in
their studies.
Unfortunately, however, the phrase "segmentary state" misled many
scholars because they took it to refer to a "state-level" society when the Alur
are no more than a rank society at best. Eventually Southall (1991 :91) himselfrecanted: "Although I called it [the Alur polity] a segmentary state, it was
hardly a state, hardly a two class society, and it could not prevent secession."
INTRODUCTION I 9
INTRODUCTION I 11
1983a; Folan, Marcus, and Miller 1995; Gledhill, Bender, and Larsen 1988;
Marcus 1973, 1992a, 1993a; Pattee 1973; Service 1975; Willey 1991;
Wright 1977, 1978). This broad menu contradicts any notion that evolutionary categories like "state" are "rigid" or "mask diversity."
Seminar participants did little to reduce the variety ofstate types in their
discussions. One source of diversity is the fact that states may have multiple
hierarchies with nonisomorphic or only partially overlapping functions (see
Marcus 1983a, 1983b). For example, a given site may be the political capital but not the economic or religious capital for its region. A second-tier political center may be a first-tier religious center; a fourth-tier politicalcenter
might be a first-tier craft production center.
To deal with such situations, Crurnley (1987) and others (Ehrenreich,
Crurnley, and Levy 1995) have embraced the concept of "heterarchy"-a
term that goes back to at least 1945, when it was used to describe the relationships among parts ofthe human brain. Then McCulloch (1945) used the
term to assert that the brain and its functions were not hierarchically structured. Heterarchy, however, is not the opposite ofhierarchy; it simply means
that different functions can exist in a system without their arrangement being hierarchical. Archaic states were both heterarchical and hierarchical.
Another comparative theme touched on during seminar discussions was
that even those aspects of the archaic state that are hierarchical can be
volatile. Although it is common for archaeologists to depict ancient states as
powerful and centralized, many rulers, in fact, had a great deal of trouble
controlling their subordinates. In this volume, Blanton as well as Baines and
Yoffee discuss the diverse strategies and arrangements that were designed to
link principal rulers with more local elites. This difficult dynamic is certainly
one reason that ancient states seem to oscillate between periods of successful
centralization and loss of control. As noted by Fortes and Evans-Pritchard
(1940: 11), it was difficult for rulers to achieve a perfect balance between
centralized authority and regional autonomy. If a king abused his powers,
subordinate leaders were liable to secede or lead a revolt against him. If a
subordinate leader got too powerful and independent, the central authority
might call on other subordinate leaders to help suppress him. Fortes and
Evans-Pritchard (1940: 11) note that kings often tried to buttress their own
authority by playing subordinates off against each other. Their ultimate goal
was to keep the power of the ruler and that of his subordinates in balance.
That balance could not be maintained if all subordinates allied against the
ruler. When they acted collectively, subordinates could pose a real threat to
the ruler, as we see in the Maya and Egyptian cases discussed in this volume.
Yet seminar participants also recognized that power in archaic states was
not always as tightly focused on individual rulers as was the case for the Maya
and in Egypt. Cross-cultural variation in the strategies ofstate rulership, and
INTRODUCTION 1 13
the relationship between these differences and variability in economic production and exchange, was often discussed. Likewise, we considered the nature of the relationship between the spatial extent of archaic states and the
degree of centralization. Many participants agreed in a general sense with
Southall's (1988: 81) observation that, other things being equal, "it is easier
for a small than for a very large state to centralize."
Finally, we discussed a number of topics at the seminar that we could not
resolve. One was the question of whether or not there were fundamental
differences between primary or first-generation states and the secondgeneration states that followed them. On the most specific level, one can see
differences ofdetail. Some first-generation states had palaces but lacked standardized temples; others had standardized temples but lacked palaces. The
second-generation states in such areas often had both, suggesting to some of
the participants a level of "maturity" in state institutions that was not found
in primary states. Other participants emphasized the organizational differences between specific states, however, and did not see these distinctions as
necessarily developmental in nature. In this organizational material, no
member of the seminar saw a universal pattern or sequence from which we
could draw unilineal conclusions. Furthermore, all participants recognized
that in any given region the degree of political integration did not increase
steadily or uniformly through time.
One reason such diachronic questions are hard to resolve is that our current chronologies are too coarse-grained to allow us to see the order in
which various institutions appeared. Archaeologists have "gross chronologies" based on ceramic phases and C-14 dates; "medium-grained chronologies" based on the stratigraphic superimposition of floors and buildings; and
(in some regions) "fine-grained chronologies" based on texts with calendric
dates. Often each of the three chronologies suggests a different political scenario for the same region, and even when all are present the three rarely fit
together as snugly as we might wish (Smith 1992).
Parenthetically, we might add that chronology can obscure the transition
from chiefdom to pristine state. At the seminar, Henry Wright pointed out
that immediately prior to state formation in southwest Iran, one ofthe area's
largest chiefdoms collapsed, creating a competitive situation that may have
led to consolidation. If the same process happened in the Valley of Oaxaca,
he argued, we cannot see it because the process of state formation was so
much more rapid there than in southwest Iran. This is probably another way
of saying that, for certain regions, we need a much more fine-grained
chronology to detect whether (and the degree to which) general processes
do indeed exist.
One impression we all took away from the seminar is that archaic states
were a lot more fragile and internally diverse than the archaeological litera-
ture would lead one to believe. Born out of tremendous effort, they also required a substantial effort to hold together. Their constituent provinces were
always more stable than the larger polities of which they were part. In virtually every region of the world, provinces kept escaping and having to be
reclaimed. Given the number of attempts at consolidation that failed, one
can only marvel at the states that lasted for half a millennium or more.
DIRECTIONS FOR FUTURE RESEARCH
Seminar participants concurred that archaic states can only be understood if
we study them at a number of scales: the household, the site, the region, and
the macroscale. In many of the regions discussed here, we are just beginning
to achieve the outlines of such a multiscalar perspective that has taken as
much as 50 years of prior research. Clearly, the research goals and questions
defined here are long term and multigenerational in nature. Meeting and resolving those aims provides a key challenge for students in the present and
the future.
Almost everyone at the seminar was struck by how much more we know
about those ancient states for which extensive ethnohistoric or historic texts
are available. There are things Yoffee can say about Babylonian states that
simply remain unknown for the Uruk state. There are things Morris can say
about the Inka state that remain unknown for the Moche and Wari states.
In such situations, there is a real temptation to push the institutions of thirdand fourth-generation states back in time to earlier states that may not have
shared them. We all agreed that we need refined archaeological tools, both
conceptual and methodological, for getting at the sociopolitical institutions
of the very first states. Flannery makes a start at this in chapter 2, but it is
only the first step of a long journey. The fact that neither Possehl (this volume) nor Webster (this volume) believes he has sufficient evidence to apply
the term "state" to the impressive society he is studying indicates how badly
we need refined archaeological tools. One of our future tasks should be to
craft them.
NOTE
1. See, for example, R. E. W. Adams 1977; Claessen and Skalnik, eds. 1978, 1981;
Claessen and van de Velde 1987; Claessen, van de Velde, and Smith 1985; Cohen and
Service 1978; Cohen and Toland 1988; Culbert 1973; Ehrenreich, Crumley, and Levy
1995; Friedman 1979; Friedman and Rowlands, eds. 1977; Gledhill, Bender, and
Larsen 1988; Kautz and Jones 1981; Larsen 1978; Patterson and Gailey 1987; Yoffee and
Cowgill 1988.
1
The 'round Plans of Anhoi( Stotes
HUT Y. nUURY
ost of the world's primary states arose at times when writing was either
absent or limited in subject matter. By the time texts become richly informative, we are often dealing with second- or third-generation states.
And in regions like the Andes, even such later states had no texts.
For this reason there has long been a need for a set of clues by which
archaic states can be identified on the basis of archaeological data. I use
the term "clue" deliberately, since archaeological evidence is circumstantial
rather than eyewitness. None of us was there to see Mesopotamia's Uruk
state, Peru's Moche state, Mexico's Monte Alban II state, or Egypt's Dynasty o. Those early states, however, left the archaeological equivalent of
fingerprints, hair and fiber samples, and blood drops with recognizable
DNA. They left settlement patterns and buildings whose ground plans
reflect the social, political, and religious institutions of the archaic state
(Flannery and Marcus 1976).
In this essay I look at the settlement hierarchies of early states, their governmental and residential palaces, their temples and priests' residences, and
their royal tombs. Had I no page limit, I would add fortifications, military
and political expansion, state-sponsored craft production, and many other
clues to the list. Since I have elsewhere given my agreements and disagreements with Childe's ten criteria for urban civilization (Flannery 1994), I will
not repeat them here.
Note that I am not proposing an archaeological definition for the state; its
definition must remain in the realm of anthropology and political science.
16 1 KENT V. FLANNERV
14
12
10
8
1'l 6
t:
ID
SETTLEMENT HIERARCHIES
At the peak of their power, archaic states were centralized systems with an
administrative hierarchy in which commands traveled downward while
tribute and information on output traveled upward (Flannery 1972, 1995).
Early states tended to have more levels of administrators and settlements than
the chiefdoms that preceded them, and these differences can sometimes be
detected from survey data.
In the Near East, Wright and Johnson (1975) have suggested that chiefdoms tended to have only two or three levels (or tiers) of settlements,
whereas states tended to have a hierarchy with at least four levels: cities,
towns, large villages, and small villages. While not a hard-and-fast "law"of which archaeology has few-this rule of thumb has proved useful in
other world regions, such as the Valley of Oaxaca (Marcus and Flannery
1996), the Maya Lowlands (Marcus 1983a, 1993a), and the north coast of
Peru (Wilson 1995).
It should be noted that "administrative hierarchy" and "settlement hierarchy" are not synonyms. The former refers to the number of tiers of administrators in the system, which may not be archaeologically detectable in
societies without written texts. The latter refers to the number of tiers of
community sizes (which may be detectable through the use of histograms
like those shown in fig. 2.1) and to the administrative institutions present in
each tier (only some of which may be detectable through the excavation of
public buildings and residences).
For example, the earliest clearly identifiable state in the Zapotec region
of Mexico (100 BC-AD 100) seems to have had a site-size hierarchy of at
least four tiers. Its heartland, the Valley of Oaxaca, had an estimated 41,000
inhabitants distributed through 518 sites (Kowalewski et al. 1989). Tier 1 of
the settlement hierarchy was a capital city covering 416 ha, with an estimated population of14,500. Tier 2 ofthe hierarchy consisted ofsix "towns"
with populations estimated at 970 to 1,950 persons, all located within 14 to
28 km of the capital. Tier 3 consisted of at least 30 "large villages" in the
5-10 ha range, with populations estimated at 200 to 700 persons. Tier 4
consisted of more than 400 "small villages" with estimated populations below 200 persons.
Middle Uruk
Tier 1
(Susa, 25 ha.)
~ 0
i-Jl''"'__......_IIL-
---lll....."--_ _- - / .liEiilI
(f)
15
ID
.0
Tier 3
14
12
10
8
6
4
2
Early Uruk
Tier 1
(Susa)
o
.5
1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0
1.5 2.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 10.5 11.5 12.5
Settlement Area in Hectares
At least where excavation data are available, these settlement tiers are also
characterized by the presence or absence of buildings related to the administrative hierarchy (Marcus and Flannery 1996: 111-12). In the Zapotec
case, the capital city had a large "governmental palace," a ceremonial plaza,
multiple residential palaces and princely tombs, multiple standardized temples,
more than one I-shaped ball court, and a building with hieroglyphic records
of subject provinces. Tier 2 sites sometimes had smaller versions of the "governmental palace," fewer temples, fewer residential palaces and tombs, and
only one ball court. Tier 3 sites usually had only one temple, no palace, no
ceremonial plaza, and no ball court. Tier 4 sites had no evidence of public
structures at all.
A four-tiered hierarchy of settlements may therefore have administrators
only at its upper three tiers. This does not mean that there are only three levels of administrators in the society. It means only that state administrators
(number of existing levels notwithstanding) are usually not present at small
villages.
In the Maya Lowlands, where hieroglyphic inscriptions are available,
textual information can be added to site-size data. For example, more than
18 I KENT V. FLANNERY
two decades ago Marcus (1973) argued that the Maya city of Calakmul was
at the head of a "central-place hierarchy." This is a term for an administrative hierarchy so well integrated that Tier 2 towns encircle the Tier 1 city
at very regular distances; in turn, Tier 3 settlements encircle Tier 2 settlements at regular (and shorter) distances (fig. 2.2a). In the Calakmul case, not
only were Tier 2 centers like Oxpemul, Sasilha, Naachtun, and Uxul spaced
regularly, 34 km from each other and from Calakmul, but also many subordinate centers mentioned Calakmul's "emblem glyph" in their inscriptions.
This was part ofa widespread pattern in which Tier 3 centers mentioned the
Tier 2 centers to which they were affiliated, while Tier 2 centers mentioned
the Tier 1 capital for their region (Marcus 1976a). Recently, Folan, Marcus,
and Miller (1995) have discovered that a series of raised roads connected
Calakmul to its Tier 2 settlements.
Four-tiered settlement hierarchies are extremely Widespread among archaic states, a pattern that has escaped many Mayanists. There may be several reasons for their failure to recognize hierarchies. For instance, some
Maya archaeologists are reluctant to see "their" Tier 2 or Tier 3 site as subordinate to any other center, apparently resenting it if their site is considered
anything less than a capital. Some Maya epigraphers, in addition, assume that
if a site possessed its own emblem glyph it could not be hierarchically below any other site. Some have, as a result, proposed that the Maya area had
as many as 800) autonomous "capitals," each ruling a minuscule territory
(Mathews1985). I know of no archaic state that had such a nonhierarchical
structure.
It is instructive, in this regard, to look at the situation in Mesopotamia.
Like the Maya region, Early Dynastic Mesopotamia was composed of distinct polities, some ofwhich maintained their integrity over centuries. Some
Near Eastern scholars (e.g., Diakonoff 1974) have compared these polities to
the hespsor nomes of ancient Egypt, and the analogy is apt: both regions had
periods when polities were unified under powerful rulers and periods when
centralized control broke down.
Consider the polity ofLagash in southeastern Mesopotamia. It contained
at least three cities, more than 20 towns, and at least 40 villages (Adams
1966: 101). Diakonoff (1974) has estimated its population of "free citizens"
as greater than 100,000, a figure that Henry Wright (personal communication 1997) considers conservative. Its two largest cities, Lagash and Girsu,
had populations estimated at 36,000 and 19,000, respectively. Though archaeological surveys have not yet determined the total area of the polity,
its irrigated fields alone are calculated to have exceeded 2000 km '. It had a
45-km-long frontier with the polity of Umma to the north, and a frontier
of unknown length with Ur to the south.
o
I
10
I
km
(j) Balakbal
10
L--L.-'-!_-'--_'
Figure 2.2. Central-place lattices around ancient cities: a, the lattice around
the Maya city of Calakmul in the seventh century AD (redrawn from
Marcus 1973); b, the lattice around the Early Dynastic Sumerian city of
Eshnunna, showing one city (Eshnunna), four Tier 2 towns (e.g., 515),
six Tier 3 large villages (e.g., 517), and a number of Tier 4 smallvillages
(e.g., 381) (redrawnfromJohnson 1972).
20 1 KENT V. FLANNERY
22 I KENT V. FLANNERV
A classic example of multifunctional complexity-combining royal residences with artisans' quarters-is the Minoan palace at Knossos on Crete
(Evans 1921-36; Graham 1987; Michailidou 1989). The remains visible today are Evans's reconstruction of the second palace, built during the seventeenth century BC. It had at least two stories, ofwhich only the ground floor
survives; it is at least 150 m on a side and covers 20,000 m 2 (fig. 2.3).
The heart of the palace is its huge Central Court (50 X 25 m). All movement to and from the various sectors of the palace originated in this court;
traffic was controlled by narrow corridors that combined privacy with a system of multiple routes from one sector to another. Small wonder that the
building's name-Labyrinth, or "House of the Double Axe"-has come to
be synonymous with "maze."
In the southeast quadrant of the palace lay the royal residential quarters,
separated from the artisans' workshops by a thick wall. According to
Michailidou, the Hall of Double Axes belonged to the king, while the
queen had her own hall, featuring a private bathroom. The five different
routes to the queen's apartment were so indirect that she could maintain her
privacy. Stairways for royal family members and servants indicate that the
rulers' quarters continued on the floor above.
To the north and east of the thick wall separating them from the royal
quarters were workshops filled with raw materials and unfinished products
ofpotters, lapidaries, and metalworkers. Surrounding them were storerooms
ofvarious sizes, filling much of the northeast quadrant of the palace. Almost
the entire western margin of the complex was devoted to storage rooms. As
in all major Cretan palaces, there were monumental storage facilities for
grain, wine, oil, and other commodities. For example, more than 400 pithoi
(huge clay jars probably used for olive oil) were found, representing a storage capacity of 246,000 liters.
Off the west side of the Central Court was a throne room, with the
king's chair carved from gypsum. Not far to the south lay a room believed
to be a shrine; numerous other rooms throughout the palace seem to have
had religious functions, but there was no formal temple.
Storage
rooms
\
o
30m
Figure 2.3. The Minoan palace of Knossos on Crete, a large multifunctional palace complex (redrawn from Michailidou 1989).
Among the New World's best examples oflarge multifunctional palace complexes are the ciudadelas or "great enclosures" of Chan Chan, a Chimu city
on the north coast of Peru. Built between AD 1000 and 1400, each of the
ten great enclosures is thought to have been the work ofa single Chimu ruler
(Day 1982; Moseley 1992).
The ciudadelas of Chan Chan range in size from 87,900 m 2 (the Rivero
Compound) to 221,000 m 2 (the Gran Chimu Compound), making them
many times larger than the Labyrinth ofKnossos. Each is enclosed by adobe
or tapia (pressed earth) walls up to 9 m in height, which separate it from the
smaller compounds of the lesser nobles and the kincha (wattle-and-daub)
24 I KENT V. FLANNERY
I
r
!i
E Storerooms
~ Audiencias
ffi]
Walk-in
wells
I
!
North
Sector
I
I
Central
Sector
I
!,
o
I
50
I
Figure 2.4. The Rivero Compound at Chan Chan, Peru, a large multifunctional palace complex (redrawn from Day 1982).
26 I KENT V. FLANNERY
rank societies of the Formative period but indisputable palaces did not become common until the state had formed (Marcus and Flannery 1996).
There are hints ofpossible Uruk palaces at Uruk and Eridu, and by Early
Dynastic times (third millennium BC) there is unmistakable evidence of
palaces in the archaeological record. It is perhaps significant that two of the
earliest examples are from Kish (Moorey 1978), since the ruler Mesalim of
Kish (c. 2550 BC) is thought to have "exercised hegemony far beyond the
walls of his own city." According to Diakonoff (1974: 10), some Early Dynastic rulers of Kish called themselves "lugal [king] of the universe" and had
supreme control over the rulers of other provinces, who were only allowed
to use titles like ensl or ensi-gar (governor).
Palace A at Kish, built at no great distance from the city's temples, consists of at least two architectural units covering an area of perhaps 90 X 60 m
(fig. 2.5). The larger of the two units was surrounded by a massive, buttressed defensive wall, which helped protect the royal residential quarters
deep in the western portion of the building. The monumental entrance was
on the southeast and led to offices and archives that had only indirect access
to the royal apartments. The smaller of the two units was separated from the
first by a narrow corridor and almost has the appearance of an annex. Deep
in its interior was a decorated reception hall with large columns.
Governmental vs. Residential Palaces at Teoelhuacan,
Basin of Mexico
Building plans at Teotihuacan provide further support for the typology of
palaces mentioned above: some large complexes appear to be administrative
or governmental; others appear largely residential. It should be noted that
many scholars have worked at Teotihuacan over the years, and not all agree
on the functions of each building.
George Cowgill (1983) believes that the seat of government at Teotihuacan may have shifted over time. Early in the city's history it may have
been near the Pyramid of the Sun; later it may have shifted to the Ciudadela
(see below); stili later it may have moved to the Avenue of the Dead Complex. I wili restrict myself here to discussion of the Ciudadela, which includes a pair ofstructures referred to as the North and South Palaces (Milion
1973).
The Ciudadela is a walled enclosure 400 m on a side, east of the Avenue
of the Dead. At least 15 minor temples frame the Ciudadela, whose interior
is a vast sunken court. At the eastern limit of this court is a stepped pyramid
called the Temple ofQuetzalcoatl, "the Plumed Serpent." This building (actually the pyramidal platform for a now-vanished temple) is one of the bestknown and most extensively explored religious complexes at Teotihuacan,
Its oldest building stage dates to AD 150-200 (Cabrera Castro 1991). Recent excavations by Cabrera Castro, Sugiyama, and Cowgill (1991) show
that more than 200 individuals, many with military paraphernalia, had been
sacrificed during the pyramid's dedication.
Already at this period there were residential structures to either side of
the pyramid. No plans are available for them, however, since they were later
covered by the residential complexes known today as the North and South
Palaces (fig. 2.6). These large twin palaces were built around AD 200-350,
when the Temple of Quetzalcoatl had a second stage attached to its west face
(enlarging it but covering its old facade).
Each of the twin palaces measures 80 m on a side and has an interior
courtyard that probably served as a light-well. Around each courtyard are
arranged five or six complexes of rooms with varying degrees of privacy.
The need for space seems to have increased over time, since an annex (known
as "Building 1C-prime") was later added in front of the North Palace.
Since Milion (1992) and Cowgill (1983) see the Ciudadela as the seat of
government at Teotihuacan, they interpret these buildings as related to the
administration of the city. If so, I believe these structures would qualify as
,
28 I KENT V. FLANNERY
t
N
North Palace
t
o
w
o
w
I
f-
LL.
o
W
>
r
BUilding
le-prime
o
Temple of
Quetzalcoatl
o
u
South Palace
Figure 2.6. The Ciudadela at Teotihuacan, showing the Temple cif Quetzalcoatl .flanked by the North and South Palaces (redrawn from Cabrera
Castro 1991). The Ciudadela is 400 m on a side. The "north" arrow actually points 15 30' east of true north (the orientation of the city's major
buildings), sometimes called "Teotihuacan north."
30 I KENT V. FLANNERY
of servants or retainers. Since one could not enter the main compound
directly from these northern room complexes, it is likely that they were for
staff members who only needed access to the interior during the day.
As mentioned above, the royal compounds at Chan Chan were surrounded by relatively ephemeral structures that housed artisans. Recent
discoveries in the La Ventilla district of Teotihuacan suggest that a similar
patron/client relationship may have existed between noble families and
commoner artisans there. La Ventilla lies not far west of the Ciudadela and
consists of a "neighborhood" in which several compounds are separated
only by narrow alleys. At least one compound, unstuccoed and less elegant,
seems to have been occupied by commoner craftsmen working with a wide
variety of raw materials. Their higher-status patrons may have lived in a
more elegant, stuccoed, red-and-white painted compound just across the
alley. I look forward to the eventual publication of La Ventilla by Ruben
Cabrera Castro, Sergio G6mez, and Nestor Paredes.
o
I
10
I
32 I KENT V. FLANNERY
Storerooms with
olive oil jars
Stairs to
upper floor
Pantry with
drinking cups
Waiting
_JJI.~!YJlIL---.!:~~Jb;~
."J
room
g.
11==-"
Queen's Megaron
N .... - - - -
Entrance
Guard room
o,
5,
Meters
storage room in the rear had vast supplies of olive oil in heavy pithoi sunk in
the ground; one of the pantries on the southwest had the broken stems of
2,853 drinking cups. We can assume a staff of household servants, but there
is no evidence of artisans' quarters within the palace.
The supposed queens apartments, occupying the eastern corner of the
building, were reached by a columned porch and a narrow corridor that
limited traffic. The queen apparently had her own megaron, a boudoir, and
a lavatory. A larger bathroom with a terracotta bathtub, not directly accessible from the queen's apartments, seems to have been for more general use
by the residents of the palace.
34 I KENT V. FLANNERV
Cist
Pillar Crypt
Great Hall
I I
Vestibule
Entry
Light-well
" .. ,
:----,---/' ../
I
,L
-----------------------_...
Figure 2.10. A small residential palace at the Zapotec city cif Monte Alhlm,
Oaxaca, Mexico (redrawnfrom Marcus and Flannery 1996).
Knossos (below). This Lowland Maya palace, like the Zapotec palaces of
Monte Alban, appears to be a royal residence largely unaccompanied by servants' and artisans' quarters.
The Zapotec Palaces of Oaxaca, Mexico
_______________________ t
L.
,:
1
; ,....
,i-r--:'
I
:
,
0
'-
IweeeeeewI
......
meters
36 I KENT V. FLANNERV
light-well. At least one room to the north of the great hall was devoted to
ritual. This was the pillar crypt or cult room, its ceiling supported by huge
split tree trunks laid horizontally and resting partly on a centrally placed
rectangular pillar. There were two cists in the floor to either side of the pillar, presumably receptacles for poured libations.
TEMPLES AND PRIESTS' RESIDENCES IN
ARCHAIC STATES
Shrines and temples were present long before the rise of the stateindeed, from the beginnings of sedentary life in some parts of the world. In
some regions, temples already had a standardized ground plan centuries before states and palaces appeared. In Mesopotamia, that ground plan took
the form of a long central room with an altar and podium, flanked by
rows of smaller rooms. On the coast of Peru, many early temples took the
form of a block U. Temples in some regions had an apparent astronomical
orientation: in Mesoamerica it was the sides of the temple that faced the
cardinal points, whereas in Mesopotamia it was the corners of the temple (see
below).
While chiefdoms might use corvee labor to build their temples, only the
most complex had full-time priests supported by tithes or tribute. Many archaic states had not only full-time priests, but also special residences for
them; the existence of such priestly residential quarters may signal the presence of a state. For example, in Mesoamerica one-room temples gave way
to two-room temples by the first century BC. Although worshipers could
come to the outer room, priests sometimes lived in the inner room and often had their own entrance or stairway, separate from that of the public
(Marcus and Flannery 1996: 181-94). In other regions, such as Mesopotamia, there were separate priests' residences (like the gipar of Ur) not far
from the temple.
Some state religions had a hierarchy in which high priests (or priestesses)
were members of the nobility, while minor priests were trained commoners. In no archaic state have we discovered the full range of residences lived
in by major and minor priests. All we can say is that convincing "priests' residences" are more likely to occur in states than in chiefdoms.
Once again, the presence of standardized temples and priests' residences
in states should be considered only a rule of thumb. Some possible archaic
states, like those of the Aegean, even lack standardized temples. They relied
instead on features called "shrines," "shrine rooms," or "cult centers" by
Aegean archaeologists.
Let us now look at a sample of temples and priests' residences from archaic states, beginning with those having the most formal ground plans.
Storage
a
o,
5,
meters
10
,
/'
o
I
~-------------~
b
Figure 2.12. Early Mesopotamian temples: a, Temple VII at Eridu, a
prestate temple <if the 'Ubaid period (e. 4000 BC) that already has the tripartite plan of later Mesopotamian temples (redrawn from Safar, Mustafa,
and Lloyd 1981); b, the White Temple in theAnu precinct at Uruk, capital <if an early Mesopotamian state (redrawn from Perkins 1949).
stage, access to the outer entry court was blocked, and a new entrance was
cut throughthe wall of the oval to the street outside.)
.
Figure 2.13 shows the initial layout of House D. Roughly 30 X 40 m in
extent (but fitted into the irregular space between the oval's walls), the house
was entered by a small door that led the visitor past a guard room to a narrow access corridor. Off this corridor were two antechambers; one was
flanked by a bath and toilet where visitors could refresh (or, more likely, purify) themselves. This done, they could enter the building's central court,
perhaps pouring libations at its offering table.
The central court was a major hub for traffic within House D. To its
south lay the priest's reception room, complete with a divan on which he
could receive visitors. Behind the reception room were his archive and
dressing/sleeping room. East of the court was a dining room, and behind it
a pantry with access to the servants' quarters. To the north of the central
court lay a storage room, the priest's private chapel, and an L-shaped passage
leading to the servants' quarters. The latter consisted of three rooms east of
the pantry, one of which contained a kiln or oven.
40 I KENT V. FLANNERY
House D looks so much like a miniature palace that it must have been
for a very high priest, possibly drawn from a family of some importance. In
some Early Dynastic states there were strong ties between the civil and religious hierarchies, and as we have seen, city rulers sometimes appointed
their heirs apparent to the position of sanga, or manager of the temple estate
of the city's patron deity (Cooper 1983; Tyumenev 1969: 9).
In later periods Mesopotamia had very large priestly residences, like the
gipar of Ur, which was set aside for the high priestess of the moon god
Nanna. The temple to Nanna lay atop a huge ziggurat, or stepped pyramid,
and the gipar built by the ruler Amar-Sin of the Ur III dynasty (20462038 BC) measured 75 m on a side. The gipar contained a temple dedicated
to Ningal-female consort of Nanna-plus an elite residence for the high
priestess, complete with kitchens, bathrooms, storerooms, ceremonial rooms,
and a burial crypt (Oates 1986; Postgate 1992; Woolley and Moorey 1982).
High priestess ofNanna was a very prestigious office, whose holder received
the title en-a word that, since Protoliterate times, had denoted nobility.
Sargon of Akkad (2334-2279 BC) sent his own daughter Enheduanna to be
the en-priestess of Nanna at Ur, beginning a tradition of royal nepotism
there that continued for half a millennium (Oates 1986: 39).
L......JL.......J.---.L---'--"
5m
c
Figure 2.14. Plans cif templesJrom Mesoamerican states: a, Zapotec temple
on Mound X at Monte Alban, Oaxaca, Mexico, discovered by Caso
(1935); b, Maya temple at Chichen Itz, Yucatan, Mexico (redrawnJrom
Marcus 1978); c, the Temple oJthe Cross at Palenque, Chiapas, Mexico
(redrawn from Marcus 1978).
42 I KENT V. FLANNERY
Temple to feline
supernatural (?)
~~
Cl
,..--....1IIl ~
I
f-
U.
"'-~IIi~w
West
temple
>
Temple to
anthropomorphic
being (?)
Altar
temples in the form of a block U. Such temples were already present in the
preceramic era (Donnan 1985), and by the time of the earliest coastal states
their location had been raised to the summit ofstepped pyramids. Often the
temple staff occupied a residential compound behind the huaca, or sacred
pyramid.
During Peru's Middle Horizon (AD 600-900), the site of Pacatnamii is
believed to have been a temple center for the north coast, both the site of an
oracle and a destination for religious pilgrimages (Donnan and Cock 1986:
19; Keatinge 1982). More than 35 truncated pyramids cover an area of
1 km 2 overlooking the sea. The largest pyramids (Type A) are up to 13 m
high and have multiple terraces with access ramps, altars, associated plazas,
and side platforms. Other pyramids (Type B) are smaller but still have platforms, plazas, ramps, and accessory room complexes. Finally, there are more
than 65 smaller structures (Type C) with U-shaped temples that can be approached at ground level. The great differences in size and complexity
among the temples and shrines at Pacatnamu recall those of some Mesopotamian cities, where the temple of the city's patron deity was the largest, the
temple of his female consort was smaller, and the temples of other deities
were smaller still (Diakonoff 1969, 1974).
Figure 2.16 gives the layout of one of the larger (Type A) temples at
Pacatnamu, with its forecourt, side platform, altar, and stepped pyramid. Behind the pyramid was the probable residential compound ofthe temple staff.
One such compound-whose inner court could be reached only after a circuitous trip down narrow corridors, and through doorways shielded by curtain walls-had two U-shaped audiencias like those of the ciudadelas at Chan
Chan. The floor of one ofthose audiencias was covered with crushed SpondyIus shell. This suggests that it had been used by a hereditary lord, since
crushed seashell dust was used to prepare surfaces for Andean lords to walk
upon (Keatinge 1982:216; Rowe 1946:47).
Minoan and Mycenaean. "Shrines"
Not every culture with large, elaborate palaces also had formal standardized
temples. In contrast to their neighbors in Egypt and Mesopotamia, the Minoans and Mycenaeans of the eastern Mediterranean region "built no great
temples and carved no large statues oftheir deities" (Graham 1987: 19). That
is not to say, however, that they did not produce recognizable and fairly
standardized ritual spaces. According to Graham (1987: 18),
we seem scarcely able to move in the Palace of Minos ... without running
into small shrines or small representations of deities or their symbols: lustral
chambers, pillar crypts, columnar shrines, temple repositories, incurved altars, sacral horns, double axes, libation tables ...
o
I
20m
I
of the acropolis
at Mycenae in
The later Mycenaeans also seem to have set aside certain rooms within
their palaces as sacred places and built recognizable but unstandardized
shrines, "cult centers," and sanctuaries (see for example Renfrew 1985).
There are hints that these features might lack standardization because each
was a shrine to a different member of a pantheon, since-as pointed out by
Taylour (1983)-some of the later Greek deities are already detectable in
Mycenaean art.
Two"cult areas" were found at the foot of the acropolis of Mycenae itself, just within the cyclopean walls of the citadel. One of these (somewhat
optimistically called a "temple" by Taylour 1983: 51), consists ofthree stone
masonry rooms: the Vestibule, the Room with the Platforms (5.1 X 4.3 m),
and the Room with the Idols (1.8 X 1.8 m) (fig. 2.17). On one of the platforms that gave the second room its name lay a female idol 60 cm high. In
the Room with the Idols, reached by a small stairway, were found more than
20 additional idols (both male and female), as well as snake effigies, libation
bowls, and drinking cups.
46 I KENT V. FLANNERV
One of the most frequently used clues to the existence of kings and queens
is the presence of royal tombs. Unfortunately, tombs are somewhat slippery
clues because there is no clear-cut distinction between the burial of a king
and the burial of a paramount chief Some paramount chiefs in Panama, for
example, were buried atop a layer of 20 or more sacrificed retainers, accompanied by goldjewelry (Lothrop 1937).
Most chiefly societies have a continuum of statuses without a clear division into class-endogamous social strata, and hence no true "commoners."
In some highly advanced chiefdoms, however, such as protohistoric Hawai'i,
a conunoner class was created by severing kin ties to the lower-ranking families (Kirch 1984: 258). In states, the gap between noble and conunoner is
larger; indeed, in some kingdoms the difference between the tomb of the
king and the tombs of the minor nobles below him can become so great that
the latter appear "middle class" by comparison (Marc us 1992b:221).
Taken alone, spectacular tombs are usually insufficient evidence for a
state. However, when they occur beneath the floor of a palace in a society
with a four-tiered hierarchy, they can represent an additional line of evidence-especially when their construction involves considerable corvee
labor and the offerings represent a quantum leap in elegance beyond those
of previous chiefdoms in the same region. We will look now at a sample of
tombs that might qualify.
Mesopotamian Royal Tombs
The earliest possible "royal tombs" in Mesopotamia come from Level X of
Tepe Gawra in northern Iraq (Tobler 1950). They belong to the Gawran
phrase, roughly equivalent in time to the Uruk period in southern Mesopotamia. Unfortunately, these tombs present many of the same problems as
the spectacular interments of the "lords ofSipan" on the north coast of Peru
(see below). That is, although they clearly reflect a society with enormous
differences in hereditary status, we still have a lot to learn about the details
of that society. Once we know more about the size of the largest relevant
sites, the number oflevels in the settlement hierarchy, and the nature of elite
residences and public buildings, we will be able to speak more confidently
about whether Gawran society was an archaic state or a maximal chiefdom.
The Gawran tombs, built of mud brick with wooden roofs, had spectacular sumptuary goods of gold, electrum, copper, turquoise, lapis lazuli,
jadeite, hematite, carnelian, obsidian, marble, steatite, and ivory. One tomb
had more than 25,000 beads, and many skeletons had ornaments on the
head, neck, hands, wrists, waist, knees, or ankles. There were ivory combs
and hairpins, possible gold handles for fans, gold rosettes with lapis lazuli
centers, and vessels of marble, serpentine, and obsidian.
48 I KENT V. FLANNERY
Young
woman
(19-25 yrs.)
Wooden litter
with queen's
skeleton
Adult man with
amputated feet,
in cane coffin
(in fill above
chamber)
-,
Row of
attendants
2m
Q~ (both me"
~ and women)
;.'.Jl' ~fu~g~~~s
~
I
'-'l
"goQQ
4J!f!f' ~
~ Q,Q
OJ~ ~ ~Ql8
()J
"The
Deat~Pit"
~.
Ox-drawn
wagons with
drivers'
skeletons
inside
6 soldiers
with helmets
and spears
Figure 2.18. The royal tombs cif King A-bar-gi(?) and Queen Shub-ad or
Pu-abi(?) at Ut, southern Mesopotamia (redrawn from Wool/ey 1934).
without a coffin, was a 19- to 25-year-old woman wrapped in a textile with
discs of gilded copper sewn to it. She also had a large copper headdress.
A number of additional humans and animals had been sacrificed at the
ruler's death. To his right lay an 18- to 22-year-old woman sprawled face
Child
(8-10 yrs.)
in cane coffin
with sacrificed
dog & snake
Figure 2.19. Tomb 2 at Sip/m, Peru, the burial of a Moche lord withpossible family members andlor retainers (redrawn from Alva and Donnan
1993).
down with no grave goods, perhaps a servant or retainer. Near her feet was
a decapitated llama with its feet tied together. A llama head (which might
have come from the decapitated animal) was found in the fill above the tomb
roof Also in this fill was a cane coffin containing an adult male whose feet
had been cut off at the ankles; he was accompanied by a feather headdress
with a large copper shaft, a copper crown, and numerous gourd containers,
all preserved by the aridity of the Peruvian coast.
Surrounding the tomb chamber were hundreds of ceramic vessels plus
offerings that included human hands and feet, "quite possibly the trophies
taken from sacrificed prisoners whose bodies were dismembered, exactly as
shown in Moche art" (Alva and Donnan 1993: 165). Thus, while the Early
Dynastic citizens ofUr thought it appropriate to sacrifice the rulers' servants
and bodyguards, the Moche included in their royal tombs the body parts of
sacrificed enemies.
While we are only beginning to glimpse the splendor of Moche tombs,
we have abundant data on the royal burials of the later Chimu. One of the
standard features of the ciudadelas or palatial complexes at Chan Chan on the
northern Peruvian coast were massive burial platforms (Conrad 1982). Each
was a truncated pyramidal mound of adobe or pressed earth, ranging in size
from 52 X 34 m to 190 X 90 m and in height from 3.5 m to more than
12 m. Walls enclosed (and reduced access to) each burial mound, and the
summit was reached by a series of earthen ramps whose entry point was a
nearby forecourt (fig. 2.20).
All the Chan Chan platforms contained a series ofburial or offering cells,
the number varying from 15 to more than 100. The principal cell, usually
T-shaped and centrally located, was probably the resting place of the Chimu
ruler (unfortunately, many ofthese had been looted). The surrounding cells,
usually smaller and rectangular, may have contained noble relatives of the
king, as well as- sacrificial victims. The majority of the latter were women,
believed on ethnohistoric grounds to include numerous "virgins of the sun,"
like those documented at the time of the Spanish Conquest. Crushed shells
of Spondylus, or spiny oyster, were placed on the floors of many cells. As at
the temples at Pacatnamu (see p. 43, this volume), such cells were probably
destined for members of royal or noble families.
Because some cells contained more than a dozen skeletons, Conrad
(1982: 99) estimates that the burial platform of one ciudadela, the Laberinto
Compound, may have had 200 to 300 burials before it was looted. The areas least affected by looting had high concentrations of prestige goods, including fine textiles and pottery, carved wood, metal, and Spondylus shell;
there were also sacrificed llamas below some cell floors.
Because of the high numbers of possible relatives, servants, retainers, and
sacrificial victims included in the Chan Chan burial platforms, they probably provide our best New World analogy for the Royal Tombs ofUr. They
were among the last features in a Chimu compound to be built, but virtually every compound seems to have space set aside for one.
Royal Tombs of the Zapotec and Maya
In both the Zapotec and Maya regions of Mesoamerica, rulers were laid to
rest in stone masonry tombs. Some of the more elaborate tombs had mural
paintings, were closed with carved door stones, and could be reached only
by subterranean stairways. In both areas, royal tombs have been found beneath buildings with a palace ground plan as well as buildings with a temple
ground plan.
Forecourt
Tomb of
oU
r-=l ~ ~rincipal
*tIEl
11 11 figure
0 1I----
aB
BB~
aa~88
gB~gg
o,
5,
52 I KENT V. FLANNERY
artificial
mound
a
blockof
Vessel 8
red pigment-CD shell-studded
Vessel 9
V.~
~~~~~
tEf
jadeite face
mask
grave pits
~
V.6
V.3
pearl
~
~
V2
o stingray spine
SJ-
Cc:
V.1
G~
Vessel 10
Spondylus shell
b
Figure 2.22. Mycenaean tombs: a, ashlar masonry tholos appropriate for
a royal family; b, cavern excavated in bedrock, appropriate for a noble or
wealthy commoner (redrawn from Taylour 1983).
"psychoduct" through which the spirit of the deceased ruler could leave and
return (Folan et al. 1995).
Mycenaean Royal Tombs
Figure 2.21. Tomb 1from Structure Ill, Calakmul, Mexico, the burial cif
a Maya lord (redrawn from Folan et al. 1995). Vertical distances havebeen
exaggerated to show the cifferings.
intrusive through the floor of Room 6 of Structure Ill, the palace already
described above. Called Tomb 1, it contained the skeleton of a man at least
30 years of age, lying fully extended in a rolled mat (fig. 2.21). His wrapped
body had been placed on five dishes that elevated him above the floor; more
elaborate vessels had been set around the skeleton. The most spectacular offerings were three jade mosaic masks- one originally made for his face, one
for his chest, and a third for his belt. Spiny oyster shell, a stingray spine, and
a pearl were included as offerings. The tomb also had been equipped with a
Aegean archaeologists are not in agreement as to whether Mycenaean society was a large state or something less (see discussion in Marcus, this volume); nevertheless, Mycenaean rulers clearly lived in impressive palaces, like
that of King Nestor at Pylos. And in their burials we see a dichotomy between ashlar masonry tombs (which could belong to kings) and bedrock
chamber tombs (which could belong to lesser nobles). The masonry tombs
were found not beneath the floors of palaces but dug into the sides of natural hills. The most spectacular (and most convincingly "royal") are the tho105 tombs, at least nine ofwhich were constructed at Mycenae between 1500
and 1250 BC (fig. 2.22a).
Each tomb consists of a conically roofed circular chamber or tholos, up
to 13.7 m high and 14.9 m in diameter, done in ashlar masonry (a construction style shared with the Inka). This chamber, deeply buried in a hillside and marked with an artificial mound of earth, was reached by a corridor
54 I KENT V..FLANNERY
or dromos up to 40 m long, also dug into the hill slope. The entrance to the
chamber was a monumental doorway up to 5.5 m high, closed with bronze
doors. After burial had taken place, the mouth of the dromos was blocked
with a stone wall (Taylour 1983: 65 - 81).
Mycenaean nobles (and perhaps some of the wealthiest commoners) imitated royal tholoi by digging less spectacular "chamber tombs" (fig. 2.22b).
These irregular caverns were dug into bedrock, given a small entry corridor,
and closed with stone walls. Such chamber tombs were apparently family
burial vaults, opened over and over again for additional burials. Drinking
cups broken in the dromos show that libations were poured for the dead (Taylour 1983: 83). These rock-cut tombs lacked the labor-intensive ashlar masonry and bronze doors of their royal counterparts. They were, however,
noticeably more elaborate than the simple grave of the average commoner.
Note that this typology of tholoi, chamber tombs, and simple graves does
not imply that the Mycenaeans had a third social stratum or "middle class."
It is likely that all members of the hereditary nobility ended up in a chamber tomb ofsome kind. Perhaps only those nobles who ascended to the rank
of king received a tholos with ashlar masonry and bronze doors.
them. With time and patient excavation, however, these missing data can be
discovered.
Archaic states often had an official religion, one featuring standardized
temples and full-time priests. We have seen archaeological evidence not only
for such temples, but for the residential quarters of the priests. Depending
on the region involved, the latter may be incorporated into the temple itself,
or situated nearby. Modest in first-generation states, the priestly residence
grew more spectacular in second- or third-generation states, as exemplified
by the gipar ofUr (Woolleyand Moorey 1982) or the high priest's residence
at Mitla in Mexico (Marcus and Flannery 1996: figs. 6, 7).
All states were hierarchical societies. As Marcus (1989, 1992a, 1993a) has
suggested, however, there were stages of development during which states
were larger and more centralized, and stages during which they broke down
into their constituent provinces. During these periods of breakdown, the
constituent provinces of a state might gain some degree of autonomy, but
this autonomy was often short-lived.
A four-tiered site hierarchy is one clue to this process, so long as one can
show administrative functions at the three upper tiers. Almost all pristine or
first-generation states, such as the Uruk state in Mesopotamia or the Monte
Alban II state in southern Mexico, show such a structure. However, when
such states broke down, it was their various provinces that emerged as the
largest stable units. Some of these individual polities might only have threetiered hierarchies (the Mixtec cacicazgos of Postclassic Mexico would be an
example). That does not mean that such provinces had reverted to being
"chiefdoms." Their societies remained stratified, their rulers still lived in
palaces and expected to be treated like kings, and many such polities kept
trying to subdue (or ally themselves with) neighboring provinces to reestablish a large regional state.
By using large-scale "total coverage" settlement pattern surveys (Fish and
Kowalewski 1990) we can document the appearance of pristine states with
four-tiered hierarchies, their breakdown into provinces with three-tiered
hierarchies, and their reintegration into a second-generation state with four
tiers. I would stress, however, that site size alone is not sufficient information: we must also excavate sites at the upper three tiers if the hierarchy to recover the
buildings that rejlect state institutions.
56 I KENT V. FLANNERY
Richard Diehl (1980) reveal that they are not in agreement as to whether
the Olmec constituted a chiefdom or a state.
Coe is convinced that the Olmec had ... a state-level organization and social stratification analogous to those of later Mesoamerican societies, while
Diehl maintains that they were organized along chiefdom lines with ranking
of individuals and kin groups but no socioeconomic classes. (Coe and Diehl
1980, vol. 11:147)
A great many colleagues working in the New and Old Worlds provided me with insights
and information for this essay. I am particularly indebted to Henry T. Wright and
George L. Cowgill for extensive constructive criticism.
J
Ihe Peaks ond Volleys of Andent Stotes
An htensioll of the DYIIDllli( Model
IOTU MUCUS
60 I JOYCE MARCUS
sometimes had settlement hierarchies with only three tiers. These provinces
were run by hereditary lords who continued to call themselves batabob or
th'ul ahauob, even though the territories they controlled were now much
smaller. Because of its repetitive cycles of consolidation, expansion, and dissolution, I referred to this framework as the Dynamic Model.
for a century, until unified by Ur-a region that had once been but a
province of Sargon's realm.
Colleagues working in the Andes wrote me of similar cycles there. The
Moche state, 550 km from north to south-so territorially extensive that
some Andeanists consider it an empire-broke down into its constituent
provinces, only to see Wari take control in the south. With the collapse of
the Wari state came the Chimu, who controlled perhaps 1000 km of Peru's
coast until they were overcome by the Inka. In many cases, the region that
rose to prominence had once been a subject province of a previous state.
Encouraged by all those colleagues who see parallels in their regions, I
will extend the Dynamic Model still further in this chapter. I do not expect
that the model will be applicable to every archaic state in the ancient world,
but it seems to be applicable to quite a few. I will begin in Mesoamerica
and extend outward to include the Andes, Mesopotamia, Egypt, and the
Aegean.
62 JOYCE MARCUS
~
ma:
WO
~~
at some sites
,
'.
Caraco I s expansion
First Maya
states form at
\
Tikal, Calakmul
. ,
sl~"FourDivisions"
Height of Puuc
)eg\on
unite at
u,
Monument "hiatus"
[jJ
Chichen ltza
ornpetmq
W
N
chiefdoms
en
W
>rn
w
Recently, it has become fashionable to refer to the highest peaks in figure 3.1 as "superstates" (Martin and Grube 1995). This term is unnecessarily hyperbolic, since the polities involved are merely typical states. We have
long known that, at their peaks, cities like Tikal and Calakmul were the capitals oflarge regional states (Marcus 1973, 1976a). The point here is that the
smaller units into which they occasionally fragmented were not usually states
at all. Like the Postclassic Mixtec cacicazgos, these former subject provinces
should be considered no more than principalities or petty kingdoms. (Some
Mayanists refer to them as "city-states," but as we shall see below, this term
has little to recommend it.)
a:
During the period 300 BC-AD 200 in the northern Peten, two large chiefly
centers arose only 38 km from each other. Calakmul and El Mirador commanded sufficiently large labor forces to construct huge pyramids and build
a major causeway linking their two towns (Folan, Marcus, and Miller 1995).
Although all the details of their interaction are not known, it appears that El
Mirador collapsed before AD 300-400, precisely the period during which
Calakmul went on to become a capital of an early state.
As is so often the case with newly formed states, one of Calakmul's periods of major expansion came early in its history. Calakmul was able to incorporate various provinces into its realm during the fifth and sixth centuries
AD. Nevertheless, its territory was a mosaic since some intervening provinces
were not incorporated. Just before the beginning of the fifth century, a royal
tomb was placed beneath the floor of a palace on Calakmul's Main Plaza
(Folan et al. 1995; Pincemin 1994). The earliest dated Calakmul stela (at
AD 431) depicts a lord who was a contemporary ofTikal's ruler Stormy Sky
and Copan's "first" ruler Yax K'uk Mo' (Marcus and Folan 1994: 22).
Calakmul formed so many alliances and incorporated so many provinces
that it became the Maya capital most frequently mentioned on monuments
outside its region. Part of Calakmul's early success was undoubtedly due to
its having been one of the earliest polities of the Maya Lowlands to achieve
statehood; it was expanding against neighbors still organized on a chiefdom
level.
Calakmul reached new heights during the reign ofJaguar Paw, who ascended to the throne in AD 686. Jaguar Paw led Calakmul on a new expansionist campaign based on two strategies: threatening attack and establishing
political and military alliances with neighbors who shared an antipathy toward Tikal (Folan et al. 1995; Marcus 1988; Marcus and Folan 1994; Schele
and Freidel1990). Such alliances, aimed at keeping a major rival at bay, were
typical of archaic states in many parts of the world (Marcus 1983a, 1989,
1992c, 1995b). Calakmul's political clout was supposed to have ended at
AD 695, when Jaguar Paw was reportedly taken captive by Tikal. Although
oW
f----------r----,--,---------,------,------,----------,------,-----,-------,,---,-------,----
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
Figure 3. 1. The DynamicModel as originally applied to the Maya Lowlands (redrawn, with modifications, from Marcus 1993a:.fig. 26).
(Marcus 1994; Sharer 1991: 196). At the very moment that cities like Tikal
were collapsing, others in Belize (such as Lamanai) and in the Puuc region
of the Yucatan Peninsula were ascending to their greatest heights (Chase and
Rice 1985; Pendergast 1985, 1986; Sabloff and Andrews 1986; Sabloff and
Tourtellot 1991). And when some cities in Belize and the Puuc began to decline, there was a new surge in state formation and expansion, particularly
in northern Yucatan-first at Chichen Itza (AD 800 -1000) and later at
Mayapan (AD 1250 -1450). Our notion that Maya civilization had totally
collapsed is due to the fact that when the Spaniards arrived in the 1530s, the
Mayapan state had fragmented into 16 autonomous provinces (Quezada
1993; Roys 1957), and many areas of the Peten had dispersed rural populations. Much of the Postclassic population had relocated to the coasts and
lakes.
My graph of these cycles (fig. 3.1) emphasizes the cyclic buildup and
breakdown of Maya polities, but the real pattern is too complex to be conveyed by one graph. It would require a series of graphs to show, for example,
that as the chiefdom at Nakbe decreased in importance around AD 100, El
Mirador reached new heights. When El Mirador collapsed around AD 200,
Calakmul-its principal rival-rose to greater prominence. After AD 378,
Tikal rose at the expense ofUaxactun, During this same period Kinal, a city
subordinate to Tikal, rose at the expense of Rio Azul, another of Tikal's
subordinates (Adams 1990). Just as Calakmul's rise came at El Mirador's
expense, so later did Tikal's rise come at the expense of other sites (Folan
et al. 1995; Marcus 1976a, 1988, 1992a, 1993a).
64 I JOYCE MARCUS
a battle with Jaguar Paw is claimed by the ruler of Tikal (as expected, this
event is not mentioned at Calakmul), the person shown as prisoner is not
Jaguar Paw, but a Calakmullord with a different name (Marcus 1976a:52,
1988 :5; Marcus and Folan 1994: 23; Schele and Freidel 1990: 206 -7).
Were it not for this claim by Tikal, we would have interpreted the period from AD 600 to 810 as an uninterrupted, nonviolent time for Calakmul's rulers. Even after Jaguar Paw's battle with Tikal, Calakmul continued
to erect buildings and carve more than 30 stelae. But around AD 850, Calakmul's government failed and the city lost substantial population. In addition
to its political and economic problems, Calakmul may have faced a severe
drought at this time (Gunn, Folan, and Robichaux 1995). Many cities that
continued to flourish were near the rivers and lakes ofthe Peten or the coasts
of Yucatan, and some were nucleated and quite unlike the sprawling, dispersed cities of the Classic period.
Cycles in the Tikal Region
Tikal's transition from a chiefly center to the capital of an archaic state apparently gained momentum with its subjugation of Uaxactun in AD 378. Its
collapse as a regional state occurred after AD 800. During the course of its
history, Tikal displayed two types of dynamic cycles that can be detected
archaeologically.
The first cycle is a clear, but incompletely documented, jockeying for
power by the sites in Tikal's regional hierarchy. As the result of this jockeying, certain sites rose or fell in the settlement hierarchy. For example, Dos
Pilas was founded as a subject polity by a relative of Tikal's ruling dynasty;
later, it became a secondary center and succeeded in breaking away to
achieve independence from Tikal. Still later, Dos Pilas became an expansionist capital itsel
Another example of this jockeying would be Rio Azul, which apparently began as an autonomous polity but was conquered by Tikal around
AD 380. Rio Azul was for a time a tertiary site in Tikal's realm; later, after
the imposition of a Tikal lord at Rio Azul, the latter site rose to become
a secondary administrative center in the Tikal state (Adams 1990). Such
cycles within Tikal's realm will, however, remain poorly known until its
huge hinterland has been systematically surveyed. Obviously this kind ofsurvey will take considerable effort, since Tikal's state may have covered more
than 20,000 km 2 at AD 500, and more than 10,000 km 2 at AD 700 (Adams
1990; Adams andJones 1981; Culbert et al. 1990; Marcus 1992a:407).
Better known than the political cycles within its realm are the cycles of
monumental construction at Tikal itself, which indirectly reveal the fortunes
of the capital. Chris Jones (1991: 121) has noted that large-scale construction projects were completed at Tikal on the order of every 200 years, and
that at least three were associated with the burials of rulers. Two of seven
projects were related to breaks in the dynastic sequence, and in both cases
the former rulers had successors who "took pains to re-establish their connections with the broken dynastic line while also honoring the intruder"
(jones 1991: 122).
The first major construction project at Tikal-the initial stage of the
North Acropolis-took place around 400 BC when a platform was built on
top of a natural hill (Coe 1965: 7- 8); it was rebuilt around 200 BC. Shortly
after AD 1, the Great Plaza was paved for the first time. By AD 75, the center ofTikal had developed its characteristic pattern-an acropolis (containing the burials of important leaders) that supported a large temple on the
north, a multiroom structure on the south, and one or more flanking pyramids. In AD 200 a broad stairway was built to allow direct access from the
North Terrace to the top of the acropolis for the first time.
Hieroglyphic inscriptions personalize the history of Tikal, indicating
that 29 rulers reigned there. The first ofthese, possibly inaugurated in AD 170,
may have been known asJaguar Paw (jones 1991: 109). Excavations by Laporte and Fialko (1990) in the Mundo Perdido complex have yielded Stela
39, dated to AD 376, bearing the name of Tikal's ninth ruler, also named
Jaguar Paw. We know little about Tikal's first eight rulers, owing to a lack
of contemporaneous texts; however, inscriptions at subordinate sites like
Uaxactun suggest that Tikal's rulers occasionally used force to subjugate other
centers (Marcus 1976a, 1992a; Mathews 1985; Schele and Freidel 1990;
Valdes and Fahsen 1995).
The ninth Tikal ruler, who might have taken the name Jaguar Paw in order to link himself to the dynasty's founder, may have been the first to construct a recognizable palace at the site. A text found on a vessel in an Early
Classic palace on the North Acropolis states that the palace was "his house,
Jaguar Paw, Ruler ofTikal, 9th ruler" (jones 1991: 111). This is a significant
event in light of the importance of the palace for identifying archaic states
(see Flannery, this volume).
Around AD 400, a ball court and the antecedent stages of Temples I and
II were built. Tikal's prosperity (and its royal dynasty) seems to have continued, even though Lord Water in his inscriptions at the rival city of Caracol
claims to have had battles with Tikal in AD 556 and 562 (Chase and Chase
1987). The ruler ofTikal at this time was named Double Bird (21st ruler in
the sequence), and his fate is unknown in spite of claims that he died in the
battle with Caracol. Haviland (1991: 5 - 6) suggests that Double Bird survived his battles with Lord Water, to die a natural death at Tikal years later.
Animal Skull, the 22nd ruler, was responsible for massive remodeling in
downtown Tikal, including a new paving of the Great Plaza. Opinions differ on whether this building program reflects a peaceful transition from
66 I JOYCE MARCUS
The Copan Valley of western Honduras has one of the longest occupation
sequences within the Maya area. As far back as 1000 BC, it had village cultures whose burial ceramics show ties with the Mexican Highlands (Fash
1982, 1991; Marcus 1993b, 1995a). Between 300 BC and AD 200 the valley
went through a period with little or no population growth (Fash 1991). This
was only one of the "fits and starts" the region seems to have experienced
over a 2000-year period (Sharer 1995).
From AD 200 - 400, the city of Copan witnessed a surge in population
and the establishment of strong ties to both the highlands and lowlands of
Guatemala. I suspect, on the basis of incomplete data, that Copan's initial
state formation took place in the Early Classic around AD 400, followed by
68 I JOYCE MARCUS
Chichen ltza
El Mirador
/Seibal /
PRECLASSIC
CLASSIC
TU~m
POSTCLASSIC
buffer zone- combined with evidence for raiding, burning ofpublic buildings, sacrifice of enemies, and one possible fortification wall-suggests a
high level of competition among several chiefdoms (Marcus and Flannery
1996: 123-24).
During the next stage of the Middle Preclassic (500-300 BC), a number
of striking demographic changes took place. A large number of villages in
the northern part of the valley (including the largest community of the previous period) were abandoned, and their populations apparently moved to
the top of a defensible mountain in the former buffer zone. The initial population of this first Zapotec city has been estimated at 5,000 persons, and between 300 and 100 BC it rose to an estimated 17,000 (Blanton 1978). During
this period Monte Alban, as the city is called today, built 3 km of defensive
walls and surrounded itself with 154 valley-floor communities (Kowalewski
1983). It appears that this dense ring of satellite communities was a deliberate concentration of farmers and warriors needed not only to support
Monte Alban with food (Nicholas 1989), but also to give it the necessary
demographic advantage to subdue rival chiefdoms in the southern and eastern parts of the valley (Marcus and Flannery 1996: 155 - 71).
Rival chiefly centers in the southern and eastern valley (one of which is
currently undergoing excavation by EIsa Redmond and Charles Spencer)
grew larger during 300 -100 BC, and in some cases sought higher and more
defensible locations. Eventually, however, they fell to Monte Alban, as did
some areas 50 to 60 km outside the valley. As early as 500 - 300 BC, Monte
Alban had erected a gallery of more than 300 carved stones, each displaying
a slain prisoner. By 100 BC-AD 200, the city had carved another 40 stones
naming subject provinces up to 150 km distant from Monte Alban (Marcus
1976b, 1980, 1984, 1992c; Marcus and Flannery 1996:figs. 234-36).
Taking advantage of the fact that it was now organized as a statewhile most of its neighbors were not-Monte Alban expanded to control
20,000 km 2 of southern Mexico by AD 200 (fig. 3.3). Its ultimate goal seems
to have been to establish a "corridor of influence" between Cuicatlan in the
north (the gateway to Central Mexico) and Tututepec and Chiltepec in
the south (the gateway to the tropics). Some subject areas, like the Valleys
of Miahuatlan and Ejutla, may have been taken over by political and economic persuasion (Feinman and Nicholas 1990, 1993). Other areas, like the
Cuicatlan Canada, were clearly kept in line by military force (Redmond
1983; Spencer 1982).
As so often happens in the cases I examine in this chapter, the Zapotec
state was most territorially extensive early in its history. It reached its limits
in the Protoclassic and by the Early Classic (AD 200 -500) had already begun to shrink. One by one its outer provinces grew strong enough to break
away, just as Quirigua had broken away from Copan (see above). During the
70 I JOYCE MARCUS
/ t("<v
/ /
<>.~ / <"
V /.
'/'
--------
/;'<v /
/1,.0/
",/"/ .
.
. '.
Quiafepec
.rtressot.....
"."
. ICATLAN
ANADA
'
..
.'
..
went a period of decentralization not unlike that seen during the "troughs
of dissolution" in the Maya Lowlands. Centers like Zaachila, Cuilapan,
Macuilxochitl, Lambityeco, Yagul, and Mitla arose to control certain sectors
of the valley, but no city ever again dominated the entire region the way
Monte Alban had during its heyday (Flannery and Marcus 1983b: 184). The
carved stone monuments of AD 600 -900 do not feature military conquests;
rather they seek to establish the genealogical credentials of each town's rulers
(Marcus 1976b, 1983d, 1992c). The Postclassic seems to have been a period
during which small principalities jockeyed for position, not unlike the situation revealed in the Mixtec codices (Smith 1983, 1994). Some Zapotec
princes sought to build power by marrying princesses from the Mixteca,
even going so far as to provide land for the tay situndayu, or agricultural
workers, who labored on the estates of Mixtec rulers (Marcus and Flannery
1983: 220).
In summary, the Zapotec show us one more case of a state that
(1) formed when one chiefdom succeeded in subduing its rivals; (2) took advantage of its superior political and military organization to expand against
less-developed neighbors; (3) reached its maximal territorial size early in its
history; and (4) later lost its outer provinces and began to contract, even as
its capital city continued to gain population.
Mouth of
Rio Verde
Figure 3.3. At itspeak in thefirst orsecond century AD, the Zapotec state
may have covered 20,000 km2 (redrawn from Marcus and Flannery
1996: ill. 242).
Early Classic there were 1,075 occupied places in the Valley ofOaxaca, but
almost two-thirds of the valley's estimated 115,000 persons lived at 38 communities in fortified or defensible localities (Kowalewski et al. 1989). No
longer the expansionist aggressors, the Zapotec were now busily defending
themselves against hostile neighbors. After reaching its peak size of 24,000
persons between AD 500 and 700, Monte Alban began to lose population
and importance. By AD 900, Jalieza, a mountaintop city in the southern valley, was the region's largest community (Finsten 1995).
During the Terminal Classic and Postclassic, the Zapotec state under-
As in the case of many pristine states, the first state in the Basin of Mexico
arose in the context of competing chiefdoms. Between 600 and 300 BC
only a few hamlets existed in the area that would become the city ofTeotihuacan, Indeed, prior to the second century BC, that arm of the Basin of
Mexico is considered a "backwater" (Millon 1992: 385). The most impressive chiefly center in the region was Cuicuilco in the southern Basin of
Mexico, which may have had 10,000 occupants (Sanders, Parsons, and Santley 1979).
While Cuicuilcos population had doubled by 100 BC, a sizable community at Teotihuacan was emerging as a rival. Almost certainly taking advantage of the irrigation potential of its springs, Teotihuacan grew to the point
where it virtually rivaled Cuicuilco; the two centers are seen as "balanced
powers in the early first century" (Evans and Berlo 1992: 7). Mother Nature
then played her hand; the eruption of a volcano named Xitle buried
Cuicuilco and its immediate sustaining area with lava. Some scholars believe
that this natural disaster gave Teotihuacan the advantage it needed to grow
rapidly to a position of political supremacy, siphoning off most of the manpower formerly under the control ofCuicuilco. Other scholars are reluctant
to see political evolution attributed to a natural accident.
The first state at Teotihuacan is believed to have emerged around the first
72 I JOYCE MARCUS
century AD. The evidence is largely based on settlement pattern survey and
intensive surface pickup of the emerging urban area; at present we have little
evidence for palaces or standardized temple plans for this early period (Marcus 1983b). Teotihuacan administered a noncontiguous territory estimated
by Rene Millon at 25,000 km 2; George Cowgill's estimate is 50,000 km 2
(personal communication 1996). Teotihuacan seems to have done this not
only by suppressing the development of competing centers nearby, but also
by deliberately bringing in rural populations to create an immense urban
center. One of 'Ieotihuacan's strategies was to extend its principal streets,
particularly the East Avenue, several kilo meters beyond the limits ofthe early
city, suggesting that it intended to bring in thousands of new farmers, artisans, and urban laborers (Charlton 1972, 1991; Drewitt 1967; Sanders
1965). Sanders (1965: 121) and Charlton (1991: fig. 15.2) note that Teotihuacan's suburbs and rural towns were aligned with the city's avenues, including sites 5 km to the south and as far away as 11 km to the east. As early
as AD 200, the city covered at least 20 km 2 . Its maximum population is estimated at 125,000 (Millon 1992: 341).
Like Monte Alban, Teotihuacan expanded its influence into provinces
beyond its immediate physiographic region. It established a secondary center at Chingu, 50 km away in the state of Hidalgo; a "military colony" at
Matacapan in Veracruz; enclaves of some kind at Mirador and Los Horcones
in Chiapas; and perhaps an elite enclave at Kaminaljuyu in the Guatemalan
Highlands more than 1,000 km away. Teotihuacan controlled the nearby
Otumba obsidian source; virtually monopolized the Sierra de las Navajas
obsidian source near Pachuca, Hidalgo; and may have had access to the El
Chayal obsidian source near Guatemala City through its enclave at Kaminaljuyu (Sanders 1974). Any designs it may have had on the Guadalupe Hidalgo obsidian source near the Puebla/Veracruz border, however, may have
been thwarted by Cantona, a non-Teotihuacin-style city on a fortified lava
flow near Guadalupe Hidalgo (Garcia Cook 1994).
Teotihuacan's area of"direct control" was irregular (Millon 1993: fig. 7),
circumventing areas powerful enough to resist. There can be little doubt,
however, that part of Teotihuacan's expansion was by force; under one of its
principal temple pyramids, more than 200 sacrificed individuals (most wearing military gear) have been found (Cabrera Castro, Sugiyama, and Cowgill
1991; Sugiyama 1989).
Like many early states examined in this chapter, Teotihuacan's greatest
expansion was early in its history. As time went on, it lost more and more
of its distant colonies, and by AD 600 the city itself was declining. With its
collapse by AD 650, its downtown area was burned, followed by a 200-year
period of decentralization. Between AD 750 and 1100 cities like Tula in Hidalgo, Xochicalco in Morelos, and Cacaxtla in Tlaxcala rose to prominence.
Aztec empire
Teotihuacan
state
Chiefly center
at Cuicuilco
XOChic:a.lco
expands
I I I
(B. c.) 500
I I I I I
I
500
I I I I
1000
I
1500 (A.D.)
All these cities lay close enough to the Basin of Mexico to have had their
growth suppressed by Teotihuacan during its heyday. Now, with Teotihuacan reduced by two-thirds, they became major urban centers in their
own right (fig. 3.4).
One of the most interesting of these second-generation states was the
one centered at Tula in Hidalgo. With Teotihuacan's collapse, the administrative center at Chingu had ceased to be important and Tula began to grow
between AD 800 and 1000 (Healan 1989). Tula became the capital of the
Toltec state, the most powerful polity in Central Mexico in its day. By
AD 1168, however, it too had collapsed, with many ofits noble families emigrating to the Basin of Mexico.
Mter a period of decentralization and considerable jockeying for power,
a third-generation state arose at Azcapotzalco in the Basin of Mexico. This
was the Tepanec state, which by AD 1300 controlled a number of less
powerful centers such as Texcoco. The hierarchy of settlements below
Azcapotzalco consisted of at least three tiers, with tribute passing from
Teotihuacan (now only a modest town) to the larger town of Huexotla,
from there to Texcoco, and then up the line to Azcapotzalco (Gibson 1964;
Hodge 1984).
Among the ethnic groups serving as vassals to Azcapotzalco were the
Mexica, the people we call the Aztec. In AD 1427 the Mexica lord Itzcoatl
allied himselfwith other disgruntled tribute-paying groups, such as the Texcocans and the Acolhua, and succeeded in overthrowing Azcapotzalco. The
Mexica then formed a political and military alliance with Texcoco and Tlacopan to create what might be considered a fourth-generation state that
dominated Central Mexico at the time of the Spanish Conquest (Berdan
1982; Bray 1968; Davies 1973; Hodge 1984; van Zantwijk 1985). This state,
like others before it, expanded rapidly during its first 90 years; but because
of the arrival of Europeans in AD 1519, its development was abruptly truncated, and it never went through the usual phase of breakdown.
74 I JOYCE MARCUS
While many Aztec cities have been referred to as "city-states" in the literature, they did not closely resemble the Greek poleis from which that term
is taken. Nor do they resemble the modem Vatican "city-state" in which the
city and state are isomorphic. Aztec cities were in fact closely tied together
by royal marriage alliances, military alliances, the sharing of tribute, and
various types of economic interdependence (Barlow 1949; Carrasco 1971,
1974, 1984; Davies 1973). In the future it might be better to use altepetl, the
indigenous Nahuatl word for such polities, rather than the term "city-state,"
because they are not like the autonomous Greek poleis (Charlton and
Nichols 1997; Hodge 1984; Marcus 1983b).
Inca
First 4-tiered
hierarchy
Wari
state
Moche
state
Competing
chiefdoms
(B.C.) 1000
I I I I
500
I I
Chimu
state
I I I I
500
I I
I I I I I I
1000
1500 (AD.)
76 I JOYCE MARCUS
ECUADOR
N
R. Motupe
R. de la Leehe
_ _~R.Lambayeque
~~-=- R. Zafia
R. Jequetepeque
R. Chieama
R. Moehe
R. Casma
R. Culebras
R. Huarmey
R. Fortaleza
R. Pativilea
Chirrui Expansion
R. Rimae
1st stage
PERU
2nd stage
100
200 km
I
Figure 3.6. The expansion of the Chimt: state involved at least two major
stages, with possibly three phases of consolidation during the second stage cif
expansion (redrawn from Mackey and Klymyshyn 1990:.fig. 1).
The Near East is yet another region where primary states formed after a long
period of competition among chiefly societies. As far back as the Samarran
phase (5500-5300 BC [uncalibrated]), competition between large villages
was so intense that some featured ditches and defensive walls, watchtowers,
and piles of sling missiles (El-Wailly and Abu al-Soof 1965; Gates 1969).
During Late 'Ubaid, sometime around 4000 BC, the archaeological record
78 I JOYCE MARCUS
has examples of raided villages, with elite residences burned and unburied
corpses left behind (Tobler 1950: 26).
While the details of their formation are still not known, the first archaic
states in Mesopotamia (Iraq) and Susiana (Iran) seem to have flourished between 3700 and 3200 BC (Adams 1981; Adams and Nissen 1972; Algaze
1993a; Johnson 1972, 1987; Wright, this volume; Wright and Johnson
1975). By Early Uruk times in southern Mesopotamia, and by Middle Uruk
times in Susiana, settlement pattern surveys and site-size histograms reveal a
four-tiered hierarchy of cities, towns, large villages, and small villages (Johnson 1973,1980).
The early state centered on Susa in southwestern Iran seems to have had
a particularly turbulent history. Collapse of one or more chiefdoms at the
end of the 'Ubaid period (c. 4000 BC) seems to have been followed by a period of economic and political reorganization during Early Uruk times. By
Middle Uruk (c. 3700 BC), Wright and johnson have detected the presence
of an early state with centralized craft production, administered local exchange, administered labor, and a four-tiered settlement hierarchy (Johnson
1987: 107). At its peak, this early Susiana state seems to have had outposts
like Godin Tepe, situated 250 km away in the Zagros Mountains (Young
1969).
During the Late Uruk (c. 3400 BC), the early Susiana state fractured into
two physically autonomous and hostile entities, one centered at Susa and
the other at its former subject community, Chogha Mish. Johnson (1987:
124) suggests that Chogha Mish broke away from Susa in order to retain sole
use of the land, water, and labor in its region. This split may have been similar to Quirigua's "battle for independence" from Copan in the Maya Lowlands.
The early state centered on the city of Uruk (Warka) in southern Mesopotamia was even larger and more territorially extensive than the one centered on Susa (fig. 3.7). The city ofUruk may have reached 200 ha by Late
Uruk, and 400 ha before the end of Early Dynastic I (Adams and Nissen
1972). Mter this point (c. 2700 BC), however, Uruk suffered a fate similar
to that of Susa in the Late Uruk period; its ascendancy was "increasingly
challenged by the growth of a number of rival cities like Umma, Shuruppak,
Zabalam, Bad-Tibira, and others" (Adams and Nissen 1972: 19). Possibly
these rival cities contributed to the collapse of the Uruk state by appropriating much of its former land, water, and labor.
At its peak, some believe, the Uruk state controlled an area stretching
from Ur to Abu Salabikh; increasing evidence shows that its "influence" extended far beyond. "Colonies," "outposts," or "enclaves" with ceramics,
tablets, administrative seals, and/or architecture in Uruk style have been
found as far up the Euphrates and Tigris Rivers as eastern Turkey (Algaze
TAURUS
MOUNTAINS
Arslantepe
Hassek
Habuba
Kabira
SYRIAN DESERT
I
0
100
200 km
Ur
Figure 3.7. The total area directly controlled by Mesopotamia's Uruk state
has not been determined, but its "outposts" or "colonies" extended asfar as
eastern Turkey (data from Algaze 1993a; Gates 1993).
80 I JOYCE MARCUS
Uruk
state
Akkad
Periodic
-,
attempts at
consolidation
Competing
Ubaid
chiefly
centers
I \
IsinLarsa
Early Dynastic
I I
I I
4000
I I I
I I I I
3500
I I I I
3000
I I I I
2500
2000
I (Years
B.C.)
1989, 1993a). Some "colonies," like Habuba Kabira South, are heavily fortified (Strommenger 1980). More distant" enclaves," as at Hacinebi, are not
(Stein et al. 1996). The parallels with Teotihuacan are clear: a "hyperurban"
capital, direct control of an irregular and noncontiguous territory, and distant "colonies" or "enclaves," some ofwhich were peacefully integrated into
the surrounding region, while others were not (for a range of such sites, see
Boese 1989-90; Frangipane and Palmieri 1988; Gates 1993; Stein et al.
1996; van Driel1983; Wattenmaker 1990).
There are divergent interpretations of these Uruk "colonies" or "outposts" on the upper Tigris and Euphrates. Gates (1993) believes that they
reflect a short Late Uruk presence (100-150 years at most), and that their
main purpose was to acquire resources that were unavailable in southern
Mesopotamia. Johnson (1988-89), on the other hand, suggests that at least
some Uruk "colonies" or "outposts" may have been founded by the refugee
elites oflosing factions in Middle or Late Uruk power struggles, accompanied by their political followers.
Whatever the case-and it will probably take decades to resolve the details of the Uruk expansion-it is clear that Mesopotamia's first state had a
demographic and political impact on an area roughly 1,000 km from northwest to southeast. This is approximately the distance between Teotihuacan
and its "elite enclave" at Kaminaljuyu. Like the Monte Alban, Copan, and
Teotihuacan states, the Uruk state first lost its more distant outposts, then
lost its grip on more nearby areas (fig. 3.8).
As this first-generation state collapsed, southern Mesopotamia entered
a 650-year period of decentralization known as the Early Dynastic (30002350 BC). This period was characterized by smaller polities, autonomous or
semiautonomous, to which a number ofterms have been applied. DiakonofI
(1974) refers to them as "nomes," using the word the Greeks applied to the
districts comprising the Egyptian state; Adams (1981) refers to them as "enclaves." YofIee (this volume) and others refer to them as "city-states."
The appropriateness of using the term "city-state" could be challenged
by data from the Kingdom of Lagash, one of the best documented ofsouthern Mesopotamia's Early Dynastic polities. It contained at least three cities
more than 20 towns, and at least 40 villages (Adams 1965: 101). Diakonoff
(1974) has estimated its population of "free citizens" as more than 100,000.
Its two largest cities, Lagash and Girsu, had populations conservatively estimated at 36,000 and 19,000, respectively; since Lagash alone covers 600 ha,
the figure could be twice as high. Although the total area controlled by the
Kingdom of Lagash has not been determined, its irrigated fields alone are
thought to have exceeded 2,000 km 2 , and it had a 45 km frontier with the
rival Kingdom ofUmma. This sounds like something beyond a "city-state."
Around 2350 BC, the ruler Sargon of Akkad succeeded for a time in
unifying all of southern Mesopotamia (fig. 3.9). Because of earlier rulers'
attempts to extend their hegemony (see below), Sargon's state must be considered "third generation." At their peak, Sargon and the later members of
his dynasty exerted control over a large area, from Tell Brak and Chagar
Bazar in Syria to Ur and Lagash in southern Iraq (Postgate 1992: fig. 2: lla).
Around 2250 BC, the Akkadian dynasty lost control of Mesopotamia.
Nomadic Gutians from the Zagros Mountains took advantage of the Akkadian collapse to loot unprotected cities. Not long after this, however, the
rulers of the Third Dynasty ofUr managed to step into the power vacuum
and establish their own hegemony over the area (fig. 3.10). Ur had spent
much of the previous 200 years as a subject province of Akkad and other
polities; between 2222 and 2118 BC, it became the capital of a fourthgeneration regional state. The territory controlled by Ur, however, was
only about half that claimed earlier by the Akkadian dynasty; for example,
it fell hundreds of kilo meters short of reaching the areas ofBrak and Chagar
Bazar (Postgate 1992: fig. 2: 11b).
In other words, Mesopotamia shows the same dynamic cycles of consolidation and collapse seen in Central Mexico and the Maya region, and the
same tendency for subject provinces to become the core provinces of later
states. When the Third Dynasty ofUr lost power, it was followed by a 260year phase of decentralization known as the Isin-Larsa period (fig. 3.11).
From 1760 BC, a fifth-generation regional state had arisen under Hammurabi with its capital at Babylon (Gates 1986; Postgate 1992:fig. 2: 11d).
YofIee's analysis provides a detailed explanation for the eventual collapse
ofthis fourth-generation Mesopotamian state (YofIee 1977). It would appear
that one ofHammurabi's mistakes was the continual bypassing oflower-level
administrators in the hierarchy, a process that archaeologists with an interest in systems theory have called meddling (Flannery 1972). Following Hammurabi's unification of Mesopotamia by conquest, local administrators were
bypassed and higher-level state administrators were assigned the duties of
policing agriculture and fishing. Taxes were collected and most land was entailed to the Babylonian state. The Babylonian state's policy of co-opting
'!j
g-
if
l::.
\C
"
c
....
'"
3
o
o
I\)
o
o
;;
19
cl
29 km
F:"
0(')0
---
Co
~~.
=tco-t'
(')
+
N
5l
0-
Figure 3. 10. Piotr Steinkeller's reconstruction of the area affected by the Third Dynasty of Ur. During thisperiod of consolidation, the "core area"of the statewasalluvial southern Mesopotamia. A larger area, from theplains
of Nineveh to the plains of Susa, was incorporated as the "periphery" of the Ur III state. Beyond that was an
even larger region of "vassal polities" in the mountains of Iraq and Iran (redrawn, with modifications, from
Steinkeller 1987 :.figs. 1 and 6).
~z
c
2
;>
.........
?/>'
c~
~'"':l
""~
N~
iX'l ~
;:s
"
s,'"
'"
'"
..... ~~
13:
;:s
-s,
oB"
l::.
:;;~
""I::...
~~
c ?/>'
'"
;:s
l:: ~
~~
~~
~.
'" 8;:s
S
..g'" '"
l::.
s ~
~~
;:s "
~"'"
g-.
\c
:s;:s~~0<;";:
e Qatna
/
UrUK e.
100
eUmma
eGirsu
Larsa
Ur e
local authority in newly conquered territory served to sow the seeds of its
own destruction (Yoffee 1977: 148). In the Old Babylonian period this
highly centralized state was administered from the capital of Babylon; because it proved to be highly oppressive to its subjects, conquered territory
quickly decomposed into its constituent elements, each of which reestablished its local autonomy (Yoffee 1977: 143). The empire built by Hammurabi experienced a long decline, not through the personal inefficiency of
succeeding rulers but because Hammurabi's system was too rigid and oppressive to adapt.
The highly centralized state had initially become prosperous from the
tribute and taxes exacted from far-flung provinces. The Babylonian state
then undertook massive building and waterworks projects. When the southern, western, and eastern provinces were able to break away, however, the
resources needed by the state to maintain its ambitious building projects
were severely reduced.
As the late Babylonian period progressed, bureaucrats not directly controlled by the state tended to emphasize loyalties to the groups from which
they had been recruited. The state was overthrown by a Hittite king, and
the bureaucracy reverted to the position of a locally based aristocracy, while
the institutional framework of the political system collapsed. The ruling
house in Babylon fell victim to its long-standing oppressive policies and the
resulting internal instability (Yoffee 1977: 149).
Whereas most Mesoamerican field archaeologists think of the "peaks" of
state consolidation as normal, and the "troughs" of dissolution as temporary
aberrations, the situation in Mesopotamia is somewhat different. There,
most epigraphers think of the periods of decentralization and provincial autonomy as normal, and the peaks of unification as rare events. The reason, I
believe, is that written documents first become abundant and decipherable
during the Early Dynastic, a 650-year period of decentralization. It stands to
reason that epigraphers-not focusing on the fact that Early Dynastic decentralization followed the breakdown of an earlier, influential Uruk statemight come to think of small autonomous polities as the normal situation.
What is often unappreciated, however, is the fact that the Early Dynastic witnessed countless attempts by the rulers of various enclaves to subjugate their neighbors. Some of these attempts temporarily succeeded in
reunifying parts of Mesopotamia. For example, the Early Dynastic ruler
Mesalim ofKish "exercised hegemony far beyond the walls of his own city"
(Cooper 1983:7). Calling himself "lugal [king] of the universe," Mesalim
left royal inscriptions as far away as Girsu in the Kingdom of Lagash; in fact,
he is credited with erecting a monument that established the border between
the provinces of Lagash and Umma. In this act he reminds us of the Maya
86 I JOYCE MARCUS
ruler Smoke lmix of Copan, who also set up boundary stelae to delimit his
"realm."
+
White Crown
(Upper Egypt)
Red Crown
(Lower Egypt)
Double Crown
(Unified Egyptian
state)
Figure 3. 12. When the Upper Egyptian chiefdom conquered the Lower
Egyptian chiefdom, the ruler if the newly unified Egyptian state wore a
"double crown" combining the white crown if Upper Egypt with the red
crown if Lower Egypt (redrawn from Emery 1961 :.fig. 68).
Egypt was then established at Memphis, strategically located near the former
border between the two chiefdoms (Kees 1961). Memphis was fortified, and
it is probably there that we see the first Egyptian palaces (although Emery
[1961 : 178] has argued that the palace might predate the founding of Memphis since a serekh, or palace facade, is present on the Narmer Palette).
The early Egyptian state was divided into districts, known as hesps in
Egyptian and nomes in Greek. Each nome had its own specific customs, traditions, and patron deity (njwty, "city or local god"). Greek historians refer
to the rulers of each nome as nomarchs and make it clear that these provinciallords had considerable power. We might see this situation as analogous
to that of the Lowland Maya, where the capital was run by an ahau but his
subject provinces were run by batabob (Marcus 1993a: 119-20).
By 2400 BC there were 22 nomes in Upper Egypt. The number ofnomes
in Lower Egypt varied over time, but eventually there were 20 (Silverman
1991 :38). The nomarch of each province had important administrative and
decision-making powers, and nomarchs sometimes formed alliances with
one another, thereby becoming powerful rivals of the pharaoh. Indeed, the
history of the early Egyptian state shows periods of strong centralized rule
alternating with periods ofpharaonic weakness; during the latter, the nomes
had a great deal of autonomy. So similar is this Egyptian pattern to that in
Mesopotamia that Diakonoff (1974) applies the term "norne" to Sumerian
provinces like Umma, Lagash, Ur, and Kish.
Egyptologists have recognized this cyclic pattern of consolidation versus
88 I JOYCE MARCUS
Unification
of Upper &
Lower Egypt
Late
Predynastic
-,
Middle
New
Kingdom Kingdom
Old
Kingdom
First
Intermediate
Kushite or Nubian
unification
Third
Intermediate
Secdnd
Intermediate
I I
I I I I
I I I I
I I I I
I I I I I I I I I I I I
3500
3000
2500
2000
1500
1000
I
500
I I (Years
B.C.)
90 I JOYCE MARCUS
Mycenaean
conquest
Minoan
Mycenaean
palaces
palaces
2000
I I I I I I I I
1500
1000
I I I I I I
500
I
o
(Years
B.C.)
focus of each community was its temple, rather than its palace (Snodgrass
1977:7-8).
Fine (1983) stresses the small size of the polis; most averaged no more
than 5,000 free male citizens. Conflict and competition among poleis were
commonplace, reflecting a degree of decentralization. However, larger units
could be created through mechanisms such as marriage alliances between the
elites of several poleis, political confederacies, and military confederacies in
which two or more poleis joined forces in order to subjugate other peoples,
temporarily expanding the territory from which they received tribute.
Two large units established in this way belonged to Sparta (which at one
point controlled 8,300 km 2 after absorbing Laconia and Messenia) and
Athens (which at one point controlled 2,800 km 2 after absorbing Attica and
Salamis). It should be stressed that these territories were made possible only
by alliance building and conquest; other poleis usually controlled as little as
80 km 2 to 1,300 km 2 and had relatively low population densities. Such territories are as small as those controlled by the cacicazgos of Mexico's Postclassic Mixteca (Spores 1967), and it is probably inappropriate to consider
them "states."
During the fifth century BC, most Greek poleis belonged either to the
Peleponnesian League under Spartan hegemony or the Delian League/
Athenian Confederacy under the control of Athens. During the third century BC, some confederacies of poleis (such as the Achaean League) had attracted as many as 60 cities into their orbit (Snodgrass 1980). Such moments
of widespread unity, however, were "transient" (Hammond 1986); they required alliances among normally autonomous small units whose only hope
of overcoming their small population sizes was through confederacy. Like
the aforementioned cacicazgos ofMexico's Mixteca, the curacazgos ofLate Intermediate Peru, or the seiiotlos that existed in the Postclassic Basin of Mexico before Azcapotzalco's hegemony, the Greek poleis represent a social
form that is unlikely to represent a first-generation state for its region.
CONCLUSIONS
We have now looked at seven series of archaic states-the Maya, Zapotec,
Central Mexican, Andean, Mesopotamian, Egyptian, and Aegean. A number of common patterns emerge, one of which is the cyclic unification and
dissolution predicted by the Dynamic Model. It should be stressed that my
seven cases do not include all archaic states, only a subset for which archaeological data are available; there is no guarantee that the patterns seen in those
seven cases are universal. I also realize that I have not distinguished between
states involving only one ethnic group (such as the Maya) and "empires"
92 I JOYCE MARCUS
that involved the conquest of foreign peoples (such as the Aztec or Inka). I
have not done this here because the differences between "state" and "empire" could fill a chapter of equal length.
Bruce Trigger (1993: 8) has recently contrasted large, unitary "territorial" states with the clusters of smaller polities sometimes referred to as "citystates." I agree with the distinction that Trigger is making, but I suggest that
these two categories were often different stages in the dynamic cycles <if the same
states, rather than just contrasting sociopolitical types. Central Mexico, for
example, gave rise to both of Trigger's types, but the clusters of "city-states"
were always the breakdown product of an earlier unitary state. I join many
colleagues in believing that we should avoid the term "city-state" whenever
possible, substituting instead a more appropriate regional or indigenous
term such as cuchcabal, ahaulel, altepetl, hesp, nome, cacicazgo, curacazgo, or
seiiorio.
I find it significant that in every case I have examined, the first state to
form in a region was unitary in Trigger's terms. I suspect that it could happen no other way. In a recent review of social evolution, Flannery (1995)
has examined two non-Western states whose origins are known in detail: the
Hawaiian and Zulu states. In neither case did a group of chiefdoms grow
into a group of "city-states." Rather, the state formed in the context of a
group of competing chiefdoms when one of those chiefdoms succeeded in
subjugating its neighbors, turning them into the provinces of a larger, unitary state. Elsewhere Flannery and I have presented evidence that the
Zapotec state formed in a similar way, when one member of a group
of chiefdoms succeeded in subduing the others (Marcus and Flannery
1996: 155).
I cannot, therefore, agree with Yoffee's sweeping statement that "all the
earliest states (Mesopotamia, Indus Valley, China, Andes, Maya, Teotihuacan: the exception is Egypt, which is a regional state) are not states at all,
they are city-states" (Yoffee 1995a: 546). In fact, all the pristine states we
have examined-Tikal, Monte Alban, Teotihuacan, Moche, Uruk, Egypt,
and so on-were unitary in Trigger's terms, with their capital cities sometimes placing enclaves as much as 1,000 km away. Nothing resembling a
group of "city-states" appeared in those regions until after the early unitary
states collapsed. Often, it is the newly autonomous former provinces of
those collapsed states that look like a group of "city-states" (Marcus 1989).
And the second-generation states in each region frequently formed when
one of those provinces succeeded in subjugating the others.
I consider it very unlikely that a pristine state could emerge as a group
of "city-states," because I see no mechanism by which a group of chiefdoms
can evolve into a group of tiny statelets. A state is not just a chiefdom that
has received an injection of growth hormones. It is a large polity, each of
whose provinces was a former chiefdom, forged together by deliberate political and territorial reorganization rather than simple growth.
As for the multiple autonomous polities called "city-states," we should
not be seduced into considering all of them states simply because their rulers
drew heavily on the ideology and symbolism of their more powerful predecessors. Both Egyptian and Maya rulers did this extensively, linking themselves to the lords of an earlier age of large states and making it appear that
there had been no "troughs" in their history. As archaeologists we know
there were many consolidations and collapses, but many royal Egyptian and
Maya inscriptions do their best to convince us otherwise (Marcus 1995c;
Rhodes 1986).
Although I am convinced that the Dynamic Model fits these seven cases,
there are interesting differences in the lengths of the dynamic cycles involved. In my review of the cycles of various Mesoamerican states (Marcus
1992a), I pointed out that most unitary states lasted only about 200 years.
The seven cases examined here suggest a much greater range of variation.
Some strong, territorially extensive states, such as that controlled by the
Third Dynasty ofUr, may have lasted only a century. Others, like that of
Old Kingdom Egypt, may have lasted as much as 500 years. Cowgill
(1988: 266) has suggested that the longevity of states may have something to
do with whether they were put together by "subjugation" or "incorporation," with the former being more "decomposable."
A similar range of variation characterizes the troughs of dissolution and
decentralization. Some periods of breakdown, such as that between the dynasty of Akkad and the Ur III state, may have lasted only a century. The relative decentralization ofEarly Dynastic Mesopotamia, in contrast, may have
lasted 650 years. However, as we have seen, this decentralization was interrupted by periods of consolidation under rulers like Mesalim of Kish and
Eanatum of Lagash; thus the Early Dynastic may represent a series of shorter
troughs, rather than a single long one. I think that a similar pattern of short
troughs, rather than one "apocalyptic collapse," followed the breakdown of
the Maya states centered at Calakmul and Tikal (Marcus 1993a).
Finally, it is worth noting that epigraphers and field archaeologists seem
forever destined to view the rhythmic cycles of states differently, In monumental public buildings and rapidly expanding pottery styles, archaeologists
see the rise of the earliest states in their regions. For them the peaks in the
Dynamic Model are the normal state of affairs, and the troughs of dissolution are enigmatic collapses. For epigraphers it is the corpus of texts that
count, whether they come from a peak or a trough in the dynamic cycle.
Mesopotamian epigraphers, for whom the Early Dynastic and Old BabyIonian periods are important sources oftexts , see decentralization as the normal state of affairs. Forgetting how extensive the impact ofUruk states was
94 I JOYCE MARCUS
in Mesopotamia and southwest Iran, they sometimes treat Sargon's hegemony as if it were a first-generation state, rather than a third-generation
phenomenon.
In much the same way, some Maya epigraphers take decentralized moments ofthe Late Classic to be the normal state of affairs, since relatively few
subject lords admit to subordinate status in the inscriptions they commissioned. The result is that when some epigraphers find evidence for the hegemonies that we archaeologists described long ago, they declare such polities
not merely states, but "superstates." I remain convinced that the Dynamic
Model can reconcile those contrasting positions by reminding us that we are
simply looking at different stages in the evolution of states.
As for why the peaks of consolidation inevitably gave way to valleys of
dissolution, I suspect it had something to do with the difficulty of maintaining large-scale inegalitarian structures for long periods of time. Large-scale,
asymmetrical, and inegalitarian structures were more fragile and unstable
than commonly assumed, even though most scholars who work exclusively
on states consider such large-scale structures durable.
4
~(Ole ond ~O(iOI
Larger, higWy centralized and urbanized states, while more visible, are not necessarily the most durable and adaptable of complex political systems.
NOTE
I wish to thank Henry Wright, who gave me extensive comments on this essay, particularly on Mesopotamia; George Cowgill, who supplied insightful and detailed comments
on Teotihuacin and other sites; Gary Feinman, who not only helped me edit all the
chapters in this volume but also did the thankless task of reformatting every paper many
times; and all the conference participants who ensured that the week was so successful.
Most participants arrived in Santa Fe loaded down with slides to accompany their excellent oral presentations, and from those talks I gained new insights into comparative states.
Last, but not least, I thank Doug Schwartz and all the staff of the School of American
Research, because they always create a marvelous environment for open discussions,
Oraanization
ofMixtec
hen I first conceptualized the topic ofscale, settlement pattern, and the
archaic state, I envisioned a rather straightforward comparative discussion of the territorial extent and population size of different preindustrial polities that are generally considered to have sociopolitical organizations
more hierarchical, differentiated, and stratified than middle-range societies
or chiefdoms (Feinman and Neitzel 1984). Now I recognize that such preliminary thoughts were rather optimistic. For the periods prior to written
histories in most geographic regions, few archaeologists have ventured to estimate the exact size or full geographic scope of bounded political domains.
Rough estimates have been attempted for regional and continental populations during these eras; however, little agreement has been reached regarding these postulations, and even less consensus has been achieved concerning
the political configurations and subdivisions of these geographic landscapes.
One can only concur with Elvin (1973: 17), who noted that "the question
of the size of political units seems never to attract among historians and sociologists the attention which it deserves."
Various evolutionary overviews have offered general dimensions for the
demographic size and territorial extent of early states, yet the scarcity of
96 I GARY M. FEINMAN
Less agreement than one might expect exists in the scholarly literature concerning the size and scale of the archaic state. Demography has often been
advanced as a prime mover of organizational change, and population size is
viewed as an implicit indicator of different levels of organizational complexity (e.g., Carneiro 1967; Upham 1987). Nevertheless, no agreement on
the demographic threshold between chiefdoms and states has been reached.
For the ancient Americas, both Sanders and Price (1968: 85) and Upham
(1987:355-56) propose a population cutoff around 10,000 people. Yet, to
their credit, neither views this figure in a rigid manner, as the former recognize that certain chiefdoms may range up to ten times that threshold value
in size, while the latter views population density, as well as size, as a key correlate of organizational complexity. The 10,000 figure also is used implicitly
by Friedman and Rowlands (1977: 220), who see early "Asiatic states" as
roughly that size. Yet Renfrew (1982) views some small states as having had
as few as 2,000 to 3,000 people. Other authors place a chiefdom/state
threshold at a somewhat larger population. Baker and Sanders (1972: 163)
argue that chiefdoms become unstable above 48,000 people. This figure
corresponds roughly with the findings from a sample of 70 middle-range
societies from the Americas in which the maximal estimates for societal
population all fall below 31,000 people, except in a single case (Hispaniola
Arawak) where estimates from 100,000 to 6,000,000 have been proposed.
In contrast, Johnson and Earle's (1987: 314) examples of complex chiefdoms (Hawai'i) range up to 100,000, while their selected examples of
archaic states begin at 14,000,000 (Inka). Flannery (1972), like Lenski
(1966:145-46), describes archaic states as having hundreds of thousands of
people. Although each of these authors may take a slightly different emphasis in their definitions of the state, the intellectual bases of these theoretical
positions stem from closely related scholarly traditions. All refer to the state
as a societal mode more complex than chiefdoms or middle-range societies,
and characterized by a multitiered hierarchy of decision making and a degree of social stratification.
Quantitative estimates of the size of state populations are rare in the
comparative literature. No specific figures are offered in the valuable synthesis by Claessen (1978a: 586), who concludes that states have more than
500 people, "a sufficient population to make social categorization, stratification, and specialization possible." Cross-cultural overviews by Narol1
(1956), Ember (1963), and Carneiro (1967) use maximal community size as
a substitute for other societal demographic parameters. Although Carneiro
(1967: 236) hoped to examine societal populations, "considerable difficulty
98 I GARY M. FEINMAN
the Mediterranean world and in western Asia than found in those areas today. For prehispanic Mesoamerica, Marcus (1992a) proposes that some later
states (investing in infrastructure) were smaller than earlier states in the same
region.
Although Carneiro (1978: 211) has always been aware of this continental-scale cycling in polity size, he prefers to emphasize the larger global trend
toward unification. This logic is somewhat analogous to the common misconception that there has been steady or uniform growth in mammalian
brain sizes because the largest mammalian brains have increased over time.
In actuality, these cranial changes are accounted for almost entirely by evolutionary transitions in select lineages (e.g., primates) of brainy mammals,
which have become brainier. Within many groups of mammals, the most
common brain size has remained unchanged over millions ofyears since the
lineage was established. The range of mammalian brain sizes has increased
over evolutionary time (as brain size in select lineages has increased), yet no
intrinsic force has pushed for larger brains across the entire taxonomic class
(see Gould 1985). Rather, one must examine the specific set of factors that
affected each lineage. Thus, returning to sizes of preindustrial empires, the
scale of the world's largest polities does increase over time. Yet that does not
imply a constant or necessary increase in the size of every state or in the size
of the largest states on each continent. As with the mammalian lineages, the
specific set of factors that influenced the preindustrial polities on each continent or in a specific ancient world would seem to be key.
Although a general worldwide trend toward growth in polity size does
seem apparent, the actual trend is more jagged (with peaks and valleys).
Fluctuations are amplified when the process is examined in specific continents and regions. From Carneiro's (1978) vantage, this inevitable global
progression requires little explication (and so is assigned to innate first principles). Yet if this process is recognized as cyclical and variable (in space and
time), it would seem to require further investigation and explanation focused at the continental and regional scales (see also Chase-Dunn 1990;
Taagepera 1978a). Such investigation should include greater attention to the
constraints, thresholds, mechanisms, and strategies associated with political
integration, expansion, and contraction (see also Marcus, this volume).
entire culture area (Blanton and Feinman 1984), yet, even so, the two
Mesoamerican episodes of relative political centralization were short (encompassing a total of roughly 750 years). In contrast, the two intermediate
periods together endured for twice that duration (Price 1976). In the Andes, a similar pattern is found, as the two horizons are thought to last less
than half a millennium, while the two intermediate periods cover 1,250
years.
This pattern corresponds with Adams's (1978) findings for Mesopotamia,
where episodes of political unification were short and regularly succeeded
by much longer periods characterized by greater local political autonomy.
For the same region, Yoffee (1988b: 59) notes that the ideal of governmental unity expressed in the native historiographic tradition does not appear to
correspond to political reality. Tambiah (1977: 84) has observed a similar
pattern for Southeast Asia, where large, powerful, centralized states were
rare and short-lived. Even the Old World's largest empires generally lasted
no more than several centuries (Taagepera 1979). As the historian Elvin
(1973: 17; see also Hartwell 1982; Skinner 1985) has written, "the Chinese
Empire is the major exception in the pre-modern world to what would appear to be the rule that units of territorial and demographic extent comparable to that of China are not stable entities over long periods of time." Yet
even in the case of China, the degree and extent ofpolitical unification may
have been smaller than certain traditional documentary accounts have suggested (Falkenhausen 1993; Keightley 1983; Rudolph 1987).
Braudel (1973: 660) has suggested that some times are favorable for small
states and others for large states, but that the period beginning in the
fifteenth century in the Mediterranean was propitious for the latter. Although that seems to be so, the conditions and circumstances present earlier
in much ofEurope appear often to have fostered smaller polities. In part, we
have not adequately recognized this pattern of small-state networks because
of the taxonomic muddle that we have long been in. Although archaeologists and historians have had few difficulties recognizing and categorizing
large preindustrial states, the smaller components of state systems (Wesson
1978), interstate systems (Chase-Dunn 1990), or peer-polity networks (Renfrew and Cherry 1986) are more difficult to classify, Perhaps that is why we
have such a plethora of terms for such societies (see Houston 1992), including peer-polities, early state modules (Renfrew 1972), palatial territories
(Bennet 1990), city-states (Bray 1972; Burke 1986; Calnek 1978; Ferguson
1991; Marcus 1989; Pounds 1969), feudal states (Mann 1987), segmentary
states (Fox 1987; Southall 1956), galactic polities (Tambiah 1977), theater
states (Geertz 1980), inchoate early states, typical early states, transitional
early states (all from Claessen and Skalnik 1978a: 23), stratified societies (Fried 1967), Asiatic states (Friedman and Rowlands 1977), cacicazgos
(Redmond and Spencer 1994; Spores 1969), and petty kingdoms (Kowalewski et al. 1989).
Certainly, many of these small polities (characterized in the citations
given above) overlap in scale and political complexity with "complex chiefdoms" (e.g., Kirch 1984). Yet it seems somewhat of a conceptual stretch to
describe many of the aforementioned polities as "chiefdoms." (1) Many of
these polities were composed of economically stratified classes (often with
slaves); (2) their rulers or kings lived in palace structures; (3) the regional
population densities were greater than is sometimes found in the same area
when that region was unified under more centralized states; (4) writing systems often were employed; (5) in certain instances market systems were in
place; and (6) at least three (often more) decision-making levels were present (although this was frequently not uniform for the individual polities that
composed a single network; e.g., Hodge 1984; Pounds 1969). Perhaps more
significantly, these component polities often were linked through marriage,
conquest, political alliance, and other ties into multitiered confederations of
greater vertical complexity (albeit ones that tended to be relatively unstable
and shifting with variable degrees of integration). Nevertheless, sometimes
the ties and connections that articulated these small states into larger confederations became more bureaucratic and centralized (as Smith 1992 describes for the concentration of power at Aztec Tenochtitlan). Finally, it is
worth stressing that the above list is not intended to suggest or promote any
simple analogies or parallels between these enumerated cases beyond the
small sizes of the component polities. In fact, from case to case, the integrative strategies and mechanisms that bound leaders to followers (and leaders
to leaders) often were distinct (e.g., Blanton et al. 1996; Lewis 1974a; see
also Blanton, this volume).
INHGRATlON, SCALE, COMPLEXITY,
AND THE ARCHAIC STATE
Social scientists (e.g., Spencer 1885: 449-50) have long been aware that a
relationship exists between societal size and complexity. A number of synchronic studies (e.g., Carneiro 1967; Ember 1963) have noted that when a
large number of populations are examined (encompassing a diversity of environments), a strong relationship is found between increases in societal scale
and more hierarchical organizational forms (vertical complexity). Carneiro
(1967: 239) has characterized this relationship: "If a society does increase
significantly in size, and if at the same time it remains unified and integrated,
it must elaborate its organization." Curiously, despite a long-standing interest in the formation and elaboration of human groups and groupings (see
Johnson 1982), most anthropologists, archaeologists, and historians have
tended to sidestep a direct examination of the size /organization relationship. Despite great environmental and technological diversity in these crosscultural samples, most theorists have evoked population pressure on food
supply as the source of stress to which organizational change is a response
(e.g., Carneiro 1970a; Sanders, Parsons, and Santley 1979).
This emphasis on population pressure as an explanation reflects several
problems that we have in trying to understand general processes that govern
the formation and evolution of human groups. As Flannery (1972) warned
25 years ago, it is not enough to examine only the socioeconomic stresses
(e.g., demographic growth, population pressure, technological innovation,
environmental change) and the specific sequences of societal transition; attention must be given to the decisions of individual actors as well as institutional structures and responses. In other words, it is difficult to model or
compare systemic or evolutionary change if there is little understanding of
the preexisting and potentially diverse structure (including the component
segments, integrative mechanisms, constraints, and thresholds) of the system
under investigation. This dual attention to the prior conditions as well as
perturbative forces is what Gould (1986) means when he says "history matters," as it does for understanding archaic states as well as biological evolution (see also Feinman 1997; Schreiber 1987: 267).
In the study oflong-term social change there continues to be little in the
way of "social physics" (Bernard and Killworth 1979). Although notable
exceptions do exist (e.g., Johnson 1978, 1982; Kosse 1990; Wright 1977),
the models and concepts that have been developed remain largely untested.
This is not an argument for the construction of general laws (see Drennan
1996). In fact, I would comfortably substitute "social change studies" for
"biology" in Mayr's (1982:37-38) statement that "there is only one universallaw in biology ... all biological laws have exceptions." Yet the time is
ripe for greater systemization, comparative application, and careful consideration of what we do know about human organizations-their scale, integration, and complexity.
Bringing Integration into the Scale/Complexity Equation
Johnson's (1982) reanalysis of Ember's (1963) earlier study noted that the
strong relationship between societal scale and complexity weakened greatly
when narrower size ranges were examined in more discrete segments. The
restudy goes on to outline three general conceptual directions (cycling,
nested structures, integration) that should be examined to unravel the weaker
fit found between sociopolitical complexity and scale (Carneiro 1967; see
Feinman and Neitzel 1984 for the recognition of similar patterns in crosscultural data). Although johnson's (1982) primary focus is not states, the
conceptual directions that he elaborates provide promising avenues to probe
the significant variation in the sizes and organizational poses of complex societies. Although each of these directions has long tap roots in anthropology, the implications have not been incorporated sufficiently in most prior
presentations of neoevolutionary thought.
Drawing an example from the !Kung San, Johnson (1982:396-403)
illustrates how household organization (basal unit size) shifts to preserve
an egalitarian organization as camp size increases seasonally. Curiously,
Carneiro (1967:241-42) aptly described similar seasonal changes in Plains
Indian organization with tribal aggregations for buffalo hunts. However,
unlike the !Kung San case, the cyclical shifts on the Plains involved the temporary institution of both vertical and horizontal decision-making structures, with an overarching chief, a council, men's societies, and a police
force. One can see how these cyclical transformations might complicate the
relationship between scale and complexity based on cross-cultural samples
of static "snapshots" (Bernard and Killworth 1979: 36).
Such seasonal fluctuations may not be a key consideration for archaic
states, yet complex preindustrial formations do show more organizational
fluidity than we generally associate with contemporary Euro-American institutions and bureaucracies (e.g., Wallace 1971). Recent studies of the late
prehispanic Inka (e.g., D'Altroy 1992) and Aztec (Hodge 1984; Rounds
1979; Smith and Berdan 1992; van Zantwijk 1985) polities emphasize both
marked temporal fluidity in their strategies of political integration (along
with the associated ideologies of incorporation) as well as significant synchronic spatial variation in strategies of administrative organization (for an
Mrican example, see Kopytoff 1987; for China, see Hartwell 1982).
The accommodation of this short-term temporal and synchronic spatial
variability in preindustrial organization requires us to move beyond the
simple polar or taxonomic divisions (e.g., kinship/territorial, city-state z'imperial bureaucracy) to more flexible analytical paradigms that examine core
features (Blanton et al. 1981; Kowalewski et al. 1983) and bundled continua
of variation (de Montmollin 1989; Easton 1959). By so doing, the Aztec
tributary domain (large in scale with a centralized settlement pattern, but
limited administrative bureaucracy) may not have to be seen as "anomalous"
or "transitional" between ideal forms (e.g., small polity vs. large bureaucratic
state). Rather, like the Aztec state, many large preindustrial states functioned
without formal, entrenched bureaucratic institutions, but their more fluid
organizational structures had ramifications for exaction, control, and stability (e.g., Hassig 1985; for African states, see also Lloyd 1965:71). To be
more explicit, our taxonomies ought not conflate large scale with imperial
bureaucratization, for in the preindustrial world it may have been easier for a small state than a very large one to centralize (see Kowalewski et al.
1983; Southall1988:81). We should not be reductionist nor view states as
monolithic.
Sizes of States
The range of estimates for archaic state sizes should now be reconsidered.
The lower values (e.g., less than 10,000) may approximate the sizes at which
marked structures of inequality, the stratification of wealth and power, and
the integrative formations generally associated with states (Carneiro 1981 :69)
may be sustained by politically autonomous polities under certain conditions (economic, interpolity) (Claessen 1978a). Such polities are most common as the component parts of larger social / cultural networks, linked
through interaction and elite marriage alliances. Conversely, the middle-tohigher population estimates (100,000 or more) may be closer to the demographic group sizes that cannot be politically assembled and integrated
without the vertically tiered decision-making complexity generally associated with states (Wright 1977, 1986a).
The importance of intersocietal interaction and intrasocietal integration
to the size / complexity relationship can be illustrated by briefly stepping
away from the archaic state and examining the "magic number" 2,500
500 (Kosse 1990). Bernard and Killworth (1973, 1979) calculate that the size
of the maximum group where information can still reach everyone (albeit
with some time lag) is roughly 2,500. This figure corresponds with the maximum sizes of folk taxonomies (where oral histories supplement short-term
recall) and estimates for long-term human memory capabilities (Kosse 1990),
notwithstanding the intensive efforts that can raise these limits further.
Empirically, in a study of the maximal community sizes found in 45 nonindustrial societies (Lekson 1985), a rather sharp cutoff was recognized
around 2,500. Although societies with communities of fewer than 2,500
people had a broad range of organizational structures, from stateless to state,
all societies with communities greater than that size had at least two levels of
decision making. In a sample (Feinman and Neitzel 1984) of 21 middlerange societies in the Americas, a similar pattern was observed, with maximal community sizes larger than 2,500 noted in only two cases. As in the
Lekson study, each of these cases had two or more decision-making tiers,
while the smaller communities were associated with one to three tiers of decision makers. None of the 46 "single-community societies" studied by
Carneiro (1967) or the 37 New Guinea societies compared by Forge (1972a)
included communities larger than 2,000. Likewise, only one of 17 New
Guinea populations had organized political groupings larger than 2,500
(Brown and Podolefsky 1976), although some of these tribal/linguistic
populations were considerably larger in total size. The majority of the New
World "middle-range" societies also were greater than 2,500 in total population (Feinman and Neitzel 1984).
Based on these findings (see also Ember 1963), it seems apparent that,
whereas communities (or politically integrated groups) larger than 2,500
500 do seem to require significant organizational infrastructure (two or
more decision-making levels), more dispersed populations or societies that
may be much greater in total size do not. These cases, from diverse environmental and economic settings, indicate that the nature of the interaction
between group members (and the kinds of integrative tasks that they must
In the Late Imperial Chinese empire, where communication and transportation, as well as agricultural production, were relatively well ~evelop~d,
provinces on the frontier were smaller than those close to the capital (Skinner 1977a). According to Skinner (1977a; see also Johnson 1982), these outlying provinces were smaller because the duties of peripheral administrators
were more demanding. These officials were responsible for both resource
exaction and boundary relations (the latter being a more significant concern
than in the core). To maintain a more-or-Iess comparable level of integration throughout the empire and to minimize potential threats to the core
from peripheral regions, provincial administrators (with greater costs and
broader duties) generally monitored smaller domains. Conversely, in the
Aztec polity (with less developed transport, communication, and agricultural technologies), peripheral provinces generally were larger than those
closer to the core (often they were not even territorially demarcated)
(Berdan 1987; Smith 1987). At the same time, political integration and tribute collection were much less frequent on the periphery than they were
close to Tenochtitlan. Clearly the degree of integration was less in outlying provinces as a concession to time and the friction of distance. This
contrasts with the Chinese case, where size was diminished at the edge to
maintain (or even intensify) political integration. In other words, for empires
(see also Doyle 1986), one cannot really understand the size/organization
relationship without considering the nature of integration as well as scale
and complexity.
.
At a still broader theoretical level, a related pattern has been recognized
by Friedman (1977; see also D'Altroy and Earle 1985), who finds that (hol~
ing other factors more or less constant) nations dependent on trade as their
primary basis for support tend to be spatially more extensive than those dependent on agriculture. Exaction of agricultural resources would seem to require greater politico-economic integration, and thus more infrastructural
support. In the preindustrial world, late Old World territorial empires may
be somewhat of an exception to this pattern, but these formations were expensive to maintain and rather unstable (see Taagepera 1978b). If we employ
Friedman's expectations synchronically, it is interesting to note that both the
Aztec and Inka polities practiced more territorial-like integration and exaction strategies near their capitals, while they implemented more hegemonic
rule and had a greater focus on tribute and trade at the limits of their political spheres (e.g., Berdan 1975; Blanton 1985; D'Altroy 1992). Yet the causal
link between organizational capabilities and economic strategies is far from
simple or unidirectional. Likewise, in both of these cases, much of the exchange at the peripheries was merchant driven, and not really a total expression of state administration.
At smaller scales, different integrative strategies also have transport and
administrative costs and requirements, which therefore entail spatial implications and limitations. We currently know much less about these than we
do about the movement of goods and armies (e.g., D'Altroy 1992; Drennan
1984; Engels 1978; Hassig 1988), although some similar constraints would
seem to apply. Yet in discussions of different integrative strategies and the
likelihood offission, the" cost-benefit" ratios cannot measure just energy or
time, since all societal bonds and interactions do not promote or diminish
the possibilities for fission in the same manner; we must go beyond an examination of economic relations, political complexity, and scale to understand this variation.
As noted above, greater attention must be given to the diverse strategies
by which human populations are articulated (see Blanton, this volume;
Blanton et al. 1996; Feinman 1995). Some integrative strategies, like kinship, tend to have lifelong durability (Barnes 1972); others require much
more constant inputs. At the same time, certain integrative strategies (kinship, fictive kinship, ritual) are more constrained by physiological limits and
distance costs. Diverse integrative strategies also have different implications
for recruitment. Sahlins (1972) systematically evaluated some of the economic ramifications for kinship relations across space; however, an expansion of this work to consider alternative integrative strategies (beyond
kinship) and other kinds of costs/benefits is in order (see Drennan 1991;
Renfrew 1974).
Spatial variation in integrative strategies also may help explain and compare the "near-decomposability" (Simon 1969) oflarge social systems. Applying general systems principles, Simon (1969; cf Yoffee 1988a) argued
that empires often collapsed or decomposed into the same smaller segments
that existed when the higher-level linkages were first forged. The basis ofhis
argument lies in the greater durability given to the horizontal linkages that
are present in the smaller component segments as compared to the more
ephemeral vertical connections hypothesized to bind these segments.
Yet despite some empirical validity, Simon's expectations (or a general
consideration of structure alone) cannot explain why some large states decompose into their prior component pieces and others break up into new
subunits, while, in rare instances, some fragment entirely. Part of that understanding goes beyond structural considerations and is a function of how
the component segments were integrated (internally), and how much these
specific mechanisms of integration were transformed during the episode
of amalgamation. For example, a lesser degree of "near-decomposability"
might be expected following early episodes of state development. During
such growth episodes, kinship and fictive kinship often were supplemented
or replaced (as the key mechanisms that bound followers to leaders) by new
economic, ideological, political, and military strategies. If the early state
THEORETICAL SYNTHESIS
The prior analysis critically evaluates key issues and conceptions concerning
the size of the archaic state. Some points of conventional wisdom have been
challenged, and directions and questions for future research have been advanced. Before proceeding to a brief examination of empirical findings from
the highlands of southern Mexico, some central themes from the above discussion are reviewed and synthesized.
Although the general increase in polity sizes since the Neolithic is recognized, this general trend was not unilineal, inevitable, innate, .or ~onst~nt
in tempo. At the continental scale (compared to a global one), shifts m polity
size were even more dramatic and cyclical. Growth in the size of specific
polities can be neither presumed nor seen as a universallawlike co~~equen~e
of evolutionary forces. Consequently, changes in the scale of political entities are seen as explicable (though historically contingent), and future investigations should focus on transport, communication, and administrative
technologies to understand historical and political trajectories. We should
systematize what we already know about the relative effectiveness of different integrative mechanisms and strategies, and how they worked (or did not
work) at varying spatial scales (e.g., Blanton, this volume; Kohr 1957).
Archaic states were extremely variable in size, with no clear threshold
separating all states from all chiefdoms (see discussion in Voorhies 1?~9).
Significantly, secondary states that formed in the context of preexisting
systems oflegitimatio.n and stratification may be much smaller and less populous than the first pnmary states that arose in the same region. As a conseq~ence, pr~ind~strial states (in toto) were most likely to be small and part of
w~de-rangmg, mterlocking state systems. These networks were dynamic,
with the similarly sized polities having variable degrees of political and economic autonomy at different times. Individual polity size alone will never
be an adequate guide for understanding either the nature of these states or
the larger networks that they formed. More broadly, and in sharp contrast
to Wallerstein (1979), the precapitalist world clearly was not composed exclusively of tribal societies, minisystems (aggregations of small autonomous
chiefdoms and states), and empires. Rather, preindustrial worlds also comprised a diverse array of complex interlocking state systems. In such networks, the political, cultural, and economic limits of specific polities were
neither always coterminous nor closed. The advent of multipolity "world"
systems long precedes the advent of capitalism (see Peregrine and Feinman
1996).
The study of long-term change requires simultaneous consideration of
at least three analytical dimensions: specific trajectories of change, socioenvironmental stresses, and the social properties of human groups and
groupings. The latter (still the dimension least considered by archaeologists
and historians) involves the reassessment of the size/complexity relationship
(seeJohnson 1982), with greater attention placed on spatiotemporal cycling,
the nested interplay between horizontally and vertically organized groups,
and a broad consideration of the diverse means and strategies of integration.
Significantly, if secondary states were frequently smaller than primary states,
c.ross-national, ethnographic comparisons (based on secondary cases) may
SIgnal relevant directional relationships but will provide little specific guidance for understanding primary states. The empirical picture for the size of
early states can only emerge from field and regional studies in archaeology.
On a more specific level, preindustrial polities were seemingly more dynamic and cyclical in scale (e.g., Fry 1990; Marcus 1993a), integration, and
even complexity (e.g., Kowalewski 1990) than our theories generally have
envisioned them. This should not be entirely surprising since their transport
and production technologies were frequently more similar to those of ancient chiefdoms (on cyclical change in chiefdoms, see Anderson 1990;
Stemper 1989) than to those of postindustrial states. At least for prehispanic
Mesoamerica, whenever historical records are available (to supplement the
archaeology and provide an additional perspective on interpolity relations),
the picture ofpolitical relations appears particularly fluid and dynamic (e.g.,
Culbert, ed. 1991; Marcus 1976a, 1993a; Martin and Grube 1995; Smith
and Berdan 1992; Spores 1974). Yet it is important to remember that, as with
the indigenous southeastern V.S. and Ecuadorian chiefdoms, dynamism and
flux at the polity scale may translate to relative persistence at more macroscopic levels of analysis (see also Appel1982: 145; Marcus 1992a).
In general, archaic states were characterized by marked spatial variation
in the strategies of interconnection !integration. These spatial differences
were not simple, direct consequences of travel time. But they do reflect the
friction of distance (the nature and degree of integration varied with space)
as modified by specific polity-to-polity (ruler-to-ruler) histories, strategies,
and interactions. A consideration of integrative mechanisms, and how they
shifted (or did not) with polity expansion, should help refine expectations
concerning "near-decomposability." The replacement oflargely mechanical
integrative mechanisms by more bureaucratic, organic strategies would diminish the likelihood that a large polity would simply break down into the
segments that were joined to compose it initially.
CENTRAL MEXICO
TENOCHTITLAN
GULF OF MEXICO
...:.::'
100
PACIFIC OCEAN
200
I
KILOMETERS
which predates the foundation of the key valley center of Monte Alban,
Subsequently, the Monte Alban polity is examined during Monte Albin
Late I and IIIA.
Late Postclassic (Monte Alban V)
I start with the Late Postc1assic period at the end of the prehispanic era, for
this phase allows for the evaluation of both written and archaeological
sources (e.g., Appel 1982; Flannery and Marcus, eds. 1983; Kowalewski et
al. 1989; Paddock 1983; Whitecotton 1992). The Southern Highlands were
fragmented politically at this time into a network of small, semiautonomous
polities that stretched from the highlands of the Mixteca Alta to the Valley
of Oaxaca and, perhaps, south to the Pacific. At the macroscale, population
densities were greater than at any time in the past. Less scholarly consensus
exists concerning the degree of polity autonomy, the relative importance of
the specific integrative mechanisms (e.g., warfare, elite marriage, marketing), and the place of these small states in the larger tributary domain established by Aztec Tenochtitlan at the close of this phase.
Specifically, this discussion examines political geography in the valleys of
Oaxaca and Ejutla from several scales. Population estimates for the area's
political segments are presented. The estimated values for central Oaxaca are
compared with those for the three smaller regions (Nochixtlan, Tamazulapan, and Cuicatlan). In general, population size estimates are based directly
on the distributions of artifact scatters (sites) (see Blanton et al. 1982; Feinman et al. 1985; Kowalewski et al. 1989; Parsons 1971). When precontact
population values were not presented for specific sites in the three smaller
regions, estimates were calculated from the settlement area using procedures
equivalent to those employed in Oaxaca/Ejutla.
The first reconstruction (fig. 4.2) divides the Oaxaca/Ejutla region into
24 settlement clusters. This figure is an extension of the map drawn previously for the Valley of Oaxaca (Kowalewski et al. 1989: 345) and uses "shatter zones" to subdivide the regional population. This method identifies
uninhabited zones between clusters of settlements on the assumption that
the boundaries between polities would be sparsely settled. This map may
represent the smallest Postclassic "petty kingdoms." A second map (fig. 4.3),
with 13 components, gives greater attention to the size and locations of the
region's largest centers as well as ethnohistoric findings (see Kowalewski
et al. 1989: 344 - 48). This reconstruction seems to reflect the regional political geography toward the end of the Postclassic period, following a degree of political consolidation.
A final reconstruction uses basic nearest-neighbor principles (see Haggett 1966:231-33) to define site clusters. Sites within 1,500 m of their
neighbors generally were placed in the same cluster. This procedure segments the region into only five clusters (three small peripheral polities and
two major ones) (fig. 4.4). The latter two are centered around the two valley communities that were known to be most powerful at Spanish contact
(Sa'a Yucu/Cuilapan and Macuilxochitl). The sixteenth-century documents
suggest competition and shifting alliances (military and marriage) between
late prehispanic Oaxaca polities, with some trending toward centralization
around these latter two sites (see Paddock 1983). The three alternative archaeological reconstructions seem to reflect the fluid, nested, and mechanical (in the Durkheimian sense, as opposed to organic) political integration
of Postclassic Oaxaca, which corresponds closely to the dynamism described
in the ethnohistoric accounts.
The assignment of population estimates to these bounded units must be
viewed as approximate. Besides the general archaeological problems of
chronological control and converting sherd scatters to people, this exercise
also is problematic because the bounds of the Oaxaca polities certainly were
not coterminous with the settlement pattern survey region. Not surprisingly,
this was evident after the Ejutla survey, when some of the Valley of Oaxaca
settlement clusters (defined previously) were shown to extend south to encompass neighboring settlements in the Ejutla region. Prior to the Ejutla
VALLEY OF OAXACA
t
10
20
I
KllOMETERS
Figure 4.2. Boundaries of Monte Alban V petty kingdoms, based on shatter zones (only sites with estimated populations cif 200 or more are shown;
the largest sites are numbered 1-20).
survey, the size ofthese clusters was underestimated. Since little archaeological information is available concerning the mountains which surround the
valleys of Oaxaca and Ejutla, the estimates presented here also probably underestimate polity extent and population. In a few cases, some frontier settlements may have been included in the estimates for valley polities when
Figure 4.3. Boundaries cif Monte Alban V petty kingdoms, based on location cif large centers and ethnohistoric accounts.
Figure 4.4. Boundaries cif Monte Alban V settlement clusters (the clusters
are separated by 1500 m or more cif unoccupied space).
these sites actually were subject to head towns in the mountains that lie outside the study region. In these cases, the population values may be too large.
Based on the shatter-zone reconstruction (see fig. 4.2), the Oaxaca/
Ejutla polities ranged in estimated size between 2,100 and 23,400 people,
with some of the smallest entities on the region's edge (table 4.1). In this reconstruction, mean polity size was close to 5,000 people in the Valley of
Oaxaca. Settlement clusters in Nochixtlan, Tamazulapan, and Cuicatlan con-
form to the same range (table 4.2). These estimates for the Southern Highlands are roughly comparable to the ranges of population size presented for
small states elsewhere (e.g., Pounds 1969; Renfrew 1982), and to the sizes
of the smallest political districts in the Basin of Mexico in AD 1568 (Sanders
et al. 1970: 443-62). At contact, these Central Mexican districts were amalgamated into larger political jurisdictions. Many, if not most, of these larger
units are thought to have been politically autonomous or semiautonomous
T ABLE 4.1. Pop ulation 01 Late Postc1assic Polities in the Central Valleys of
Oaxaca
TABLE
PolitylArea
Population
Shatter Zone
Centers/Ethnohistory
Large Clusters
Macuilxochitl
23,400
23,400
81,000
Mitla
19,500
21,600
Sa'a Yucu
18,800
32,400
60,000
16,600
20,800
20,500
Eastern Etla
9,900
9,900
Teitipac
9,300
15,000
Tlalixtac
9,200
11,700
Jalieza
8,800
8,800
Yagull Tlacolula
8,300
12,500
Ixtlahuaca
5,900
Yanhuitlan
Quialana
5,700
Chachoapan
8,500
Taniche
5,000
Nochixtlan
10,750
Coyotepec
4,600
Polity/Area
Cuicatlan
Ejutla
4,300
4,200
El Choco
4,200
El Vergel
4,100
Huitzo
3,500
Coatecas
3,500
Tlapacoyan
3,300
3,000
Tule
2,500
Zautla/ Tejalapan
2,400
Matatlan
2,100
Yogana"
Median size
4,300
Dominguillo
2,300
Quiotepec '
400
Tamazulapan
Tejupan
10,900
Tamazulapan
6,600
Teposcolula b
2,300
Nochixtlan
Etlatongo
12,900
12,900
4,200
7,500
jaltepecMedian size
20,500
9,500
700
10,125
5,900
3,300
250
4,800
Cuicatlan
12,500
20,500
until the last two centuries before Cortes, although the nature of the externallinkages are thought to have been dynamic and cyclical. Interestingly, the
range of demographic estimates for these larger Basin of Mexico jurisdictions (generally between 10,000 and 30,000, excluding Tenochtitlan) prior
to contact (Bray 1972; Sanders et al. 1970) overlap the population figures for
Oaxaca/Ejutla that were derived with closest attention to ethnohistoric accounts (see table 4.1; fig. 4.3). These larger units may reflect the smallest late
prehispanic Oaxacan political entities with considerable autonomy.
Early Postclassic (Monte Alban IV)
In the remainder of this chapter, the same basic site clustering procedures
that were employed for Monte Alban V are applied to earlier phases. Given
the absence of historical information, clusters were formed considering both
By Monte Alban IV, the monumental hilltop center of Monte Alban (so
important in the immediately prior Classic period) had lost much of its
population. Monumental building at the site ceased. In contrast to Monte
Alban V, settlement during this earlier period was much more nucleated, and
there were sizable open areas between substantial occupations. Because of
these "shatter zones," the definition ofsite clusters was fairly straightforward
in most instances. This spatial separation of population clusters, several of
which were positioned in defensible hilltop locales, may reflect tension and
reorganization following the collapse of the Monte Albin-centered polity
that dominated the region for more than a millennium (Finsten 1983;
Kowalewski et al. 1989). Intraregional integration appears to have been
lower at this time than it was in either Monte Alban V or during the previous (Classic) period.
Significantly, the spatial distribution of most of the Monte Alban IV
settlement clusters seems to correspond closely to the small polities of the
next epoch (see figs. 4.2, 4.3). The most consistent difference between the
two phases is that additional Monte Alban V polities emerged in areas that
had been Early Postclassic (Monte Alban IV) shatter zones. As in Monte
Alban V, Macuilxochitl (see fig. 4.5) was one of the region's two most populous polities; however, the largest Monte Alban IV cluster (jalieza) lost most
of its population by the end of the prehispanic era. Jalieza was occupied in
Monte Alban V, but it is not mentioned in ethnohistoric accounts, so its
significance may have been minimal by the time the Spaniards arrived
(Kowalewski et al. 1989).
In terms of population, two Monte Alban IV clusters were much larger
than any others. Nevertheless, despite this marked bipolarity, the estimated
median polity size for period IV was similar to that found for the Monte
Alban V clusters (table 4.3; see table 4.1). The overall size range in Monte
Alban IV also was comparable, but the small size of some of these units may
indicate that the period IV polities stretched into the mountains. If this pattern holds, it would conform with the poor valley integration and centrifugal
trends that followed the political demise ofMonte Alban. Greater valleywide
integration (although not political centralization) was achieved in the Late
Postclassic (Monte Albin V) with population growth and filling-in, and the
replacement ofJalieza as a key center by Sa'a Yucu/Cuilapan, situated near
the centrally located former valley capital of Monte Alban.
site spacing and the distribution of the region's largest settlements. There
was no single nearest-neighbor delimiter for the definition of clusters,
although these boundaries were always drawn through sparsely occupied
areas. Clusters always were defined to include at least one major center. The
cluster map (fig. 4.5) for Monte Alban IV indicates both key differences with
the subsequent era as well as some significant continuity.
PolitylArea
Population
]alieza
19,300
Maeuilxoehitl
17,200
Animas Trujano
6,000
Mitla
5,900
El Choco
5,000
Monte Alban
4,300
3,900
Yagul
3,700
2,000
Mixtepee
1,600
Ejutla
1,400
Tlalixtae
1,100
Coateeas
1,000
Huitzo
Median size
750
3,800
r
10
20
I
KILOMETERS
Figure 4.6. Boundaries of Rosario phase settlement clusters, based on shatter zones and the location of large sites (only sites with estimated populations of 20 or more are shown; the largest sites are numbered 1-7).
."
iN
0
10
20
I
KILOMETERS
Figure 4.7. Boundaries cif Monte Alban Late I settlement clusters, based on
shatter zones and the location cif large sites (only siteswith estimated populations cif 100 or more are shown; the largest sites are numbered 1-15).
interconnection may have varied in conjunction with the distinctions discussed here.
Given the spatial (core zone vs. Tlacolula/Ejutla) differences in integration and organization, it also is interesting that funerary urns, closely associated with death ritual (Marcus 1983e) at Monte Alban, were found
(in survey collections) to be proportionately well represented in the Valle
Grande core and at the largest sites in Tlacolula and Ejutla. But otherwise
these artifacts were underrepresented in the latter two regions. In general,
Monte Alban as well as Valle Grande and Etla centers also had greater access
than did Tlacolula and Ejutla centers to the wide assortment of ceramic
goods available in the region (Kowalewski et al. 1989: 240). Since Monte
Alban had just about the full range of ceramic variability, this pattern may
be another indicator of lower interconnection between these latter areas
(Tlacolula and Ejutla) and Monte Alban than existed for the core cluster.
These findings would seem to foreshadow the spatial patteming of Monte
Alban's collapse by the end of the Classic period. In this balkanization process (that culminated in the fragmented political landscape of Monte Alban
IV), Monte Alban was separated politically from both Tlacolula and Ejutla
(see fig. 4.5).
CONCLUSION
N
0
1
10
20
I
KILOMETERS
cluster. Yet, two lines of defensible sites were located inside the valley. One
of them subdivides Tlacolula from the eastern Valle Grande (Elam 1989:
406). The other separates Ejutla from the southern edge of the Valle Grande
core. Both of these lines of defensible sites are almost coterminous with (and
parallel) the divisions drawn between settlement clusters.
generally small. Often, but not always, these small states were the breakdown products of larger, earlier states (as in Monte Alban IV in the Valley
of Oaxaca; Marcus, this volume). In a given region, the size ofstates did not
necessarily increase linearly over time. There is no simple numerical threshold that demarcates all states from all chiefdoms. The ways in which ancient
states were integrated and interconnected often varied markedly over space,
and differences in organization and integration have profound implications
on state size.
It is evident that the theoretical issues investigated here require regionalscale data. The kinds of broad-scale and diachronic information considered
for the valleys of Oaxaca and Ejutla are unfortunately rare in archaeology
and history. Yet it is just this kind of information that is necessary to dissect
the complex relationships between polity size and organization. We will not
really understand the complicated linkages between human demography
and organization until archaeologists, historians, and geographers more
regularly extend their investigatory perspective away from ancient capitals
and monumental centers. At the same time, and as noted above, more work
on the different kinds of integratory connections that joined followers to
leaders and hinterlands to capitals is needed.
No effort has been made here to advance a new theory of the state. As
North (1986: 248) aptly has written, "the long path of historical research is
(already) strewn with the bones of theories of the state." Likewise, no new
"magic" thresholds have been proposed. Archaic states were diverse in scale
and integration, and past considerations of the scale / complexity relationship
require greater examination of variation in strategies of integration. This
work has challenged several simplistic positions, linear principles, and taxonomic schema. An understanding ofsocietal change and preindustrial states
requires much more than a single theory, evolutionary scenario, or set of
prime movers. Rather, this discussion has offered several complex considerations and long-term research directions for modeling and studying the size
and organization of ancient states. By endeavoring to offer bridges between
history and process, this essay hopefully nudges us a small step along a long
and complex road. For as Wallerstein (1992: 6) has noted, "History and universalizing social sciences are the identical activity. The object is not to
reduce complex reality to simple generalizations, either middle-range or
large-scale. The object is to construct an interpretation of complex reality
by surpassing simple generalizations, interweaving them, and defining the
degree of their relevance."
NOTE
I am deeply grateful to the School of American Research and its president, Douglas W.
Schwartz, for the opportunity to exchange ideas with such an exceptional and stimulating group of colleagues. Our stay in Santa Fe was magnificent and much appreciated.
Thanks are due all of my collaborators on the Valley of Oaxaca and Ejutla Vall S tl
ment Pattern Projects. A special debt of gratitude is owed Richard E Bl ey et e;;
Step hen A. Kowalewski, who directed the Oaxaca research and Linda M
hanlton ahn
has be inst
al i he Ei tl
, . IC 0 as, W 0
en InS rument In t ~ ju a investigations. Linda generously prepared the figures
that are ,present~d here, while .Lane Fargher assisted, I appreciate the essential Support of
the National SCIence Foundation and the important assistance of the Instituto N ' al
.
d h
acion
de Antrop I H'
,
0 ?gra e
istona an t e Centro Regional de Oaxaca. I wish to thank all the
seminar participants for their camaraderie and the free exchange of ideas that occurred
dunng our week In Santa Fe. ]oyce Marcus, Richard E. Blanton, George L. Cowgill, and
an anonymous reviewer are especially acknowledged for the extremely helpful comme t
that they offered on early drafts of this work.
ns
N'
5
Beyend (entrolizol:i &11
Steps To"onl 0Iheory of Igolitorion BehoYior in Archoic Stotes
RI(HARD LBlANTlIN
n their proposal for a School of American Research symposium on the archaic state, Joyce Marcus and Gary Feinman asked participants to accomplish a number of tasks, two of which I thought would be particularly
apropos of certain contemporary theoretical issues in anthropological archaeology and adjacent fields. Participants were asked to make use ofa comparative approach to explain variation between archaic states, and we were
asked to rethink and critique the typological evolutionary approaches like
those of Fried, Service, and Polanyi. In this chapter, I tackle the assignments
by elaborating on a topic only infrequently addressed in the literature on archaic states, which I refer to as "corporate political economy." I take a first
step in the direction of comparative analysis by describing the varied forms
of corporate political economy in selected examples from several world areas. I believe the results of this preliminary comparative effort constitute a
useful starting point for future research.
In applying this approach, I found it possible to clarify a number of issues that previously had clouded my thinking about archaic states. Most importantly, I found the approach useful in its ability to summarize many of
the major conclusions about archaic states made by previous researchers,
while at the same time pointing to several topics that require a fresh look, as
well as new concepts, theory, and analysis. To take this new direction I had
to deviate considerably from theories currently in favor among anthropological archaeologists, including those inspired by Marxist thought and systems analysis. The springboard for my discussion is a critique of the systems
approach.
their control mechanisms, arranged hierarchically. Control mechanisms assess the output of their respective subsystems, comparing them with "goal
values" that indicate acceptable levels of functioning. Goal values include
"ideological values," which are the human population's "cognized model of
the way the world is put together" (1972:409). Government is a higherorder control system able to assess the functioning oflower-order (i.e., less
inclusive) subsystems. By assessing the output of subsystems in relation to
output goals and ideological values, governmental institutions are able to assess deviant values in subsystem output, and then respond to them adaptively
so as to preserve overall system functioning (Flannery 1972: 409; cf Rappaport 1977:53-58).
According to this adaptationist approach, governmental controls are
strengthened (centralization) when socioenvironmental stresses such as warfare or population pressure provide an adaptive milieu in which evolutionary
mechanisms are triggered. The basic mechanisms resulting in centralization
are "promotion," in which a subsystem is raised to a higher, more inclusive
position in the control hierarchy, and "linearization," in which control of
lower-order subsystems is captured by higher-order subsystems (Flannery
1972: 412). These adaptive responses to socioenvironmental stresses preserve system functioning, but system collapse, fission, or decentralization
(devolution) may result from an inability to respond appropriately to deviant
subsystems states. In addition, excessively high levels of centralization may
eventually result in "hypercoherence," a systems pathology that may lead to
collapse or devolution owing to excessive integration (1972:420-21; Rappaport 1977: 60).
An important element of systems analysis is the idea that an increased
quantity of information is generated within and between subsystems as society becomes more complex. Therefore, a main trend in cultural evolution
is the elaboration of institutions that process, store, and analyze information
(Flannery 1972: 411). Gregory Johnson and Henry Wright have elaborated
on this latter point in several papers, showing, in particular, the advantages
of an increased hierarchical structure of control systems under conditions of
increased demands on information-processing institutions (Johnson 1978,
1982, 1983; Wright and Johnson 1975). From their information-processing
perspective, the state is defined as having minimally three levels of administrative hierarchy (e.g., Flannery 1972:418; Wright 1969; Wright and Johnson 1975).
Systems analysis contains within its theoretical position a useful strategy
for integrating cognitive and ecological (or cultural materialist) theories,
since, as stated by Flannery (1972: 400), "In an ecosystem approach to the
analysis ofhuman societies, everything which transmits information is within
the province of ecology" (e.g., as discussed in Wilson 1992).This admirably
approach points to the fact that social life consists of a discursive interaction
of social structure, carried intergenerationally, and social actors pursuing
varied goals. Political actors are influenced by the prevailing social structure,
and make use of it, but do not always accept the prevailing system; they may
struggle against it, or even elaborate on it, in accordance with their political
aims. Detailed knowledge of a social system and its culture is thus a political
resource, not simply a determinant of be havior.
In the approach developed here, I start from the position that a state is
the major governing institution within a particular geographic territory
(e.g., Collins 1981 :71; Mann 1984; Weber 1978: 54). While we may define
a state (as opposed to a chiefdom) by reference to the presence of an administrative hierarchy of minimally three levels, the direction I take does
not view the state primarily as a hierarchically structured control mechanism. Rather, my behavioral approach views the state, wherever it is found,
as the major social arena within which the competition for power is played
out in society. This approach leads the researcher to focus on political process, in other words, the competition among various individuals and groups
attempting to exert power in such a way as to shape or control the state (c
Bright and Harding 1984:3-4). Thus I concur with Sewell (1992:22)
when he writes that states are "consciously established, maintained, fought
over, and argued about rather than taken for granted" (compare the comments in Hindess 1982: 500 -501; S. F. Moore 1993; M. Estellie Smith
1985). Political actors with varying aims, such as secondary elites (e.g., local
patrimonial nobility or middle-level bureaucrats) versus central elites (who
control the central institutions of the state), often employ diverse, often
conflictive, political strategies, frequently making use of differing material
and symbolic resources to further their aims (see below). In this framework,
the state is not necessarily the kind of highly integrated informationprocessing subsystem the systems theorists would have us believe. Instead,
the formal, functional, and dynamic properties of the state are outcomes of
the often conflictive interaction of social actors with separate agendas, both
within and outside the official structure of the decision-making institution.
These social actors attempt to exert power in order to influence the state, its
form, policies, and activities.
My expansion of a systems approach to incorporate political behavior
does not deny a simultaneous administrative role of the state as a functional
control mechanism. But I would maintain that to understand both crosscultural variation and temporal change in states, a politically based theory
will be far more productive and realistic than systems and adaptational analyses alone, because it goes directly to the issue of human behavior rather than
basing its analysis on an abstract, idealized, reificationist analysis of system
functioning that suffers from many of the epistemological difficulties of
functionalist theory (Vincent 1986). In the next section I outline an approach to a behavioral theory of archaic state politics that can substitute for
a management-inspired systems analysis.
TYPES AND SOURCES OF POWER
IN ARCHAIC STATES
Political processes are analyzed here by reference to the relationships found
among types ofpower, on the one hand, and sources ofpower, on the other,
an approach I will call "political economy." I illustrate some of the major
kinds of relationships found among types and sources of power, with selected examples, in the conceptual scheme detailed in tables 5.1 to 5.4 and
discussed below. Each table lists several categories of political economy and
selected examples illustrating a type of power (see below), subdivided into
upper and lower cells to illustrate the dual dimension of material (or objective) versus cognitive-symbolic sources of power. I refer to each of the categories listed in the table's cells as a political-economic strategy (e.g., debt
peonage in table 5.1).
These tables were constructed to illustrate the basic elements of political
economy found in archaic states; they do not contain exhaustive lists of even
the major types ofpolitical-economic strategizing that conceivably could be
identified. To enumerate that list would be a far larger project than I have
attempted here. In developing the tables, I included those types and sources
of power that have been discussed at length in the political economy literature, but I add several categories, relevant to my particular interest in corporate political economy, that require additional commentary. The tables
are arranged so that those political-economic strategies that have been subject to wide treatment in the literature are found in the upper part of each
cell, while those that are less well understood, or less frequently discussed,
are placed lower in the cell spaces. Owing to space limitations, most of the
frequently discussed political-economic strategies are not addressed in this
chapter beyond their citation in the tables. Several political-economic strategies requiring extensive commentary are followed by numbers in brackets
that are keyed to numbered paragraphs found below. In the next section,
I discuss the logic of the four tables by explaining the system used here to
categorize types and sources of power.
Sources of Power
Sources of power (or "funds of power" in Wolf 1982:97 and Brumfiel
1992: 554 - 55) are those resources that can be mobilized by political actors
as they pursue their political agendas. I divide those political-economic
strategies found in tables 5.1 to 5.4 into two categories to illuminate one
Bureaucratic administration (Weber 1978: chap. 11) supported by "free-floating" resources not tied to traditional ascriptive domains (Eisenstadt 1969:47, 1980: 850)
Development ofcommercial systems in which "broad markets cut across ascriptive units"
(Eisenstadt 1969: 47, 1980: 850)
Free peasantry and mobility oflabor weakens traditionallascriptive domains (Eisenstadt
1969: 34, 136)
Curtailment of traditional/ascriptive domains through the development of a transcendent cognitive code emphasizing mechanical integration of society at large and group
representations; social groups defined in "universalistic terms" (Eisenstadt 1986: 9)
Structurally integrative classification of social statuses (D. Miller 1989: 70)
Liturgies and euergetism ("civic gifting") in the Greek polis (Veyne 1990:72-77; Millett
1989) [2] a
Decentralization of prestige-goods systems ("Distributive Metal Age") [2] a
Ritual sanctification of corporate cognitive code and ritualization of political communication [4] a
Imperial ideology (Veyne 1990: 377); promulgation of interest ideology (Carlton 1977:
158-65); hierarchical solidarity supported by heroic history idealizing rulers (Sahlins
1983); hegemonic ideology (Hoare and Nowell-Smith 1971)
Types of Power
Viewed cross-sectionally and diachronically, archaic states exhibit a wide
range ofpolitical-economic strategies. The main patterns ofvariation can be
illustrated using a scheme that groups strategies by what I call types ofpower.
A principal dimension of variation captured by the resulting typology is the
distinction between person-centered networks of power, on the one hand,
and the varied expressions of what I will refer to as "systemic" power, on
the other. The tables break into two parts when viewed in these terms, consisting of the "intermernber" power type (see table 5.1) and the various
systemic power types (see tables 5.2-5.4). This dichotomy makes use of
Lehman's (1969: 455 -56) terminology identifying intermember power as
the form of superordination found within individual-centered social networks developed on the basis ofpersonal social ties. European feudal society
and similar social formations dominated by varieties of patron/client relations ("noncorporate groups" in Schneider, Schneider, and Hansen 1972)
are built around intermember political-economic strategizing. I include in
the intermember category clan systems and similar social structures in which
bureaucracy is not entirely subordinated by the ruler and, instead, is constituted so as to ensure that persons in power, including the ruler, conform to
established political practices. In addition, the particular practices utilized
for recruitment into the state's administrative apparatus may be designed to
level social differences in society (Weber 1946: 224 -28), as"! discuss below.
For example, in contrast to the pattern of aristocratic dominance of the
Shang dynasty, by the first century BC, Chinese government officials were
drawn from a wide range ofsocial strata (Creel 1970 :7). I mention these examples preliminarily here to make the point that in a behavioral theory of
the state the importance of the bureaucracy is not to be found only in its
information-processing efficiency, as systems analysts would emphasize, but
also in its potential role in political processes whereby intermember power
is constrained by systemic power, and whereby corporate political strategies
constrain exclusionary power.
Having laid out my conceptual scheme of political economy, I next turn
to the main focus of this chapter, which is a discussion of the elements of
corporate power, presented in a conceptual framework that facilitates comparative discussion. First, though, I wish to reiterate several points regarding
my scheme and the contents of tables 5.1 to 5.4. The purpose of the tables
is to facilitate a comparative analysis of political processes in archaic states by
pointing to the highly variable nature of political-economic strategies found
in them. In part this variation reflects cross-cultural and diachronic variation
between and within states, but it also may describe the variant strategies of
political actors who compete for power within a social formation at a single
point in time (e.g., Rowlands 1985). Given the possibility of coexisting but
distinct political-economic strategies (e.g., Barth 1959), the tables aim at the
elucidation of the nature of political process and cannot be expected to produce a neat political-economic classification in all cases. M. Estellie Smith
(1985: 98) is correct in her estimation that the political anthropology literature has too many typologies of states, and certainly tables 5.1 to 5.4 are not
meant as one more typological scheme.
Social actors with conflictive aims make use of various power sources
and types to those ends, so tidy categorization of political systems often may
elude the researcher. Further, while a given social formation at a particular time may represent a preponderance of some particular dimension of
political-economic strategizing and thus, perhaps, be construed as a particular political type, such arrangements are always contingent and subject to
change, shifting, in essence, that society's major form ofpolitical action from
one area of my tables to another. Cycling between types of power may be a
common historical pattern, as my colleagues and I point out for prehispanic
Mesoamerica (Blanton et al. 1996).
cultural legitimation of cognitive codes that, as we put it, "emphasize a corporate solidarity of society as an integrated whole, based on a natural, fixed,
and immutable interdependence between subgroups" (Blanton et al. 1996:
6). Further, we point out, the ecumenical viewpoint of the corporate orientation supplants or replaces the ancestral ritual that legitimated the control of society by a few high-ranking individuals or households and replaces
it with collective representations and the accompanying ritual emphasizing
universalistic themes offertility and renewal in society and cosmos (c Pasztory 1992). Corporate strategy thus transcends the scale limitations and narrower scope of kinship rhetoric, which exclusionarily empowers particular
individuals or groups based on gender, generation, and primacy of descent
from common ancestors. In addition, corporate social formations constrain
the exercise of power by subsuming it within hierarchically graded sets of
roles and statuses, what D. Miller (1989: 70) refers to as "structurally integrative classifications."
In our conceptualization, corporate power is like Manus (1986: 22-23)
concept of "ideological power," a "transcendent sacred authority" (p. 22)
(note that in what follows in this chapter, unlike Mann, I use the term "ideology" to refer specifically to cognitive codes that mystify exclusionary
domination; where Mann uses the phrase "ideological power," I would substitute "corporate cognitive code," as I discuss below). Ideological power,
according to Mann, is based on the promulgation of norms that are "shared
understandings ofhow people should act morally in their relations with each
other" (p. 22). Thus it is similar to Durkheim's concept of mechanical integration, but operationalized for the governance of complex society. Further,
and related, I would argue, to Renfrew's observation that the grouporiented chiefdoms include large architectural spaces suitable for group ritual, Manu's ideological power is supported through the institution of what
he calls "aesthetic/ritual practices." I discuss the importance of ritual integration in corporate political strategies below.
By pointing to the integrative importance of group ritual and mechanical integration based on a corporate cognitive code, I do not mean to imply
that corporate political economy is necessarily a "primitive" social form,
well suited only to small-scale social formations but unsuited to larger-scale
societies where only centralized administration would suffice as a control mechanism (see below). Nor would I regard corporate organization as
simply a transitional evolutionary stage of the archaic state preceding larger
and more complex societies integrated through the centralized power of
rulership. Over a period of3,000 years, for example, regions of Mesoamer ican civilization displayed a complex pattern of cycling between more corporate and more overt ruler-centered social formations, rather than a simple
ton 1977: 33), peasant resistance and class consciousness (e.g., Kahn 1985;
Scott 1985, 1990), popular collective action (B. Moore 1978; Tilly and Tilly
1981), and other aspects of resistance to elite hegemony (Castells 1983) and
male dominance (e.g., Maher 1984).
The ritual expression of egalitarian behavior is found in rites of reversal
and symbolic inversion (summarized in Miller 1989:67-68; c Kertzer
1988: 131, 145; Scott 1985: 331), and in other aspects of the role of ritual in
political competition (Kertzer 1988). The roles of witchcraft and sorcery in
social leveling are discussed in Besnier (1993), Rodman (1993), Rosaldo
(1968:534), and Rowlands and Warnier (1988). Forge (1972b:533-34) reminds us that reciprocal gift exchanges do not necessarily result in social
equality among exchanging parties. Instead, in reciprocal gift exchanges, "to
be equal and stay equal is an extremely onerous task requiring continued
vigilance and effort"-in other words, egalitarian behavior.
Few sources directly address the evolution of corporate political economy in archaic states. Haas (1982: 112), for example, mentions the paucity
of studies of resistance to authority that can be found in the archaeological
literature. One possible reason for this lack of interest is the persistent idea
that the European enlightenment's social contract theory is largely a recent
and culturally specific phenomenon, with little relevance to non-Western
states. Lamberg-Karlovsky (1985: 23), for example, argues that
The "moral obligations" imposed on rulers for constituting freedom, equity
andjustice are asold asthe ancient palaces and temples ofthe Near East. Such
"moral" concepts continued to receive affirmation whether in the Acropolis
of democratic Athens, the Magna Carta of King John, or the principles of
Liberte, Egalite, and fratemite of the French Revolution. As certain as these
concepts are pivotal to Western Civilization, they are foreign to the political
ideology of ancient Egypt, China, and India.
My understanding of the nature of egalitarian behavior leads me to propose
that Lamberg-Karlovsky's contrast of Western and others is overdrawn, although one of the societies he mentions, ancient Egypt, did not, in fact, develop corporate political economy (e.g., see Humphreys 1978).
The form of egalitarian behavior developed in the stream of Western
civilization, extending in time from the ancient Near East to the modem
democracies, is unique in some important ways, a point discussed below.
A less sharp distinction between the Western pattern and others is evident,
however, when the researcher adopts a more broadly conceived notion of
egalitarian behavior in archaic and modem societies governed by states (e.g.,
Scott 1990: 81). In the following, I discuss several of the elements making
up such an expanded conceptualization, but it should be kept in mind that
this subject matter is exceedingly complex and involves a higWy varied domain of human experience. The suggestions I put forward constitute only a
small step toward a lengthier discussion of this seminal topic that I hope will
develop in the neoevolutionist literature in the coming decades.
Classical Athenian democracy (the polis state) is the best known example
of assembly government in an archaic state society (Humphreys 1978). Evidently, owing to this example's direct historical connection to the rise of democracy in western Europe, the United States, and elsewhere, it is thought
to be uniquely Western and hence has not figured importantly into the
broader neoevolutionist literature on early state evolution until recently
(Ferguson 1991; I. Morris 1987, 1994:43; Runciman 1982; Small 1994;
Snodgrass 1986). But the Athenian experience is only the best-known example of a more general type of political structure found in other eastern
Mediterranean polities that included representative assemblies, such as the
Syria/Palestinian region in the second millennium BC, early Israel, Phoenician city-states, and Carthage (summarized in Humphreys 1978: 181- 82; cf.
Snodgrass 1980:32). As Humphreys (1978:181) points out, the presence of
assemblies in these cases can be traced historically to the governing assemblies of pre-Early Dynastic Mesopotamia, as first described by Jacobsen
(1943,1957; cf Lamberg-Karlovsky 1985).
It should be, but unfortunately has not been, of considerable theoretical
interest to anthropological archaeology that Mesopotamian social formations appear to have emphasized corporate forms of government from an
early period, including forms of assembly government that evidently had developed by the Uruk period (3800-3100 BC) (Jacobsen 1943, 1957). Even
for the preceding 'Ubaid period (5500 -3800 BC), Stein (1994b) found it
necessary to develop what he called a "revised chiefdom model" because the
usual attributes of chiefdoms as described in the neoevolutionist literature
are absent. As he points out, 'Ubaid lacks elite burials (p. 39), iconographic
evidence of chiefs (p. 39), as well as "exotic, rare, or high-status trade goods,"
or elite craft goods (p. 40). Instead, the 'Ubaid people developed "an egalitarian ethic ofshared group membership and ideology" based on an elaboration
of temple ritual (1994b: 43; cf Oates 1977: 475). Although the evidence indicates state formation during the subsequent Uruk period (3800-3100 BC)
(Johnson 1973), there is considerable agreement, following Jacobsen, that
assembly government existed during this period and that rulership and
palace institutions developed only later in the Early Dynastic periods of the
third millennium BC (Adams 1966: 138 - 41; Oates 1977: 475 - 76; Postgate
1992: 80 - 81, 137). The contrast ofUruk with the Predynastic and Archaic
periods ofEgypt, given the abundant evidence in ancient Egypt of elaborate
elite tomb burials and early forms of pharaonic and palace symbolism, is
striking (Kemp 1989: chap. 1; see Baines and Yoffee, this volume). Two
highly distinct patterns of sociocultural evolutionary change leading to early
state formation are suggested for these two regions, with Egypt's built principally around exclusionary political economy while Mesopotamia's was
more corporate.
Linduff (1988: 317-37) note that during the Western Zhou period (1100770 BC), a bronze technology developed that was in wider use, displayed a
greater secularization of design motifs, and was more commercialized, by
comparison with the bronze-based prestige-goods system of the preceding
Shang dynasty. A debate between state authorities and Confucians during
the reign of Chao Ti of the Han dynasty (in 81 BC), reported in the "Discourses on Salt and Iron" (Esson 1931), affords a glimpse into the social contexts of early Chinese iron production. During this debate, the Lord Grand
Secretary notes several reasons for maintaining the state monopoly of iron
production: (1) to increase the productive efficiency of agriculture (e.g.,
p. 7); (2) to generate revenues for the state (e.g., p. 41); and (3) to facilitate
an equitable distribution of goods (pp. 6-8). The latter is threatened, he argued, when powerful producer families and "vassal houses" (the aristocracy)
gain control over industries such as iron (pp. 54, 87).
In addition to the materialist factors discussed to this point, cognitive
codes also facilitate corporate regulation of sources of power. One important strategy is giving persons of a wide range of social statuses access to positions of power-what Weber called "democratic administration" (1978:
948). In this and similar situations, a cognitive code is developed in which a
prime determinant of access to positions of political power is exemplary
moral and/or ritual behavior (similar to Kelly's [1993: 14] "prestige-stigma
hierarchy"). As expressed by Dumont (1980:56), in these cases, the "distinction of purity is the foundation of status." Cognitive codes of this type
reflect egalitarian behavior in that positions of power in society may be
opened up to persons of a wider range of social ranks, rather than being ascribed or restricted to those who control material resources.
Moralistic and ritualistic determination of social status is an outcome of
egalitarian behavior also in cases where the system of moral rules applies to
all persons in society, including rulers and other power wielders (e.g., see
Norbeck 1977: 73). In Eisenstadt's formulation (1980: 851, 1981: 157), such
moralistic cognitive codes are predicated on the development ofa "symbolic
problematization." Here, what I am calling the corporate cognitive code envisions a "basic tension between the transcendental and mundane orders" in
which "political order is conceived of as lower than the transcendental one
and . . . had to be structured according to the precepts of the latter"
(1981: 159). Moral orders so constructed are corporate in that political actors, including rulers, can he called to judgment and deposed for failure to
comply with the dictates of the moral code. For example, according to the
Confucian system, the Chinese ruler's behavior was "open to official reprimand" (Weber 1951 :31-32; c c.-Y. Hsu 1988: 180-81; Wei-Ming 1986).
According to c.-Y. Hsu (1986:308),
the significance of the Chou conquest was the emergence of the Mandate of
Heaven which endowed the supreme being with an authority as well as the
responsibilityof ensuring the people to be governed by pious and righteous
rulers. I think such a transformation of characteristics of supreme being in
ancient China should be regarded as an important breakthrough.
Such corporately inspired mandates, whereby the ruler is the upholder
of the "purity of the collectivity" (Eisenstadt 1981: 175), are distinct from
cases of "spiritual domination" (Weber 1978: 54-56), where divine kingship (e.g., in ancient Egypt) is an exclusionary political strategy predicated
on control ofaccess to the supernatural. Ancient Egypt is an example ofWeber's (1978:54-56) "hierocratic organization," where there is monopoly
control ofwhat he terms "religious benefits." The Classic period Maya concept of kingship (ahau) provides an additional example of spiritual domination. This concept placed rulers at the center of the temporal cycles of
history, shamanistic prophecy, curing, and divine ancestor worship (Freidel
and Schele 1988; Schele and Miller 1986).
Although it is not my purpose here to elaborate on the varied material
manifestations of corporate political economy, note that the form of the major Egyptian temples, with their sequence of hallways, passages, and rooms
arranged hierarchically along a central axis (e.g., Badawy 1968: chap. III;
Kemp 1989: 94), places the image of the temple's deity in a secluded back
space accessible only to the pharaoh and a few high-ranking priests or
temple officials. This placement is consistent with exclusionary control of
supernatural forces. Such exclusionary architectural strategies date to the
earliest periods of Egyptian temple construction (Kemp 1989: 66 - 83).
These architectural forms can be contrasted with the plan for the Islamic
temple ofKufa, on the east bank of the Euphrates, built in AD 636. Its more
corporate aim was to band "together disparate groups through a collective
variety of worship that knew no priestly hierarchy, mysteries, or hidden
sanctuary within the precincts of worship itself" (Campo 1991: 60).
In addition to architectural layout of the temple, the potential for monopoly control of supernatural forces also can be limited by extending important ritual practices to households ofall social statuses. A decentralization
of ritual emphasizes the commonality of all of society's constituent elements
and the mutual interdependencies of varied social sectors in the reproduction of nature and society (Miller 1985: 61). The gradual evolution of the
pattern of ancestor worship in China is a case in point. Earlier forms of ancestor worship revolved around images of the prominent ancestor gods of
clan groups; their use in sacrificial rituals was a "prerogative of the aristocracy" (Laufer 1965 : 449). In the Confucian reforms, by contrast, clan structure was de-emphasized and an elaborated code of household behavior
substituted for it, informed by a concept of filial piety that extended to fam-
mentioned above. Below I describe several of the variety of reflexive communicative channels that developed in archaic states. But in archaic states,
generally, this kind ofbe havioral monitoring is problematic. Institutions capable of correctly identifying breaches of norm-governed political behavior,
and of communicating information about it that makes possible an appropriate redressive action, are typically not well developed. One of the major
differences between the archaic states and modem democratic states is to be
found in the complexity and effectiveness of the instrumentalities for reflexive communication (Apter 1965: 456; Wirsing 1973: 163). Typically, in archaic states, mechanisms of enforcement of the cognitive code are only
weakly developed (Eisenstadt 1980: 847).
In the Axial Age civilizations, reflexive communication was vested primarily in a cultural elite who served as carriers of the moral order, such as
Jewish prophets, Greek philosophers, Chinese literati, Hindu Brahmins, and
Islamic Ulema (e.g., Eisenstadt 1980). The role of the vizier, minister, or
similar high-ranking administrator or advisory council (e.g., in Benin; Bradbury 1969) is often pivotal in the reflexive communicational capacity of
archaic states. Persons in these roles maintained considerable political autonomy, were highly versed in society's moral precepts, and were able to observe the ruler's behavior in specific political and ritual situations closely.
Thus, they could authoritatively detect and report deviations from accepted
practice. (By contrast, in ancient Egypt's highly centralized political system,
the vizier was little more than an administrative assistant to the pharaoh;
Eisenstadt 1969: 279.) Moral evaluation of those who govern also may occur more broadly in society, for example, when Roman emperors were
judged by their demeanor in public venues such as the Circus, or by reference to the "sobriety of their private lives" (Veyne 1990:304). Roman imperial edicts, Veyne (1990: 300) points out, often evinced a concern "not so
much to be obeyed as to prove to his people that he shared the principles
and sufferings of his subjects; as if the law was not essentially imperative but
aimed also at bearing witness." In the next section, on ritual sanctification
of egalitarian cognitive codes, I discuss additional elements of reflexive communication as it is manifest in the ritualization of political communication.
4. Ritual Sanctification of the Corporate Cognitive Code
and Ritualization of Political Communication
Ritual may serve as a kind of "primitive government" by making its underlying propositions appear "powerful and holy" (Bloch 1974: 57). According
to Bloch, it is "precisely through the process of making a power situation
appear a fact in the nature of the world that traditional authority works"
(1974:79; cf Bloch 1977a; Rappaport 1979:197). This view is like the
publicly proclaimed. Thus ritual expression may carry with it what Rappaport (1979: 196) calls the "notion of obligation." For example, oaths of
office in Classical Athens "marked the assumption of political roles and reiterated the norms governing behavior in political contexts, particularly the
obligation to leave aside all personal interests" (Humphreys 1977: 348). As
Rappaport puts it (1979: 196), "the formal and public nature of liturgical
performance makes it very clear that an act of acceptance is taking place, that
the acceptance is serious, and what it is that is being accepted."
This argument can be further elaborated by recasting Maurice Bloch's
(1974, 1975) interpretation of "political language." Political language is a
particular kind offormalized language that includes elements such as the extensive use of homilies, patterned timing and emotive levels, and separation
of the speaker from invariant text that has been "handed down from ancestors" (Bloch 1975: 8). Bloch argues that this formal communicative mode
enforces a hierarchy of social statuses by placing speaker and hearers in a
"highly structured situation" (p. 9) that provokes acceptance, what he calls
"illocutionary force" (1975: 22; c Bloch 1974, 1977a, 1980). From the
perspective of corporate strategy, however, I see political language serving as
a channel of reflexive communication. Its precise form and often lengthy,
complex ritualization require skill acquired through extensive training and
rehearsal (Bloch 1975: 22; cf Geertz 1980: 130). As such, it is metacornmunicative; even though political language may not transmit much actual information (see below), its demanding requirements allow viewers and
hearers to assess the degree to which an act of acceptance of the cognitive
code underlies the formalized behavior. It is reflexive communication because it allows, as Bloch himself notes (1975: 23-24), an evaluation not so
much of the message, but of the speaker (as illustrated in the case of Confucian China, see Creel 1970:96; Elvin 1986:328-31; Weber 1951 :31-32).
In addition, ritualization of political communication, as Bloch (1977b) puts
it, disconnects power from rank in the sense that "If the ruler totally adopts
the mode of communication of ritual for carrying out his will he at the same
time loses his ability to affect events for his own personal ends as opposed to
the ends of his office" (p. 331).
Metacommunicative ritual and political language are important dimensions of communication between those who rule and their subjects in all societies governed by corporate codes, but they are particularly prominent in
archaic states. Certainly not all political communication relates to adaptive
control functions, as the systems analysts would have us believe, nor would
it necessarily consist of heroic history or other interest ideology that is expressed in narrative form. In political language, the message may not be expressed in ordinary language, and thus is not capable of communicating
much information ofany kind. The communication taking place is, as Bloch
(1975: 13-15) puts it, "impoverished." Owing to its formalization, the ritualized communication of political language is what Bernstein (1965, 1971)
calls a "restricted code" (cf Humphreys 1977:360-67, 1978:262-75). It
is a type oflanguage whose connection to reality is more tenuous than that
found in ordinary language.
Applying the concepts of metacommunicative ritual, political language,
and restricted code may help us to understand the enigmatic mode of expression of the masked and elaborately costumed figures in murals and other
artistic media at Teotihuacan. Although writing is minimally used, the portrayed figures often show elaborate, largely undecipherable speech scrolls. As
Pasztory (1992: 305) notes: "At Teotihuacan, writing exists, but it is speech
that is glorified in art. This speech is entirely local in character; it is among
members of the group who know what they are saying, and it is not clarified
for an outsider. One can decipher Maya hieroglyphic writing, but one cannot decipher most speech scrolls."
The pervasiveness of ritual and ritualized political communication in
such societies, however, entails a significant dilemma. According to Bloch
(1980: 94), "Formalized-ritualized communication ... does not bear direct
relation to events; new experience neither modifies nor contradicts its content directly because its form rules out such modification and contradiction." The dilemma that arises in the context of many archaic states results
from the time and resources that officials of a state- if they are to maintain
their legitimate hold on power-must devote to forms of communication
not involving ordinary language. Particularly in cases where political metacommunication and ritual sanctification are greatly elaborated, such as in
nineteenth-century Bali, the demanding requirements of exemplary ritual
behavior "imprisoned" the king in the ceremony ofrule (Geertz 1980: 133).
The inefficiency and costliness of this is apparent, especially near the top of
the governmental hierarchy, "where the 'far beaming blaze of majesty' consumed so much more fuel, the necessity to demonstrate status warred with
the necessity to assemble the support to make the demonstration possible"
(Geertz 1980: 133). In a similar vein, Kilson (1983 :421) notes that Ga ceremonial kingship "consumes energy, time, and money but it does not
produce solutions to poverty, urban congestion and disease. Ceremonial
kingship recreates a heroic tradition, but fails to create new worlds."
5. Semiautonomy of Lower-Order Subsystems
In the previous two sections, I argued that the implementation of corporate
political economy is predicated on the development of additional communication channels that allow an assessment of the degree of compliance of
power holders to the expectations of the corporate cognitive code. The
strategy I refer to as semiautonomy also requires alterations in communica-
tions channels, but along other lines. In this strategy, lower-order subsystems
are reconfigured so as to operate more autonomously. This allows a reduction in the amount ofinformation flow between lower-order subsystems and
central control systems. The sociocultural evolutionary outcome of semiautonomy is a reduced degree of centralized control in society, the opposite
ofFlannery's (1972: 413) process oflinearization, in which lower-order subsystems lose autonomy. Although not all governing functions will be vested
in such low-order institutions, the greater degree of control situated at lower
hierarchical levels necessarily reduces the scope of centralized administrative
action in society. From the systems-analysis perspective (e.g., Rappaport
1977 : 54, 61), gains from decentralization include information-processing
efficiency and adaptive advantage, but in my perspective, semiautonomy
may reflect the outcome of political processes as well.
Semiautonomy is predicated on the development and promulgation of a
cognitive code that provides a detailed program for the conduct of everyday
life that can make subsystems, such as households, more self-governing; Islam and Confucianism (discussed below) are two major examples. However,
not all such codes are corporate in nature. Blau (1968: 465) describes a strategy, similar to Fesler's (1968: 373) "pseudodecentralization," in which there
is the promulgation of "impersonal control mechanisms that constrain operations to follow automatically the policies and programs specified by top
executives" (Blau 1968 :465). Control mechanisms of this latter type are
egalitarian only if the prescribed code applies equally to the behavior of
rulers and other officials as it does to ordinary households or other social entities and individuals, and if the intent of the code is to decentralize political controls. Blau's pseudodecentralizing codes are, in contrast, strategies for
the "subordination of the bureaucracy," as indicated in table 5.3, and are
manifestations of exclusionary systemic domination. In corporate systems, it
is not likely that new codes will be developed and promoted by "top executives," as Blau indicates, since the aim of these codes is to constrain central
authority, not augment it.
The devising of new social codes to promote semiautonomy in lowerorder institutions need not imply a decomposition of society into completely autonomous local systems; in fact, no net decline in integration need
occur. Corporate society may be highly integrated, but the modalities ofsocial integration, control, and communication are changed. Some aspects of
central control are kept in place, owing to needs such as external offensedefense management. But as an outcome of semiautonomy, higher-order
controls have fewer functions and will be less able to exercise centralized
power in at least some domains of societal governance. In addition, semiautonomy is not complete social decomposition in the sense that the code
and its accompanying ritual expressions often are found in the practices of
CONCLUSION
My comments in this chapter represent a limited effort in light of both the
complexity and the significance of corporate political strategy in the sociocultural evolution of archaic states. In these brief comments, I have tried to
argue convincingly that our discipline needs to move away from the excessive emphasis on political centralization that has grown out of its use of
Marxist and systems approaches. I have also tried to illustrate the difficulties
inherent in seeing the accountability of rulers as solely a product of Near
Eastern and Western political evolution (based on the "significant breakout"
of the ancient Near East). Corporate behavior is found outside the Near East
and the West, and it takes many forms; the accountability of rulers is only
one dimension ofa far more complex political process that at this time is not
well understood or very much studied.
Rather than sharply contrasting "the West and the rest," a productive
avenue ~f compara~ive inquiry will be to trace the evolution of corporate
systems III all of their considerably varied cross-cultural and temporal forms,
and to compare them with situations in which exclusionary political strategies dominated. Why do some centralized social formations, like ancient
Egypt,. evolve w.ith little corporate structure? Why are some corporate
st.rategIes emphasI~ed and not others? China's "significant breakout," beginmng before the middle of the first millennium BC, was built around an elabor~te theory relating a system of ethics to ritual. In my conceptualization,
this brought with it a cognitive code emphasizing ritual sanctification and
ritualization of political communication, reflexive communication, status
determination through moral/ritual exemplarity, and semiautonomy, as
we~ as a distributive Iron Age. An older Mesopotamian pattern, probably
datmg to the Uruk period and perhaps with antecedents in the comparatively egalitarian 'Ubaid period, was carried forward into the eastern
Mediterranean and beyond. In it, by contrast with China and many other
archaic states, modalities of assembly government were central to the evolution of corporate strategy.
. These specific cases require explanation with corporate strategy kept in
mind, but general trends in sociocultural evolution also need to be addressed
by comparative research. When viewed in the most general evolutionary
sense, a trend is evident in the evolution of archaic states, and in the evolution from archaic states to modem states, from intermember to systemic
power types, from material to symbolic bases of power, and from exclusionary to corporate power, but these broad trends have not been simple unilinear changes or simple evolutionary stages. Such broad patterns of political
ch~nge .c~nnot be .explained by an anthropological archaeology that only
builds rigid evolutjonarv stage typologies and that sees cultural evolution
primarily in terms of one political process-centralization.
At present we have few theories to explain the rise of corporate governmen.t or the cross-cultural variation therein, since most of our discipline's attentron has been paid to identifying the socioenvironmental stresses that
bring about centralization (Flannery 1972). Which socioenvironmental
stresses brought about the evolution of corporate strategies in the archaic
states? Discontent with the outcomes of exclusionary power may prompt
change, but we need to know more about the specific sources of stress and
how discontent is articulated, so as to reconstruct the social order (Eisenstadt
1977). A growing wealth inequality may have fiieled the prevailing clientalism in Attica prior to the revolutionary rise of the assembly-governed
polis state about the middle of the fifth century BC (1. Morris 1987: 202 - 5;
van der Vliet 1987). Powerful tyrants accumulated fortunes in part through
their control of the silver trade, and silver also was used by them as a prestige good in politically charged exchange (van der Vliet 1987 :73, 75).
A central aim of the polis system was the regulation of the politicalgifting and client-based intermember politics. Millon's view of the revolutionary rise of the 'Ieotihuacan system, beginning in the Tlamimilolpa
phase, sees behind it an established, highly centralized state rather than a
tyranny of clientship based on wealth inequality. In Teotihuacan, he (19~2:
400) hypothesizes, "Effective means were institutionalized by a collective
leadership to prevent rulers from transforming executive authority into personal rule. The rejection ofpersonal rule was so thorough and so lasting, and
the checks on rulers so effective, that it placed its stamp on Teotihuacan's history for another 500 years, profoundly affecting its political system, its religion, and its art." From Stein's (1994b) discussion of the highly egalitarian
and evidently not politically centralized 'Ubaid period, it seems unlikely
that either of these two situations-wealth inequality or a powerful statecould have prompted the development of corporate political economy in
early Mesopotamia. I hope that Near Eastern archaeologists will address this
perhaps unique and important sequence from the perspective of the evolution of egalitarian behavior, but anthropological archaeologists working in
other areas should devote increased attention to corporate political economy
in the many forms it has taken.
NOTE
I am grateful to Doug Schwartz and the staff of the School of American Research, who
provided a comfortable and stimulating environment during our weeklong meeting.
Comments I received from the other session participants helped me think more carefully
about some of my topics and issues; I also acknowledge the helpful comments on an early
draft of this essay made by Stephen Kowalewski, Peter Peregrine, and two anonymous
reviewers. My wife, Cindy Bedell, helped me solve several terminological problems. I
particularly thank Gary Feinman and ]oyce Marcus, who carefully and patiently read
through and edited several earlier versions of this chapter; however, all errors are my
own.
,
Uruk ~totes in ~ollthwesterlllrOIl
H~NRY T. WRI6HT
nthr?pology develo~ed within the institutions of state polities. The perceptions and th~oretIcal perspectives ofanthropologists have been forged
by the contradIctory demands of our research disciplines and of the
peoples, living and dead, whom we have come to represent. It is therefore
not surprising that we want to know as much as possible about the kinds of
entities that give contexts for our work, including how states came to be and
?Ow they came to differ from the diversity of other kinds of polities. Such
mterests lead to a focus on the earliest states. There is, however, much disagreement among scholars about the characterization of early states. Are
they theater polities, mere symbolic games in the minds of self-appointed
princes? Or are they apparatuses that controlled the material resources and
lives of their participants? Such disagreements benefit from consideration of
the evidence of past states.
. Prim~ry states, those that arose from interaction between prestate societies (Wnght 1977), rarely had the capacity to record discursive accounts of
their ge~esis and operation. The task of understanding how such polities
worke~ IS largely the responsibility of archaeologists, who must proceed by
ev~luatmg the consequences of different ideas about the past with material
~vIdence, criticizing the assumptions of proposals that fail, generating new
Ideas, and evaluating these yet again. The process is tedious, but it has enabled archaeologists to go beyond idealist and racist conceptions of cultural
formation, beyond elementary material/energetic theories of change, to the
more complex systemic and existential conceptions that are the focus of research today.
The application of such approaches during the past few decades in
widely separated parts of the world by the small cadre of anthropological archaeologists has generated far more insight into operation than genesis. The
reasons for this disparity are various. On the one hand, the periods in which
primary states originated were times of conflict, characterized by the movement of communities and by general human misery. The relevant archaeological layers are often ephemeral, badly eroded, or deeply buried by the
debris oflater, more prosperous periods; the modest discarded material remains have attracted less interest than they merit. On the other hand, even
if we could obtain the archaeological record needed, we would still lack the
theoretical structures essential for talking about the complex social, symbolic, and material transformations that seem to have been important in the
origins of states. It may be useful to consider the better-known operation of
developed primary states, where in some cases we do have richarchaeological records of material remains and even documentary records of individuals' actions, beliefs, or ideological claims. In this essay, I take a systematic
perspective to discuss the operation of one of the regional states of the midfourth millennium BC in Southwest Asia.
Scholars' assessments of mid-fourth-millennium polities are evolving
rapidly. Four decades ago, when I first read about this period, it was viewed
as a time of urban growth. Each developing town was centered on its temple,
whose priests and dependent craftsmen were among the few specialists in a
largely egalitarian world of farmers. The most developed governmental institutions were thought to be assemblies of prominent citizens (c Frankfort
1956: 52-78). Rulers, dynasties, and state administration were thought not
to have arisen until the middle of the third millennium BC. This assessment
was based upon a concordance of three classes of evidence, each of which is
open to question.
First, the earliest known written records of rulers and dynasties were of
the mid-third millennium BC. However, such political statements are only
possible with writing systems able to represent language. Mesopotamian
writing prior to this time appears to have been graphic systems for representing numbers and objects, not language; it could have conveyed only indirect economic information about politics.
Second, the earliest architectural evidence of palatial residences separate
from temples also dates to the mid-third millennium. This seemingly late appearance, however, results from the erosion at many southern Mesopotamian town sites that has exposed third-millennium layers. Early in the
twentieth century excavators were able to expose a number of mid-thirdmillennium towns and to identify temples and palaces. Since temples tend
t
1
Ur
Eridu
SouthwestIran and
AdjacentRegions
100
L----l-_I
km
-:
l~ ... ~ /
I I
\ \
"\
,./
I
I'
II
<::::::::,
I I
Pkm
~-~~,
.~~
'I
~I~
~
"-
7-
..........
\ \.
'
I
~I\'./
",I
"
)I
I~ ~ (:
:( ----1-.,
KS-36
0, ~
->
; ,\
/ Ii
.
I'"?
I "\ \ ,,4:..'"- ~",
Chogha
/; ~
'.,,:a...
MishI
I
,I
:
...\ ' , "4
~/,,-i',
KS-1~1'l'
I, \~\('" v.I S
'
,,'.,
- ' I . 1.1 "
-- 1'.,---- I
\
\
... / - . / 'I
"
...
,
1
\"".
usa
.,
--... 11
I
,.-., ~
.I.
,
'-:DJ "
,.
II'!'
.. ,~. I
I,
l' ~ ~.... \:... f i \ i
",
\~.
' "" ,
- 'I
I'
\ , \to
\
~'" ,/,I
.
'
, ,\ }
I}
"
'.i
.I ,; \) ...," ... ''. ,.. ... ,',_/
.
.-L ,
--
\ ,- ....<
Abu Fanduwehe
I
I
1
I
i""
'i l'
I.
~-
'
' '\~
...
\\
1\.
"
....,)1
"
-, '_.~ i!
i
"" \
.::-...
'" \ ,
~
Largecenter'
Smallcenter
...
Large village
Small village
!
I
~".--r"
'
/1/I
r" . /
.
~J
l,~ i
i (
~,,_~:,,--
~i,%.;\
~ .l,
r-____ -1I
I
. ) {
,.._Y,''------.~
~,
(;}
-~".
.,J (
..,
"
~DeZR:'\i
L _ ---'- _ -'I
.'
'\./
.KS-54'~.,~"
-.',
~~_,
Hypothetical route
Most of our evidence of Middle Uruk village life comes from a large pit
filled with seasonally stratified garbage (Wright, Miller, and Redding 1980;
Wright, Redding, and Pollock 1989). Miller's analysis of the plant remains
from this pit indicates that the people ofSharafabad cultivated fields of barley, wheat, and lentils and had access to irrigated flax. However, we cannot
estimate proportions grown. or eaten, since barley may enter the record in
the form of carbonized seeds through the burning of animal dung as well as
the accidental burning offood, whereas wheat is evidenced primarily by the
burned by-products of grain winnowing and cleaning. There is no evidence
of other plant foods such as orchard crops or wild foods. Redding's analysis
ofthe animal bones has shown that the inhabitants culled local herds ofsheep
and goats, primarily the former, killing lambs and kids in late summer and
older female animals in late winter. In the first agriculturally successful year
represented by the pit, animals of all types and ages were culled, following a
strategy that optimized the reproductive capacity of the herd. In the following year of agricultural crisis, the number of sheep and goat killed
dropped to 26 percent ofits former level, and there was a focus on the killing
of younger animals, even of females that would normally be saved for future
breeding stock. The villagers also butchered an occasional cow or pig,
hunted waterfowl, and fished.
Sharafabad, however, did not have a self-contained village economy.
Many of the things people used came from outside, and some of their activities were coordinated by outside authorities. Most of their durable
goods-pots, chert cores, grinding stones, and so on-probably came from
elsewhere. There is evidence that most jars and bowls came from one or the
other of the larger towns (Johnson 1973: 113-29; Wright and Johnson
1975 :279 - 82), perhaps depending on market opportunity. The other
goods could have come from specialists elsewhere or been obtained by the
villagers themselves. For the most part, vessels were discarded at similar rates,
regardless of season or of whether or not the year was prosperous. A cylinder sealing of nonlocal clay on a jar rim shows that at least some of the pots
came with goods sealed in them (James Blackman, personal communication
1980). Some sealings on bales and baskets, particularly several with impressions of a cylinder seal with a beer-drinking or "symposium" scene, may
represent the closure of goods by officials at another place and their subsequent dispatch to Sharafabad. Other seals-stamp seals with iconically similar representations of a canid and a person, stamped on both bale seals and
on the seals of storage room door locks-are those of an official or several
officials who either were residents or visited the site to authorize closure of
its storerooms. Regardless of whether the year was agriculturally good or
bad, the same number of bales and jars seemed to have been opened at
Sharafabad.
It is notable that storehouse doors were opened on the site, most often
in winter or early summer. While few discarded door sealings were found
in debris of the first prosperous year, many were found in debris of the second year, that of agricultural crisis, when we might expect institutional
storehouses to have been opened as domestic food supplies ran out. Conversely, in the good year, there is more evidence of involvement by
Sharafabad people in institutionallabor. The crude bevel-rim bowls, probably made for the issuing of institutional rations or meals (Johnson 1973:
129-39; Le Brun 1980; Nissen 1970: 137), were recycled in domestic tasks
and discarded at least twice as frequently in the prosperous first year as during the second year, indicating either that local people participated in more
institutionallabor or that more outside laborers came in when times were
good. Also, what are believed to be fragments of unfinished clay envelopes-"pherical bullae" wrapped around clay tokens indicating quantities of items (Amiet 1972; Schmandt-Bessarat 1992) and probably recording
goods leaving the site-were discarded twice as frequently during the prosperous first year, suggesting double the rate of transport from the site that
year. Sharafabad people seem to have profited from their labor in the good
year, discarding four times as many sheep and goat bones as well as more
bones of pig, cow, and wild animals. Duringthe bad year they lived less well
but received some supplementary help from institutional storehouses in jars
and bales, as noted above.
The large village site ofKS-54, the second rural site for which we have
a range of information, covered up to 2.5 ha during the Middle and Late
Uruk periods and perhaps had as many as 500 inhabitants. It is located in a
dense settlement concentration on the long-occupied, but more arid, southern edge of the Susiana Plain, west of the Dez River. Unfortunately, twothirds of the site was leveled before investigation could take place. Our
evidence comes from a variety of contexts, mostly of the Late Uruk period,
from a series of small excavations along bulldozer-cut faces transecting the
settlement (Johnson 1976). These soundings revealed social differences in
architecture, including the remains of a large domestic building with massive mud-brick walls in the center of the settlement and more modest structures elsewhere. Differences in possessions discarded around these two types
of architecture indicate marked social differences. Stone vessels, decorative plaques, and beads of gold and lapis lazuli were associated with the
elaborate building, but no such embellishments were found around modest
buildings.
Subsistence patterns at KS-54 were broadly similar to that at earlier
Sharafabad. Naomi Miller (personal communication 1974) examined the
plant remains from samples taken along this section at the time of initial
survey and found carbonized grains ofdomestic barley, glumes ofwheat, and
]ohnson (1973: 73, fig. 10) classified four of the larger settlements on the Susiana Plain as small centers from 4 ha up to 8 ha. None ofthese sites has been
excavated; we can only note that three of these interesting settlements are
on the margins of the cluster of settlements and may have been involved in
frontier activities.
The larger centers or towns (settlements larger than 8 ha) on the plain
seem to have been organizationally similar in their spatial structure, the production activities pursued in them, and their span of control, as indicated by
seals and sealings. All show a distinct separation between a smaller, upper
town situated on the mounded remains of an earlier settlement and a larger,
newly founded or refounded lower town. All have discarded sickle blades,
indicating participation of townspeople in the harvest, when ripened crops
must be brought in as quickly as possible. All have evidence of the production of pottery and stone tools. All have indications of the administrative
technology that in Mesopotamia implies control of the storage of goods.
However, the three known larger towns differ in their historical trajectories,
in the maximum size they reached, and in the particulars of their economic
and political organization.
Susa, on the west edge of the plain, was by far the largest; its 9-ha upper
town was established during the Early Uruk period on the remains of the
great ritual platform of the late fifth millennium after a brief abandonment.
At the beginning of the Middle Uruk period, the lower town extended several hundred meters to the north and covered at least 25 ha. Evidence of
larger buildings (Dyson 1966; Steve and Gasche 1973) and administrative
activities comes from the upper town (as detailed below). However, evidence of a large pottery kiln is well reported from the lower town (Miroschedji 1986), and there may have been larger buildings there as well (Steve
and Gasche 1990). There is little evidence this lower town was occupied
during the Late Uruk period.
Abu Fanduweh (KS-59), on the south edge of the plain, was much
smaller. Its 4 ha upper town was established during the Early Uruk period
on the remains of a platform of the late fifth millennium; its lower town extended to the north during Middle and Late Uruk times to cover an additional 6 ha or more. Excavation was limited prior to the cessation of
fieldwork in 1978; the only evidence ofadministrative activity is a door sealing from the surface of the upper town. On the northern periphery of the
lower town, however, a large series of small pottery kilns has been recorded
(Johnson 1973: 107-8).
.c.....
11'
~""~~~
,.'
.'11.&'0
~,,/
. .?&~.--_._"
/'
in the region, such as alabaster, calcite, or bitumen. Just outside the door of
this small room was a possible kitchen area (Locus 830) with a single spherical bulla and an oblong bulla with several impressions of a single cylinder
seal, which could be used either for verification of other sealings or validation of a messenger.
The other unit is L-shaped and somewhat smaller. Its three rooms have
exterior access but no access to each other. The largest room has a sunken
oval fireplace and a raised fireplace, but only a few sherds and a toothed
mandible of a sawfish on its floor (Locus 759). Another room with an oven
and mortars may have been a kitchen (Locus 773). The L-shaped unit also
has a small room with dense concentrations of discarded material on its floor
(Locus 797). There are domestic artifacts such as a bottle and bottle sealing,
many cups, bowls, and grinding stones, and a spindle whorl. Administrative
artifacts such as a cylinder seal and a numerical tablet were recovered. There
also are four alabaster and calcite maceheads of a type fixed onto a handle
with straps, some finished and some unfinished, perhaps for social coercion
or display. Oblong pecking stones may have been used to manufacture these
maceheads and other stone items.
The assemblages on the floors of these units are probably items not
deemed worthy of systematic disposal or recycling when the buildings were
leveled. Scattered items are difficult to interpret, but the concentrations in
small rooms are stored items that either spoiled or were broken in place. The
regular occurrence in both storerooms ofbottles or tall jars and cups or small
bowls probably indicates routine serving of liquids in each unit. The seals
and numerical tablets probably indicate the general involvement of the occupants in record keeping. It is notable that both ofthe cylinder seals in these
two units are small and share spiderlike motifs, but they were not used to
impress any of the other items found. The numerical tablets in these units
had impressions of large cylinders with flies of animals, or with scenes in
which human figures manipulate jars, or with both. The numbers recorded
are small, ranging from 3 to 22, but we do not know what commodities
were monitored and whether the tablets record past deposits, promised future transactions, or both (Le Brun and Vallat 1978). The concentration of
macehead production debris in one unit and social display items in another
indicates individual specializations just before rebuilding of the complex.
But similar concentrations occur on floors of the succeeding Layer 17B 1, albeit in different rooms, so these specializations were enduring and not
merely a result of unique circumstances. The debris pattern suggests these
special activities-presumably in support of social rituals-were performed
only episodically by the inhabitants of this residential area.
From bullae found long ago at Susa (Amiet 1972), we know that during
the Middle and Late Uruk periods summary documents reached Susa, were
assessed by a higher authority, and then were stored or discarded there.
However, we have very few well-excavated contexts that can be informative
about the organization of the Middle Uruk control apparatus. Nevertheless,
still-incomplete excavations on the Northern Acropolis, the "Acropole Ill"
excavation, provide useful details of office life (Wright 1985). The building
in question, that of Layer 2, seems to have faced north within a rectangular
space between a vacant area (fig. 6.4, Locus 157) and the court of another
building. One crossed a courtyard (Locus 163), entering the building
though a vestibule on the left (Locus 160), and turned right into a large
room (Loci 131/154) with only a simple surface hearth in the center (Locus
159). From this room one could enter a corridor (Locus 106) which gave
access to several small rooms at the rear of the building, now largely eroded
away. The single surviving nearly complete rear room (Locus 102) had very
small hearths, one deep and one shallow. This building had been carefully
cleaned prior to its finalleveling; only a small stone vessel in a niche by the
t
o
~---
(-
-~.
@J
\
\
-,........ /
112
\
\
\
door (1021.01), a bevel-rim bowl sherd, and a cup sherd remained on its
floors.
Excavation revealed a pit containing debris probably discarded by workers in the immediate predecessor of this building, which was very similar to
the building described here. This pit contained many nearly complete
bevel-rim bowls and some small jars, sheep/goat bones, charred debris, and
various artifacts resulting from administrative activity. Among these were
bale and basket sealings with the impressions of perhaps eight different seals,
most frequently of two small cylinder seals, one representing a spider and
one with rows of small figures, presumably removed from incoming packages. There also was a sealing from the opening of a storeroom. In contrast
were items probably indicating outgoing summary information, both unfinished bulla coverings and tokens representing various quantities ofitems. Although we can presume no necessary relation between the opening of the
packages or storerooms and the preparation of the summaries-the packages might have been incidental to the central activities of what we might
term an "office"-numerical records were prepared here. Whether these
records were summaries of prior transactions, as argued for Sharafabad, or
agreements for future transactions, as may have been case for the Late Uruk
Southern Acropolis, their preparation indicates the recording ofprecise numerical information for future reference.
We learn almost as much about the organization of the Middle Uruk
control apparatus from the way this public building was constructed and repaired as we do from the evidence of activities within it. It was laid out as a
rectangle of18 by 15 units (each measuring .65 m), similar to the larger unit
in Acropole I, 17B2. The use of similar units of measure indicates that masons had strict standards, as did other crafts such as potting and stone working, but this need not imply centralized control of these crafts. The walls of
the Acropole Ill, Layer 2 building were not laid precisely upon the footings
of the earlier buildings in the area, so the footings settled irregularly into the
softer sediments of former room fills. To remedy this, the mud-brick walls
were cut down almost to floor level and relaid with a brick of slightly different size, suggesting that a different work crew was called in to deal with
the problem. Finally, thick mud-plaster floors had to be put in place to cover
additional irregularities. Thus, the building hierarchy was able to correct for
the errors of imposed architectural design.
After some period of use, the building required minor replastering and
reflooring. A pit was dug inside the building, blocking a key doorway (see
fig. 6.4, Locus 117), and presumably used to make mud plaster. Later, slabs
of old plaster were dumped into the pit. Then a set of complete bevel-rim
bowls (in which rations or institutional meals of stew or bread could have
been served, as noted above), a few small jars and their jar stoppers, and some
sheep/goat bones-the durable remains of a meal for a number of people
very similar to meals consumed by the building's normal occupants (as represented by the external pit)-were thrown into the internal pit. A large
mass of evidently unneeded still-wet mud plaster was dumped in, followed
by a load of miscellaneous stone and battered sherds. Finally, its purpose as
a source of mud and as a receptacle for the debris of the repair operation
finished, the pit was sealed beneath a new mud flooring. Thus, when the
building needed repair, it was evacuated, a work crew subsidized by an institutional kitchen was called in, and the work was finished in a brief period
oftime. In sum, when maintenance was needed, a higher-level organization,
whether the same organization that built the building or not, could see to
its repair with dispatch. At least one modest urban organization was able to
deal routinely both with its basic operations over time and with errors in the
construction and operation of its material infrastructure.
In the six or more centuries during which Susa reached a peak of prosperity, declined as Chogha Mish grew to become a competing center, and
finally was abandoned, the peoples of the marginal valleys developed varying relations with larger centers. The Deh Luran Plain is centered 110 km,
or three to four days' travel time, to the northwest of Susa in the foothills of
the Pusht-i Kuh, the front range of the central Zagros (see fig. 6.1). The
plain is a relatively large foothill valley, covering about 940 km>, watered by
the perennial Mehmeh and Dawairij (or Ab Danan) Rivers. Most of its
rolling surface is difficult to irrigate, though well suited to grazing and dry
farming. Deh Luranis had direct access to oak forests on the flanks ofKabir
Kuh to the north, bitumen seeps, and gravel beds rich in cherts and other
useful stones. There was easy access not only to the Susiana Plain, but to the
Mesopotamian plain to the southwest; it is not surprising that the small communities of the Deh Luran area were often dominated by other areas. We
know of fourth millennium events on this plain through detailed archaeological survey (Neely and Wright 1994) and through excavations at the small
center of Farukhabad (Wright 1981).
The Deh Luran Plain was resettled during the Early Uruk period after
several centuries of near abandonment. In Middle Uruk times there were
about 770 people on the plain, only a third of the population maximum
reached a millennium earlier. Settlement hierarchy had only two tiers-the
little center at Farukhabad and four or five small villages (Neely and Wright
1994: 208, table VI. 16). Small excavations in the center revealed fragments
of modest mud-brick houses, ovens, and the familiar technology of blades,
sickles, grinding stones, bowls, and jars. The striking feature of this assemblage is that the pottery is not a development of the local Susiana traditions
but is in the "chaff-faced" tradition that became established during the late
fifth millennium in the mountain and foothill valleys of the Zagros, antiTaurus, and Amanus Mountains.
Our small soundings did not reveal evidence of elaborate Middle Uruk
buildings, but the presence of mother-of-pearl furniture inlay (Wright
1981: 156, fig. 75j, pl. 19a) hints at such buildings elsewhere. The occurrence of a sealed spherical bulla made of clay probably from elsewhere on
the Deh Luran Plain (lames Blackman, personal communication 1980) indicates the receipt of messages by administrators in the little center (Wright
1981: 156, fig. 75d, pI. 16e). About a quarter of the animals consumed
around modest housing in this center were hunted gazelles and onagers.
Sheep and, to a lesser degree, goats provided the bulk ofthe meat in the Deh
Luran diet; cows were rare and pigs were absent. About half the sheep and
goats, probably the males, were butchered between the ages of one and
three, a prudent strategy for herders interested in the reproduction of their
flocks.
The preponderance of barley in the preserved Middle Uruk plant remains, a reversal of the fifth millennium pattern, may indicate increasing
salinization (Helbaek 1969) but may equally result from an increased focus
on herding and increased use of dung as fuel (Miller 1984). There also is
much evidence of spinning and weaving technology, suggesting that cloth
production, presumably from animal fiber, was a local industry. Other industries are the preparation ofbitumen and large coarse-grained chert blade
cores, apparently for export (Wright 1981 :266-70, 275-77). Imported
durable goods include copper, fine-grained cherts, and marine shell for ornaments (Wright 1981 :272-75). In sum, Middle Uruk Deh Luran had a
small, prosperous population involved in craft production and export of animal products, bitumen, and chipped stone. From the broader Uruk world,
they received modest supplies of exotic commodities.
During Late Uruk times, the organization of the Deh Luran Plain
changed, eve~ though the basic two-tier settlement hierarchy was maintained. There were about 400 people, most in the center at Farukhabad, the
rest scattered among two or three small hamlets. The distribution of settlements was similar to that during Middle Uruk times, but there seem to have
been fewer people-less than a tenth the number that would occupy the
plain at the beginning of the Early Dynastic period, three centuries later
(Neely and Wright 1994:208, table VI.16).
Our excavations in the center revealed an area with fragments of modest mud-brick houses (fig. 6.5a) with the usual blades, sickles, grinding
stones, bowls, and jars. A notable feature of this assemblage is that the local
chaff-faced wares were used only for some of the cooking jars; the serving
and storage vesselswere made with the sand-tempered, often wheel-thrown,
Uruk Wares typical of the larger centers of Mesopotamia, such as Nippur,
Uruk, and Susa. In the trash around these modest residences, the density of
the bones of hunted animals drops off and that of domestic animals increases
in comparison to that ofMiddle Uruk times (Redding 1981: table 69). Goats
appear to predominate over sheep. One-third of the kids and lambs were
killed before reaching the age of one year, a less-than-optimal butchering
pattern that may be a result of restricted access to summer pasturage (Redding, personal communication 1996). As during the Middle Uruk, barley
predominates in the seed samples. There is evidence of pottery firing and
spinning and weaving around these simple structures, and a lightly baked
sealing, probably from a bale of skins, was discarded near them (Wright
1981: 156, pI. 16a).
Our excavations also documented an area of nonresidential activities and
elaborate buildings with massive walls of miniature mud bricks (fig. 6.5b),
circular platforms (fig. 6.5c), and ceramic drains. There are high proportions
of bevel-rim bowls around all of these features, and a high proportion of
cups and spouted bottles around the platforms. A concentration of bitumen
and chert-processing debris was found around these buildings. Whether
these were residences or institutional buildings, the occupants apparently
were involved in mass labor and in the preparation of items for export. The
discard ofwaste products relative to the actual local use of these materials increases greatly, indicating increased production for export. In spite of a continued emphasis on herding, evidence of spinning and weaving decreases,
suggesting that animals on the hoof or unprocessed fibers, rather than cloth,
were exported (Wright 1981 :266-72). Notwithstanding the evidence of
_1
....L... _ _
hearth
51.05
54.OB
I,
1,
J,
--,<
!/
a.
F23
10
J~
---1_ -
~--
... -
b.
-"""=,,,,,,-
---'----1
54.40
1,
,I
__)1J
\,:2~
~~--
"y
1
;0 -
~i
"
J~
I,
KEY
I@J
stones
i,
I
I,
-l,
I
( ) hearth
( ) intrusive feature
54.42
cylinder seal
1,
J
,
F21
-;
c.
numerical tablet
_. sealed bullae
.~
10
~~
54.05
elevation in meters
Figure 6.5. Late Urukfeatures at Farukhabad (adaptedfrom Wright, ed. 1981): a, Excavation A, Layer 21; b, Excavation B, Layer30; c, Excavation B, Layer 29.
increased production for export, import of durable goods such as finegrained cherts, copper, carnelian, and other material for ornaments decreased (Wright 1981 :272-75).
In sum, Late Uruk Deh Luran had a small population involved in the
production and export of quantities ofunfinished animal products, bitumen,
and chipped stone, perhaps under the control of a central institution. From
the broader Uruk world, they received modest supplies of consumer goods
but few exotic commodities. It seems likely that an outside polity, whether
Susa or some other, took direct control ofDeh Luran and its strategic route
and exploited local sources of stone, bitumen, and perhaps wood, as well as
the pasturage of the area.
On the Plain ofIzeh or Malamir, 155 km east-southeast, or five to six
days' travel from Shush (see fig. 6.1), a very different situation developed.
The Izeh Plain is a basin in the midst of the central Zagros, only 750 m
above sea level, though surrounded by peaks reaching as high as 3,100 m. It
covers only about 135 km-, including two seasonal lakes on its floor. It receives ample rainfall throughout the winter and has excellent land for grazing and dry farming but only a few springs that provide water during the dry
summer. Oak forests cover the mountain slopes and may have spread onto
the plain in the past, but there are few mineral resources. We know of events
on this plain through detailed archaeological survey and through soundings
at the small Uruk center of Tepe Zabatjad (Wright 1979, 1986b).
The Izeh Plain seems to have been settled continuously from the early
fifth millennium into the Early Uruk period, when a small center and a large
village indicate that the maximum population of the plain was about 2,400
people (Wright 1979:59, fig. 26). Neither settlement was close to a permanent water source, suggesting they may not have been fully inhabited during the summer. The absence of smaller settlements, which are common in
several earlier and later periods, suggests that Early Uruk settlements were
not secure. We have few excavated samples from Early Uruk layers, but the
ceramics from the surfaces of sites are similar to the earlier Uruk ceramics
that developed on the Susiana Plain. This contrasts to those ofDeh Luran,
which are in a northern tradition.
Only the small center at Tepe Zabatjad survived into the earlier Middle
Uruk period, and it diminished in size, with at most 500 inhabitants (Wright
1979: 59). Small excavations in the center revealed the familiar technology
of sickles, grinding stones, bowls, and jars. The pottery is identical to ceramics from contemporary sites on the Susiana Plain (Wright 1979: 69-81).
Our small soundings revealed only a remnant of the stone footing of a small
wall. However, a mass ofpartially burned mud brick and a group ofsmashed
bevel-rim bowls in a pit indicate that mud-brick buildings had been built
and repaired. The lack of evidence of administrative technology such as sealings or counters may be a result ofthe intensive weathering of the upper layers of the site.
At least a quarterof the animals consumed around modest housing in the
center were hunted pigs and gazelles. Domestic sheep and goat bone were
the most common bones. Although we lack sufficient samples to construct
age curves and sex ratios and assess exploitation, all elements ofboth old and
young animals are present, suggesting local herding. Cow bone, however,
munities whose archaeological traces are poorly known. While the campsites and cemeteries ofsuch people Seem rare and show none ofthe evidence
of hierarchical organization notable after 3000 BC (Wright 1986b: 147- 49),
they do exist. Smaller interstitial mountain valleys, for example the Hulailan
Plain just southeast of Kuh-i Dasht (Mortensen 1976; Wright 1986b: 14647), have campsites probably of Middle or Late Uruk times. Similar campsites have been found on the arid lowland plains elsewhere in Mesopotamia
(Bernbeck 1993); it seems likely that future survey in the arid foothills will
reveal similar evidence in southwestern Iran. Although nomadic people may
not have posed a military threat to larger polities, they probably sought
goods from farmers and townspeople in return for animal products.
common at Abu Salabikh may have been brought down the Euphrates from
Syria, and the quantities of bitumen found at the site could have come either from southwestern Iran or the middle Euphrates.
With such mass movement of materials, however it was organized, it is
not surprising that some larger Late Uruk polity reached out to control
sources of chert, bitumen, goat hair products, and woods on or near the Deh
Luran Plain. The evidence does not indicate whether this polity was centered at Susa or at a center on the Euphrates (Wright 1981: 187-88). The
presence of a major Middle Uruk trading partner on the Euphrates and the
evidence of conflict within the Susiana Plain, discussed below, suggest that
the center of control was on the Euphrates. This economic takeover may
not have involved overt conflict; given Deh Luran's tiny settled population,
mere threat may have been enough to bring it under the control of stronger
neighbors.
It is clear that Middle Uruk Susiana had a direct interest in areas of the
Zagros to the east. As discussed above, while the little settlement cluster in
Deh Luran to the west maintained its own local ceramic style, those in Izeh
and other valleys to the east used ceramics identical to those ofSusa. There
is no evidence that relations between southwestern Iran and areas to the east
involved the mass movement of materials. It is possible that Susa was interested in low volumes of high-value goods such as copper. It is also possible
that its interests were not economic at all but simply involved securing its
frontier against expanding plateau communities. In any event, the abandonment of mountain valleys and the settling of Chogha Mish on the eastern
Susiana Plain at about the same time is difficult to understand unless external force was being exerted.
Given the pressures from both the east and the west, it is not surprising
that problems within the Susiana polity were exacerbated. Johnson (1973:
143-56, 1987, 1988-89) has argued cogently that Late Uruk settlement
patterns and iconography indicate conflict between the western plain under
Susa and the eastern plain under Chogha Mish. Chogha Mish was abandoned before the end of the Late Uruk period, and Susa itself was abandoned near its end. Their populations could have moved as refugees to other
parts of Southwest Asia, leaving the rich Susiana Plain with few inhabitants
(Alden 1987).
LESSONS LEARNED
The diverse, if far from comprehensive, evidence assembled here allows a
qualitative evaluation of the operation ofthe first states in southwestern Iran
and their impact on the lives of their inhabitants. It is clear that goods and
labor were extracted from rural producers, probably to support rural elites,
and both elite and ordinary town dwellers. Figures with complementary hierarchical roles in a system of control acted to control the movement of labor or the flow of goods, as well as to keep records of transactions. They also
acted to maintain their own institutional apparatus, building and maintaining installations, manufacturing appropriate paraphernalia, and no doubt using it in the impressive rituals portrayed in some of the cylinder seals. In bad
years, some ordinary producers did receive assistance from central institutions, but an ordered life of reciprocal relations may have benefited only
some of the participants in the state and for only limited periods of time. At
other times conflict broke out, polities were shattered, and people became
refugees. A more measured assessment ofthe relations within these states and
of how these relations developed over the several centuries of the Uruk period in Mesopotamia will be possible only after future cycles of fieldwork
have been completed.
1
Order, leailimo(y, nod Weollll in Anlient (gypt
and Mesopotomia
JOHN BAINIS no NORMAN YIlun
"Hattusha
Kanish
ANATOlIA
.Kadesh
MEDITERRANEAN
Byblos
SYRIA
.Oamaacus
SEA
PHOENICIA
t. Cs"",1
'Arunah
PALESTINE oM iddo
.,et
Tall G t J rusalem
Zawiy81 Umm
ar-Rakharn
PHllISTlA)Gaia
Aaphl
_ 'Ar:d
l-,"'->--~""-';haruhen
ISTHMUS OF
SU!Z
LIBYA
200 rn,
100
160
I
200
I
2bo
aOO km
ARABIA
300
'km
Figure 7.1. Mesopotamia (after Postgate 1992; Edzard, Farber, and Sollberger 1977).
Figure 7.2. Egypt and Syria-Palestine. From "Egypt" (author and map
consultant John Baines) in Encyclopaedia Britannica; reproduced with
permissionjrom Encyclopaedia Britannica, 15 th edition, 1988 by Encyclopaedia Britannica, Inc.
o
o;I
o
o
o
If)
o
o,....,
)
o
o
;-
o
o
00
00
N
I
N N
I
I
0'
\0
00
00
,....,
r--
OL.{)OOOl.f)~
If);-;-)))
\0
N
r-.
If)
o
,....,
o
If)
0'
N
00
00
0
00
00
00
o If)
\0 r-.
,....,
,....,
NN
I I
If)
If)
000000
000000
00
0
,...., 0
If)
If)
)
,....,
NN
0'
,....,
\0
,....,
If)
o
,....,
o
r-o
,....,
r-.
If)
0
00
0'
If)
If)
00
,....,
0
)
,....,
o
)
\0
,....,
If)
,....,
If)
,....,
r-.
I
o
r-o
,....,
;\0
I
If)
0'
)
Cl
-<:
dimensions and for what reasons institutions-and what may presumptuously be termed the spirits of both civilizations-differ, we learn more
about their structure and character.
We do not, then, compare the two civilizations to enumerate similar
traits or to establish the core principles of an abstraction, the "archaic state."
Rather, through this controlled comparison in time, place, and historical
contact, we seek to identify major axes of variation and to advance an important anthropological principle: by knowing what is institutionally and
structurally dissimilar in one society judiciously compared with another, we
can begin fresh investigations of the principles of organization and change
in either society, or in both. Our larger intention is to contribute to the
set of comparisons of archaic states or early civilizations in general, to see
what organizational principles are widely shared, what, if anything, is truly
unique, and what general societal and transactional models can address data
from a wide range of societies (cf Trigger 1993).
SOME PRINCIPAL DIFFERENCES BETWEEN
ANCIENT EGYPT AND MESOPOTAMIA
We review differences between Egypt and Mesopotamia in terms of the relation between political and cultural systems, kingship, and urbanization.
Egypt exhibits nearly total convergence of polity and culture. The establishment ofthe unified and centralized polity was characterized far more by territorial extent than by urbanization, although settlement sites, which might
round out the picture, are almost inaccessible. From ancient Egyptian civilization into modem times, there has been a clear definition of the extent of
the country. The principal change has been that, whereas in antiquity the
Nile Valley and Nile Delta made up the area ruled, lines have recently been
drawn on a map and used to justify a geographical claim to rule adjacent
deserts with their transit routes, as well as resources of a few oases, a very
small number ofnomads, and significant mineral deposits. In antiquity, these
regions (except for the nearer oases), although exploited, were treated as being "abroad." In some periods Egypt conquered large sectors of the Middle
Nile and of Palestine and Syria, but these were never held permanently. Attempts to integrate the southern domains into a larger conception of
"Egypt" did not succeed in the very long term.
This congruence of country and region also was cultural: Egyptian civilization extended to its southern border at the First Cataract of the Nile. At
times, Egypt had great influence farther south, but generally less in SyriaPalestine.
The most apparent of all differences between Mesopotamia and Egypt is
that there never was any enduring political unification in Mesopotamia
until the Persian conquest in the mid-first millennium BC; if there must be
a political definition of the state, then there was no "Mesopotamian" state.
In a wider comparative perspective, however, it is Egypt that is exceptional
in displaying convergence between a polity and a more abstract civilizational
boundary. Mesopotamia consisted politically of a congeries of city-states and
culturally of an overarching cultural tradition.
In later periods, after 1600 BC, there were trends toward the formation
ofAssyrian and Babylonian regions, which in the south were especially loose
fitting. Although Assyria did conquer Babylonia in the seventh century BC,
absorbing it into its "empire," that unification diverted military authority
and resources from other imperial ventures. Ultimately, Assyrian rule over
Babylonia was successfully resisted, and this led to, or was combined with,
other missions toward independence by former Assyrian subjects. These
struggles were followed in rapid order by military defeat and the demise of
Assyria (Dalley 1993; Postgate 1993; Yoffee 1988b).
Kingship and Other Forms of Rule
on earth. Official forms displaying his qualities related him to the gods, but
he was not the same order of being as they-more central and salient for
human society, but oflesser status and potential.
The two basic terms for "king," njswt and bjtj, related to hierarchically
ranked aspects of kingship and, in dynastic times, were connected with Upper and Lower Egypt (roughly equivalent to the Nile Valley and the Nile
Delta). This characteristically Egyptian dualism held that only entities
formed from dualities were meaningful (e.g., Hornung 1982:240); by implication, the unity of the country-typically known as the "Two Lands"
and long lacking an overall proper name-was vested in king and kingship.
Neither the country of Egypt nor full rulership could be imagined without
kingship, because the king was the sole formal intermediary with the gods.
Only around 750 BC, toward the end of several centuries of the Third Intermediate period, did significant numbers of regional leaders emerge who
did not claim the title of king.
The king's role in relation to and in combination with the gods perpetuated the fragile order of the cosmos, offering a central legitimation that
overrode the "moral economies" of smaller social organizations (Baines
1995b). This principal ritual requirement remained in force into Roman
times, when the emperor, who could have known little of what he was subscribing to, was represented in temples in forms that conveyed essentially the
same message as the key originating works of the late Predynastic period
(Derchain 1962).
The ritual and cosmological aspects of kingship are embodied in much
of the country's vast monumental legacy, but also in royal action and in foreign relations. Missions abroad were undertaken to bring back materials
needed for king, cult, and the dead. Conquest was an "extension of the
boundaries" that built upon the idea of maintaining the cosmos. The basis
of kingship was not, however, strongly military, and for much of the third
millennium the country seems to have lacked a standing army.
Within Egypt, royal authority was underpinned by the king's theoretically absolute ownership of the land and rights over his subjects. Even in
Greco-Roman times, streets running past private houses were termed "the
street of Pharaoh" (e.g., Smith 1972: 711). Kings appear to have asserted
these rights in early periods by redefining landholding patterns on principles
defined at the center and by constructing many new settlements, imprinting their requirements on the fabric of the land (Helck 1974:49-53;
janssen's reservations [1978: 226] seem excessive-such phenomena are
known elsewhere).
The king's most powerful influence was probably on the elite. Their status and wealth depended on him-often on his personal favor and caprice.
The palace was the central institution that mobilized the country's resources,
although in most periods there also were significant "secular" and temple
administrations. The term "pharaoh," regularly used for kings of Egypt by
foreigners at least since the first millennium BC and by Egyptians from
around the time ofAkhenaten (c. 1350 BC), derives from the ancient "Great
Estate (pr- 3)" that focused on the institutional and economic aspects of
kingship.
Mesopotamian kingship contrasts strongly with that of Egypt. Without
an overarching political state, its forms of kingship were markedly different.
Kingship acquired its character in the endemic struggle among the Sumerian
city-states in the time before Sargon of Akkade (c. 2350 BC; Cooper 1983).
It seems that kings were at first elite landowners, perhaps important figures
in community assemblies, who progressively assumed more power as war
leaders and who bought land from corporate landholding groups (Diakonoff
1969; Gelb 1979; ]acobsen 1957). In pre-Sargonic land-sale documents
(Gelb, Steinkeller, and Whiting 1991), the buyer of the land is often a ruler
or high official; the seller is denoted both in the body of the document and
by the list of his relatives who are recorded as the witnesses and who receive
gifts. While the texts do not indicate what happened to these newly landless
people, it is assumed that they did not actually move from their land, but acknowledged its new owner and paid him both taxes and obligations of service (Yoffee 1995b). These documents show the strong difference from
Egypt in how early Mesopotamian kings were able to gain power, labor, and
resources. The Mesopotamian king was a local lord whose acquisition of
power was internal and unrelated to conquest outside his own state.
Rulers of pre-Sargonic city-states were variously called en, ensi, or lugal.
Although these titles have different etymological meanings, and some have
tried to see a progression from priestly to secular kingship as reflected in
their evolution, they can all be translated as "ruler" for pre-Sargonic times.
With the conquest ofSargon, however, lugal (Akkadian sarrum) became the
accepted title for "king" and ensi was reserved for the governors of city-states
(and en; originally "lord," became a title of the priesthood). In the Old
Babylonian period ensi became further devalued, meaning "manager of an
agricultural field."
Early Mesopotamian city-states were arenas for a normative and constant struggle between the burgeoning royal authority and the power of the
temple estate. The so-called reforms ofUrukagina of Lagash (c. 2400 BC) in
southern Mesopotamia indicate that there the temple was able to stage a
coup d' etat against the kings who were seizing its land and privileges, but
that the coup was only a minor interruption in the trend toward increasingly
centralized power vested in the royal government (Nissen 1982).
In the succeeding Akkadian period, Sargon and his grandson Naram-Sin
reorganized administration, founded a new site as capital of a regional state,
C
and established new titles in order to imply that the House of Akkade was
not just another powerful dynasty: it was the legitimate political center ruling over all Mesopotamia (see Liverani, ed. 1993). Naram-Sin himself became deified, thus reinforcing his imperial status over the Mesopotamian
city-states (Glassner 1986). When the Akkadian dynasty fell, city-states
reemerged, as they did also after the short-lived regional state of the Third
Dynasty ofUr. The kings of these newly independent city-states once again
began the struggle with their neighbors, just as had their predecessors in the
days before Akkade.
In sum, while Mesopotamian kings were powerful leaders in war and in
civil administration, they never achieved the same position as the foci ofideology, economy, and social life as the kings of Egypt did. In some periods
the Mesopotamian king shared power with temple estates and local assemblies. Furthermore, the palace often contracted with, and sometimes depended upon, private entrepreneurs to supply its local subsistence needs,
as well as its desire for distant luxury goods (see section on "Economy";
Yoffee 1995b).
Urbanization
Mesopotamia and Egypt contrast strongly in the vital area of urbanization.
City-states were the major arenas for the interplay of characteristic
Mesopotamian institutions. This statement can be defended, even against
the charge that written sources and archaeological investigations are utterly
biased toward urban places (a bias that intensive field surveys seek to correct,
e.g., Adams 1981). It is not that villages, nomadic, seminomadic, or cryptonomadic pastoralists, and de-urbanized bandits were not integral to the
Mesopotamian scene. However, it is in the comparison with Egypt that one
can see the significance of city wards (Gelb 1968; Yoffee 1992), local assemblies, resistance to urban rulers, a temple-versus-palace struggle, an urban
prejudice against the countryside, and the superior ability of nonurban
people to use their extensive ties to seize political power (Kamp and Yoffee
1980). All of these features ofMesopotamian civilization are conspicuous by
their absence in Egypt, or are extremely attenuated in the context of the
central symbols of Egyptian civilization.
The urban implosion oflate-fourth- and early-third-millennium Mesopotamia resulted in a massive population shift into large sites (Nissen 1988).
These new city-states, consisting of one or more large sites, such as Lagash and Girsu of the city-state of Lagash, Uruk and Kullaba ofUruk, Kish
and Hursagkalama of Kish, and attendant towns and villages (for thirdmillennium Lagash, see Gregoire 1962), set the pattern for Mesopotamia as
"the heartland of cities" (Adams 1981). For as long as Mesopotamian civilization remained independent, with multiple polities, it retained not only
the configuration of city-states and countryside, but also the ideology of the
city-state (Postgate 1992). Rulers were mainly defined in connection with
the city-state from which they ruled; even those associated with extensive
conquests focused their domains on a core city (e.g., Hammurabi of BabyIon, 1792-1750). Major reorganizations of empire, however, from Sargon
of Akkade and notably including Tukulti-Ninurta of Assyria (fourteenth
century) and various Neo-Assyrian rulers in the first millennium BC, were
often accompanied by the establishment of new capitals. These new cities
served to dislocate and disenfranchise old elites and bureaucratic networks,
and they also were monumentally emblematic of changes in administrative
power and purpose. From the end of the Uruk period to the conquest of
Cyrus the Great of Persia (539 BC), city-states were an irreducibly essential
quality of Mesopotamian civilization. In the Sumerian King List, a historiographic text relating the birth ofMesopotamian political systems (Michalowski 1983), kinship descended from heaven to cities: without autonomous
cities, a Mesopotamian way of life was unthinkable.
For Egypt, central places were important on a number oflevels; the idea
of a walled, nucleated settlement goes back into prehistory. Certain crucial
towns, such as Buto in the Nile Delta, Hierakonpolis in the south, and Elephantine at the First Cataract, played key roles in defining the extent of
Egypt during the period of state formation. Nonetheless, only scholars who
appear to feel that urbanism is a sine qua non of civilization (e.g., Kemp
1977) are prone to maintain that the city was a primary motor of development or strongly characteristic of Egypt. In early times the Egyptians seem
to have been almost more interested in their frontiers than in their center
(e.g., Seidlmayer 1996); government policy toward regions and settlement
patterns appears to have disfavored cities in certain respects, notably by
using an estate-based system of redistribution. The elite's ideology had a
rural tinge-rather like that of the English country gentleman-despite the
pattern of land tenure, which was theoretically insecure because rights to
land were based upon holding administrative office.
These biases changed in periods of insecurity and decentralization, and
more profoundly in the New Kingdom (c. 1520-1070 BC) and later, when
the ideal of the city was well established along with the notion of city as cosmos (Kozloff, Bryan, and Berman 1992: 103-4; O'Connor 1998). From
early Islamic times to the present day, the country's common name, Misr,
has been the same as that of the capital city; this congruence also can be observed for the first millennium BC. The city was the country.
Moreover, when the Assyrian king Assurbanipal described Egypt around
660 BC, he did so in terms of cities and their rulers, most of whom he designated with the Akkadian word for "king" (sarrum/lugal). This was a period
when the Delta in particular had moved toward something like city-state
forms, but one wonders whether his approach owed more to his background
and that of his recording officials in a city-state civilization than to what was
observed on the ground.
glimpses of activity only rarely attested in Egypt, principally for the Ramessid
and Greco-Roman periods (c. 1300-1075 BC; third century BC-AD fourth
century).
Egyptian. The terms of order, the negotiation of order, and its appropriation by elites are defining activities of civilizations. Order cannot be taken
for granted.
The elite appropriation of order is one of many legitimations of inequality, which was perhaps most extreme in Egypt. It was far from natural
or necessarily easy for ancient elites to achieve legitimacy in the eyes of others and of themselves. Elites sustain their self-image and transmit it down the
generations both through their pragmatic actions in maintaining inequality
and through their understanding of their own position and mission. These
legitimizing activities and attitudes encompass the mission, one that the elite
take upon themselves, to achieve and maintain order in their societies. Legitimacy, however, is most strongly expressed in the "dialogue" between the
ruler and his superiors or peers, the gods. Ancient Near Eastern rulers or
elites did not have exclusive access to religious life, but they did have access
to more grandiose varieties of it and to more of its profound meanings,
while others were excluded from some of its domains. A major thrust of
religion, on which so much of society's resources were expended, was
legitimation.
In complex societies, wealth, especially conservable wealth, is a vital feature that sets elites apart from others. The division and administration of
society enhance enormously the potential ofwealth to be produced, differentiated, stored, and negotiated, while the organizational capacities of the new
social forms allow great distances to be exploited in order to move goods and
people so as to generate and mobilize wealth. All this is administered by the
elite or their employees; so far as our sources allow us to gauge, these activities seem principally to benefit the elite. Yet wealth is probably not the
prime motive force in the development and maintenance of complex social
forms; rather, it is an enabling factor, one that has an extraordinarily powerful communicative and persuasive potential. Wealth and legitimacy are
almost inextricably linked. Wealth, controlled and channeled, can sustain
order. Destitution of wealth spells disorder or a reversal of order.
Thus, the three interrelated aspects-order, legitimacy, and wealth-cover
much ground in the study and comparison of features distinctive of early
(and other) civilizations, of their emergence, persistence, and eventual collapse. Below we survey the evidence according to more traditional subject
divisions instead of these rather abstract ones, but in singling them out we
emphasize the active role of the elite in constituting and, especially, transmitting the characteristics of a civilization. Our longest case study, of high
culture, addresses most directly the nexus between these factors.
Finally, these three terms have the advantage ofbridging analysts' and actors' categories. Although such terms and topics as politics and economics
have no counterparts in the ancient evidence (which is not to say that they
are invalid as fields and methods of study), order is a central ancient idea,
wealth is much mentioned in the texts and displayed in the record, and the
theme of legitimacy has manifold and close correspondences in verbal and
iconographic sources.
ofsuch features as cylinder seals, monumental art and architecture, and writing are hardly prefigured by the 'Ubaid.
The origins of writing are a good case in point for the dazzling innovations that accompanied the rise of the Uruk (and other) city-state(s). Although Schmandt-Besserat (e.g., 1992) has assiduously shown that a system
of "tokens" preceded writing by millennia, many have criticized her argument that the tokens evolved directly into writing (Friberg 1994; Le Brun
and Vallat 1978; Lieberman 1980; Michalowski 1990, 1993b; Zimansky
1993). The shapes of most tokens bear no relation to later cuneiform signs;
the tokens are distributed over a much wider area than that in which writing later developed. In Elam, where some of the tokens and bullae enclosing them were found, the form of the writing and language were not the
same as in Mesopotamia. And at Tell 'Abada, one of the few archaeological
contexts from which we have tokens, the small clay objects were found in
children's graves, not a likely locus for trade and business records, which
Schmandt-Besserat has argued was their primary function.
Michalowski (1990, 1993b, 1994) has emphasized that the earliest preserved written signs are extremely complex and abstract, bearing little resemblance to the tokens. Tokens are part of a long process of signification
that includes glyptic arts, pottery decoration, and potters' marks, but they
cannot explain the nature or form of the writing system. Indeed, writing
seems to have originated through invention (see Boltz 1986 and 1994 for
similar views on the ancient Chinese script), perhaps the product of a single
individual's work (Powe1l1981 :419-24). In Warka, the reasons (or at least
the context) for the appearance of writing are reasonably clear. Upon the
formation of a city-state with a central core of around 300 ha and a suggested population ofmore than 20,000 (Nissen 1988)-Warka was only one
of a number of city-states in southern Mesopotamia-the ability to manage
a burgeoning economy was greatly facilitated by a new system of record
keeping and communication that could name names, specify obligations,
and count resources. While most early tablets consist of such economic accounts, a significant percentage (c. 15%) are lists of professions and other
matters that were aids for teaching the new scribal arts; these demonstrate
the institutionalization and cultural import of this new technology. Writing
is, however, only one of a series of rapid and dramatic innovations that occurred at the end of the Uruk period.
As noted, the nucleation of settlements at this time represents a demographic implosion in which the countryside was progressively depopulated
over about 500 years while large urban sites grew. This process of implosion
also led to a significant explosion, since, it is argued, the southern city-states
sent forth expeditions to establish colonies up the Euphrates into Syria and
Turkey, and also into Iran (see Yoffee 1995b). Although these colonies,
whose purpose was to serve as access and transshipment points, were easy to
found in a countryside of relatively low political centralization or organized
resistance, they were impossible to maintain in the medium term, and many
disappeared within 50 years or less.
Since no political unity was present in southern Mesopotamia and neighboring Khuzestan, Algaze (1993a: 115-18) suggests that each independent
city-state established its own colonies, as in the case of early Greek colonial
expansion (Schwartz 1988). We know little about other city-states in the
Late Uruk period. Furthermore, according to some (e.g., Stein et al. 1996),
"Uruk colonies" may date to the Middle Uruk period. Stylistic criteria have
been inferred to show that sealing motifs from these colonies resemble those
from Susa as well as, or rather than, those from Uruk (Pittman in Stein et al.
1996). Although it is hard at present to test the hypothesis that individual
city-states in middle- and late-fourth-millennium Mesopotamia and Khuzestan established distant colonies, it is clear that, aside from ephemeral conquests and alliances, no political unity existed in Mesopotamia before the
imperial successes of Sargon (c. 2350 BC), despite the region's self-image as
belonging to a single civilization. If third-millennium city-states, thus, are
the logical outcomes of rapid social evolutionary trends at the end of the
fourth millennium, their destiny was to compete unceasingly for the best
agricultural lands and for access to trade routes. Although political unification was a likely result of such endemic conflict among city-states, it was
equally improbable that the independent traditions of city-states could be
overcome and that they could be easily integrated into a regional polity.
Egypt
A quite different evolutionary story can be seen in Egypt. Around 4000 BC,
the material culture and social forms ofsedentary groups in different regions
of Egypt and the Middle Nile was of a fairly uniform Neolithic/Chalcolithic character (Wetterstrom 1993); few signs of social complexity are to be
found (Midant-Reynes 1992). Nonetheless, some material and ideological
elements of inequality typical of later periods can be seen in the southern
Nile Valley, in the Naqada I culture (e.g., Bard 1994:68-75). Notable
among them is the emphasis on elaborate burials and the realm of the dead.
While the apparent prominence of this sphere owes much to the siting of
cemeteries in the desert, where they could be excavated, the expenditure of
resources is striking. Moreover, the crucial site of Hierakonpolis contains a
small group ofNaqada I tombs that are distinctively larger than anything else
of that date, suggesting the prominence of a single leader and providing a
topographical marker that was significant for the later, Naqada II inhabitants
of the area (Hoffinan et al. 1982:38-60).
The Naqada II culture (from c. 3500 BC), which originated in Upper
III
il,i!
!'
Much else is beyond recovery; settlement sites are virtually inaccessible beneath the floodplain silt, whereas cemeteries were sited for preference on
the low desert. Cemeteries evince massive consumption of luxury goods,
involving the mounting of expeditions into the deserts for minerals; longdistance transport of goods, including many Palestinian imports and probably delivery of basic foodstuffs; and the presence of specialized artists and
craftsmen alongside the nascent scribal group. The inner elite was integrated
into a small group of administrative officeholders near the king. These
people, who were almost certainly literate, were bound to the king both by
office and perhaps by membership in a distinct group (the pet) that would
have consisted of notional or real kin and who were qualified for the highest office. Even if this group once existed as a distinct entity, it quickly became a retrospective fiction (Baines 1995a: 133).
Elite and other cemeteries became numerous in the Memphite/Cairo
region-which has been the focus of population ever since-and also are
scattered through the country. These cemeteries demonstrate that in Dynasties 0-2 resources were by no means so narrowly concentrated on the
king and inner elite as in the following period, and they probably attest to a
gradual erosion of elite privileges that existed before and during unification
and centralization. They set the stage for the foundation ofthe city ofMemphis at the start of the First Dynasty, when administration and the requisite
people were focused there, while royal burials and perhaps the ceremonial
center remained at Abydos. In this period, burial and the realm of the dead
consolidated their position as a principal mode of display and signification,
as well as a consumer of resources. Burial sites were basic to Egyptian society, and especially to the elite who could aspire to a privileged afterlife denied to those who had no proper burial.
The Egyptian state's characteristic territoriality is evident at the frontiers.
The First Cataract region was annexed during Naqada II and was henceforth
the southern boundary. In the north, Egypt asserted a brief hegemony in
southern Palestine, probably founding some small colonies there, but withdrew during a recession or consolidation in the mid-First Dynasty. In Lower
Nubia, the royal cemetery of an A-Group polity around Qustul that imitated
the style and iconography of Egypt was thoroughly vandalized (Williams
[1986] sees Nubia as the source rather than the recipient of these styles), and
the A-Group itself disappeared, leaving an archaeological vacuum-but
probably not a complete habitation blank-spanning more than 500 years.
It seems that Egypt wished either to incorporate and exploit a politically
weaker culture in the surrounding area or to set up a cordon sanitaire,
within which its unitary civilization long stood in isolation (for a Nubiancentered view, see W. Y. Adams 1977; O'Connor 1993). These features
contrast strongly with Mesopotamia's treatment of its neighboring cultures.
Although the Neolithic progression from the first sedentary agricultural villages in which sites and social institutions became progressively differentiated and stratified was protracted, political centralization emerged rapidly in
both Egypt and Mesopotamia. Apparently, the various aspects ofrivaling and
cooperating groups, both occupationally specialized and socially distinct; the
complex routes of circulation of goods, services, and information from both
local and long-distance ventures; and conflict with neighboring cities and
regions all built up a head of sociopolitical and ideological steam manifested
in the emergence of new leaders, new forms and symbols of centralized authority, and new demographic shifts. In detail these processes look very different in the two areas, and their phasing was complex. In Egypt, a cultural
and political development with the elaborate forms of state and kingship
emerged toward the end of the process; in Mesopotamia, forms of social and
political struggle were never quite resolved by kings, and regional states were
atypical and unstable.
POI.ITICS AND ECONOMY
The earliest political system in third-millennium Mesopotamia was the network of city-states that endemically vied for arable land and access to trade
routes. Before the unification of Sargon of Akkade (c. 2350 BC) some
ephemeral hegemonies were achieved by successful city-state rulers. Thus,
it is not in Sargon's conquest but in his administrative and ideological innovations that a territorial Mesopotamian state was created (see below in the
section on high culture as a vehicle for political and cultural change). The
early Mesopotamian city-states, however, were also later the normative
products of imperial breakdown and were the loci of political struggle, especially between the royal and temple estates (see earlier section on "kingship and other forms of rule"). In this section we delineate the contrast
between the politics of early Mesopotamian city-states and the Egyptian
tendency toward territorial centralization. Apart from temple and royal
estates, community assemblies are prominent integrative institutions in
Mesopotamia. Furthermore, the diversity of ethnic groups in Mesopotamia
profoundly affected the modalities ofsocial organizations and social and political struggle.
The centralized Egyptian state left much less evidence of how its politicallife and its regions were organized than can be gleaned for many periods in Mesopotamia. Internal conflict and disorder were largely suppressed
fro~ the record of all but decentralized periods, and signs of dissent that can
be identified relate most often to personal antagonisms among the elite and
to the disgrace kings inflicted on people or on their memories. But the
Mesopotamia has produced important but controversial evidence for the organization oflocal power, in terms of controlling both institutions and social groups. In one of the classic articles on early Mesopotamian history,
Jacobsen (1957) considered that secular kingship arose not from sacral auspices, but from community assemblies. His argument was that incessant warfare in the earlier third millennium, as documented especially from the
Lagash archives, required the election or appointment of a war leader. Other
sources used for his argument were epic compositions (especially the tale of
"Gilgamesh and Akka" [Katz 1993; for further notes on Enmebaragesi of
Kish, see Katz 1995; Shaffer 1983]) and myths (especially Eniima elit, the
"Epic of Creation"; Foster 1993), in which assemblies (or councils) are mentioned. Critics (e.g., Evans 1958) rejoined that these poetic works composed
in the second and first millennia were too late to refer specifically to thirdmillennium events, and furthermore, that it was naive to think that what
happened in heaven reflected what was happening on earth. In the "Epic of
Creation," moreover, it is at a banquet assembly of the gods that Marduk
performs the magic trick ofmaking a constellation vanish and then reappear,
thereby convincing the drunken deities to choose him as their war leader.
Whereas Jacobsen (1943) viewed Mesopotamian assemblies as a form of
"primitive democracy," others thought that they were residual institutions
of tribal, nomadic groups that were being progressively assimilated in urban
Mesopotamia.
jacobsen's argument received support from those studying land-sale documents (see earlier section on "kingship and other forms of rule"), in which
the sellers of property were thought, especially by Diakonoff (1969), to be
the "elders" of inferred third-millennium assemblies (see also Westenholz
1984, who discusses ab-ba urn, "elders of the city"). In Late Uruk and Early
Dynastic lexical texts, the term "leader of the assembly" appears, and the
sign for "assembly" occurs in economic texts from Uruk, Jamdat Nasr, and
Ur (Englund, Nissen, and Damerow 1993; Green and Nissen 1987. According to Englund, however, the sign conventionally assumed to be "assembly" is no more than the representation of a pod).
Although references to assemblies are otherwise rare in third-millennium
texts (Wilcke 1973), which mainly come from temple and palace archives,
a term in the Ebla vocabulary texts has been interpreted as referring to an
assembly (Durand 1989). In the early second millennium, from which there
are many private documents, references in the texts to assemblies, elders,
mayors, and judges are legion. Local authorities decide cases of family law
and other matters not requiring royal intervention; headmen notarize the
hiring of community laborers on palace estates; and it has been suggested
(Yoffee 1988b) that the babtum, interpreted as a "city ward" by the Chicago
Assyrian Dictionary (University of Chicago 1992), might be a patrilineage. In
a brief essay on the ledrum, Kraus (1982), like others before him (Walther
1917), noted that the term was not simply a collectivity of merchants, as it
was in the Old Assyrian texts, but functioned as a judicial assembly. In the
Old Assyrian texts, it is clear that there were "assemblies/councils of big
men and small men" and a "city hall" in the city-state ofAssur (Larsen 1976),
and that the council shared power with the king, at least until the time of
Shamshi-Adad's seizure of the kingship in the eighteenth century BC.
In sum, we may infer that political integration in Mesopotamia was not
solely encompassed by the formation of centralized governmental institutions. Indeed, the evolution of the state government in Mesopotamia, which
came to hold ultimate jurisdiction in matters of dispute among existing corporate groups, did not mean that the political functions of these groups
ceased to exist. Such local organs of power typically represented both opportunities for the centralized state to channel local resources to its own advantage as well as arenas of resistance to the goals of the state-and thus
were an essential locus of political struggle.
In Mesopotamia, the role of ethnic groups and their ability to mobilize
personnel across the boundaries of city-states was one of the most important
factors promoting political change (Kamp and Yoffee 1980). Thus, after Sargon's coup in Kish and his foundation of the new city-state of Akkade as his
capital of a united Mesopotamia, the Akkadian language was employed in
place ofSumerian as the primary language of administration. This linguistic
switch, formerly interpreted as evidence of a new group of people-Akkadian speakers-entering Mesopotamia, is now seen as a mechanism to privilege scribes who could write in Akkadian and who were trained in the new
royal court. Akkadian had been spoken in Mesopotamia for hundreds of
years before Sargon's conquests, as is seen, for example, in the Akkadian
names of scribes who copied Sumerian texts (Biggs 1967). In the Ur III
I
222 I JOHN BAINES AND NORMAN YOFFEE
period, when the imperial bureaucracy was swollen to unprecedented numbers, scribes conversely were trained in Sumerian and owed their positions
to the new royal system in Ur (Michalowski 1987).
In the second millennium, with the political ascendancy of various
Amorite groups in Mesopotamian cities, the inference is again no longer that
there was an Amorite invasion: "Akkadeians" were in "Sumer" well before
Sargon, and Amorites were in Mesopotamia before 2000 BC, interacting in
a multilingual, multiethnic population. Amorites made alliances with-and
against-other Amorites, mobilizing kinsmen across the countryside in the
struggle for power within various city-states (Whiting 1987). Down to the
middle of the second millennium, civil order in Mesopotamia was fragile,
and it was negotiated both within city-states and among them. During periods of extreme political decentralization, "solidarity" within ethnic groups
could be the decisive force in the struggle for regional power. In this instance, language change did not accompany dynastic political change: no
document was ever written in Amorite, and Amorite rulers were careful to
present themselves as reproducing venerable Mesopotamian cultural traditions in order to legitimize their rule.
In Egypt, the center air"ued to control local offices in a way that was
hardly attempted in Mesopotamia. During most periods down to the GrecoRoman, the goal of government administration in Egypt was generally to
take as much power and as many resources as desired for the personnel and
projects of the center. But the long, thin form of the country, with
significant concentrations of specific resources scattered over its length, required much transport of goods, as well as organizational structures that
could handle both local and central affairs (Fischer 1977; Helck 1974, 1977).
How far the concerns ofthe provinces or ofprovincials mattered to tne center no doubt varied, but the lack of a developed urban ideology at the center in earlier periods may have militated against extremes of neglect of the
countryside and its inhabitants. The frequent presence ofa southern national
center at Abydos or Thebes, in addition to the political and economic center in the Memphite area, may have had a similar effect. Suggestive of the
opposite possibility (neglect of the provinces) is that in Egypt today-a
country whose basic orientation is toward the north (in antiquity it was toward the south)-few venture away from the dominant direction and travel
south of Cairo.
A local administrative structure of nomes (provinces), of which there
were nearly 40, was set up in the first few dynasties. This system appears to
have respected traditional settlement patterns and loyalties to some extent,
but it was a centralizing creation. Most nomes were similar to one another
in size, arable extent, or population, and in many periods they were not
favored as major administrative or political units, except for the densely
populated Greco-Roman period. Some nomes therefore acquired significant roles while others are like ciphers. In several periods nome organization seems to have been passed over in favor of larger, more centralized
groupmgs.
Two tendencies could disturb the efficacy of a centralizing administration: regional dissolution of the country, as in the First Intermediate period,
and an overproliferation of central bureaucracy that stifled activity, as in the
late Middle Kingdom. These two factors came together at times, but also
could be distinct.
The state's essential strategy in creating the economic basis for administration was to found estates that were attached to central institutions or
offices but geographically scattered (Jacquet-Gordon 1962). This pattern
had the advantage, perceived by medieval centralizing rulers, of avoiding a
concentration of landholdings controlled by a single beneficiary in a single
place. It also probably brought economic and ecological benefits by exposing only small holdings to local risks of failure and diversifying forms of exploitation across regions. Important offices in the central administration
(Strudwick 1985) were concerned with gathering and redistributing harvests and controlling animal wealth (less significant economically than in
Mesopotamia). High officials held large numbers of titles, all of them providing sources of income or, as "ranking titles," marking positions in the
elite hierarchy that were no doubt at least as important in the eyes of some
holders as were substantive offices.
In addition to functions concerned with products ofthe land or ofworkshops and specialized production, important central officials had purely administrative duties, for example, running royal bureaus. There was much
ceremonial and some seeming caprice, as during the mid-Fifth Dynasty,
when a whole hierarchy of titles of "palace manicurists" briefly became
prominent (Moussa and Altenmiiller 1977 :25 - 30). These functions or ceremonies centered on the person of the king and fostered contacts and networks that he may have exploited to shortcut elaborate administrative
structures. In the later Old Kingdom there were frequent rearrangements of
the hierarchy ofrank (Baer 1960); the introduction ofthese no doubt helped
the king maintain his dominance.
During the earlier Old Kingdom (c. 2600-2350 BC), this central, redistributive administration, one of whose principal concerns was to organize
enormous building projects, was supported by a provincial administration
(Martin-Pardey 1976), but the principal officials seem to have resided, when
possible, near the capital-or at least built their tombs there. Little survives
from the nomes themselves. This changed around the end of the Fifth Dynasty, when some officials began to be buried near nome capitals; toward the
end of the Old Kingdom they increasingly displayed local loyalties. Unified
rule collapsed around 2150 BC. Although the semblance of a single kingship
was maintained, regional centers competed until rival dynasties (the
Ninth / Tenth and Eleventh) were set up at Memphis (deriving from Herakleopolis) and Thebes; this development is known as the First Intermediate
period.
The consequences of late Old Kingdom decentralization seem clear in
retrospect, but the processes leading to it are disputed. Earlier writers argued
that there was a positive weakening and loss of control, whereas some recent
scholars have seen the innovations as a deliberate central response to changing conditions (e.g., Kanawati 1980). These approaches have contrasting
weaknesses, the older one in working largely from hunch and the newer one
in keeping close to the inscriptions and probably taking too much at face
value their assertion that all was well. It is uncertain whether the nascent regionalism of the late Old Kingdom derived from a politically motivated
identification by members of the central elite with local areas, or whether
the leaders genuinely originated from the areas they came to champion and
use as power bases. Two such strategies could have coexisted.
Although the dissolution of the Old Kingdom must be ascribed to some
extent to regionalism, that of the Middle Kingdom (c. 1980 -1640 BC)
seems to relate more to bureaucratic proliferation and stasis at the center.
The Twelfth Dynasty kings gradually suppressed nome organization and
nomarchs in favor of a division of the country into four large units. Midranking administrative offices began to multiply vastly, and a few leading
officials acquired great power. Under the Thirteenth Dynasty, about 60 kings
ruled for an average of around two years each, while officials held office for
much longer. Prosperity was maintained initially, but there followed a political decline that led finally to the division of the country between ethnic
Asiatics (the Hyksos) in the north and a local dynasty (the Seventeenth)
based in Thebes.
These intermediate periods exemplify local regionalism and the breakdown of the country into two units (cf Franke 1990). Later periods show
other patterns of struggle for power more clearly. In the New Kingdom
(c. 1520-1070 BC), the priesthood and the military emerged as distinct
forces. The military acquired their position through imperial conquest, and
subsequently through a response to invasion and immigration that ultimately brought ethnic Libyan groups to political prominence (Baines 1996;
Leahy 1985). The priesthood derived their influence from enormous royal
dedications to the temples of the fruits of conquest in the form of sacred
buildings, goods, and land for endowment.
These more recent foci of power acquired a dominant role in the first
millennium BC, when political life was more fragmented than that of earlier
times. By the late eighth century the country was divided into numerous
domains, only some of them ruled by kings. Yet Late Period (664-332 BC)
and Greco-Roman (332 BC-AD 395) rulers were able to revive centralization and the nome structures, the latter perhaps through study of old records
rather than through experience on the ground. Here, the maintenance of
high-cultural and "scientific" traditions may have aided pragmatic government. The history of Greco-Roman Egypt, although known principally
from the rural provinces of the Nile Valley and Fayyum, illustrates most
strongly the ability of the center to dominate the country in its own interest (Bagnall1993; Bowman 1996); in Roman times that center was Rome
rather than Alexandria.
Economy
In the earlier Mesopotamian states (until c. 1600 BC, roughly the end of the
Old Assyrian and Old Babylonian periods), the major economic units were
palace estates and temple estates. For all periods, however, there is evidence
of "community" and/or "private" organizations and families that owned
land, the chief form of enduring wealth; mercantile associations and entrepreneurial traders contracted with temples and palaces to supply distant
goods and to manage facets of their economies. The economic history of
Mesopotamia must be written in terms of the dynamic forces of struggle
among these economic sectors, and of degrees of intersection and cooperation among them.
Both the palace estates and the temple estates were, in essence, households. They consisted oflarge tracts ofland, numbers oflaborers and managers of labor, residential and ceremonial structures, and facilities for the
storage and manufacture of goods. Older interpretations held that the early
third-millennium temple estate was the primary focus of economic, social,
and political activity-the so-called Tempelstadt theory (Falkenstein 1974
(originally published in 1951, refuted by Diakonoff1969; Gelb 1969)-and
gave way to totalitarian control of the economy by the state under the Third
Dynasty ofUr; recent studies find the nature of economic activity to be far
more complex.
In the early third millennium, for example, the physical structures of
palaces and temples were separate, as were the units of land and personnel
managed by them; there also were endemic antagonisms over the wealth of
these estates. The trend through the third millennium was the familiar encroachment by the royal sector on sacral property (an opposite tendency to
that observed in New Kingdom and later Egypt). This struggle, however,
was a subtle one: kings required ideological support from the clergy and were
important players in religious ceremonies (although the evidence for this
comes from much later documents, such as the New Year ceremony texts
preserved from the later first millennium BC [Black 1981; Thureau-Dangin
226
real-estate entrepreneurs of the Old Babylonian period. Freed from the authority of husbands and fathers, they presumably led richer and more interesting lives than other women in Mesopotamian antiquity.
The increasingly wealthy "private sector" was further drawn into the
economic activities of the royal and temple estates. Although no full picture
of these interactions has yet been drawn, a number of detailed studies have
appeared (Charpin 1980, 1986; Diakonoff 1985, 1990; Renger 1989; Stol
1982 [reinterpreting Koschaker 1942]; Stone 1987; van de Mieroop 1992;
Yoffee 1977, 1982). While there were strong differences between north and
south Babylonia (which cannot be enumerated here), an important similarity is that the great estates employed large numbers of outside people in addition to their own staff of dependents. These private contractors, members
of the" community" rather than ofthe temple estate or the royal estate, supplied the estates with food (from fish to meat and wool products) and notarized the hire of laborers on the estates' fields. In times of political
centralization during the reigns of Hammurabi and Rim-Sin in the eighteenth century BC, the state naturally tried to control this independent sector, but in the time of weakness towards the end of the Old Babylonian
period, the power of these private contractors grew enormously.
In the Old Babylonian period, the economic resources of the great estates were not small. On the basis of van de Mieroop's study (1992) of the
texts from Ur, dating to c. 1984 -1864 BC, some figures can be cited to
illustrate this point. Tablets from the warehouses attached to the Ningal
temple complex record (mostly annual) deliveries (from various years) of
140,000 liters of grain, 50 tons of dates, a group of 31 shepherds managing
about 20,000 sheep, 16,803 cattle inspected, 18,710 liters of ghee, 16,200
liters of cheese, and 1,498 kg ofwool. The temple storehouse also purchased
9,600 liters of bitumen from a private businessman for 1 kg of silver. In one
text a group of merchants delivers 4,123 kg of copper to the palace. During the same early Old Babylonian period, the royal estate controlled some
23 km 2 ofland.
Such enormous quantities of goods were produced not only in the "redistributive" sectors of the temple and palace estates. In Ur, private entrepreneurs organized the fishing industry, engaged in long-distance trade,
supplied bread to the palace, and functioned as money lenders. One individual loaned 1.03 kg of silver to a colleague at 20 percent interest that was
due in one month! An individual sent 14,700 liters of bread or barley to the
palace. Another businessman rented a boat with a capacity of9,000 liters for
a business trip.
It is not necessary here to repeat the importance of the private sector to
the Assyrian economy in the Old Assyrian period. As numerous studies have
shown (e.g., Larsen 1976,1977,1982, 1987a; Veenhof1972, 1980), private
merchants transported tons of tin and textiles to central Anatolia and made
huge profits on the silver and gold markets there. The Assyrians did not control access to any resources, but they were expert in moving goods from
where they were plentiful to where they were scarce, transacting business,
formingjoint banking partnerships to accumulate capital, and taking advantage of the lack ofpolitical centralization in the areas they exploited. Finally,
it is worth repeating Larsen's (1976) judgment that the merchants were important players in the Old Assyrian state, members of councils in Assur, and
provided reasons for state military intervention in foreign lands.
We have unavoidably drawn a superficial and too coherent picture of
economic behavior in earlier Mesopotamia. Research on the relations
among the various sectors of the economy, especially the private economy;
how these relations changed through time; and how economic activities
were restrained and/or facilitated by political processes has changed our understanding not only of the production and distribution of goods and services in Mesopotamia, but of the structure of Mesopotamian society itself
The Egyptian economy is neither as well documented nor as well understood as that of Mesopotamia. The best known periods are the late New
Kingdom (c. 1300-1100 BC) and the Greco-Roman period, but the monetization and "colonial" character of the latter differs from the situation of
earlier times. The general picture is an extreme one of a centralized, command-driven economy (e.g., Janssen 1975b), but one that, contrary to today's wisdom, worked acceptably for long periods (see Kemp 1989 for a
contrasting interpretation). Much interpretation has been in the shadow of
Karl Polanyi, but there is no consensus as to how viable his approach is. Both
the overall context and the detail ofits operation are poorly known (see, e.g.,
Helck 1975). In particular, the proportion of economic life that is covered
by the sources cannot be well estimated-as is true also for today's command economies-and this unknown leaves the picture of subsistence
strategies and private enterprise uncertain. Because of these difficulties, discussion tends to focus as much on issues of social organization and administration as on economics more narrowly defined.
There was no "money," although various units of account and exchange
were used. The highly administered sector of the economy may have
touched the lives of most people relatively little, except to the extent that
they had to pay rents or taxes. The fact that most organization was in terms
of goods and the appropriation of labor, rather than of credit and such abstractions, may have restricted what the center and, in particular, what entrepreneurs could do. (This is not to say that the Egyptians could not work
with abstractions: legal documents often record regularizing fictions [e.g.,
Eyre 1992; Lacau 1949], while grain was lent at interest in a local context
[e.g., Baer 1962:45].)
The state's basic economic interest was in ensuring that the land was cultivated and in exacting taxation or rents from the produce. The state was
then responsible for storage and redistribution, notably of grain, in particular to those who did not produce for themselves. The state and temples made
many craft goods in their own workshops. Specialized workers were paid essentially in emmer wheat for bread and barley for beer, the two staples ofthe
Egyptian diet. Much production was channeled through state institutions
(e.g., Posener-Krieger 1976). The elite appear to have received their remuneration primarily in the form ofland, from which they could derive an income, and of other productive elements such as herds. The Old Kingdom
elite presented itself in tomb decoration as enjoying vast estates that produced most of the necessities of life and many luxuries (e.g., Harpur 1987).
This picture is idealized, but it is one pointer to how the monolithic character of the command economy might be tempered by a more complex
reality.
Apart from securing what was needed for the daily life of the center and
of specialists, major building projects, with their attendant requirements for
expeditions into the desert to extract raw materials (including gold), were
an important part of economic life and often of international relations.
There is a clear correlation between monuments and centralization; hardly
any major monuments were constructed in decentralized periods, but when
the country was centralized the amount of construction varied in both the
short and the long term. This pattern is anything but economically "rational" and clearly obeyed other dictates (e.g., Morenz 1969). Two periods
when the resources invested in construction were at their greatest were the
Fourth Dynasty, with the building ofthe largest pyramids, and the late Eighteenth and early Nineteenth Dynasties (c. 1400 -1225 BC), with vast temple
and tomb building by Amenhotep Ill, Akhenaten, Sety I, and Ramesses II
(as well as major private monuments). Even during these periods, there were
significant interludes without major construction. (Theories that the great
pyramids, and the ziggurats of Mesopotamia, were constructed as some sort
of unifying project for the country-e.g., Engelbach 1943; Mendelssohn
1974 -founder on this difficulty.)
Land was held on a use-value rather than an absolute basis of tenure, although parcels might remain in the same nonroyal hands for centuries. Generally, the cultivator was not the owner / tenant; most land belonged to large
institutions, including royalty, high officials, or perhaps wealthy individuals.
Cultivators were not free to leave their land. Ifland, whoever controlled it,
fell out of cultivation, the state assigned it to a new responsible tenant and
collected revenues from that institution or person (e.g., Gardiner 1951).
Those who fled and left their land uncultivated seem to have become vagrants who were then organized for labor by a state works department (e.g.,
Quirke 1988) and put to essentially the same tasks as those they had abandoned. In Greco-Roman times, it was temples that performed this resettlement function (Posener 1975). The reason for this regime-which is perhaps
characteristic of command economies while having obvious analogies with
feudal patterns-was probably that people were in shorter supply than land
(see Baer 1962); such evidence as can be gathered suggests generally low levels ofpopulation density and life expectancy (Bagnall and Frier 1994; Baines
and Eyre 1983 :65-74).
Salient questions raised by this rather bleak picture include how major
institutions meshed their economic activities together, how far the command economy could provide the requisite range of goods, and the extent
to which there was an independent "private sector"; the latter two are
closely related.
Relations between institutions have been discussed primarily for the
New Kingdom (e.g., Janssen 1975b), from which numerous economic documents are preserved (Gardiner 1941-52; Gasse 1988). These sources suggest
that the principal crown and temple institutions were not economically distinct, and that temples, in particular, could provide storage and supplies for
state concerns and interests. The state also could use temples as administrators or as intermediaries in the transmission and import of goods. Nonetheless, the basis of temple power, which was in landholdings, allowed the high
priest of Amun in Thebes to become politically autonomous at the end of
the New Kingdom (c. 1070 BC; Jansen-Winkeln 1992; Kitchen 1986:24854). The region in which the temple of Amun was the principal landowner,
which stretched from the First Cataract to about 150 km south ofMemphis,
with its northern border fortress at el-Hiba, became effectively independent
during the Third Intermediate period.
Representations of marketplaces, where small numbers of perishable
goods were sold, are found in Old and New Kingdom reliefs and paintings
(e.g., Altenmiiller 1980; Hodjash and Berlev 1980). A late New Kingdom
administrative papyrus records the voyage of a ship belonging to a temple
along the river. The ship dispenses clothing and honey, probably from the
temple's estates and workshops, to women on the river bank; in return the
women give these and other goods, the latter presumably ones the temple
did not produce itself (Janssen 1980). This is one of the few clear cases of
an interaction of "state" institutions and the private economy (on transport,
see Castle 1992).
More detailed material, which shows the privileged artisans who built
the New Kingdom royal tombs trading among themselves and selling their
services, derives from papyri and ostraca (inscribed flakes of limestone and
sherds) from their desert settlement of Deir el-Medina (Janssen 1975a).
Among the most revealing aspects of their lives is that some of the artisans,
who were amply salaried state employees, owned land in addition and
much for such local grandees as nomarchs taking over traditional royal functions of administration, military action, storage, and largesse. One also can
contrast dynastic times with the progressively monetized Greco-Roman
economy, from which there are attested such features as banking, forward
sale of standing crops (Pierce 1972: 81-93), and elaborate internal accounting on great estates (Rathbone 1991). Nonetheless, the essential thrust ofthe
Ptolemaic economy, like its predecessors, was state control, which extended
through taxation, ownership, or regulation to the most minor activities
(e.g., Bowman 1996:56-121).
Insufficiencies of Political and Economic Analysis
Perhaps most clearly for Egypt, we have encountered areas where the data
resist analysis in primarily political and economic terms. In a socioeconomic
perspective, the essential difficulty is created by the scale of inequality in the
ancient social systems and by the vast expenditure ofhuman and material resources on such projects as pyramids and major temples, or simply the burial equipment of a minor king like Tutankhamun. In Mesopotamia, where
the economy was more diversified and contested than in Egypt, the relation
between politics and economy was itself very complex. Political goals were
shaped through the forces of production, consumption, and distribution;
rulers alone did not dominate these economic spheres. Also, while the economic activities of members oflocal groups were affected by policy goals of
the state, one still needs to ask to what extent these political goals were motivated by economic factors.
Inequality, such as existed in both civilizations, created a large surplus for
a small elite-the ruling group of high officials in Old Kingdom Egypt
numbered perhaps 500 people (Baines and Eyre 1983: 66); this required
legitimation to the people from whom the surpluses were exacted, or so
modem analysts tend to suppose. Although state formation created great
economic potential, its consequences may have left those below the elite, after the exactions required of them, in an economic condition similar to that
oftheir prestate forebears. Throughout the history of the early state, the majority ofpeople hardly had alternatives or points of comparison beyond their
own societal environment; this limited perspective would have reduced the
requirements of legitimation, in comparison with those the outsider may
feel to be necessary. Data on "lower-class" residential areas, nonelite burials,
and the material inventory of people who are absent or depersonalized in
writing (all of this more accessible for Mesopotamia than for Egypt) provide
some evidence for the status of such people and of their social groups and
the way in which they were integrated into society as a whole.
While traditional forms of local social organization and their "moral
economies" may have retained some validity for the nonelite, precisely be-
cause the state (and/or large manorial estates) removed from them the means
of storage and provision against misfortune, the state appropriated the salient
discourse on the constitution of social order. Although we should not assume that those outside the elite always accepted the rhetoric of their superiors, state legitimations were generally designed so that elites could exploit
rather freely the resources available to them. Elites were able to be profoundly separate from the rest of their societies. This separateness extended
to the system of values, which was hardly accessible to those outside an inner social layer.
Despite the residual survival of the moral economy on which their inferiors relied for a legitimation of their dependence, elites had little regard for
the human lives of those whose efforts they were eager to utilize for their
own grand plans, taking huge disparities of circumstances between groups
for granted. Only rarely and mainly in the later stages of these two ancient states did the moral economy appear to a significant extent in the texts.
Since much of society was involved in the execution of the grand plans, additional values and interests must have held societies together both in these
goals and more generally. But the analysis ofthose plans needs to focus principally on the elite groups and on the ways in which they created and sustained among themselves the mechanisms for supporting and ensuring the
success of specific types of goals. These elite values were not only political;
economics were a means more than an end. Political and economic analysis
only partly addresses elite motivations. In the next section we outline a different approach to these issues, which are common to the study of many
civilizations.
HIGH CULTURE
Context and Definition
The inner elite controlling ancient Near Eastern (and presumably all other)
states and civilizations were few; during early postformative periods their
numbers became further reduced. In Egypt, this process culminated in the
Fourth Dynasty (c. 2500 BC), when a high proportion of the country's resources was devoted to the king's funerary monument. The number of the
surrounding, less grandiose tombs of the inner elite suggests that this group
consisted of no more than a few dozen male officeholders, in a population
of perhaps 1-2 million for the entire country. These men formed the central decision-making group, who together with their families controlled and
enjoyed the fruits of the country's labors. The group was larger in other periods, but it can never have numbered far into the thousands.
The more populous and numerous city-states in Mesopotamia did not
gather resources to a single center in the same way, but in Mesopotamian
civilization, too, wealth and status were highly concentrated. The urban
implosion that began in the late fourth and reached its apogee in the midthird millennium accentuated the formation of an urban inner elite that is
reflected in extravagant practices, such as the mid-third-millennium royal
tombs of Ur and other cities. The lexical "list of professions" shows that
many bureaucratic titles already existed in the late fourth millennium, as
well as names of occupational specialists and community officials (Civil et al.
1969; Englund, Nissen, and Damerow 1993; Green and Nissen 1987). Few
of these, however, can be described as privileged elites. In all periods, these
Mesopotamian elites, which included high temple officials, private landowners, community elders, and wealthy traders as well as high military and
administrative officials, however numerous they may have been, formed a
minuscule percentage of the population, as can be seen by comparing lists of
officeholders in later third-millennium texts with the vastly greater numbers
of people who received rations during that period.
The formation and maintenance of elites, and then of elites within elites,
lie at the heart of civilizations: inequality is fundamental. For these two ancient civilizations, the option of equality or of a serious search for an integrating "moral economy" hardly existed (contrast with Classical Greece;
Morris 1997). Cosmological elaborations and "political economies" are
among the features that can distinguish civilizations from noncivilizations.
The formation and entrenchment of such inequalities set the evolutionary
trajectory toward civilization apart from trajectories that led to less differentiated and stratified societies.
In the most ancient civilizations, elites controlled material and symbolic
resources but were scarcely subject to cultural requirements to disburse them
in fulfillment of social obligations. The distinctive achievement of archaic
civilizations is as much to transform the meaning ofwealth as to create more
wealth. Elites control symbolic resources in such a way as to make them
meaningful only when it is they who exploit them. This appropriation of
meaning is complementary to, and at least as important as, other legitimations available to controlling individuals or groups. There is also the "religious" affirmation that cosmic order is maintained only by the activities of
leaders, typically of the king and the central priestly officiants or, if religious
imperatives are acted out in the wider world, of the military.
These elite activities are characterized by the massive appropriation of
material resources, which are put to use in the enduring forms characteristic of ancient states. Such resources are due to the ruler and elite because
they are the carriers of exclusive and expensive cultural meanings that require such exactions for their maintenance and development. Elites, as the
principal human protagonists and prime communicants to the deities who
are the supreme members of the total society, require the highest products
of culture (cf Chang 1983). High culture, therefore, is one of the essential
loci, even the essential locus, in which order exploits wealth for legitimacy.
Here, high culture becomes self-motivating and self-sustaining, while its
meaning-bearing acquires a measure of autonomy through the expertise and
internal discourse of the specialists who maintain it.
These points are not new. As is widely accepted, if not in precisely these
terms, high culture is a central phenomenon of most civilizations from the
ancient Near East until today. Large-scale democracies and social movements, among others, redefine high culture in terms different than those of
the ancient civilizations we are considering; few dispense with it. Cultural
pluralism, however, turns the question of what constitutes high culture into
an issue that appears to have been largely absent in our cases. This point
will become salient for Mesopotamia, where the existence of many ethnolinguistic groups tended to promote rather than fragment high culture.
Despite the significance and centrality of high culture, it often does not
receive its share of attention as a factor in the creation and maintenance of
elites and civilizations. It is ironic that, while archaeologists acknowledge the
importance of high culture when they recover elements of it in the physical
record, they tend, for understandable reasons, to place more value on evidence for less exclusive social phenomena. The general public, with its interest in "treasures," may here be closer to the ethos of ancient elites than
are socially aware archaeologists.
We take high culture to be characteristic of civilizations rather than simply of states, and we see the boundary between one form of high culture
subscribed to by local elites and another as the boundary between one civilization and another. We define high culture as theproduction and consumption
if aesthetic items under the control, and for the beneju, if the inner elite if a civilization, including the ruler and the gods. The phrase "aesthetic items," rather
than "works of art," is intended to encompass a wide range of domains including visual art, visual, verbal, and musical performance, garments, perfumes, and the most highly prized food and drink. The phenomenon also
extends to such extravagant forms of "traditional" practice as big-game
hunting, the keeping of exotic animals, and the breeding of higWy specialized ones. At the extreme, the whole lives of the ruler and elitesare aestheticized, as well as strongly ritualized-the two aspects being mutually
supportive. The range of high culture is such that it can accommodate a diversity of interests and aptitudes among rulers and inner elite.
The aesthetic character of high culture does not imply that works of art
are ends in themselves. Many scholars object to the term "art" as applied to
non-Western cultures, and we do not wish to engage this issue here (see
Baines 1994; Hardin 1993). Works and practices ofhigh culture are strongly
aesthetic, but the aesthetic element is mostly integrated into some broad
context, such as the conduct of royal and elite life, religion, festivals, or provision for the dead.
came before, the transformation of art, the introduction of writing, and the
centralization of symbolic structures all suggest that new meanings and values were arising from the cauldron of state formation. We further infer that
the evolution of centralized government and an inner elite was seen as having the mission of enhancing the new order through its exploitation of the
wealth it created. In the new social and cultural hierarchy, the invention and
elaboration of high culture become self-legitimizing. We explore below
how this happens and how the high culture is maintained.
This legitimation also relates to the issue of change versus stability, since
both are goals of elites. Egyptian and Mesopotamian civilization endured for
longer than any of their successors and achieved great consistency and stability of primarily high-cultural style. This style was adhered to by ruling
groups, and it was maintained and transmitted by them and by specialists on
the edge of the elite. In considering Egypt one can set aside short-term political instability, which often affected only the inner elites. The pace of
change in any period could have been almost imperceptible to the actors
(whose generally short lifespans render change more difficult to perceive
than it is for us); this imperceptibility was no doubt deliberately sought, for
it reinforces the image of the high-cultural order as given and immutable. In
Mesopotamia, the high culture was maintained in the face of numerous
episodes ofpolitical and other change, and it was reproduced to render such
changes orderly and legitimate.
An illustration of the value of restricted transmission incorporated in
works of art is the early Egyptian complex of representation and writing
(Baines 1988a, 1989b). Egyptian writing divided from the beginning into a
cursive variant used for administration and the "monumental," hieroglyphic
form used in works of art. In early times neither encoded full syntactic forms
of the language, yet the limited forms endured for some centuries. Writing
was an adequate and valuable tool of administration, even with little expression of syntax; in art it essentially supplied captions and was not needed
for continuous text. Captions were integrated with figural representation to
form a genre of record and display of such things as royal exploits; in combination with the power of the visual image, this created a form that was
probably more effective for being laconic. Moreover, the visual qualities of
the hieroglyphs, and their interaction with fully representational images and
with an intermediate emblematic mode of representation (Baines 1985: 4163, 277-86), created an enormously powerful complex; the best parallels
may be in Mesoamerican artistic practices (e.g., Marcus 1976c, 1992c;
Reents-Budet 1994). These features were embedded in the system of decorum already mentioned.
Most of the surviving early objects that bear hieroglyphs and pictorial
representation are quite small (e.g., Adams 1974), many of them dedicated
Nonetheless, the high-cultural complex was so important that the main elite
must have participated in changes.
The general development ofthe First-Fourth Dynasties was toward ever
more extravagant and elaborate monuments for the king and for a diminishing proportion of the inner elite. There is a striking contrast between the
extensive and Widespread cemeteries of the First/Second Dynasties and the
central Fourth Dynasty, when almost all major tombs were at Giza, surrounding the king's massive pyramid (for a provincial exception see Garstang
1904).
The Giza tombs were broadly separated into two groups. The highestranking tombs were sited between the Nile Valley, on the east, and the Great
Pyramid, while the slightly less prestigious were sited to the west of the
pyramid (e.g., Dunham and Simpson 1974-80). All were massive and, for
the first time, constructed in stone, and they partook in a grandiose undertaking that projected elite hierarchies into the next world. The secondary,
western group lacked public decoration and, thus, marked a step back from
the finest tombs of the previous generation. The concealed stelae and sculptured heads found in the tombs (e.g., Smith 1949:pls. 5-9, 32) show that
art of the highest quality was available to these people (see also Russmann
1995: 118). Even though relief decoration may not have been carved in earlier tombs of the same status-which belonged to extremely few peopleits absence in these Giza tombs constituted a severe restriction of choice in
the context of a generally increased potential ofwriting and an existing decorative repertory. This restriction conveyed in high-cultural terms the appropriation of all major symbolic means to the king and his immediate
group, most of whom were members of his family. It also may have related
to a wider use of such media in lost temple contexts. An aesthetic aspect of
the development is visible in the contrast between the full decoration of the
temple of Snofru at Dahshur (Fakhry 1961) and the remarkably austere and
abstract style of the Second Pyramid complex of Re'kha'ef (Khephren) at
Giza (e.g., Arnold 1992:198-202; Baines 1994:77-78).
This development was reversed by the end of the Fourth Dynasty, and
more clearly in the early Fifth, when both royal and nonroyal monuments
became smaller but richly decorated, and the beginnings of a nonroyal "literature" appeared in the form of inscriptions in elite tombs containing extended titularies and brief narrative passages (Baines 1998). This change was
political as well as cultural. A narrative preserved in a literary work from
a millennium later implies that the transition from the Fourth to the Fifth
Dynasty had the sanction of the sun god Re', the principal deity, and was
in some sense subversive (Lichtheim 1973:217-22). Yet both dynasties
accorded Re' great prominence, and an explanation in such terms does not
appear evident. Rather, there may have been struggles between elite groups
that mildly disfavored the kingship, which never again had quite the same
control of symbolic and material resources, and favored a strongly entrenched elite that was not closely bound to the king by kinship. This development restricted the role of the royal family, while probably enlarging
the general ruling group. A crucial point of contention appears thus to have
been the restriction of display and high-cultural resources. Those associated
with the Fifth Dynasty spread these resources a little more widely, at least in
semipublic display on the monuments.
The broader society did not participate to any great extent in such transitions; rather, high-cultural resources were exploited both for their own
sake and for their political potential. The nature of the material at issue is
also significant: the eventual development toward a more widespread literate high culture was very difficult to reverse. No later reversal comparable
with the progressive restriction of the far less literate First-Fourth Dynasties is known, although there was much variation; the indigenous temple
culture ofthe Greco-Roman period, which was significant in many ways for
the whole country, was accessible only to tiny numbers.
As before, those outside the inner elite appear not to have had a
significant historical role. What becomes visible is the existence of different
factions within the elite; one of these factions evidently offered benefits to
somewhat more people than did the tiny central group of the Fourth Dynasty. Later in the Old Kingdom people exploited their old and secret religious knowledge in the play of status (Baines 1988b), again indicating a
diversity in the inner elite, which was then probably larger, if not wealthier.
Such displays, which rely upon shared high-cultural values, are part of the
currency of status competition in all ancient states.
In Mesopotamia, writing lies perhaps still more at the heart of high culture than it does in Egypt. Mesopotamian high culture crystallized at about
the time of the formation of the first city-states, but its raison d'etre cannot
be explained simply as part of the political and social transformations of the
Uruk period (Yoffee 1993b: 64-68, 1995). Writing was from the start concerned not only with economic recording but also with the re-creation and
standardization of Mesopotamian cultural "encyclopedism"-the description and systematization oftitles ofpeople and things that became part ofthe
high-cultural complex. Although writing as a semiotic system was invented
in the late Uruk period (Boltz 1994; Michalowski 1990, 1993b), and "lexical lists" appear among the first tablets (Englund, Nissen, and Damerow
1993) and were used as part of the scribal curriculum, Sumerian writing did
not achieve much standardization until the mid-third millennium BC. It is
remarkable that such standardization owed little to political developments,
since the early third millennium was a time of endemic warfare among citystates. It was in a multilingual environment of independent city-states that
systems of measurement and mathematical notation became regularized
found in private houses (Charpin 1986), to the first millennium, when private libraries were owned by priests (Parpola 1983) and schools were affiliated with temples, so the context of the texts themselves changed. For
example, epic compositions of the early second millennium such as "Atrahasis," which contains a creation story, were part of a ritual in the first millennium, when the text was used to cure barren women (Lambert and
Millard 1969). Although social and economic records and certain correspondence (especially in the Old Assyrian and Old Babylonian periods,
when the writing system was the simplest; Larsen 1987b: 219 - 20) were
written at the behest of elites, the art of writing was inaccessible to the
population at large. Written texts spoke to other written texts, and a highcultural corpus of written matter reinforced the separateness of the inner
elite and scribal class. Although scribes themselves did not often achieve the
status of elites, they were themselves not independent of the institutions of
palaces and temples and never became semiautonomous guilds of literati, as
Hsu (1988) describes for Han China. As in Egypt, some written texts faced
walls or were placed in mountain aeries: these impediments to sight, however, mattered little to the intended divine readership.
Again as in Egypt, high culture was mobilized in, and formed a leading
part of, major political and cultural changes. It is a commonplace in Mesopotamian scholarship to note that Sumerian texts were learned well after the
time when Sumerian was actually spoken (c. 2000 BC), and indeed until the
end of Mesopotamian civilization. This continuity of an aspect of high culture illustrates well the power of the textually denoted Mesopotamian discourse community (Cooper 1993). By this term, at least for Mesopotamia,
we obviously do not mean the vast majority of Mesopotamians, who could
neither write nor understand the arcane language of belles lettres, religious
texts, or royal pronouncements. Rather, it was the inner elite (and their
scribal dependents) who sponsored and reproduced the texts that delineated
the hierarchies composing the world and defined the critical roles of rulers
and gods. There were, however, enormous changes in political systems
throughout Mesopotamian history, and it was the role of high culture to be
flexible enough to legitimize and naturalize those changes.
In the middle and late third millennium, official writing systems changed
at least twice. With the conquest ofthe House ofAkkade, Akkadian became
the normal language of the administration, and scribes I bureaucrats were
trained in new ways in Akkadian as well as Sumerian. In the Ur III period,
Sumerian was again the language of the bureaucracy, and scribal schools
were reoriented in the new imperial structure (Cooper 1973; Michalowski
1987). Ur III rulers, however, sought to depict themselves as descendants
from the heroic past, avowing their kinship with Gilgamesh and other kings
oflegend from the city ofUruk. In scribal schools, as well as in the incuIca-
tion of new administrative language, tales of these kings were copied and lor
composed. High culture was thus manipulated in order to legitimize new
kings, a new bureaucracy, and new imperial rule.
In the second millennium BC, after the fall of the Third Dynasty of Ur,
the governments of city-states in Mesopotamia were progressively seized by
Amorite leaders, and the number ofAmorite personal names in the texts also
increased. By the end of the seventeenth century BC, one royal edict describes the population of Babylonia as "Amorites and Akkadeians" (Kraus
1984). Naturally, scholars (beginning with Clay 1909) have sought to identify Amorite social institutions and systems of beliefs that ought to accompany these political and demographic changes. However, ascriptions of new
levels of private enterprise as a reflection of the business mentality of Semites
(Pettinato 1971)-whatever that might be-or new marriage customs deriving from a tribal past (Falkenstein 1956 -57) are oddly juxtaposed against
the absence of any texts written in Amorite. Indeed, it is the lack of largescale culture change in early second-millennium Mesopotamia, other than
those economic and social changes that can be accounted for as internal historical developments (Yoffee 1995b), that can be explained as a conscious
policy of the new rulers of Mesopotamian city-states.
Seizing political power in city-states, Amorite leaders contested not only
with local elites but also with other Amorite elites. One mechanism that advantaged Amorite leaders was their ability to mobilize support across the
countryside, that is, beyond the borders of the autonomous city-states. In a
letter from the king ofUruk to the king of BabyIon, the former appeals for
help against enemies because the kings of Uruk and Babylon are of "one
house" (Falkenstein 1963: 56), presumably of one particular Amorite group
(the Amnanum Amorites). These Amorite rulers, having successfully gained
power in Mesopotamian city-states, made sure that Mesopotamian high culture was reproduced faithfully. The Akkadian language of Hammurabi of
Babylon (descendant of the same Amorite ruler of Babylon mentioned in
the letter cited above from a king ofUruk), as embodied in his "Law Code,"
is regarded to this day as a classical text to be assigned to first-year Akkadian
students; most of the Sumerian poetic compositions that were presumably
composed in the late third millennium are known from schoolboy copies
discarded in Nippur and Ur in the Old Babylonian period.
After the collapse of the Old Babylonian order in the mid-second millennium, the Kassites, who spoke a non-Semitic language (an isolate, possibly evolved in remote valleys in the Zagros Mountains), established a new
dynasty in Babylonia. Their family structure, composed of landholding
lineages (Brinkman 1980), seems distinctive in this period. The kudurrus
(boundary stones) ofthe Kassite period are similarly distinctive, and the Kassites were endogamous (Maidman 1984). Yet the literary and high-cultural
objectification that archaeologists mark in the material record. Such traditions are made feasible by writing and material high culture, which recursively enhance their meaning and which can be shaped by inner elites. In the
cases we have chosen, then, change is particularly warranted in times when
new political leadership requires stabilization and legitimation.
Major innovations in high culture were fundamentally important for the
self-image of Egyptian civilization; later times looked to the Middle Kingdom as the "classical" epoch. Despite the continuing significance of the visual arts, the clearest later focus of this classicism was language and literature,
so that the Middle Kingdom reforms ofwritten language and genre acquired
a retrospective conservatism that protected the continuity of Egyptian selfdefinition. In, Mesopotamia, scribes in Hellenistic and later times were
doggedly half-learning languages and textual traditions that were of little
meaning or importance to anyone but themselves. Singing the praises of antediluvian sages (Klochkov 1982; Reiner 1961), they appealed to a tradition
that now was totally divorced from any inner elite or high culture and was
soon to disappear even from that marginal condition.
CONCLUSION: COMPARISONS,
CONTRASTS, COLLAPSES
Order, Legitimacy, and Wealth: Coda
In this comparison of Egypt and Mesopotamia, we have attempted to delineate some critical institutions of these civilizations and to investigate what
made these civilizations distinctive and what allows them to be compared.
We have devoted little space to origins. In conclusion, we review our findings and consider still more briefly the collapse of the civilizations.
In the rapid crystallization of states in both regions, we see the rise of a
new kind of order that reformulated the cosmos so that a new form ofleadership and the principle of hierarchization were proper to the continuance
of that cosmos. In all aspects, from the material and economic to the religious, the institutionalization, continuance, and, on occasion, expansion of
the new order are the essential tasks of the leaders.
Although this ideological principle of order evolves or is invented in
both civilizations, in Egypt the cosmos is firmly connected to one head of
state and one organized system ofvalues and beliefs; office and values are inextricably linked. Even the much discussed diversity of such conceptions as
creation myths proves to apply principally to later periods (Bickel 1994).
The manifestations of values and beliefs, which in the archaeological record
begin with artistic forms, spread very slowly to extensive verbal forms and
to texts ofthe type known from many civilizations (but are hardly preserved,
for example, from Mesoamerica) as transmitting central concerns; the artistic forms are maintained for as long as the civilization.
In Mesopotamia there was no single political system, but a strong cultural sense of unity was manifest in the material culture and in standardized
language and school curriculum. In part, this order was originally imposed
because of the necessity to keep track of people and commodities, which is
the logical outcome of trajectories toward differentiated societies that result
in centralized administrative institutions. Nevertheless, the early trends toward standardization of written expression across separate and independent
city-states and the counterfactual conception (especially vivid in the Sumerian King List) that there should be political unity in the land bespeak the existence of an overarching cultural sphere of interaction (Yoffee 1993a) that
cannot be reduced to economics.
Order is more than a political necessity. It is the logic of a new way of
thinking about society and about the cosmos, one that justifies the association of people who are not kin, especially those in the service of the inner
elite, and establishes the principle of stratification and of limited access to
wealth. Insofar as order itself creates a "natural" progression toward increased order, complexity, and hierarchy, especially in its high-cultural manifestations, it must exploit wealth for this self-enhancement; it cannot be
generalized to everyone.
Wealth, together with its restriction to certain groups, is one of the most
obvious facts of civilization. The acceptance of agreed measures of wealth
and the creation ofstorable and to some extent convertible forms ofit transform its social potential. The move toward imperishable forms is especially
significant, even if fashion may bring an opposite effect by devaluing the
wealth of the recent past. Wealth is displayed, and such displays require further stratification because it is obtained through the labor of others, from
networks in which the negotiants are not kin, or from organizing the procurement of materials, some of them from remote regions, on a scale not
feasible for kin groups alone. Wealth therefore requires new codes of communication that establish the ability to trade with foreigners and connect
these distant people in a community of interests. Such activities assume a
scale and importance hardly seen outside complex societies. "Interaction
spheres" in the archaeological literature usually denote such elements of elite
negotiation that interconnect people and regions above any commonality in
ethnicity, language, or politics. Alternatively, wealth may involve "raiding"
on a scale that is feasible only if highly unequal relations can be imposed
on surrounding groups. In either case, civilizations contend in their search
for wealth with other groups whose values and organization may be quite
different from their own. In doing so, they must achieve more than a simple
material domination; in ambivalent fashion, they must moderate and enhance their comprehension of those with whom they deal, assigning negative values to the world defined as outside and thus further limiting access to
their own values.
The internal elaboration of order and its exploitation of and expression
through wealth involve significant legitimation. The ruling elite must return
benefits to the rest and to posterity. This return is made in matters of war
and defense, economic security, and legal procedures, but above all in perpetuating the cultural patterns that establish and maintain order. These patterns include sacred rites, which may sometimes be the irreducible focus of
elite activities, but they also extend to the entire high-cultural complex
within which the rites have meaning. Kings and rulers must show their concern with the whole population by promulgating "laws" and edicts that present such concerns and by staging events in which spectacle and ceremony
define the state's role. Through the common definition and labor of ruler
and ruled, the arenas of temple, palace, city, and country in which the ruling elite act out their role, their concerns, and their privileges are constructed as interlocking representations and enactments of the cosmos and its
maintenance. All these activities emphasize the dispersive ties between ruler
and ruled, interweaving order, wealth, and legitimacy into the civilization's
fabric. What they do not do is enact any legitimizing requirement that the
elite redistribute the wealth created by order throughout society.
in the Hellenic world, but within Egypt the kings worked through the traditional elites, temples, and general cultural forms. The Greco-Roman period saw the greatest expansion of high culture in the numerous large and
small temples constructed throughout Egypt and Lower Nubia (e.g., Arnold
1992), and for Ptolemaic times in widespread artistic and literary production. This cultural provincialization-since native culture was not strongly
represented in the capital Alexandria, and Hellenistic culture was in second
position outside it-had a partially leveling effect, because the entire native
population could be set as provincials against the Greek incomers. There also
was broad religious participation of a kind poorly attested from earlier times.
Yet the main temples were even more elite products than their predecessors,
with inscriptions carved in an arcane elaboration of the script that few could
read. In relation to central patronage, it is characteristic of this development
that the peak of temple construction in the south-financed by the ruler,
not the local population-was around the time of the Roman conqueror
Augustus (30 BC-AD 14). The Ptolemies and the Roman emperors made
Greek the official language, and the Romans suppressed the traditional elite
politically, but they had to use the core symbolic forms of the ancient civilization (it is said that only Cleopatra VII [51-30 BC] ever learned Egyptian,
but also that she learned a host of other languages).
If we are not to follow Voltaire's Brahman interlocutor (see MiillerWollermann 1986: 1) and say that Egyptian civilization died because it had
lived, it is difficult to explain its demise and inappropriate to attribute this to
any single cause. So far as one can speak of demise, the notion that a civilization is an interconnected entity remains valid. While the world within
the temples never accepted the Ptolemies as fully as it had accepted native
kings (Quaegebeur 1989; Winter 1976), the changes imposed by the new
rulers were not irreversible and are insufficient to have set the scene for dissolution. What seem more serious are the disappearance of a wealthy native
elite in early Roman times, reduced integration of the temple high culture
into local communities (Baines 1997a), and the difficulties of the Roman
Empire in the late second and third centuries AD, which led to a worsening
of general conditions (Bagnall 1993). This was when Christianity began to
spread, being taken up in Egypt by both the elite and others (Frankfurter
1998). At this point, a reversal of interests is very suggestive. Although traditional Egyptian religion maintained a hold in the extreme south into the
mid-sixth century, when pagan cults had already been suppressed in the empire, Christianity gained ground by accompanying a revival of the Egyptian
language written in Greek letters (now known as Coptic) and was thus
both a religious and a cultural rallying point for Egypt against the foreign
rule that was now almost a millennium old. Thus, Egyptian civilization was
long able to maintain itself in the face of foreign rule through its focus on
NOTES
We are grateful to Jeremy Black, Marianne Eaton-Krauss, GeoffEmberling, and Andrea
McDowell for commenting on drafts, and to Alan Bowman, John Davis, and, especially,
Geoff Emberling, for patching holes. John Baines's initial formulation of the ideas on
high culture was presented in a lecture to the School of American Research, Santa Fe, in
April 1992. Norman Yoffee wishes to acknowledge the NEH fellowship that enabled
him to be a resident scholar at the School of American Research, 1991-92, when this
essay first took shape.
This essay is argued principally on a theoretical level. In the frequent absence of syntheses, the argument could not be supported with references without extending the bibliography beyond reasonable bounds. Citations are selective; where possible, we have
chosen items that lead to additional material. Our text was composed in 1992 for the
seminar at the School of American Research and revised for publication in 1993. Since
then we have not extended the number of citations significantly.
8
Sodoculturol(omplexit, Without the Stote
Ihe Indlls (jYilizotioll
GRlGORY l. POSSlHl
TAJIKISTAN
CHINA
AFGHANISTAN
INDIA
Arabian Sea
research has led a number ofscholars to question the proposition that the Indus civilization was the kind of complex sociocultural system that is usually
associated with the archaic state. If this is the case, it suggests that the internal workings of the Harappan civilization were significantly different than
those found in Mesopotamia and Egypt at a comparable time. This essay examines this critical question.
Harappa
o
I
100
!
WHAT IS A STAH?
My own thoughts on the nature and definition of the state are found
throughout this chapter. No claim is made that there is a consensus among
archaeologists who study early sociocultural complexity on the definition of
the archaic state, although there seems to be some agreement on a few
points. The state is a form of political organization. It developed among
peoples with large-scale economies, having considerable specialization in
craft and career tracks. The administration of the state is a prominent feature of this form of political organization, and some portion of the career
specialists formed a bureaucracy. The craft and career specialists, and the
other peoples of their society, were arranged in a hierarchy of classes. The
state monopolizes the legitimate use offorce as a means ofsocial control and
an agency to protect, if not expand, the sovereignty of the people it encompasses. The state form of organization places the management of diplomacy
and warfare in the domain of a strong, forceful leader. There is an imperative among states to focus on individual leaders, usually called "kings,"
which reflects the presence of unequivocal political leadership.
Kings can be seen as the symbolic embodiment of the people and their
sociocultural system. They are often deified, or allowed to flirt with notions
of human deification. From the point of view of the citizenry, the
deification of one's king might be thought of as the deification of oneself
Such matters of ideology and self-identification mean that the ties between
the state political apparatus and state ideology are intimate and involved with
the idea of political legitimization. These ties may lead to considerable tension between the state as a political form and religion as an institution ofideology, unless, as was often the case in ancient Egypt, the "king" (pharaoh)
was also the chief religious personage. The economy of the archaic state is
partly shaped by the political form to protect and expand its sovereignty. The
king, as the leader in war, even cold wars, was in constant need of resources
to serve his political ends. He strove for an economy that was responsive and
productive in a quantitatively superior way. Historically this is accomplished
best through centralization, and all that the centralization ofa political economy implies in terms of a bureaucracy.
The political form we have come to call the archaic state has a strong focus on kingship, or centralized leadership, that is in all likelihood given to
the aggrandizement of the individuals who rise to this office. The economies of states tend to be centralized, heavily (not exclusively) controlled
from the office of the king, so that it can effectively serve the diplomatic and
military needs of the political apparatus. This implies a staff of functionaries
(a bureaucracy) to implement and monitor the economic decision making,
as well as to collect revenue and produce for the use of the center.
There are many conceptual and definitional problems with the archaic
state (e.g., Cohen 1978), and there is still no sound definition useful in archaeological terms (Crumley 1976). It is often suggested that states are composed of people whose allegiance is determined territorially, rather than by
their kinship. Those who have perused the ethnographic record know that
this view is not well-founded; many people in very small, simply organized
societies use territoriality to define group membership, such as the Punan of
Borneo. The records of disputes among Native Americans are filled with
evidence for this point.
For classical unilineal evolutionists, the state is a level ofsociocultural integration where the means of integration is "that special control, the use of
force and constant threat of force from an institutionalized body of persons
who wield the force. A state constitutes itself legally: it makes explicit the
manner and circumstances of its use of force, and it outlaws all other use of
force as it intervenes in disputes between individuals and groups" (Service
1963 :xxvi). An archaic "state differs from a chiefdom most strikingly because it is integrated by a special mechanism involving a legalized monopoly
offeree" (Service 1963:xxviii).
Other thoughts on the nature of the state suggest that it is "the organization for the regulation of social relations in a society that is divided into
two emergent social classes, the rulers and the ruled" (Skalnik 1978: 597). In
strict terms this definition is as appropriate for tribes and chiefdoms as it is
for the state; it is a good example ofsome of the fiizzy thinking that emerges
from many discussions of the archaic state.
The state is held in opposition to the chiefdom, which is seen as a step
lower on the evolutionary scale.
The rise of centralized leadership involves a rise in the prestige of the
person holding the office of chief. This prestige attaches to relatives of the
chief, depending on genealogical nearness; and as the status becomes more
permanent: hereditary. Thus, we find not only that unlike parts have arisen
in society but also unlike persons. That is to say that chiefdoms are typically
nonegalitarian, that they are characterized by differences in the hereditary
rank, higher and lower status of persons and their families, in addition to the
universally human age-sex status differentiation. (Service 1963: xxiv-xxv)
goods and production, but with no true private property. Social stratification is present, but true socioeconomic classes are not.
Robert Carneiro, a unilineal evolutionist who published an important
paper on the origins of the state, emphasized that states are able to "draft
men for work, levy and collect taxes, and decree and enforce laws" (Carneiro
1981 :69). The problem with the words used in this kind of definition is
that, in developmental terms, they lack precision. When does the force of
persuasion in interpersonal relationships give way to being able to "draft
men for work"? How does a "tax" differ from the kind of tribute that "big
men," sometimes only "charismatic men," are able to coerce out of their
followers? What is a "law" in developmental terms, and how can we differentiate it from a pronouncement? Moreover, true law codes appear many
centuries after hypothesized states appeared. Some hold that laws are made
by states, and lesser organizations on the continuum of sociocultural development do not formulate them. But laws define the state and are not simply
an attribute of them.
In reviewing the nature of states, especially from the point of view of
Mesopotamia, Charles Redman (1978: 280) suggests that they (1) hold a
concentration of economic power; (2) are organized along political and territoriallines; (3) are hierarchical in terms of differential access of resources;
and (4) hold a monopoly of force. One might wonder how"concentrated"
is a "concentration of economic power"? Try to find a society that is not organized in some way and to some degree along "political and territorial
lines." Social hierarchies are found in many societies that are not states.
What is a monopoly offorce, when perfectly legitimate states are forced to admit (do not merely allow) that every citizen has the right to bear arms, and
turn a blind eye to physical combat between urban gangs and citizens?
The state is seen as functioning with an administration headed by a sovereign-a form of administration that involves what is generally called a
"bureaucracy," with full-time administrative specialists who are on a permanent career track different from other specialists. The bureaucracy is
arranged hierarchically and charged with the functioning ofgovernment according to accepted norms and with ensuring that the wishes and desires of
the ruler are carried out, down to the lowest levels of society. Because bureaucracies handle the collection, processing, and dispersal of information,
and the organization of settlement patterns, from urban centers to village
and hamlet outliers, they are thought to be a proxy through which the
growth or evolutionary stage of a sociocultural system can be measured
(Wright and]ohnson 1975).
In Claessen and Skalnik (1978b) we learn that the following four factors,
regardless of their historical sequence, had a relatively direct influence on
the formation of the early state: (1) population growth and/or population
pressure; (2) war, the threat of war, conquest, or raids; (3) conquest; and
(4) the influence of previously existing states.
Regardless which of these factors actually triggered off the development, the
ultimate result was always very much the same, namely the early state. The
above four factors promoting the rise of the early state do not replace the
essential conditions of a producing economy with a surplus and the consequent conditions of social inequality and the beginnings of social stratification, without which the early state would be inconceivable. (Claessen and
Skalnik 1978b: 642)
These causal factors are in line with the well-developed theory of state
formation put forth by Robert Carneiro (1970a). Other general findings by
Claessen and Skalnik (1978b) on formative processes for the early state can
be summarized as follows. (1) Historically, social stratification began much
earlier than the emergence of the state; however, the early state took this social differentiation and expanded and codified it (Claessen and Skalnik
1978b: 642-43). (2) Conquest can lead to the formation of the early state,
although this was not well represented in their sample (Claessen and Skalnik
1978b: 643). (3) Population growth and/or pressure was found to lead to
more complex political organizations (Claessen and Skalnik 1978b: 644).
And (4) urbanization was found to play a decisive role in the formation of
the early state (Claessen and Skalnik 1978b: 644).
Unfortunately, the role of trade and commerce, especially in the formative contexts of early states, was poorly handled by Claessen and Skalnik
(1978b). However, an independent review of the individual studies in their
book shows that trade and commerce act as stimulants to the growth ofmore
complex management and governmental institutions. They can be accompanied by raids, warfare, even outright conquest, which also are stimulants
to the differentiation of sociocultural systems.
An aside seems to be in order here. In a very useful review of the chiefdom, Earle and his colleagues (Earle, ed. 1991) investigated those factors that
seem to lead to the emergence and integration of this form of sociocultural
organization. They found that neither ecological factors nor population
pressure were useful explanatory concepts (Earle 1991a: 5). In contrast to
Claessen and Skalnik, Earle and his colleagues focused more on issues ofproduction (debt, creating infrastructure, means for powerful people to cause
others to work), political configuration (circumscription, power, force, forging external ties), and ideology (seizing and/or creating means and signs of
legitimacy). The importance and utility of ideology emerged as a powerful
force in culture change and the stability or collapse of chiefdoms.
The state is a highly successful form ofsociocultural organization. In spite
of bureaucratic inefficiencies and sometimes wasteful competition between
TABLE 8.L Site Counts, Total Settled Area, and Average Site Size for
Some of the Indus Age
Site
Counts
Settled Area
in Hectares
Average
Size
Post-Urban Harappan
c. 1900-1400 BC
1281
1837
3.50
Mature Harappan
c. 2500 -1900 BC
1022
3846
7.16
Early Harappan
c. 3200-2600 BC
477
2113
4.50
256
1336
5.21
84
295
3.51
Burj Basket-marked
c. 5000 -4300 BC
33
85
2.58
20
53
2.65
Togau
c. 4300-3800
c. 7000-5000
BC
BC
Notes: The Early/Mature Harappan transition, which occurs c. 2600-2500 BC, is not included in
this table. Many sites are not measured; therefore average size cannot be computed by dividing the
settled area listed in the table by site counts.
Period Il at Amri, the transitional phase between the Early and Mature
Harappan, ended with signs of a significant amount of burning. The excavator has observed: "The upper levels are blackish and ashy, but they are
mostly so near the surface that it is difficult to say whether this occurrence
should be interpreted as evidence of some sort of violence or of fire" (Casal
1964: 7). A large fire was also evident in Period ID, associated with a building that Casal believes was a godown, or storage facility. The walls of this
structure had been reddened by fire, and parts of the superstructure had
fallen to foundation level.
The excavations at Nausharo have not yet been completed, but the excavation team, under the direction ofDr. Jean-Franyois Jarrige, has found
evidence of extensive burning associated with Period ID, a transitionalphase occupation at the site. He has noted: "The two architectural complexes of Period ID in sector NS.G and NS.L have been heavily burnt and
the walls have turned red due to heat. At Kot Diji, too, the final phase of the
pre-Indus period has been destroyed by fire. This could support the romantic hypothesis of the Harappans destroying by fire the pre-Harappan settlements. But fires can be as well accidental" (Jarrige 1989:64-65).
Renewed work at Harappa has uncovered a settlement ofthe transitional
phase of some 24 ha, covering most of what is called Mound E; however,
there are no signs of disruption or burning (R. H. Meadow, personal communication 1993).
An Evaluation of the Burning at the End of the Early Harappan
It is difficult to evaluate this kind of archaeological evidence. jarrige's caution that fires have many causes is true. Attributing burning to warfare or
raiding is out offashion in archaeological explanation, no matter how prevalent these activities were in the lives of people during historical times in the
Near East and South Asia. These disruptions also have been noted by S. P.
Gupta (1978: 142), who tends to think of them in terms of natural calamities, which might well be true for Kalibangan (Raikes 1968). Nonetheless,
there is a conspicuous amount of evidence for burning at the historicaljunction where the Early and Mature Harappan meet. This is in contrast to earlier times when there is little evidence for large-scale burning in the greater
Indus region. For example, there is little, if any, evidence for large-scale fires
in levels at Amri, Gumla, or Nausharo preceding the transitional period.
The same is true at Kalibangan, Balakot, and other sites in the greater Indus
region,
If fires are only accidental domestic events, one would expect a random,
or haphazard, pattern to their occurrence, an occasional "blip" in the archaeological record resulting from the careless handling of a lamp, o~ a
child's play getting out of hand, or a cow knocking over a lamp and the milk
stool. The scarcity of evidence for this kind of day-to-day tragedy in the
lives of the ancient inhabitants of the region is, therefore, significant. It leads
one to think that much of the evidence for the little fires, the kind that destroy one or two buildings and are the unhappy grist of daily life, is erased
by the process of cleaning up the mess and rebuilding. Some evidence is
probably still there, but it is a detail of the archaeological record; it does not
jump out at the excavator as a catastrophe and become a significant observation in the site report. When signs of large-scale burning are noted in an
excavation, as at Kot Diji, Gumla, and Nausharo, and to some degree Arnri,
it is likely to denote a large conflagration, the damage from which is so massive that it cannot be diffused by cleanup and rebuilding.
There is a pattern in these conflagrations. First, they seem to have been
relatively large-scale-not little domestic fires, but events that consumed a
significant portion of a settlement, ifnot all of it. Second, they are associated
with the Early/Mature Harappan conjunction. Such fires are rarely documented at sites prior to this time. Still other observations relating to the conjunction of the Early/Mature Harappan have a bearing on the development
of the Harappan civilization.
Further Suggestions of Disruption
Archaeological exploration has proceeded to a point where we can see that
many Mature Harappan sites were founded on virgin soil, and many Early
Harappan sites were abandoned and not reoccupied during the Mature
Harappan. This is well documented in the deep settlement histories ofSindh
and Cholistan (see fig. 8.1). Table 8.2 presents site counts from Sindh and
Cholistan (Mughal 1982, 1997) to illustrate the abandonment of Early
Harappan sites and the founding of Mature Harappan settlements.
The numbers are striking. In Cholistan, 33 of the 37 Early Harappan
(Kot Dijian) sites were abandoned by the end of this period. For the Mature
Harappan, 132 of 136 sites were established as new settlements on virgin
soil. A similar, if less dramatic, pattern is evident in Sindh. The peoples of
the Mature Harappan clearly preferred to find new places to live, cutting
their historical ties with the older Early Harappan settlements. This is another significant disjunction between the Early and Mature Harappan. It
complements the signs ofburning and strengthens the notion that important
changes took place at this junction, which emerges as a period of both disruption and of renewal-a time when the peoples of the Harappan civilization seem to have severed ties with older places and sought fresh, new places
to establish themselves and their new way of life. While the burning could
represent one of the consequences of war, or raiding, it also could result
from an act of renewal.
TABLE 8.2.
Cholistan
Site
Count
Total
Hectares
Average
Site Size
Site
Count
Total
Hectares
Average
Site Size
Early Harappan
54
79
2.85
40
256
6.40
Mature Harappan
86
436
7.95
176
877
5.62
23
42
3.48
12
6.25
Early Harappan
sites abandoned
27
37
2.48
38
244
6.42
Mature Harappan
sites on virgin soil
63
415
9.43
170
947
5.64
Note: Many sites are not measured; therefore average size cannot be computed by dividing the
settledarealistedin the tableby site counts.
SOCIOCULTURAL COMPLEX
Possehland Rissman 1992) suggests a rapid origin of 150 yearsfor the Harappan. We suggestit was even less, or 100 years, ca. 2600 - 2500 BC. Within the
next 100 years, the Harappan became the largest ethnic group within the Indus Valley. This rapid distribution rate was matched only by Harappan abandonment of large sections in the Indus Valley which was under way by ea.
2000 BC, a process intensified by later hydrological changes. Whatever the
Harappan group's organizational complexity, it was a cultural system promoting rapid territorial expansion. (Shaffer and Lichtenstein 1989: 123)
Afghanistan
This short period of rapid, paroxytic change seems to occupy the century from about 2600 to 2500 BC. Such precision probably goes beyond the
discriminating powers ofradiocarbon dating. The transitional stage is a good
example of a time when a major acceleration in the rate of culture change
led to the development of a new level of sociocultural complexity.
Arabian Sea
o
I
200
I
400
kilometers
In this sense, urbanization is a hallmark, a symbol of the internal differentiation and structural specialization of Harappan life. It is these sociocultural
variables, and the sociocultural processes they engender, that allow us to
define and differentiate complex societies from others.
This new way of life involved urbanization-the building of large,
multifunctional, internally differentiated settlements where many of the
various classes and occupational categories of people lived and worked.
Many smaller settlements were integrated into the whole as well; they preserve signs of their own level of sociocultural complexity. Two of the most
prominent institutions of the archaic state-temples and palaces (Flann~ry,
this volume)-seem to find little or no expression in the archaeological
record of the Harappan civilization.
There Are No Harappan Temples or Palaces
There is no monumental religious architecture associated with the Mature
Harappan: none, for example, at Mohenjo-daro (Mackay 1937-38; Marshall 1931), Harappa (Vats 1940), Chanhu-daro (Mackay 1943), Lothal (Rao
1979, 1985), or Surkotada (Joshi 1990). The pyramids of Dynastic Egypt
and the ziggurats of Mesopotamia have no parallel in the Indus Valley.
A succinct definition ofthe Harappan religious institution or institutions
is elusive, but it is clear that these were peoples with strong religious beliefs
on several dimensions, which they expressed in a totally different way than
their neighbors to the west. The construction of huge, physical monuments
to the Mature Harappan gods and goddesses was apparently inappropriate.
The capability in terms of engineering skills and the ability to mobilize a
work force are apparent in the archaeological record of the large urban centers, but no such tasks were undertaken.
Another contrast between Egypt, Mesopotamia, and the Harappan civilization is the absence of palaces-the large abodes of heads of government
and their powerful associates charged with managing the fortunes of the
Harappan political apparatus. Perhaps the Harappan political system did not
focus on a single personage, like a king or a pharaoh. But whatever role the
powerful may have played in Mature Harappan life, the manner in which the
sociocultural system expressed it is significantly different from that of the ar-
chaic state.
The tension and competition between the religious and political institutions of Egypt and Mesopotamia were a source ofsome creativity. The construction of ever larger temples and ziggurats, funerary monuments, palaces,
and city walls was fueled by this rivalry. Competition was not the only
source of inspiration for the construction of such monuments, but it was
an important factor. Their absence in the Harappan civilization suggests
that the fundamental organizing and operational principles of the Harappan
civilization were different than those ofEgypt and Mesopotamia. It does not
necessarily mean that the Harappan civilization was any less complex or developed in a sociocultural sense. The refined nature of Harappan city life and
management, the mastery of complex technologies, the writing system, and
the ability to mobilize resources from distant lands all argue for a very high
level of sociocultural development.
Some structures at Mohenjo-daro are proposed to have been used by the
Harappan religious establishment. Marshalllists several such structures, none
of which, on detailed examination, turns out to be a good candidate for a
temple or place of worship (1931 :22-23; see also Wheeler 1968:51-53).
The best candidates for buildings with religious functions include three
buildings on the Mound of the Great Bath (fig. 8.4): the Great Bath itself,
the so-called College of Priests, and an "Assembly Hall," in L Area. In the
lower town, Structure 10 in HR-A Area and two buildings in DK-G Area
have also been discussed in this context (Wheeler 1968: 51-53). There are
structures that have been called "fire altars" at both Lothal and Kalibangan,
along with the "ritual structure" at the latter site, but the city of Harappa is
conspicuous by the absence of architectural remains associated with religious
activities. None of these buildings, except possibly the Great Bath, seems to
fulfill the expectations of a religious or ritual structure; this possible exception does not seem sufficient to invalidate the hypothesis advanced here concerning the absence of monumental expression for the Mature Harappan
religious institution.
The Indus Civilization Is a Faceless Culture
Almost certainly related to the absence ofmonumental architecture expressing the Mature Harappan political and religious institutions is the lack of
representations of individuals in Mature Harappan life. There are many human figurines in terracotta and other materials. They offer some insights
into daily life and the diverse dress ofmen and women, but there is very little
that could be called portraiture. The bronze dancing girl from HR Area of
Mohenjo-daro could be a portrait (fig. 8.5). On a slightly larger scale is the
famous "priest-king" from DK-B Area of the same site (fig 8.6). Although
it is not certain that he was either a "priest" or a "king," let alone someone
who held dual offices, it does seem that this artwork could be the portrait of
an upper-class Harappan, and that he might have held an important office,
or offices, in the political and/or religious institutions. Three or four other
pieces of sculpture from Mohenjo-daro might fit the same description.
None is inscribed or identified in any way, nor are there representations in
other forms of art that would seem to identify or single out the individual
in Mature Harappan life, whether of the upper classes or lower down on the
social scale. Moreover, it cannot be confirmed that any of them are portraits
Scale.
20
50
40
100
60
200
Dealing with this form of question concerning the peoples ofthe ancient
Indus region has not always been seen as important. For the early investigators of the Harappan civilization, the case was clear.
Marshall, Mackay, Wheeler, Piggott
For Sir John Marshall and those who followed him in the interpretation of
the Harappan civilization- Ernest J. H. Mackay, Stuart Piggott, and Sir
Mortimer Wheeler-the level of development of this culture was not in
question, or at least was never specifically addressed. They were dealing with
a civilization, and that was enough for them. The Wheeler IPiggott position
on the Harappan sees an especially strong government with "rigorously authoritarian rule" (Piggott 1950: 140) and a "rigid and highly evolved bureaucratic machine capable oforganizing and distributing surplus wealth and
defending it" (Piggott 1950: 154, quoting Wheeler).
1991). The diversity that is being found in the form, adaptive capacities, ~nd
evolutionary position of the "chiefdom" suggests that more than one SOCIOcultural and evolutionary form is represented.
Not much of the Harappan civilization appears to fit the definition of
the state as developed by Claessen, Skalnik, or Service. Nor do the archaeological markers of the archaic state as outlined by Flannery (this volume)
seem to fit the Mature Harappan. Where is the evidence for a prime sovereign and the bureaucracy that supports this kind of administration? There
are no accounting records; there is no evidence, even a hint, of a series of
specialized administrative positions, carrying out orders from the .c~pital~ via
the nested hierarchies of officials that characterize state-level administration.
Where is the evidence for a state religion, high priests, or royal graves?
Where is the face of the individual, the highest of the elite, revealed to us
as the supreme authority, or self-proclaimed hero of Harappan life? Many
early states have these features, but not the Harappan civilization. Therefore,
I am inclined to think that it had a different form of organization.
The Allchins
Bridget and Raymond Allchin (1968,1982) have written two syntheses of
South Asian archaeology that attempt to place the Harappan civilization in
a broader context of prehistoric life in the subcontinent. They found little,
if any, support for "twin capitals," an Indus "empire," or "priest-kings"
(Allchin and Allchin 1968: 129, 1982: 169). They do not, however, directly
address the question of the level of development for this civilization: "Even
if the political and economic unity is admitted, there remain profound and
tantalizing problems of how it came about and how it was maintained."
Many of these issues have yet to be tackled successfully (Allchin and Allchin
1982: 169).
The Indus Civilization Was Not a State: Recent Views
More recently, several scholars have presented evidence that the Harappan
civilization was not a state. They have challenged the "traditional" view of
the level of sociocultural differentiation of the Harappan civilization. The
first serious questions came from Walter A. Fairservis, jr., in two seminal papers published in 1961 and 1967 (reprinted in Possehl 1979:49-65 and
66-89). In a reexamination of the evidence, Fairservis begins to erode the
testimony supporting the infrastructure of the Indus civilization as seen by
Wheeler and Piggott.
I submit that Mohenjo-daro was almost purely a ceremonial center and that
its functional intent was similar to the Old Kingdom Egyptians and the
Mayans. Consequently, I think there are grave doubts that trade had much
influence on its location or its upkeep. I suspect that the political or military
orientation involved in the term "empire" is unwarranted by the present evidence. (Fairservis 1961 :18)
plex than villages but not like Ur, Uruk, or Nippur, which were centers of
vast political power with large temples dedicated to their civic deity.
This proposal that the Harappan civilization was neither sufficiently internally differentiated nor structurally specialized enough to be considered a
state is shared by S. C. Malik (1968), who wrote a thorough review of what
he called the "Formative period" ofIndian civilization. Malik (1968: 99) approaches the level ofintegration of Harappan culture from the standpoint of
societas (kinship society) and civitas (civil state society) and concludes that,
from the point of view ofunilineal evolution (he cites the work of Elman
Service), the Indus civilization was at a level of "specialization, redistribution, and the related centralization of authority which can integrate still
more complex societies" (Malik 1968: 103). This is just below the "state"
and, while he does not use the word, would be the equivalent of a chiefdom. Malik reiterated his position in 1979.
While there is an appreciation here of many of the points raised by
Fairservis and Malik, their view of the level of complexity of the Harappan
civilization is not the one I hold. Nor is it the position taken by J. Shaffer.
In 1982 Shaffer published a broad review ofthe nature ofthe Harappan civilization, intended to revitalize our examination of it:
For too long one has let preconceived notions about the nature of Harappan
Culture dictate what the archaeological data will reveal. It is time to view the
archaeological data for what it is, and not what one thinks it is. Recent studies are just beginning to indicate the real importance of Harappan studies,
showing that in South Asia a unique experiment in the development of urban, literate culture, was under way. Such a culture was highly attuned to local conditions and not a mirror image of Mesopotamia's urban experiment
or, for that matter, any other region which witnessed the development of
comparable cultural achievements. (Shaffer 1982: 49)
In his recent book on the Indus script, Fairservis elaborates this position
with the proposition that the "civilization" was a hegemony of chiefdoms
with paramount chiefs and subsidiary chiefs who were responsible to them.
The latter's responsibilities included administration as called for by the paramount chiefs and providing leadership in Harappan settlements or settlement areas (Fairservis 1992: 133).
Fairservis constructs an image of the Harappan civilization as a chiefdom, an evolutionary step below the state, and yet he retains the idea of a
civilization. He advocates a political structure lacking strong central rule, either secular or sacred. There was no bureaucracy. The writing system was,
by and large, used to identify the holder of a seal. While he admits to the
possibility of written records, direct evidence for an accounting system is
not known. The cities-like Mohenjo-daro, Harappa, and Ganweriwalawere the seats of the chiefs but had the character of supervillages, more corn-
form of organization in the sense that there is no close parallel in the archaeological, historical, or ethnographic record.
The Indus Civilization Was a State
Three recent reviews of the Harappan civilization attempt to sustain the
proposition that the civilization does represent an archaic state. Jerome
Jacobson, whose primary interests are in Palaeolithic archaeology (Jacobson
1986, see also Jacobson 1979:487-88), considers 12 sociocultural features
of the Harappan civilization: (1) urbanization; (2) writing; (3) population
size and geographic range; (4) social stratification; (5) full-time centralized
government or bureaucracy; (6) police power; (7) communication, market system, trade, and commerce; (8) taxes; (9) sovereignty; (10) religion;
(11) militarism; and (12) expansion and colonization.
The discussion of social stratification is critical. His first index for this
variable is architecture (Jacobson 1986: 146-47). He notes the presence of
very large structures at Mohenjo-daro and Harappa (citing Allchin and
Allchin 1968: 242 and Dales 1982: 199) and the alleged presence of residential units on the "citadel" ofKalibangan (Lal1979:85-86). The presence of
small houses next to larger buildings at Lothal also is noted (Rao 1973). His
point is linked in a theoretical way to Claessen and Skalnik (1978b: 639),
who state: "There is a tendency on the part of the emergent class of rulers
or the ruling stratum to live more or less apart from the other citizens of the
state." However, the data from Lothal compare domestic structures with
other kinds of buildings that are thought to have served public, nonresidential functions. Moreover, the large buildings at Mohenjo-daro are integrated
into the lower town and not isolated on the Mound of the Great Bath. Thus,
at least a portion of the elite population of this city was not living "more or
less apart from the other citizens."
In his extended discussion of government and bureaucracy, Jacobson
brings in the issue ofsettlement hierarchies and differentiation in the Harappan civilization (Jacobson 1986: 152-53). In discussing "artifacts" (Jacobson 1986: 149), he agrees with Shaffer's (1982: 47) analysis of the
distribution ofluxury items at Harappan sites, which concludes that there is
no evidence for social differentiation based on this criterion. But this is moot
since he also notes that according to Skalnik (1978:614) there is little economic differentiation within any archaic state; the Mature Harappan pattern
is what would be expected.
Attention also is drawn to physical anthropology (Jacobson 1986: 14950). It is widely known that Harappan skeletal remains are remarkably free
of trauma and differences in terms of nourishment and general health. This
lack of difference might indicate a bias in the archaeological record; perhaps
only elites were buried in the cemeteries that have been excavated, and the
poorer classes were disposed ofby other methods (Kennedy 1981 :24). In his
discussion of "disposal of the dead," Jacobson also notes another widely recognized feature of Harappan interments: they rarely contain artifacts of
significant value. If there were wealthy people at Harappan sites, it was not
appropriate for them to take artifacts with them to the grave. His discussion
of why this might be is useful (Jacobson 1986:315), but he concludes that
it makes little difference because of Skalnik's (1978: 614) dictum that there
is generally little economic differentiation among the citizens of archaic
states.
Jacobson (1986: 150) deals with writing and social stratification as "documentation." No one doubts the presence of writing, but its role in the life
of Harappans is not well understood.
The final proxy for social stratification is "art" (Jacobson 1986: 151).
The Indus seals are discussed in terms of their possible use as identifiers for
elites. Terracotta female figurines and the well-known bronze dancing girl
from Mohenjo-daro (fig. 8.5) are noted as possibly indicative of a class of
performers as opposed to the fully clothed "mother goddesses." These observations are all interesting and worth noting, although Jacobson correctly
recognizes that there is no overwhelming case here.
Overall, Jacobson presents a strong case for his position, but he does not
deal with the key issues of centralized leadership-bureaucracy and central
government. In the end we need to know, "Where are the Harappan kings
and the rest of the state apparatus?" Jacobson does not answer this question.
J. Mark Kenoyer (1992, 1994) has written two articles that relate directly
to this topic. The first is a well-reasoned review of craft specialization and
social segregation that is the best statement on this important feature of Mature Harappan life. The second deals directly with the question of whether
the Mature Harappan political system was a state (Kenoyer 1994).
The position Kenoyer takes on the issue of whether the Mature Harappan was a state revolves around what he terms the "state-level" of a society
(1994:71,73,75). He correctly assesses Fairservis's position that the Mature
Harappan was a step below the complexity we would generally associate
with a state. Kenoyer (1994: 75) observes that "In a similar vein, Shaffer argues that the urban centers of the Indus region were a unique feature of an
equally unique form of pre-state society (Shaffer 1982, 1993; Shaffer and
Lichtenstein 1989)." This does correctly represent Shaffer's work (personal
communication 1997). Shaffer's position is that the Harappan civilization is
not "prestate" in its form because he does not hold stock in evolutionary
theory of this type. Since the Harappan civilization is sufficiently different
in sociocultural form that it does not compare well with Mesopotamia or
Dynastic Egypt, which are by convention called "states," Shaffer is reluctant
to use this term to describe Mature Harappan organization. Shaffer is somewhat reserved in his assessment of the level of sociocultural complexity witnessed in the Mature Harappan, but he does see a society based on class
coupled to craft and career specialization. The level of urbanization and the
size of the Harappan civilization also are indicators of considerable complexity for him. Although Shaffer does not know how the Harappan civilization was organized, he is willing to say that it was as complex as
Mesopotamia.
Shaffer's position is in some ways the one held here. The Harappan civilization is sufficiently different from Mesopotamia and Dynastic Egypt that
it should be separated from these "states" of the archaic world. The issue is
not evolutionary in the sense of being "prestate" or even "poststate." The
Harappan civilization was a "nonstate."
In the end, whether or not we call the Harappan civilization a state is of
less importance than the central issue of recognizing the diversity of organization and form among the sociocultural systems that are associated with this
term. All societies are unique, and yet they share some set of features that allows us to call them "societies." On the other hand, terminology has its
practical utility. When one begins to understand a complex sociocultural
system like the Harappan civilization and finds that using models from other
places has a dismal record of failure-witness Wheeler's priest-kings, granaries, and implications ofMesopotamian origins-one may wish to set the
Harappan civilization apart, conceptually and terminologically, from other
complex societies called "archaic states." It is at least an act ofintellectual renewal and a candid admission that the "state" paradigm does not work; it
could be time to attempt to understand the Harappan civilization by using
another conceptual framework.
Shereen Ratnagar is a historian of ancient India at Jawaharlal Nehru
University whose first work (1981) on the Harappan civilization dealt with
the third-millennium maritime commerce between the Indus Valley and
Mesopotamia. Her interests have since turned to the political organization
ofthe Indus civilization (Ratnagar 1991). Ratnagar proposes that the Harappan civilization was a state. Her position poses an interesting, not fully resolved question. Are "urbanism" and the "state" linked in a causal way? If
they are, she makes her case by definition, because no one seriously questions the status of Mohenjo-daro and Harappa as "urban." But if urbanism
and the state are not necessarily linked, if historically documented cities exist outside the structure of a state, she may be on the wrong track. In spite
of her opinion on the differences between chiefdoms and states (1991: 1618), this point is never resolved; the contention that "urbanism is not viable
in the context of chiefdoms" is never demonstrated.
In fact, it is widely believed in anthropological archaeology that the city
or urbanization is impossible without the state. Bruce Trigger says that:
"The state is a necessary condition for the development ofurbanism" (1972:
592). Nothing in the archaeological or ethnographic record would lead one
may have been disposed of in this manner. It appears unlikely that burial in
cemeteries was the norm for the whole of the Harappan population. Class
also emerges from an examination of craft specialization in the Indus civilization, which is sophisticated and well represented at many sites. The absence
of palaces and temples, and the faceless quality of the Harappan civilization,
are the contrasts that emphasize my point on the kind (not level) of development in the political and religious institutions of Indus civilization.
The image of the Indus civilization that appears as a result of this study
places little emphasis on political and/or economic centralization, which
seems to have been at a minimum. There is no evidence for kingship; the
ideology does not express itself through the centralization of resources and
the construction of monuments. The faceless nature of these peoples is in
keeping with this position. Centralization is also one of the hallmarks of the
archaic state. Lacking evidence for strong, centralized institutions of the
type characterized by the archaic state, one might logically begin to look for
an understanding through a model using strong, decentralized institutions.
The "nonstate" model has to include the notion that Mature Harappan
institutions were strong, since there is evidence for strength in Harappan
culture. There is a suite of artifacts and an associated style that covered about
1,000,000 km 2 for a period of about 600 years. It includes such things as the
writing system, the system of weights and measures, and a host of artifact
categories like ceramics, beads, figurines, and metal objects. Architectural
standards are apparent, as well as the ability to build and maintain a city like
Mohenjo-daro with its grid town plan and a hydraulic system, the Wasserluxus of Michael Jansen (1993). The control and manipulation of water
within settlements is a notable feature of many Mature Harappan sites.
These peoples were also adventurous seafarers and overland explorers, venturing into the Arabian Gulf and to the mouth of the Red Sea. They knew
eastern Iran, Afghanistan, Central Asia, and the western Ganges, as well as
central India. Some of them also lived and worked in Mesopotamia. Strong
institutions were needed to create and maintain this sociocultural system.
Therefore, one has to eschew the notion that centralized is automatically
"strong" and that decentralized is therefore "weak."
The political strength of the Mature Harappan, or the emerging Mature
Harappan, can also be seen in the Early/Mature Harappan transition. This
might have been a period ofpolitical unrest involving some level of combat,
raiding, and conquest, with the ancestors of the peoples we see documented
at a place like Mohenjo-daro and other Mature Harappan sites being the
winners. The absence of centralized leadership and a proposal of multiregional political success may seem contradictory, but it is possible to imagine strong temporary alliances among a number of groups. Political issues
may then have been resolved through decentralized councils and negotiation. The sense that the defining quality of the Mature Harappan was a
distinct, new ideology leads one to suggest that these beliefs may have been
used as the force that bound these decentralized units together, giving them
a powerful common ground.
The same kind ofdecentralization also might be true for the religious institution. Worship or devotion seems to have been a personal matter, carried
out in the home for the most part. The Great Bath at Mohenjo-daro suggests that there was a civic level of worship, but this is not the predominant
theme, and other prominent sites (e.g., Harappa, Kalibangan, Nausharo,
Chanhu-daro, Dholavira) do not seem to have comparable religious architecture. One of the emerging themes within the study ofIndus civilization
is that it represents an ideology, a way of life in a holistic sense, much like
Islam. This ideology was more than a simple religion; it is better understood
as a genre de vie (Sorre 1962). Michael Jansen has come to believe that Mohenjo-daro was a physical manifestation intended to symbolize the Harappan ideology and way of life (Jansen 1983, 1985, 1987, 1989, 1992). This
line of thinking is relatively new and immensely interesting. J ansen also has
proposed that Mohenjo-daro was a founder city, established by the peoples
of the Mature Harappan as their premier urban environment.
Whatever the ideological life of the Harappans may have been, there is
reason to believe that their political system was segmented and decentralized, lacking a "king." Considering what is known of the Harappan civilization, the system might have looked something like the following. (1) There
is no evidence for a central government, or bureaucracy, implying that older
"tribal" organizations wielded political power within regional contexts.
(2) Territorial sovereignty for the Harappan civilization, as a whole, would
have been weakly developed compared with that of the regional centers.
Lacking a strong center, this system would be vulnerable to collapse without the presence ofpowerful integrating forces. The whole was held in place
through a strong Harappan ideology, which crossed the segmented, regional
political boundaries, reaching into every Harappan family (Possehl 1990).
Other forms ofsolidarity, such as trade, ecological interdependence, and intermarriage, would have augmented the ideological basis of Harappan civilization. (3) There were few, if any, bureaucrats and public administrators.
Crafts and trades were highly developed. Private administration by merchants, traders, "money" managers, religious practitioners, bards, and others
could be admitted as possible career specialists, given the level of sociocultural development and the absence ofa strong central government. (4) The
monopoly of force, so often associated with states, was weakly developed at
both the civilization level and that of the regional polities as well. It would
not be surprising to find that it was a subject of controversy and contest
within regional contexts. (5) There seem to be two levels of allegiance built
into this kind of system, one for the region and another for the civilization
as a whole. To some degree, the foundations for this allegiance are different;
one is a kind of "national" ideology, and the other is related to lineage, territory, and history. (6) Given this duality, the system would have been in a
constant battle against fissiparous forces: on one side, ideology, trade, ecological integration, and marriage; on the other, lineage, brotherhood, territory, history, one's occupational locus, and pride in one's social group and
the memory of ancestors who helped to make it great. This seems to have
been a lopsided battle, and over the long term the forces of division and disintegration had the upper hand. From the point ofview of cultural ecology,
it was not a satisfactory form of adaptation. It succeeded for a while, probably because the ideology was new and strong, and the manner of socializing young people to the norms ofthe civilization as a whole was undertaken
effectively. But sooner or later this too failed.
The actual form of organization for the Mature Harappan is obviously
not well understood. In fact, there may be considerable debate about certain
key points presented here: ruling councils during the transitional period, settling disputes through decentralized debate, the lack of strong central government, diffused territorial sovereignty, control over force, and the like.
This all sounds rather different than the usual thoughts on the organization
and operation of a civilization, but then this one failed; the Harappan civilization was a fleeting moment of Bronze Age urbanization. As such, the
proposal that the founding principles of its organization were "different,"
flawed from the standpoint of an ecological or adaptive perspective, makes
perfectly good sense.
CONCLUDING REMARKS
It is not possible in a short essay to develop adequately a number of important themes characterizing the Indus civilization as an example of ancient
sociocultural complexity without the archaic state form of political organization. Nor has the case been proved, in the sense oflogical finality, leaving
much to be done on this issue. But one thing is clear. The Harappan civilization has been the subject of an immense amount of thought, most of
which has been couched in conventional comparative terms. One of the
most important themes has been an assumption that it was a state, which has
led to a host of comparative assumptions and no small amount of interpretation of this civilization in terms of its contemporary neighbors to the west,
in Mesopotamia. There are many examples of this practice, some not to be
taken seriously (Pran Nath 1931, 1932; Sankarananda 1962; Waddell 1925).
But there also are serious, even eminent, scholars who have failed to understand the Harappans on their own terms and relied too much on what we
know, or think we know, of ancient Mesopotamian life. One of these is Sir
Mortimer Wheeler:
.If such a line of thought remains the prevailing mode to understand the
ancient peoples of the Indus, is there any wonder that so little is known for
sure? With so much evidence against it, is there any reason to continue to
see the Harappan political system as a copy of the Mesopotamian archaic
stat~ and, in so doing, miss one of the truly important aspects of the ancient
Indian experiment with urbanization?
. What we have learned from the new studies of the Harappan civilization
IS t~lat ancient civilizations, or complex societies, are far more variable in
their form and organization than the typological schemes of traditional unilineal evolution can accommodate. Concisely stated, well-conceived definitions have an important role to play in archaeology, along with typological
schemes. For the ~ost part these tools can assist us in reaching our ends, but
they are not ends In themselves. Whether we call the Indus civilization an
arc?ai~ .sta~e or something else is of less importance than recognizing the
vanability Inherent in the experiment with the level of sociocultural complexity that it represents. Much work is left to be done in terms of describing an~ defining this diversity of complexity. But we already know
some~hing.about the diversity; the time has come to phrase our analytical
questions In a way that tries to understand it. Why, for example, did
MesopotaIlll~ and. the Indus take such different courses to civilization? Why
are these sociocultural systems so different in so many ways and yet seem to
~hare a level of sociocultural integration that brings them together on other
Important terms? Such a research agenda is entirely different from the one
that is n~w in place. Its implementation would represent a significant step
forward In the study of Harappan civilization as well as sociocultural complexity in general.
.......
QUITO
i-
1._.,,_.~J
..,:
"l
-v
'.
Inkal1akta
t
l00nilles
_...... - .....
'"
PACIFIC
OCEAN
BOLIVIA
"
"
\
CHILE':
Figure 9.1. Map of the Inka empire in western South America, showing
principal sites and places mentioned in the text.
undergoing can therefore be instructive in understanding sociopolitical
growth in the Andes in general, including those processes that centuries before might have led to pristine Andean states. Indeed, if we could trace the
entire prehistory of the Inka as they evolved from a small regional polity
(probably at the level that some would call a maaoetnla and others a chiefdom), we would be able to see the relatively uncontaminated birth of a state,
though obviously not a pristine state (Rostworowski 1988).
Reliable data on the initial stages of Inka prehistory are still virtually
nonexistent. My strategy for evaluating the Inka expansion is to look briefly
at a few of the better understood examples ofthe regions and kingdoms they
incorporated. In these cases we can see markedly different strategies of incorporation and governance. These differing strategies were shaped mainly
by the different circumstances of the local regions at the time they were
added to Tawantinsuyu, but they also probably reflect that the annexed regions were added at different phases in the process of Inka political development. The results of this examination reveal, I believe, a need to shift the
focus somewhat from military coercion and administrative bureaucracy to
an increasing stress on the mechanisms whereby new levels of sociocultural
identity, political collaboration, and economic growth were attained. Both
military force and a very complicated administrative bureaucracy were indeed involved, but many analysts have tended to view these in terms of a
European model (D'Altroy 1992; Patterson 1991). Although the Inka may
have taken some steps toward such a model in their final decades, at the time
of the Spanish invasion they can still be best understood in terms of some of
the principles of traditional, small-scale societies-in spite of the vast size
and enormous complexity that had been achieved.
Three features or tactics characterized the Inka's rapid march to empire:
first, an ideology of rule communicated and instilled through a set of elaborate ceremonial activities and visual symbols; second, the provision of notable economic rewards; and only third (and probably late in the process),
the use of coercive military force. These features or tactics were used in
varying degrees and with great flexibility, depending on the local circumstances the Inka encountered.
part on the degree to which the Cuzco rulers were ~ble to .achieve a degree
of peaceful collaboration or were forced to deal with resistance and confrontation. One strategy involved the utilization and enhancement of the
extant structure of authority. This classic strategy of indirect rule through
the existing power structure appears to have been followed only rarely by
the Inka, in large part because few really large centralized states existed in the
Andes at the time of the Inka expansion. In part it also was the result of
strong resistance to Inka rule by some large kingdoms, such as t?e Chimu
on the Peruvian north coast. A prominent case of the peaceful mcorporation of a centralized polity was the Chincha kingdom on the south-central
coast of Peru. There, our current research is beginning to provide a picture
ofhow the Inka inserted themselves into an important state and manipulated
its internal workings.
Maria Rostworowski's (1970) studies of the written sources on Chincha
suggest a gradual incorporation in which Chincha was permitted to retain
its rich trade-apparently including metals, textiles and, especially, Spon~y
Ius shell-by sea with the coast of what is now Ecuador and overland with
the altiplano regions in the south. Incorporation was i~tiated by the I~ka
with the exchange of gifts and women. As the relationshIp proceeded, Chincha gave lands that were cultivated for the Inka, and women for making
cloth and beer, two of the state's most essential products. They also agreed
to an Inka request to construct a palace in the Chincha capital, now the archaeological site of La Centinela (Morris 1988b).
La Centinela is a group of massive compounds that were the focus of
public activities of the groups, perhaps based on economic specialization and
descent that constituted the Chincha kingdom. In their midst, the Inka
palace compound was constructed of adobe bricks, in contrast to the earlier
Chincha construction of tapia (walls of pressed earth). The Inka compound
used the trapezoidal niches and doorways characteristic ofInka architecture.
But alongside the buildings that appear to have served as a typical Inka palace
was a truncated pyramidal platform reminiscent of Chincha architecture. It
was constructed of adobe bricks and was a formal part of the Inka compound. Yet its shape almost seems to mimic the central features of the Chincha compounds. It is almost as if the Inka had recognized and legitimized
local patterns within the compound from which people closely identified
with Cuzco-controlled Chincha.
As revealing of Inka tactics for controlling the Chincha capital as their
own administrative compound were the modifications-at the same time
subtle and powerful-they carried out in the existing Chincha architecture.
Several of the tapia compounds built in Chincha times were altered by the
Inka. Adobes were added to tapia construction; niches were incorporated
i~to s~me. public spaces; traffic flow of at least one compound was altered to
~~ It with the Inka compound. However, the evidence suggests that actrvities were not drastically ~odified by ~he Inka overlords. They changed
t?e fac,:a~e but not the functions. The tapia compounds with their pyrarnid-
like architecture were clearly public and ceremonial in nature, but they seem
to have been related to relatively small and probably fairly tightly structured
~oups. It ~~y be s?~culated that these compounds were the backdrops for
ntes of political legitimation and social reproduction, such as initiation and
marriage, of the groups that composed the Chincha kingdom. If this interp~e~ation is correct, the Inka essentially used the compounds to promote indlVI~U~S o.f their choosing within the local Chincha hierarchy. Their
administration was based on selecting and manipulating administrative per~onnel while leaving the existing structure of positions and statuses basically
mtact.
by Garci Diez de San Miguel shows part of the region of Lake Titicaca to
have been the relatively centralized Lupaca kingdom under the dual leadership of Cari and Cusi. Based on the written source alone, it is reasonable to
conclude that the Lupaca was a pre-Inka entity (Murra 1964), not unlike
Chincha in scale and centralization. But my reading of the still limited archaeological evidence leads me to suspect that the Lupaca kingdom may at
least in part have been the result of successful Inka efforts at local-level centralization. Hyslop's (1977: 170) emphasis on the great elaboration of burial
towers (chulpas) for the local leaders during Inka times is a case in point. According to his analysis, the power of certain cooperative local leaders was enhanced as a system of indirect rule was put into place. In the Yanamarca
region of the central highlands, the Wanka also seem to have been moving
toward some degree of organizational complexity, though not necessarily
loyalty to the Inka rulers. The peoples of that region promptly allied themselves with the Spaniards in the aftermath of Cajamarca (D'Altroy 1992:
47, 186).
In the three regions mentioned, the Inka had built major provincial administrative centers: the Yanamarca area was under the settlement of Hatun
Xauxa; the Lupaca part of the Lake Titicaca area was under Chucuito; and
the upper Huallaga Valley was under Huanuco Pampa. One of our best avenues to understanding Inka rule is to examine the functioning of these
provincial capitals. However, both Xauxa and Chucuito are now largely
covered by modern towns; only Huanuco Pampa has received extensive archaeological study.
HUANUCO PAMPA
Huanuco Pampa probably had a relatively small permanent population; its
nearly 4,000 structures served instead to house many thousands of people
who came periodically for state festivals and to work for the state as part of
their mita labor tax. The buildings also provided various kinds of public
space, housed production of certain arts and crafts, and served as warehousing for the supplies that made the other activities possible.
Our research in the 1970s (Morris 1972, 1982; Morris and Thompson
1985) has shown that the great central plaza and the buildings around it witnessed public ceremonies, probably including the festivals of the elaborate
Inka ritual calendar and possibly also ritual battles, such as those known for
Cuzco (Cobo 1990[1653]: 135, 215, 245), which helped define relations
among groups in the region. Royal dwellings and temples formed a compound at the city's eastern edge. The thousands of corn beer jars and plates
found in buildings around smaller plazas between the royal compound and
the central plaza suggest that public feasts were prominent in these areas.
More than 500 warehouses held food and other supplies. A compound of
50 houses provided work and living space to a group of weavers and brewers who turned out the cloth and maize beer that were such fundamental
parts of the Inka's gifts to their subjects, on the one hand, and their hospitality obligations on the other (Morris 1974, 1982; Murra 1962a, 1980'
Rostworowski 1988). Most of the activities identified archaeologicall;
thr~ugh excavations at Huanuco Pampa fit with the picture of a regional
capital where the Inka could fulfill obligations of asymmetrical reciprocity.
It was a provincial administrative center, but one that fit into a system of
reciprocal administration rather than being based on bureaucratic and military mechanisms. These patterns of royal hospitality based in asymmetrical
reciprocity functioned through a structure of dual divisions that offered the
building blocks of a hierarchical political and economic organization.
A second set of evidence from Huanuco Pampa that is pertinent to Inka
political processes is somewhat more abstract. It comes from an analysis of
architecture and the way space was organized in the administrative center's
architectural plan (c. Morris 1987). Huanuco Pampa is divided into two
main divisions by the Cuzco-Quito road that runs through it from northeast to southwest. One of these divisions contains more elaborate and finely
made buildings and can probably be considered Hanan, or the "upper" half,
in Inka principles of organization as expressed at Cuzco. The two main divisions are clearly subdivided, creating a quadripartite pattern around the
large rectangular plaza (fig. 9.2). The four divisions may have been further
subdivided into three divisions each, creating twelve zones.
It seems likely that the city plan of Huanuco Pampa reflected basic Inka
organizational principles that permeate several aspects of Inka ideology and
practice. The best documented expression of these principles is in the organization of the shrines around Cuzco (Cobo 1990[1653]:47-84; Rowe
1979; Zuidema 1964). There is no written evidence as to the meaning of
the complex set of divisions at Huanuco Pampa; however, the similarity
to the patterning of the Cuzco shrines and the way they were worshiped
seems too close to be coincidental. Similar principles of division and combination pervade other aspects ofInka organization and ideology, such as the
calendar.
The elaborate set of divisions expressed in the plan is probably related to
Huanuco Pampa's role as an administrative center. For Inka rule to function
effectively, it would have been more efficient to establish a hierarchical order among the small local groups comprising that region, so that the state
would have to deal with fewer local leaders. It is intriguing to speculate that
the city plan might have been used not just as an abstract expression of
important ideological principles, but actually to place groups of people in a
physical relationship to each other that mirrored the hierarchical relationships
the Inka state wanted them to assume. These positions could then have been
reinforced (and occasionally rearranged) through participation in the ceremonies prescribed in the ritual calendar. Seen in this way, the city plan becomes a diagram of sociopolitical ideology and a template for the creation
of actual social, political, and economic organization. It is a way of defining,
maintaining, and, if the need arises, sometimes restructuring the social segments that are the building blocks of state society. Relationships among and
between the various divisions were probably mediated and manipulated
through reciprocity; in this way the various small polities could be converted
into a coherent hierarchical local organization by creating new levels of
sociopolitical scale that the Inka manipulated.
From the Huanuco Pampa example we can begin to formulate an idea
of an Inka strategy for converting groups of relatively noncentralized polities into small kingdomlike units that they could administer through indirect rule-by manipulating personnel within a hierarchical and centralized
system, similar to the way they controlled Chincha. Even though we are still
far from confirming the identities or intergroup activities of the peoples
who used the various sectors ofHuaimco Pampa, it is reasonable to suggest
that public architecture was used as a means of encoding and teaching an
overall political structure. In a society that lacked writing, architecture and
other media carried an especially heavy burden in facilitating communication and assisting the inculcation of new beliefs and behaviors.
POLITICAL GROWTH PROCESSES
IN THE INKA STATE
I mentioned at the outset that the fundamental processes in the creation of
Tawantinsuyu were political. They involved the inculcation of a complex
ideology of rule instilled through elaborate ritual, as outlined for Huanuco
Pampa, and communicated with visual symbols. They also were economic,
involving the provision of important sumptuary and subsistence goods
through the redistributive economy. These two processes were intertwined
by the important social and ideological symbols carried by the sumptuary
goods that the emerging complex economy produced and utilized in highly
visible and ritually important ways (see Morris 1991 for fuller presentation).
These goods-along with the symbols encoded in architecture and site
planning and the khipu-formed a technology of communication that enabled a highly differentiated political and administrative structure to function, even without writing.
The examination above of cases of the Inka incorporation of hinterland
regions suggests some of the ways the economic and political processes
worked in actual strategies of annexation and governance. The flexible
nature of Inka rule put great emphasis on the adjustment of state policy to
the details of local realities. But overlaying this flexibility was an insistence
on the actual physical presence of governmental installations and personnel
in the hinterlands. This took the form of the well-documented system of
state installations linked by roads (e.g., Hyslop 1984; Morris 1972, 1981).
The ceremonies and the symbolic paraphernalia of the central authority
were carried to the outlying "capitals" of the state in places like Huanuco
Pampa, La Centinela, and dozens ofother administrative centers. In this way
reciprocal administration could be dispersed and differentiated but would
still function almost as if the highest level of the Inka elite (perhaps even the
ruler himself) was everywhere present. This same system redistributed the
key sumptuary and subsistence goods, thereby conferring prestige, assigning
roles and positions, building support for state authority, and generating state
revenues. Huanuco Pampa and La Centinela, each in its very different way,
fulfilled these primary ceremonial and economic functions that were the
foundation of the state.
This brief examination of Inka governing processes allows a somewhat
different perspective on some of the features that we have often thought to
be essential defining characteristics of state societies. The Inka certainly possessed most of these characteristics, including control over a very extensive
territory and a well-developed structure of centralized authority. While the
centralized rulership was not strictly hereditary, there were well-established
mechanisms for selecting the new ruler from among a limited number ofpotential successors in the event of the death or deposing of the sitting ruler.
Rulership was an institutionalized and permanent position.
There was also a well-established ruling elite, even though this elite
group was probably viewed more in ethnic terms (i.e., Inka vs. the various
non-Inka groups) than in terms of social class in a strict sense. Many of the
members of this elite were organized into an impressive and apparently
highly effective administrative bureaucracy. There were certainly record
keepers (khipu kamayuq) and other administrative officials, as we would expect in any bureaucracy. But just as much of governance was based in an administration of reciprocity and ceremony, the bureaucracy, too, was a
bureaucracy of reciprocity that managed the feasts of an elaborate ritual calendar and saw to it that the goods and people were in the right place at the
right time throughout a vast administrative network. The more we begin to
grasp the details of what was accomplished, the more impossible it seems
given the communication technology then available.
The Inka also managed the economy to an unusual degree. This mainly
consisted of the large-scale mobilization of labor to produce goods to support the re distributive economy and the elite. As we have seen, the functions of goods produced in this redistributive economy went well beyond
the economic, being critical to communication as well. The vast labor resources were also used for enormous capital projects. On the one hand, they
built the elaborate system of state cities, way stations, and roads that constituted the infrastructure of administration; on the other, they were used for
investments in irrigation and terracing systems to increase production and
for warehousing systems to even out seasonal and annual variations in the
availability of goods.
A complex hierarchy of settlements, taken by archaeologists to be essential material indicators of state-level organization, is easy to discern in the
Inka case: there is a hierarchy of administrative centers within the installations built by the state (Guaman Poma 1980[1613]: 1000-1007; Hyslop
1990; Morris 1993). In addition, there is the superimposition of Inka over
non-Inka settlements (Morris 1972); in some of the cases where preexisting
complex societies were incorporated by the Inka into Tawantinsuyu, the
pre-Inka settlement hierarchies may have continued to function. This seems
to have been the case in Chincha. The settlement system was thus extremely
complex, both in terms of hierarchy and in terms of horizontal functional
specificity.
From all this evidence there is no doubt that Tawantinsuyu was one of
the most vast and important of ancient states. There are, nevertheless, some
incongruities that our picture of the Inka raises for our traditional conception of ancient states. I briefly emphasize two. The first is that principles of
economic reciprocity remained central to the nature of authority and to the
functioning and growth of the state, even at its impressive scale. This emphasis on bringing the ruler and the elite, with all the attendant hospitality
and generosity, into direct contact with large numbers of people mimics simpler levels of political organization (Morris 1982; Murra 1960: 400;
Rostworowski 1988). The second incongruity has to do with the kinds of
conflict and warfare that were part of the political process. The topic ofInka
warfare is far more complex than the studies of it have implied. The comments below raise questions that will hopefully encourage more research
aimed at producing new information and a greater understanding of the
Issue.
While there is no question that elements of force were involved in Inka
rule (Murra 1986), I suspect that we have tended to exaggerate its coercive
role (D' Altroy 1992; Hyslop 1990; Morris 1982). Much ofthis exaggeration
derives from the early Spanish sources, which tended to assume that the Inka
expansion had been based on imperial conquests not unlike the one in
which they were engaged (see Webster, this volume, for an alternative perspective). They also assumed that most of the elaborate royal retinue of the
Inka were soldiers. Most important, they did not grasp the distinction between ritual battles and battles of coercion. Although ritual battles were
planned events, usually tied to the ceremonial calendar, they usually produced injuries, and even fatalities. Hartmann (1972) has noted their possible
ritual role; blood was spilled, in a kind of substitute for sacrifice, to ensure
the fertility of the earth. Victories in ritual battle conferred prestige (Gorbak, Lischetti, and Mufioz 1962; Hartmann 1972) and helped establish relationships ofdominance and subjugation resulting in a political hierarchy, but
they were not intended to inflict real military defeat leading to coercive control supported by a continuing military presence.
My evidence for altering the role we have traditionally assigned to warfare is almost entirely archaeological. It is, in addition, heavily influenced by
personal research. In the 3,000 or so provenience units excavated in imperial Inka contexts at several sites by projects I have directed, such as those
mentioned in the cases summarized above, we have recovered fewer than
50 artifacts that can be classified as weapons. Furthermore, all of these could
just as well be attributed to ritual warfare as to coercive warfare. I emphasize that the sites we have studied have been basically administrative in nature, and none was located in a defensive position. Nevertheless, we fully
expected to find military remains at Huanuco Pampa; the first excavations
we conducted there were in an area I suspected to be military barracks
(Morris and Thompson 1985:70). In the past I have offered elaborate rationalizations for the rarity of weapons in the city (Morris 1982: 158, 160);
many ofthem were made ofperishable materials (Rowe 1946: 27 4 -78), and
they were probably not the kind of artifacts abandoned, except on the
battlefield (Morris 1982: 160). I also argued that military activities were
probably focused on the frontiers, and that they would have been reduced
at sites like Huanuco Pampa as such sites became increasingly devoted to administrative matters as the frontier moved far away. Some of the same factors may have accounted for the absence of weapons in the warehouses
(Morris 1967), yet some Spanish writers insist that weapons storage was one
of the functions of the warehousing system (Cieza 1959[1553]).
The excavated archaeological evidence for Inka weapons elsewhere is not
much greater. In his assessment of Inka military sites, John Hyslop (1990:
154) noted: "The identification of military settlements is compounded by a
curious lack of evidence of arms in the few sites that have been excavated or
where extensive surface collections have been made." But he went on to observe that no sites thought to be of a primarily military function have been
thoroughly excavated. Small excavations in the Quitaloma section of the
enormous walled hilltop site ofPambamarca, about 30 km from Quito, have
yielded sling stones (Oberem 1968: 338 - 42). Pambamarca is one of a number of heavily fortified sites near the part of the Inka frontier that was still
expanding in 1532.
Most of the archaeological attribution of military functions is based on
because of the obligations of reciprocal administration and the need to produce symbols ofprestige and markers of social position and status. The state
was the primary beneficiary of overall increases in production, but the economy as a whole probably benefited from the increases as well as from improved redistribution. The motivations for the workers' contributions, thus,
came in part from economic benefits, in part from increased prestige, and
probably in smaller measures from threats of governmental or supernatural
punishment. The rapidity of these changes perhaps resulted from the society's sudden capacity to move to a new level and scope of resource management in dealing with the multiple ecologies that formed its environment.
The mechanisms of Inka rule were based in the redistribution of economic goods, of power and prestige, and of the potential of supernatural favors with their incumbent feelings of security. They were backed up by the
threat, implied and real, of both physical and supernatural punishment.
These mechanisms of Inka expansion and rule were based in strategies and
innovations that had begun centuries earlier. The underlying economic and
adaptive benefits of a coordinating central authority that built an umbrella
over extreme diversity also had been known for centuries and probably
made people, whether consciously or not, more willing to accept a central
authority that was at times distant and benign, and at times direct and even
oppressive (see also Mareus 1987c). The advantages of access to a wider variety and greater quantity of goods outweighed alienation of certain lands
and the displacement oflarge amounts oflabor from the community economy to maintenance of an elite and a bureaucracy, as well as the substantial
program of investment in infrastructure and in capital improvements.
Research on the Inka state is still in its infancy. Recent work has shown
that the situation was extremely complex and that real understanding can
come only from detailed information. It also has shown that our two primary sources of evidence are not always in agreement. The written record
and the archaeological record can give quite different perspectives; interpretation based only on written sources can lead to overly simple, Europeanized
models. A fuller body of archaeological information is urgently needed to
give a more complete and reliable picture of how the Inka state grew and
governed. With its juxtaposition of characteristics associated with simpler
societies and those associated with more modem ones, and the concomitant
suggestion that the various subsystems of archaic polities may evolve at different rates, the Inka case can help move us toward a better general understanding of archaic states and their growth.
10
Worfnre ond Stotus Rivolry
lowland Moyo lInll PolYllesian (omparisolls
DAYID WIBmR
arfare was associated with the origins of virtually all preindustrial civilizations. My original assignment for the School ofAmerican Research
seminar was to make a detailed comparison between Maya warfare and
that of one other ancient archaic state. This approach turned out to be unsatisfactory because, despite its ubiquity, early warfare is poorly understood.
Fortunately, the "archaic state" concept also logically includes more recent
and better-documented examples of state formation, and another opportunity suggested itself while I was reading about Polynesia. I found the Polynesian data opened up stimulating comparative perspectives, and since I
recently lamented the lack of comparative studies of Maya war (Webster
1993), it seemed only fair that I should undertake one.
Of particular comparative significance are patterns of status-rivalry warfare that both Marcus (1993a) and I (Webster 1993) have inferred for the
Maya (see also Houston 1993: 130). Status-rivalry war is accordingly the
subject of the last and most important part of this essay. I emphasize it in an
attempt to show that Mayanists do not have to envision a dichotomy, as
much of the literature seems to suggest, between ritual conflict on the one
hand and conflict motivated by population pressure on the other. Rather,
ritual war and territorial aggrandizement complement one another in a syndrome ofcompetition that was widespread in many cultures, including those
of Polynesia.
Polynesia may seem an inappropriate choice. The knee-jerk anthropological reaction envisions a region characterized by chiefdoms or rank
societies when first encountered by Europeans. Indeed, our classical conceptions ofthese societal "types" derive strongly from Polynesia. Behind the
widely shared facade of ranking, however, was an enormous range of sociopolitical complexity, from the essentially egalitarian Maoris of New Zealand
to the stratified systems of Tonga, Tahiti, and Hawai'i.! The most complex
Polynesian societies have, in fact, been characterized as or compared with
"archaic" or "incipient" states or civilizations by many anthropologists (e.g.,
Bellwood 1979; Cordy 1981; Goldman 1970; Hommon 1986; Oliver 1974;
Suggs 1960). Others, such as Earle (1978) and Kirch (1984), retain the term
"chiefdom" while describing sociopolitical patterns that are in many respects
"statelike." General comparative treatments (e.g., Claessen and Skalnik, eds.
1978) of early state formation commonly include complex Polynesian societies such as Tahiti (Claessen 1978c) and Hawai'i (Seaton 1978), emphasizing their "stratified" organization.
Such confusion of terminology itself is revealing, since it signals the existence of complex sociopolitical forms for which our comparative or evolutionary terminology is inadequate. No issue discussed during the seminar
generated more heated and diverse opinion than the utility of such labels and
their associated evolutionary models. Whatever our personal views on this
controversy, I think we would all agree that the concept of archaic state, the
theme of the seminar, includes the evolutionary processes by which the basic institutions and attendant "great traditions" of the earliest states emerged
from preexisting forms. This very transformational dimension is what makes
the concept so interesting.
One fruitful perspective is that of Fried (1967), who envisioned the state
as a set of institutions that uphold an order of stratification. Incipient states,
and probably some archaic states, as I understand these terms, should exhibit
some forms of stratification. But the institutions and great traditions that
support these systems of stratification may be only incompletely evolved.
The most complex Polynesian societies certainly accord with this definition.
Gailey's (1987) very detailed analysis of Tonga, for example, concludes that
processes of class formation began long before European contact, although
kinship modes of organization were not replaced by mature state institutions.
My intention is not to suggest a Polynesian model for the Maya, nor that
the Maya were in detail "like" any particular Polynesian society. Emphasis is
instead on Polynesian warfare in its wider sociocultural contexts and the insights that it can provide about Maya war in its own distinctive sociocultural
setting. In particular, I use Polynesian data to discuss three hotly debated issues: (1) how were Maya wars conducted? (2) who fought Maya wars? and
(3) what were the purposes and functions of Maya war? Of these, the last is
most important and relates to the subject of status-rivalry war. As we shall
see, Polynesia offers Mayanists not only interesting insights about these issues, but also some cautionary tales as well.
War was a conspicuous feature of Polynesian society: "Warfare-ranging from simple raiding aimed at restoring the honor of the insulted party
to major wars of territorial or inter-island conquest-was ubiquitous in
Polynesia" (Kirch 1984: 195). Evidence concerning war is abundant, since it
was featured in Polynesian oral histories, observed and participated in by
Europeans, and represented in the archaeological record. Although only occasionally treated as a specialized topic in the Polynesian literature, references to war pervade more general accounts. Perhaps for no other region are
conflict and competition so consistently emphasized as essential for understanding how societies were structured and how they changed. As we shall
see shortly, one of the more elegant and sensible general anthropological
models of warfare is partly based on Polynesian data.
Although I have been erroneously characterized as an advocate of warfare as a "prime mover" in cultural evolution (Sharer and Ashmore 1987:
539), I have in fact taken pains to emphasize war as only one variable in much
more complex evolutionary processes (Webster 1975). The following discussion reflects this perspective, which also is consistent with general analyses of Polynesian war by scholars such as Earle (1978) and Kirch (1984).
Most of my comparisons involve complex Polynesian societies, but important lessons can be learned from simpler ones as well.
WARFARE
Although warfare is much discussed by anthropologists, it is difficult to
define because of the many forms that conflict can take. Hassig (1992), for
example, offers no general discussion of the concept in his recent synthesis
of Mesoamerican warfare (which includes the most up-to-date and exhaustive compilation ofsource material on Maya war). Below is a brief overview,
partly abstracted from another publication of my own (Webster 1993).
Definitions of War
Following Otterbein (1973) and Berndt (1964), warfare consists of planned
confrontations between groups of people who conceive of themselves as
members of separate political communities. Political communities are composed of individuals who possess and recognize a high level of common interest, have the organizational capacity to cooperate effectively, and are
prepared to defend or augment their collective interests through the use of
violence. Such groups are often, but not always, territorially defined. Complex polities may consist of multiple, differently constituted political communities that might be better characterized as factions. Members offactions
may not share equally in the conduct, risks, or rewards of war.
Warfare is organized and sanctioned group violence that involves armed
conflict, including confrontations that combatants recognize may result in
deliberate killing. Conflicts are organized and carried out by at least one of
the factions with the intent of maintaining the status quo or bringing about
a shift of power relations, usually the latter.
Finally, Otterbein (1968) makes the distinction between internal war,
which consists of conflicts between culturally similar groups, and external
war between culturally dissimilar groups. This distinction is important because some treatments of Maya warfare (Demarest 1978; Freidel 1986) assert that Maya internal war was usually limited in its political, social, and
economic consequences because of shared ethics or charters of permissible
behavior.
War as a Process
Vayda (1976), partly basing his ideas on Maori warfare, has avoided the strict
definitional problem by emphasizing war as a long-term, graded process
with a series of distinct but related phases. The simplest phases might be
confrontations between individuals (who inevitably represent factions),
which then escalate into larger-scale conflicts, eventually resulting in all-out,
deadly competition between political communities. It is these most intense
phases that we generally recognize as warfare, but Vayda's point is that this
process of escalation is not inevitable. Each phase generates information in
cultural systems concerning the efficacy ofescalation as opposed to other behaviors. Another implication is that some phases of warfare may not be
highly organized and planned by combatants, or even highly lethal. Confusion results if we focus too specifically on one particular phase of the warfare process-something we are very prone to do since some phases are
much more visible in the archaeological record than others.
These definitions ofwarfare cover a wide range ofbehavior that includes
individual acts of mayhem, small groups independently raiding a neighboring community, and full-scale engagements between large, highly organized
armies. They also incorporate intrapolity conflicts, such as struggles over
succession or civil wars. Finally, they include acts of violent treachery with
overtly political ends.
Analyzing Warfare
Any effective analysis of warfare must take into account the following factors, which partly serve to structure my comparative discussion:
the lines. There is the danger ofallowing a single triumphant event to colour
our perception of an entire disastrous, or essentially peaceful, reign. The
situation is made worse because there is little in the way of compensating
documentation of an archival kind: the obsessive bureaucracy of the Mesopotamian palaces did not up sticks and tag along on campaign; and we don't
have archives of any military quarter-master. Hence we are unable to gauge
the extent of the communal investment in offensive and military measures,
or the impact oforganized campaigns or civil disorder on the lives ofthe bulk
of the population.
Much of this dismal prognosis derives not from our inability to investigate war using archaeological methods, but from the fact that these methods
have not been effectively applied on the necessary scale. On the level of culture history or particular events, such application still has much to tell us. To
continue the Mesopotamian example, whether or not the city ofUruk was
walled before Early Dynastic times is surely a resolvable issue. So, too, are
the effects of war on a settlement system, although this issue presents much
greater practical problems. Diachronic processes and functions of war are
much more difficult to understand, particularly through the lens of what
passes for history in most archaic states. Postgate's apt comments have general relevance and should be borne in mind as we consider the Maya and
Polynesian records.
Lowland Maya Archaic State Warfare
The Lowland Maya, paradoxically, offer significant insights into the role of
warfare in the comparative process of archaic state formation. I say paradoxically because most Mayanists have emphasized the unique aspects of
Maya culture at the expense of comparison. Also, until very recently, the
Classic Maya were envisioned as peaceful theocrats who lacked not only
warfare but also the general political institutions characteristic of early states.
They have accordingly often been omitted from comprehensive surveys of
early state formation (e.g., Claessen and Skalnik, eds. 1978), though almost
invariably being included in comparative discussions of the decline of civilizations (Tainter 1988; Yoffee and Cowgill 1988). Particularly prescient
among those who recognized early on the importance ofwarfare among the
Maya was Robert Rands (1952), although his work has not been widely
cited or appreciated.
Our perception of Maya warfare has been revolutionized during the
past 20 years by the presence of historical information preserved on Classic
(and sometimes Postclassic) monuments, along with associated iconographic
depictions (fig. 10.1). It is largely due to our increased understanding of inscriptions that many Mayanists have abandoned their long-cherished notions of the "peaceful Maya" and swung around to an almost polar opinion
Figure 10.1. This lintel from the Classic Maya center cif Yaxchilan shows
the capture of enemies by the ruler BirdJaguar (right) and an associated elite
warrior (drawing byJohn Klausmeyer, courtesy cifJoyce Marcus).
that emphasizes the Maya as compulsively warlike (in my opinion an equally
unlikely idea). Inscriptions have revealed the dates, participants, and at least
some of the claims of victory resulting from many military confrontations.
These typically are found at the centers ofputative victors, but in at least one
case a monument of the victor (Caracol) was set up at the defeated site
(Naranjo).
Most current debates among Mayanists about the nature of Maya society in general, and warfare in particular, stem from different interpretations
ofinscriptions. These interpretations in turn often reflect divergent opinions
about the degree to which inscriptions represent reliable historical accounts,
myth, or self-serving elite propaganda (e.g., Coe 1992; Marcus 1992c).
What is clear is that the epigraphic record is extremely spotty in time and
space; limited in content, given current knowledge, to essentially elite/royal
concerns; and still imperfectly understood. This is especially true for warfare
(D. Stuart 1993). On the one hand, I agree with Norman Hammond that
Maya inscriptions have "brought the Maya from the margins of prehistory
into merely liminal history"(1991 :256). On the other, the enormous importance of emic epigraphic data is apparent.
To date (interpretational difficulties aside), the epigraphic record has primarily provided culture-historical information about Maya war (see Stuart
1995 for the most useful and up-to-date overview). As archaeologists have
long insisted, culture history is not explanation, but rather what needs to be
explained. Combining epigraphic data and robust archaeological approaches
is the effective road to explanation (Webster 1989: 1-4).
My discussion of Maya warfare focuses heavily on the Late Classic, a period (AD 600 -900) for which dates, inscriptions, and art are most abundant,
but clearly such a view distorts evolutionary perspectives. The Classic period was preceded by at least 1,000 years of complex sociocultural development during which there is evidence for war. There are Preclassic and Early
Classic fortifications, and the iconography ofthe latter period (AD 250 - 600)
is "replete with martial imagery" (Houston 1993: 138). Unfortunately, associated inscriptions referring to warfare are virtually absent. Only after
about AD 650 did such inscriptions become numerous, especially proliferating in the eighth century (Marcus 1976a; D. Stuart 1993, 1995). After
AD 800, dated inscriptions fell off markedly but warfare continued. Our effective historical window on Classic Maya war is thus a scant 150 years long.
Archaeology apart, our understanding of Post classic Maya war (AD 9001520) derives mainly from a few native books written shortly before or just
after the arrival of Europeans, from the accounts of Spanish explorers and
soldiers (Cortes 1986; Diaz 1963), and from reconstructions ofpreconquest
Yucatan (e.g., Landa 1941; Roys 1943, 1957). In many respects the Maya
Lowlands and Polynesia present us with mirror-image data sets concerning
warfare-heavily archaeological and epigraphic for the Maya, and heavily
ethnohistoric for Polynesia.
1988), probably began with fairly organized formations and then after an exchange of missiles degenerated quickly into general melees, including single
combat between prestigious warriors. Both raids and pitched battles could
include ambushes, feints, false retreats, and other stratagems to confuse and
disorganize the enemy.
Maya rulers fought personally in some battles, obviously supported by
lesser warrior elites, but there is little unambiguous evidence of battlefield
organization or command. Notable warriors (holcans) were important in
Postclassic conflicts, but there is no strong evidence for the existence of elite
military orders such as are found among the Aztec. My own guess is that
most Maya war parties had minimal organizational structure and that a ruler,
a few elite captains, and a stiffening of experienced warriors sufficed for directing combat, especially once battle was joined.
The same basic range of conflicts is found in Polynesia, although here
again the relationships among the various phases is generally not clear.
Pitched battles were conducted in much the same fashion as among the
Maya. Maori chiefs typically participated personally in combat, leading by
exhortation and example but not giving orders (Vayda 1960). In more complex Polynesian polities the highest-ranking leaders were present at battles
but sometimes remained aloof Special Tahitian "battle shapers" and "exhorters," often attired in gorgeously cumbersome military costumes reminiscent of those worn by Maya elites, helped direct battles but apparently
with more psychological than organizational effect (Oliver 1974:390-95).
Apart from special bodies of slingers, the only other distinctive participants
were the aito, described by Oliver as especially prominent and experienced
warriors; similar high-status warriors existed elsewhere as well, such as the
toa of the Marquesas (Kirch 1991: 126).
Polynesian war canoes seem to have been used most often to disembark
warriors on enemy beaches, where conflicts then occurred. According to
some accounts, naval battles required considerable discipline and concerted
maneuvering. None was ever witnessed by Europeans, although Captain
Cook observed an enormous muster of Tahitian warcraft and associated
mock engagements.
FORTIFICATIONS. One form of conflict-sieges-Ieads us to the subject of asedio
I 325
200 m
Figure to. 3. Plan of the Becan fortifications; the inner parapet of earth is
not shown (drawing courtesy ofJoyce Marcus).
zanjas, terraplenes
t---1-------~
metres
approx
Figure 10.5. Reconstruction of earthworks, palisades, and fighting platforms associated with a Maori pa (from Fox 1976: 14). Reprinted with
permission of Penguin Books (NZ) Ltd.
populations. Chiefs might have their houses permanently within the pa, but
the most important nonmilitary constructions were storage facilities for
staple foodstuffs of the community at large. Well-established Maori pas often had multiple lines of defense, tricky entrances, and elevated fighting
stages that gave advantage to the defenders' missiles. Apart from pas themselves, outlying scattered residences were protected by light stockades
(Vayda 1960: 10-11) that would probably be archaeologically invisible.
Once effectively manned, pas were difficult for enemies to take by
storm, since attacking war parties generally had no significant numerical advantage over the defenders and usually could not even invade the defensive
perimeters adequately. This is one reason why Maori warfare depended so
much on surprise raids; major damage could only be done if victims were
caught unprepared outside the pas, or the pas themselves were seized
quickly. Although sieges sometimes occurred, defenders had the advantage
ofstored food supplies. Only when attackers could live off the land (warfare
tended to be scheduled during seasons when foraging was possible), or when
undefended stores could be seized, were prolonged sieges possible. A related
factor was the vulnerability of the attackers' own communities if warriors
were absent for too long.
Maori forts have several implications for Maya war, given generally similar weaponry and logistical capabilities (see below) of the two cultures. One
during the eighth century, a general atmosphere ofhostility might have been
constant, as it seems to have been during the Late Postclassic. At the very
all other agrarian states; the Polynesian data on interpolity war suggest the
same broad picture. On the other hand, intrapolity conflict would not have
been similarly constrained by the seasons.
What Hassig (1988) calls the "friction ofdistance" severely limited Aztec
campaigns, and would also have constrained Maya war. Except where water
transport was available, all food and other supplies were carried on human
backs. Even if parties of warriors were accompanied by noncombatant
porters, food supplies would have been extremely limited. Stephen Houston
has pointed out to me an apparent war-preparation scene on a polychrome
vessel (Reents-Budet 1994: fig. 1.8) that seems to show elite warriors picking up their burdens to set off on a campaign. In Polynesia the friction of
distance was to some extent overcome by the use of canoes, which could
quickly deliver warriors to the beaches where raids or battles took place and
also cheaply carry sizable amounts of supplies.
Given the dispersed nature of rural settlement, offensive Maya forces
would even have found it difficult to requisition food from their own populace as they moved through their own rural hinterlands. Once in enemy territory, their ability to live off the land would have been extremely limited,
especially during the end of the dry season when there were no crops in the
fields and household stocks that could be looted were low. This is one reason why duration of conflict must have been short and why operation of
large forces must have been the exception rather than the rule. Protracted
sieges of fortified towns probably were impossible.
Dispersed rural settlements around centers undoubtedly affected Maya
military tactics. Although much of the current literature emphasizes the
small-scale raid as a basic tactic (consistent with elite warfare), such raids
would have been highly dangerous, except when carried out by very small
groups moving over very short distances. Roys (1943: 67) says that the Postclassic Maya posted guards near polity boundaries to give early warning of
attack; a dispersed agricultural population would have served the same function. The deeper the penetration into enemy territory, the greater the danger of early detection. Major political capitals oflarge regional states (if these
existed), surrounded by subsidiary centers and far-flung rural populations,
would have been virtually invulnerable to surprise or to attack by small raiding parties of elite warriors. This may be one reason why defenses were typically light.
ARENA. Approximately 20 known Maya centers/polities are recorded as having engaged in wars or conquests, or in their ritual (usually sacrificial)
aftermaths (Hassig 1992:219-21; Schele and Mathews 1991). The named
centers themselves (leaving aside the extents of their territories) range from
huge and venerable ones like Tikal down to small centers that emerged late
and had comparatively short dynastic histories, such as Dos Pilas. Some of
th~ oth~r centers referred to in the inscriptions remain archaeologically
unidentified. Large centers predictably defeated small ones, but small centers also claimed to have defeated, or broken away from, larger ones.
Using a sample of 16 paired real or putative antagonist centers, I calculated straight-line distances between protagonist centers. The range is from
19 to 109.5 km; the mean distance is 57.5 km. Assuming that attackers actually traveled from their own centers to those of their enemies, the spatial
arenas ofMaya wars were often surprisingly large. In some cases (e.g., TikalYaxha, Dos Pilas-Seibal), however, Maya wars were very localized. But
these distances are generally minimal ones, since Maya war parties would
have had to negotiate both terrestrial and aquatic obstacles (e.g., hills and
swamps), even on their comparatively low-relief landscape. Only on the
western margins ofthe lowlands and in Belize would river transport have facilitated travel (e.g., in the Dos Pilas/Yaxchilin and Piedras Negras conflicts). War parties also could have utilized the long chain oflakes across the
central Peten, although some portages would have been necessary.
In all but two or three of the cases, combatants moving on a relatively
straight line between opposed centers would probably not have had to trespass on the territory ofan independent third center. If opposing centers each
possessed large contiguous territories, preparation and staging for attacks
might have taken place at subsidiary places near territorial boundaries, thus
overcoming some logistical constraints.
Unfortunately, these figures are not very useful because we generally
have no idea where Maya battles were fought. Only where fortifications exist can we be certain that conflicts occurred or at least were anticipated. According to Roys (1943: 67), contact-period Maya armies often met on the
roads or trails between warring provinces. If Classic period combatants met at
or near the midpoints between their centers, then travel would be effectively
halved, to an average of about 28 km. Hassig estimates that sizable bodies of
men with weapons, perhaps accompanied by porters carrying provisions,
could travel overland about 19 km per day (1992: 44). The exceptional
maximal distances would thus necessitate round-trip journeys of ten to
twelve days, and the shorter distances three to four days, if only the straightline distances are calculated. Time must be added for additional distance, but
we cannot know how much. At the outside, any recorded Maya campaign
(assuming battles themselves were short) could have been accomplished in
two weeks or less; most probably lasted only a few days.
Arenas of Polyne sian war were quite different. Except in New Zealand,
enemies on any particular island were closely juxtaposed because ofthe small
sizes of terrestrial environments. For example, Hawai'i, the largest of the
Polynesian islands after New Zealand, has an area equivalent to a zone with
a 58-km radius around Tikal. High island environments often were much
more rugged than the Maya landscape, but both small, sharp raids and major engagements were facilitated by rapid water transport. Distance was of
little relevance both in intra- and interisland warfare, and canoe transport
also lessened logistical constraints as already mentioned. Enemies approaching by sea were much more difficult to detect, and both attack and withdrawal were more rapid. Such mobility allowed effective interisland conflict
over comparatively short distances (e.g., between the main islands of t?e
Hawai'ian archipelago) and sometimes expansive conquest warfare over dIStances of 800 to 1,000 km (e.g., Tongan conquests of parts of Fiji and
Samoa).
Another difference is the comparative lack of fixed elite places. Households ofparamounts were often shifted around dependent territories, in p.art
to take more efficient advantage oflocal subsistence resources. Protagomsts
in Polynesian wars, like the Maya, probably always knew where the establishments of enemy paramounts were located, but the ease ofrelocating such
places in much of Polynesia must have created more complex and dynamic
political arenas.
Mobility of warriors rendered the arenas of Polynesian war spatially
larger than those of the Maya, with associated differences in strategies and
tactics, but in another sense they created similarities as well. At least in the
large archipelagos, mobility allowed war to extend over complex political
landscapes and involve multiple polities, as it did among the Maya.
Warfare in all archaic states was an elite enterprise in the sense that it was initiated by elites; commanded, and at least partially fought, by elite contingents; and served elite interests most directly. The only exceptions were
when outsiders with forms of nonhierarchical political organization intruded on archaic state territories, or when nonelite political factions rebelled. Neither the Maya nor the Polynesians were much affected by either,
with the possible exception of the peasant rebellions that J. E. S. Thompson
(1954) believed responsible for the Classic Maya "collapse" ofthe eighth and
ninth centuries AD.
Mayanists agree that Classic rulers and elites initiated warfare, organized
campaigns, and participated in battles and their ritual aftermaths, but there
consensus ends. Freidel (1986: 107), summarizing a view held by many, asserts that Maya war was "a prerogative of the elite and fought primarily by
the elite, [and] the bulk of the population was neither affected by, nor participated in, violent conflict"; he offers no evidence for this conclusion. I
personally (Webster 1993) advocate more general and pervasive dimensions
for Maya war, as we shall shortly see.
ENTITIES AND GENERAL MILITARY ORGANIZATION. Maya wars could involve several
kin~~ of entities, all of which represented elite factions. First, independent
polities of equal or unequal rank engaged in warfare on behalf of their sovereign rulers and associated elites. Such confrontations also might include
temporary allies. Second, factions internal to a polity fought for control of
its offices and titles. Finally, territorial segments of a larger polity fought to
break away under strong local elite factions and assert their independence.
All these sorts of warfare probably occurred among the Late Classic and
Postclassic Maya (Marcus 1992c, 1993a).
There were many forms of archaic-state elite warfare in terms of the recruitment and organization of combatants. Probably none (except the Eastern Zhou, a Chinese state, 770-256 BC) had a professional army in the
~uropean ~ense of the word-in other words, a body of strategically stationed, trained men of different ranks, permanently mobilized and with its
own command structure, whose members identified themselves as full-time
military specialists, used special weaponry, and were recompensed for their
services to the state or ruler. On the other hand, elite males in some archaic
states were highly trained as warriors, and participation in combat was essential to their status and role. In some cases, warfare seems to have involved
fairly small numbers of elite warriors, perhaps supported by lower-ranking
servants (e.g., Bronze Age Greece, early Japan, early feudal Europe),
whereas in others there appears to have been situational conscription of
commoners to fill out the ranks (Sumer and probably Egypt).
Though Maya inscriptions and monumental public art predominantly
f~at~rerulers and their defeated opponents, there also are many group de~lctIOns ofMaya warriors, usually on wall murals, ceramic vessels, figurines,
lintels, and wall panels (most portray war-related rituals rather than combat)
that refer to other individuals. The latter are usually unidentified by title or
name but are typically shown armed and costumed in elaborate fashion including what may be emblematic or heraldic devices (e.g., the Bonampak
murals; see Miller 1986). Most of these supporting personages seem to be
elites, judging both from their accouterments and their close association
with impressive rituals.
. Apart from rulers themselves, there is only the sparsest evidence for any
kind of specialized military leadership before the Postclassic. David Stuart
(1993: 330 - 31) notes that the sahals, subordinate lords who figure prominently on monuments in some regions, often have overtly military roles.
Schele's (1991) reconstruction of what she believes is the first historically
documented Maya war includes a personage, Smoking Frog, brother of the
Tikal ruler, who possibly functioned as the Tikal war chief If her interpretation is correct (and it is disputed by many archaeologists, e.g., D. Stuart
1993: 334), there were specialized Classic elite military roles in some sense
analogous to that of the contact period war chief, or nacom. Also present
during the Late Postclassic were "braves" or notable warriors called holcans
(Landa 1941; Roys 1943). The next generation of decipherment will probably provide more detail on Classic military roles and offices.
Was Classic Maya warfare restricted to elites, or did it involve large numbers of commoner soldiers? Unfortunately, the Maya generally avoided producing large-scale narrative scenes that might help answer this question. To
my knowledge, there is no depiction of a battle or other war-related event
that clearly shows elite warriors leading large contingents of commoner
soldiers. Exceptions might be the large figural compositions at Terminal
Classic/Early Postclassic Chichen Itza, although by this time the organization of warfare might have been rather different than in the Classic period
Maya heartland to the south. The iconographic/epigraphic data can thus be
used to promote a view of warfare that involved only elites. This view is
conspicuous in much of the current literature, and it inevitably leads to considerations of military scale and entities.
SCALE. If Classic Maya polities recruited their combatants only from those
segments of society who were hereditary elites, the forces involved must
have been tiny. Hassig, who thinks Maya warfare was mainly conducted by
elites, calculates that, out of a core population of 49,000, Tikal could have
fielded an offensive force of only 619 to 1,055 combatants (1992:77), perhaps supported by additional noncombatant burden bearers. At Copan, the
highest elite segment of the population probably represented no more than
10 percent of the total of about 28,000 (Webster 1992; Webster, Sanders,
and van Rossum 1992). If one-fifth of these were adult males ofmilitary age,
Copan's "army" would have numbered between 500 and 600 men. Many
Maya polities were smaller than Tikal and Copan, with correspondingly low
elite military potential. Regardless of the imprecision of these numbers, it is
obvious that if a strict "elite warfare" model is adhered to, the forces involved from any particular center were minuscule. In fact, the scale ofMaya
war would be almost that ofthe essentially egalitarian Maori. If, on the other
hand, larger regional states consisting of multiple centers were the entities
and mobilized their collective elites, forces several times this size could have
been mustered. The same is true if there were military alliances among several independent centers (e.g., between Calakmul and Caracol, or Calakmul
and Dos Pilas).
Alternatively, Maya rulers may situationally have drafted much larger
contingents of commoners into their military forces. A Copan king under
these circumstances might in a dire emergency have scraped together an
army of3,000 to 4,000 menjust from his own kingdom, but this would have
been an all-out effort. Large regional states or alliances might have mustered
20,000 or more, although assembling them would have been much more
difficult; forces of this size were more feasible for defensive than offensive
operations.
There are reasons for supposing that the Classic Maya did in fact sometimes assemble sizable armies. First, we know that the contact-period Maya
did so. Roys (1957; 6), summarizing the early colonial evidence, notes that
rulers could summon all adult males of their provinces for war. Such mobilization is consistent with eyewitness Spanish accounts (e.g., Cortes 1986;
Diaz 1963) suggesting that, even allowing for customary exaggeration, very
sizable Maya forces, probably numbering at least in the several thousands,
could be assembled quickly from the local populace. Not all participants
could have been elites; Roys (1943) clearly identifies low-ranking and highranking warriors. When advance warning ofthe Spanish approach occurred,
Cortes claimed not only that many thousands of warriors were arrayed
against him, but that they were assembled from eight provinces and even had
written lists of their contingents (Cortes 1986: 22). Unless there are extremely serious discontinuities between Postclassic and earlier military traditions, Classic (and perhaps Late Preclassic) Maya warfare was at least
sometimes fought on a scale beyond that of elite contingents. The implication is that Maya males of all statuses were prepared to fight and probably
possessed weapons.
Also, some fortifications are much larger than would be necessary to repel small elite forces. The most obvious cases are Becan, fortified at the very
end of the Preclassic period, and the somewhat later boundary defenses of
Tikal (Puleston and Callender 1967; Webster 1976b). Advocates of large
Maya conquest states involving numerous large centers seldom discuss military organization in detail, but the clear implication is that forces must have
been at least situationally large ones.
In suggesting that Maya warfare involved sizable armies, I am not implying that it always did so, nor that only large conflicts had important consequences. Maya war probably had many phases or dimensions, as suggested
by Vayda's model. The problem is that only some of these are represented in
Maya art and epigraphy, and even then they are unclear. Only a very intense
phase of warfare could have necessitated defenses like those at Becan. But
what phase is represented by depictions of kings shown dispatching royal
captives? It could be anything from a Becan-scale engagement to a much
more restricted ritual/royal duel.
We have already seen that warfare in complex Polynesian polities involved paramounts and lesser elites, including some with special roles and
military experience, and thus was reasonably similar to that of the Maya.
Some paramounts appear to have maintained entourages of experienced
warriors as parts of their households (Oliver 1974:878); Cordy (1981:20)
refers to the outright hiring of warriors by elites in Hawai'i. Crucial in
warfare, nor its character" (D. Stuart 1993: 333). I have recently summarized
much of the varied opinion on these issues (Webster 1993) and do not repeat the details here. Briefly, one pole of opinion is that Classic Maya war
consisted primarily of highly choreographed, ritualized conflicts between
rulers and elites, with few territorial, economic, political, or social purposes
or functions apart from maintaining boundaries and the political status quo.
Only when the ethics or charters that maintained equilibrium broke down
did unlimited, destructive war occur (Demarest 1978; Freide1 1986). My
own opposing perspective is that, in addition to its ritual dimensions, war in
all periods had larger political and economic purposes and consequences, especially related to status rivalry.
Epigraphers and iconographers continue to identify new dimensions of
Maya war that relate to material concerns. Stuart (1995), for example, summarizes new information that identifies exaction of tribute and the presentations of status objects apparently made by rulers to their victorious noble
warriors. I think much more fine-grained information of this sort will
emerge in the near future, and that Classic Maya warfare will increasingly be
seen to resemble Aztec war on some levels, particularly in its meritocratic
dimensions.
WAR AS RITUAL. Everyone agrees that Maya war was heavily ritualized, not
only on the battlefield but also during its preparation and aftermath. This
much is conspicuous from artistic depictions alone, in which military
themes are extremely frequent. Such depictions, along with associated inscriptions, allow us to reasonably infer some of the major emic purposes or
motivations for war.
First, Maya rulers sought to project the image of powerful and successful warriors both to other elites and to commoners. Participation in war was
part of the royal role and one way in which the efficacy ofruling personages
was made manifest, particularly at critical junctures in their lives, such as accession and heir designation. Second, war provided victims in the form of
illustrious warrior captives (most desirably, rival rulers) for sacrifice in postbattle rituals, which might include ballgame contests. One obvious purpose
of such sacrifices and contests was to publicly demonstrate the captor's
prowess and the humiliation ofthe victim. Step hen Houston (personal communication 1995) notes that the long hair on a figure shown on a Tonina
monument probably indicates a noble female captive (Villar 1996: 71).
Because Maya rulers were ideologically and ritually responsible for the
well-being of their subjects and the harmony of the Maya cosmos, it is possible that war-related sacrifices were seen as essential to cosmological order,
as they were elsewhere in the Mesoamerican great tradition (although the
Maya seem to lack a major deity specifically identified with war). Heads and
other body parts of sacrificial victims or slain enemies were apparently kept
as trophies, and these, or facsimiles thereof, formed parts of elite ritual costume and regalia. Much of the imagery of Classic period rulership, including warfare roles, was borrowed from non-Mayan Mesoamerican societies,
particularly Teotihuacan. Resorting to such "foreign" symbolism was obviously impossible for Polynesian rulers. Interestingly, contact-period Maya
rulers seem to have been much less fixated on public display of military roles
than their Classic counterparts.
Presumably some phases of war were highly choreographed and ritualized, perhaps even prearranged with an eye to auspicious dates and convenient places. Classic and Postclassic ritual warfare seems to have been
particularly identified with Tlaloc/Venus imagery (Schele 1991) and may
have been to some degree regulated by the cycle of that planet. Unfortunately, the only direct evidence we have for ritualized battlefield behavior is
the gorgeous and distinctive attire of participants.
Although most overviews of Maya war pay lip service to multiple levels
or functions, by far the most conspicuous dimension emphasized is war as
royal ritual. For example, Schele and Miller (1986:220) say that "while the
Maya made war as a ritual of intrinsic interest, and one that was necessary
for the functioning of their world, they also fought to achieve the material
ends that mankind has always sought in war." Yet these "material ends"
scarcely figure in their preceding discussion of Maya war, so heavily does it
focus on the dramatic depictions and events of royal victory and defeat, torture, and sacrifice. To the extent that Maya warfare was motivated by ritual/
religious purposes, it was about irresolvable and nonnegotiable issues. The
extreme view that Maya war was mainly ritual but occurred in a landscape
of political fragmentation characterized by constant war is a far cry from the
"peaceful Maya" perspective.
Polynesian warfare also was highly ritualized. One of the more detailed
summaries is given for Tahiti by Oliver (1974) and broadly reflects patterns
of warfare in the most complex polities. We already saw that preparations
for war included auguries and divination by special priests and pronouncements by shamans. Other ceremonies included human sacrifices to the war
god Oro at principal ritual centers (marae). These preliminary rituals were so
lengthy and public that often enemies were forewarned, and some rulers became so preoccupied with them that they ignored more secular preparations
and lost the war. Protracted rituals did, however, provide time for forces to
be mobilized, allies to be recruited, and deliberations to take place. Some
sham wars were fought that had distinct sporting overtones.
Wars were scheduled to avoid first-fruits and harvest festivals. Pitched
battles, which usually occurred on beaches, were preceded by martial music played on drums and conch shell trumpets. Armies were accompanied by
priests and religious images, and ceremonies were carried out even as battle
was joined. As already noted, gorgeously dressed notable warriors were conspicuous in battle, and success was reckoned as much by the killing or capture of distinguished enemies of high rank as by inflicting heavy overall
casualties. Loss of important warriors could quickly dishearten their forces
and cause a rout.
Elaborate peacemaking ceremonies sometimes took place after battles
when there was no clear winner, but since large battles usually resulted in
the resounding defeat of one side, ritual celebrations of victory were more
common. Captives were sacrificed, corpses of distinguished enemies were
interred in ceremonial buildings, and skulls, jawbones, and even whole bodies were kept as trophies.
Inherent in ritual were important ideological concepts. A general comment by Earle (1987: 298) expresses a Polynesian ideal: "Chiefs rule not because of power, but because of their place in a sacredly chartered world
order." Military success exemplified the supernatural efficacy and sanctity of
elite persons and was a physical manifestation of mana, concentrated most
potently in persons of the highest rank with the most legitimate claims to
highest political office. But the structural logic in this association was double
edged. Successful usurpers of highest titles and offices obviously manifested
superior mana, so their behavior was supernaturally legitimized while the
mana oflosers was devalued.
VIOLENCE. Although violence is part of the conduct of war, I deferred discussing it until this point because of the common assumption that ritual or
conventionalized Maya war must have been relatively nonviolent. Polynesian war, despite its ritual dimensions, often was, in fact, extremely violent.
When one Tahitian army was decisively routed or forced to surrender, an
orgy of slaughter and destruction often followed that included not only
combatants but nearby communities and their inhabitants as well. Houses
and crops were destroyed, and people of all ages and sexes who could not
flee might be killed or taken captive for later sacrifice (Oliver 1974:39598). Tongan war was similarly lethal (Gailey 1987: 168 - 69). Even essentially
egalitarian Maori war seems to have featured the butchery of surrendered or
fleeing enemies, including noncombatants of all ages and sexes (Brailsford
1981; Vayda 1960).
Such violence, however, was limited to the immediate aftermath of
battles. Once defeat was acknowledged, peaceful relations between conquered and conquerors were quickly established, even including the reinstatement of defeated enemy notables (as in Tahiti) who might survive to
start new wars. While the internal character of Polynesian war thus failed
to inhibit extreme violence, it did allow rapid reconstitution of peaceful
relationships.
The current emphasis on the ritual aspects of Maya war reflects a much
older anthropological tradition. As Vayda (1960: 3) noted long ago, many
discussions of "primitive" war seem to "disproportionately emphasize various ideological and non-utilitarian concomitants of war ... treating only
sketchily the environmental, economic, and sociopolitical setting ofmilitary
activity." Ritual dimensions of Maya war are so obtrusive that it is difficult
to place them in their larger sociopolitical and economic contexts. This we
can do for Polynesia, and it is accordingly the focus of the most important
part of this chapter.
Polynesian society was strongly hierarchical. Conical clan organization was its fundamental structural principle, creating both a common thread of relationships and the basis for segmentation
as populations grew in the larger island systems (Kirch 1984: 31-36). Segmentation, along with polygyny, the high status ofwomen, and bilateral descent and inheritance, created ambiguities ofpower and pretexts for conflicts
of many kinds.
Although expressed in extremely varied ways, hierarchy stimulated intense status rivalry among highly ranked individuals and their supporting
factions for titles, offices, privileges, supporters, labor, status objects, and basic resources. All accounts of Polynesian war stress status rivalry, and some
(e.g., Eade 1978) single it out as the major cause of war. Great internal tensions plagued the most stratified Polynesian political systems because of the
combination of"respect and reverence for hereditary rank via seniority with
necessary concessions to political and economic power" (Goldman 1970:
20). We saw such contradictions in the concept of mana. To paraphrase
Oliver (1974: 376), serious armed conflicts, whether between or within
complex Polynesian polities, expressed the ambitions of socially powerful
individuals and their attempts to extend their political power and curb that
of competitors.
Status-rivalry war is by definition internal war. It is fueled by the shared
understandings of the participants concerning their respective positions
within a traditional network of hierarchical social and political relationships, the associated structure ofoffices and titles, the behaviors and symbolic
expressions appropriate to them, and the effective means by which augmentation of position and power may be sought. Such shared cultural understandings operate most immediately among people of high rank within
existing political entities; the prize of successful status rivalry is control of
these entities. Where such understandings are shared across political boundaries, status-rivalry war has wider political arenas; this was the situation in
both Polynesia and the Maya Lowlands.
Where independent polities existed on the same or nearby islands,
conflict involved status rivalry between paramounts of separate ruling lines.
Competing entities were thus both elite and territorial factions. At European contact, for example, there were four sovereign Hawai'ian polities, one
on each major island. Status rivalry also was expressed in intrapolity civil
wars involving elite factions with competing claims to paramount titles and
offices, or fighting for independence. Very occasionally, distant island groups
were conquered, as Tonga conquered parts of Fiji (one of the few possible
instances of Polynesian "external" war in Otterbein's sense).
The most complex Polynesian societies
(Hawai'i, Tahiti, Tonga) were socially stratified. Members of the highestranked stratum were variously called ariki, ari'i, ali'i, and so forth (note the
possible correspondence to the Postclassic Maya almehenob, or nobles descended from nobles, and the Classic ahau). They were potential "players"
in the political game, themselves collectively ranked by conical clan structure with respect to titles and offices. Codes of elite comportment were
widely shared, as were status symbols, so war was internal war, as we have
seen. Highly ranked individuals were very jealous ofinsults, which could be
pretexts for aggression.
Truncated from this elite stratum were the "nonplayers"-commoners
(variously manahune, tu 'a, etc.) who were essentially disenfranchised from
traditional rights to land and from elite kin organization and its associated
status rivalry. In Hawai'i, which has been best studied (Cordy 1981; Eade
1978; Hommon 1986; Kirch and Sahlins 1992), membership in commoner
communities (ahupua'a) was largely defined by residence, but also by attachment to local notable individuals. Commoner families had very short
genealogical depth and no important corporate functions. Interestingly
though, commoners seem to have been able to change their residences and
political affiliations. Various grades of minor elites and managers were intermediaries between these strata (see Spriggs 1988 for the emergence of this
system).
Highborn and lowborn were ideologically linked only by the most
metaphorical concepts of very distant common descent. Despite this genealogical gulf, Polynesian elites never claimed different origins from commoners, as occurred in some other ancient complex societies, including the
Maya, Mixtec, and Zapotec.
"War, as the overt form of competition between greater and lesser powers, higher and lower status, indigene and invader, was thoroughly ingrained
in Polynesian concepts of society" (Kirch 1984: 195). Motivations for war
included "struggles of rival chiefs for power, disputes as to succession, insubordination or rebellion of a subchief, or his resentment at the domination of his superior, lack of respect to a chief, disputes as to land, boundaries
and property generally, the need of food and subsequent raids, abduction of
or elopement with women, vengeance for the death of a relation or insult or
WAR AND THE SOCIAL HIERARCHY.
injury inflicted upon him, and the need for human sacrifice to the gods"
(Williamson 1937: 3 as quoted in Kirch 1984: 197).
Status-rivalry war had two basic dimensions: it continually reinforced
the gulf between elites and commoners and differentiated among elites
themselves. Since achievement could counterbalance hereditary rank (Earle
1978; Goldman 1970), the single most tangible expression of power was the
mobilization of supporters, especially for war. From the paramount perspective, war served to maintain the status quo or to extend political
influence. Lesser elites who supported paramounts sought rewards and
confirmation of their positions. When lesser elites (especially junior collaterals) initiated war themselves, they sought to usurp offices of superiors or
to establish independent polities. For them war was thus a vehicle of social
advancement and mobility.
Similar opportunities were available to successful commoner warriors.
Widespread in Polynesia was the role of distinguished warrior, known by
various honorifics (koa, toa, aito). Although open to elites and commoners
alike distinction in war was for commoners almost the only avenue to social advancement. Privileged warrior status brought with it prestige, as well
as tangible rewards and possibly the right to marry otherwise unattainable
women. Distinguished warriors were much sought after by paramounts and
often were attached to their households. In some of the less centralized
island systems, such as the Marquesas, distinguished warriors (along with inspirational priests) "wrested substantial control of both ritual and production from the hereditary chiefs" (Kirch 1991: 121).
MATERIAL REWARDS OF WAR. Much more was at stake in Polynesian war than
titles, offices, privileges, and prestige. Status rivalry and the assumption of
offices were inextricably linked to control of material resources. The production and redistribution of resources among elites were essential in forming, enlarging, and retaining the allegiance of political factions. Paramounts
had to maintain large and impressive households for themselves and their entourages, and they also had to reward powerful subordinates or lesser relatives. All elites had to exhibit status symbols appropriate to their stations; one
source was booty from successful war, which in some societies included
slaves.
Earle (1978: 168) has differentiated between two parallel economies for
Hawai'i: the subsistence economy and the political economy. Consistent with status rivalry, he sees the commoner subsistence economy as minimizing and
self-dampening, whereas the elite political economy was maximizing and
self-amplifying. Into the political economy flowed the goods, labor, and services required by paramounts to fund their own establishments and to reward their followers. Material rewards might be redistributed directly by the
control (Cordy 1981: 20), delegating the rest to followers. Recipients often
were close relatives, and personal "estates" were given both to elite supporters and to distinguished commoner warriors. Special managers (konohiki) were appointed to supervise estates. Imposition of elite proprietors by
victors severed relationships between commoners and their traditional local
elites, with obvious evolutionary effects. Conquest warfare in the most complex Polynesian polities essentially created a system of paramount patrons
and elite clients with respect to land and labor (see Webster 1975 for a more
general evohitionary discussion of this pattern). Warfare i~ Polyn~sia t~us
functioned to redistribute resources and to restructure SOCIal relationships.
The concept of "territorial conquest" is a slippery one under conditions
of status-rivalry war. Paramounts do conquer territory, in the sense that they
replace a competing set of elite proprietors and thus acquire the ability to
dispose of the products of an enlarged political landscape and the labor of
people on it. On the other hand, such conquest may not involve s~gnifi~ant
disruption ofbasic producers or the influx oflarge numbers of o.utsId~~s ~~to
conquered areas, with obvious implications for the archaeological visibility
of this process.
How was it possible that commoners shared in the risks of war but not
in its rewards? What rendered a lopsided system like this stable with regard
to commoner unrest, especially when elites held no demographic or technological advantage of force? I think there are some obvious answers apart
from the sacred and genealogical charters of Polynesian hierarchical relationships, especially for Hawai'i. First, as I have suggested for the Maya
(Webster 1985), the per-capita extraction of goods and services was probably not onerous. Even if it had been, commoner communities were d~s
persed and often spatially isolated from one another and could ~ot. easily
form collective political factions to resist elite demands on any significant
supracommunity scale.
The presence of nonlocal elites and managers, along with very limited
kinship depth, reduced the organizational potential for local corporate action of any kind. But more importantly, commoners had more to lose by resisting participation in elite wars than by compliance. Any small commoner
group that resisted could easily be punished and dispossessed by the mil~tary
force available to paramounts, limited as it was. On the other hand, neither
victory nor defeat changed commoner fortunes very much, aside from the
occasional violence inherent in battles and their immediate aftermaths.
Peace was quickly reestablished. Victory of one's own paramount reconfirmed existing relationships and maybe even rewarded some commoner
warriors. Defeat resulted in a new set of proprietors, culturally like the last,
whose interests were consistent with the well-being and productivity of
question that wars were fought between small Polynesian polities for possession of particularly productive parts of the landscape, as Vayda (1976) has
maintained for the Maori and Buck (1934) for Mangaia. In Hawai'i only
limited parts of the environment could be modified for intensive agriculture
or fishponds, and no matter why this was initially done, these areas, along
with their comparatively dense associated populations, became particularly
attractive objects of competition.
POPULATION GROWTH AND WAR. The role ofpopulation growth is less controversial. It led to the effective in-filling of most Polynesian islands and the multiplication of potentially competing political entities in the larger systems.
Political arenas thus became more complex and circumscribed. Enlargement
of the elite stratum increased demands for resources, so both production and
conflict intensified. During this process the character of war probably
changed markedly. Based partly on game-theory simulations of island evolutionary processes, Gumerman (1986) suggests that conflict tended to become more aggressive, treacherous, less profitable, and more self-defeating
as circumscription increased. In highly circumscribed environments the result might be a pattern of dynamic equilibrium characterized by cycles of
expansion and fragmentation, as described by Earle (1978: 174) for Hawai'i
and Marcus (1993a) for the Maya. Such political oscillation may have characterized large systems such as Tonga and Hawai'ijust before European contact changed the equation and helped "kick" both systems into more
statelike forms. As Gailey (1987: 250) notes for Tonga, conquest abroad, far
from indicating effective internal political centralization, was sometimes undertaken to avoid rebellion.
MAYA/POLYNESIAN COMPARISONS:
SOME IMPLICATIONS
What have we learned from this comparative exercise? I think several implications are obvious, for the study of both Maya war and war in general.
1. War has many phases, not all of which are equally visible to archaeologists or equally significant in their effects.
2. Absence of fortifications cannot be assumed to indicate the absence of intense phases of war, including territorial conquest.
3. Dispersed settlement can be maintained even in contexts oflethal,
destructive, territorial conquest.
4. Warfare fought primarily for self-serving elite purposes may involve large numbers of commoners.
5. Even small-scale war may have important political, social, and economic consequences and functions.
Mayanists differ on the utility of calling the crisis that overcame Maya
polities in the eighth and ninth centuries a "collapse." Whatever we call it,
warfare almost certainly contributed heavily to social and political disruption during that period. If population pressure was a factor, as it certainly
was at Copan (Webster, Sanders, and van Rossum 1992), it would have
exacerbated earlier patterns of status-rivalry war and created the kind of
bottom-up stress that could destabilize commoner communities, adding an
unprecedented element of political volatility and destructiveness.
A detailed consideration of status-rivalry war and its effects on the distinctive political landscape of the Maya Lowlands is beyond the scope of this
essay. War, however, would have operated very differently at various times
and places, within the contexts of different political systems. For example,
the Copan polity was small enough so that some kin linkages might have integrated the whole society right up through the eighth century. The much
older and larger Tikal polity might well have been more stratified, as Haviland (1992) believes.
The hypothesis that the Maya engaged in something reasonably comparable to Polynesian status-rivalry warfare raises many questions. To what extent did Maya elites share descent on an interpolity basis? How did status
rivalry work in an environment of fixed royal centers and greater political
centralization? To what extent did Maya lords claim rights to dispose ofland
and labor, and how did they reward their elite supporters? What were the
dynamics of ascription and achievement in Maya society? How best do we
define a Maya polity-territorially, or in terms of political allegiance and
elite factions? IfMaya elites pursued status-rivalry war, how is it reflected in
inscriptions? All these are questions for future research.
The archaic-state concept, as noted at the beginning of this chapter,
applies more problematically to the complex Polynesian societies and the
Classic Maya than to the other societies discussed in this volume. Ifthe Polynesians and Classic Maya were in some ways institutionally and organizationally more "chiefdornlike" than Egyptians or Mesopotamians, were their
warfare patterns markedly different from those of archaic states as well?
More broadly, is the character ofwarfare different in chiefdoms as contrasted
with states? I think exploration of these questions requires case-by-case
comparisons because the sociopolitical characters of archaic states are themselves so variable, as are their ecological settings, their associated political
economies, and their degree of urbanization. Such an effort is clearly impossible here, so I will restrict my comments to the issue of status rivalry.
To the extent that some sort of kinship ranking dominates an entire system (i.e., links people from the lowest to the highest social ranks) and access
to rulership is theoretically determined by some kind ofprinciple of descent,
I would expect more intrapolity political fragility for two reasons. First, as
we have already seen, political pretensions and ambitions of multiple contenders for paramount offices are fueled by ambiguities of kinship. Second,
contenders for power have natural constituencies-their kinspeople-who
potentially form the factions or political communities central to the definition of warfare as given above. Many scholars have noted such fragility, and
the really interesting question to me is: Under what conditions does frequent, internecine warfare stimulate the development of state-type institutions? This is a subject I have explored elsewhere (Webster 1975).
In the complex Polynesian polities, kinship ranking (for all practical purposes) included only a small component of the population; the majority
were effectively disenfranchised as political players (although not as warriors). In these societies the ambiguities of kinship still fueled competition
at the top, but the creation and mobilization oflarge political factions must
have proceeded quite differently.
Clearly, elite competition was a potent factor in the more stratified and
bureaucratic archaic states as well, but I think the factions were more likely
to be powerful families (or other institutions) and their economic or political clients, who acted largely outside a pervasive web of ranked kinship
relations. To the extent that reasonably mature institutions of the state suppressed the more destructive levels of internal conflict, competition increasingly involved struggles between territorial entities, which in turn favored
the development of new kinds of recruitment of soldiers and new kinds of
formal military institutions.
My final observations have to do with even more basic issues relating to
the archaeological record. During our seminar Flannery noted that some
chiefdoms and archaic states are extremely flamboyant in their politicoreligious displays while others are not. Unfortunately, archaeologists see
only the most durable dimensions of display and symbolic information. Obviously, the Maya created flamboyant and durable public monuments and
symbolic systems that projected royal/elite information on a scale unmatched in Polynesia. Why such contrasts exist is a significant problem in
its own right. But despite this difference, I suspect that Maya and Polynesian
warfare and its sociocultural settings shared much more than we presently
recognize.
What really worries me is a question that ran repeatedly through my
mind while reading many accounts of Polynesian war. Despite its ubiquity
and importance, how would we perceive such warfare archaeologically, except where durable fortifications existed? My guess is that if we had to rely
only on archaeological materials, we would dismiss as inconsequential one of
the most important components in the structure and evolution of Polyne-
sian society. This is perhaps the most significant lesson of all for Mayanists,
and for students of prehistoric war in general.
NOTE
1. Although the Maori retained principles of kin ranking developed before their migration to New Zealand, and certainly had chiefs, their basic sociopolitical and economic
behaviors appear to me to be markedly egalitarian, not the least in warfare.
Referenes
1984
1992
Heartland if Cities: Surveys if Ancient Settlement and Land Use on the Central
Floodplain of the Euphrates. University of Chicago Press, Chicago.
Mesopotamian Social Evolution: Old Outlooks, New Goals. In On the Evolution if Complex Societies: Essays in Honor if Harry Hoijer, edited by T. Earle,
354 I REFERENCES
REFERENCES I 355
Apter, D. E.
1965
The Politics of Modernization. University of Chicago Press, Chicago.
Arnold,D.
1992 Die Tempel Agyptens: Giittenvohnungen, Kultstatten, Baudenkmiiler. Artemis and
Winkler, Zurich.
Asselberghs, H.
1961
Chaos en Beheersing: Documenten uit het aeneolitisch Egypte. Documenta et Monumenta Orientis Antiqui 8. E. ]. Brill, Leiden.
Assmann,].
1990 Maat: Gerechtigkeit und Unsterblichkeit im alten Agypten. C. H. Beck, Munich.
1991 Stein und Zeit: Mensch und Cesellsihoft im alten Agypten. Wilhelm Fink, Munich.
1992 Das Kulturelle Gediichtnis: Sthrift, Erinnerung und politische ldentitat in fruhen
Hochkulturen. C. H. Beck, Munich.
Badawy, A.
1968 A History of EgyptianArchitecture. University of California Press, Berkeley.
Baer, K.
1960 Rank and Title in the Old Kingdom: The Structure of the EgyptianAdministration in
the Fifth and Sixth Dynasties. University of Chicago Press, Chicago.
1962 The Low Price ofLand in Ancient Egypt. Journal of the American Research Center in Egypt 1 :25-45.
Bagnall, R. S.
1993 Egypt in Late Antiquity. Princeton University Press, Princeton.
Bagnall, R. S., and B. W. Frier
1994 The Demography of Roman Egypt. Cambridge Studies in Population, Economy
and Society in Past Time, No. 23. Cambridge University Press, Cambridge.
Bailey, F. G.
1965 Decisions by Consensus in Councils and Committees, with Special Reference
to Village and Local Government in India. In Political Systems and the Distribution of Power, edited by M. Banton, pp. 1-20. Tavistock, London.
Baines, ].
1985 Fecundity Figures: Egyptian Personification and the Iconology of a Genre. Aris and
Phillips, Warminster, and Bolchazy Carducci, Chicago.
1988a Literacy, Social Organization and the Archaeological Record: The Case of
Early Egypt. In State and Society: The Emergence and Development of Social Hierarchy and Political Centralization, edited by ]. Gledhill, B. Bender, and M. T.
Larsen, pp. 192-214. One World Archaeology 4. Unwin Hyman, London.
1988b An Abydos List of Gods and an Old Kingdom Use of Texts. In Pyramid Studies and Other Essays Presented to 1. E. S. Edwards, edited by]. Baines, T. G. H.
james, M. A. Leahy, and A. F. Shore, pp. 124-33. Occasional Publications 7.
Egypt Exploration Society, London.
1989a Ancient Egyptian Concepts and Uses of the Past: 3rd to 2nd Millennium B.C.
Evidence. In Mo Needs the Past? Indigenous Values and Archaeology, edited by
R. Layton, pp. 131-49. One World Archaeology 5. Unwin Hyman, London.
1989b Communication and Display: The Integration of Early Egyptian Art and
Writing. Antiquity 63 :471-82.
1990 Restricted Knowledge, Hierarchy, and Decorum: Modem Perceptions and
Ancient Institutions. Journal of the American Research Center in Egypt 27: 1-23.
1994
On the Status and Purposes of Ancient Egyptian Art. Cambridge Archaeological
Journal 4:67-94.
1995a The Origins of Kingship in Egypt. In Ancient Egyptian Kingship, edited by
356 I REFERENCES
REFERENCES I 357
Beaulieu, P.-A.
1989
The Reign cif King Nabonidusof Babylon, 556-539 B.C. Yale University Press,
New Haven.
Bellwood, P.
1971
Fortifications and Economy in Prehistoric New Zealand. Proceedings cif the Prehistoric Society 37: 56 -95.
1979 Man's Conquest cif the Pacific. Oxford University Press, New York.
Bender, B.
The Dynamics of Nonhierarchical Societies. In The Evolution cif Political
1990
Systems: Sociopolitics in Small-Scale Sedentary Societies, edited by S. Upham,
pp. 247-63. Cambridge University Press, Cambridge.
Bennet,J.
1990
Knossos in Context: Comparative Perspectives on the Linear B Administration
ofLM II-III Crete. AmericanJournal cif Archaeology 94: 193-211.
Bentley, G. C.
1986
Indigenous States ofSoutheast Asia. Annual Review cif Anthropology 15 :275 - 305.
Berdan, F. F.
Trade, Tribute and Market in the Aztec Empire. Ph.D. dissertation, Department
1975
of Anthropology, University of Texas, Austin.
1982
The Aztecs cif Central Mexico: An Imperial Society. Holt, Rinehart, and Winston,
New York.
1987
Political and Economic Geography of the Eastern Aztec Realm. Paper presented at the 86th Annual Meeting of the American Anthropological Association, Chicago.
Bermann, M.
1994
Lukurmata: Household Archaeology in Prehispanic Bolivia. Princeton University
Press, Princeton.
Bernard, H. R., and P. D. KilIworth
1973
On the Social Structure of an Ocean-Going Research Vessel and Other Important Things. Social Science Research 2: 145-84.
1979
Why Are There No Social Physics? Journal cif the Steward Anthropological Society
11 :33-58.
Bernbeck, R.
1993
Steppe als Kulturlandschaft. Das 'Agig-Gebiert Ostsyriens vom Neolithikum bis zur
Islamischen Zeit. Dietrich Reimer, Berlin.
Berndt, R.
1964
Warfare in the New Guinea Highlands. American Anthropologist 66(4) : 183 - 203.
Bernstein, B.
1965 A Socio-Linguistic Approach to Social Learning. In Penguin Survey of the Social
Sciences 1965, edited by J. Gould, pp. 144-68. Penguin Books, Baltimore.
1971
Class, Codes and Control, Volume 1: Theoretical Studies Towards a Sociology of Language. Routledge and Kegan Paul, London.
Besnier, N.
1993
The Demise of the Man Who Would Be King: Sorcery and Ambition on
N ukulaelae Atoll. Journal cif Anthropological Research 49: 185 - 215.
Bickel, S.
1994
La cosmogonie egyptienne avantle Nouvel Empire. Orbis Biblicus et Orientalis 134.
Editions Universitaires, Fribourg, and Vandenhoeck and Ruprecht, Gottingen,
Biggs, R.
1967
Semitic Names in the Fara Period. Orientalia 36:55-66.
358 I REFERENCES
Bisht, R.
1991
Dholavira: A New Horizon of the Indus Civilization. Puratattva 20: 71-82.
Black,]. A.
The New Year Ceremonies in Ancient Babylon: "Taking Bel by the Hand"
1981
and a Cultic Picnic. Religion 11: 39-59.
Blanton, R. E.
1976
Anthropological Studies of Cities. Annual Review ofAnthropology 5:249-64.
1978 Monte Alban: Settlement Patterns at the Ancient Zapotec Capital. Academic Press,
New York.
1985
A Comparison of Early Market Systems. In Markets and Marketing, Monographs in Economic Anthropology 4, edited by S. Plattner, pp. 399-416.
University Press of America, Lanham, Maryland.
1989
Continuity and Change in Public Architecture: Periods I through V of the
Valley ofOaxaca, Mexico. In Monte Alban'sHinterland, Part II: The Prehispanic
Settlement Patterns in Tlacolula, Etla, and Ocotlan, the Valley ofOaxaca, Mexico,
by S. A. Kowalewski, G. M. Feinman, 1. Finsten, R. E. Blanton, ~nd 1. M.
Nicholas, pp. 409-47. Memoirs 23. Museum of Anthropology, University of
Michigan, Ann Arbor.
1994 Houses and Households: A Comparative Study. Plenum Press, New York.
The Basin of Mexico Market System and the Growth of Empire. In Aztec
1996
Imperial Strategies, by F. F. Berdan, R. E. Blanton, E. Boone, M. Hodge,
M. E. Smith, and E. Umberger, pp. 47-84. Dumbarton Oaks, Washington, D.e.
Blanton, R. E., and G. M. Feinman
The Mesoamerican World System. American Anthropologist 86:673-82.
1984
Blanton, R. E., G. M. Feinman, S. A. Kowalewski, and P. N. Peregrine
1996
A Dual-Processual Theory for the Evolution of Mesoamerican Civilization.
Current Anthropology 37: 1-14.
Blanton, R. E., S. A. Kowalewski, G. M. Feinman, and]. Appel
1981 AncientMesoamerica: A Comparison of Change in Three Regions. Cambridge University Press, Cambridge.
1982 Monte Alban': Hinterland, Part I: Prehispanic Settlement Patterns ofthe Central and
Southern Parts of the Valley of Oaxaca, Mexico. Memoirs 15. Museum of Anthropology, University of Michigan, Ann Arbor.
Blanton, R. E., S. A. Kowalewski, G. M. Feinman, and 1. M. Finsten
1993 Ancient Mesoamerica: A Comparison of Change in Three Regions. 2nd ed. Cambridge University Press, Cambridge.
Blau, P. M.
1968
The Hierarchy of Authority in Organizations. American Journal of Sociology
73:453-67.
1970
A Formal Theory ofDifferentiation in Organizations. American Sociological Review 35: 201-18.
Bloch, Marc
1961 Feudal Society. University of Chicago Press, Chicago.
Bloch, Maurice
1974
Symbols, Song, Dance and Features of Articulation: Is Religion an Extreme Form of Traditional Authority? European Journal of Sociology 15: 5581.
1975
Introduction. In Political Language and Oratory in Traditional Society, edited by
M. Bloch, pp. 1-28. Academic Press, London.
REFERENCES I 359
The Past and the Present in the Present. Man (n.s.) 12: 279-92.
The Disconnection between Power and Rank as a Process: An Outline of the
Development of Kingdoms in Central Madagascar. In The Evolution ofSocial
Systems, edited by]. Friedman and M.]. Rowlands, pp. 303-40. Duckworth,
London.
Ritual Symbolism and the Nonrepresentation of Society. In Symbol as Sense:
1980
New Approaches to the Analysis ofMeaning, edited by M. 1. Foster and S. H.
Brandes, pp. 93-102. Academic Press, New York.
Boehm, e.
1993 Egalitarian Behavior and Reverse Dominance Hierarchy. Current Anthropology
34:227-54.
Boehmer, R. M.
Uruk 1980 -1990: A Progress Report. Antiquity 65 :465 -78.
1991
Boese,].
1989-90 Tell Sheikh Hassan. Archiv fiir Orientforschung 36/37:323-31.
Bohannan, P.
1958 Extra-Processual Events in Tiv Political Institutions. American Anthropologist
60:1-12.
Boltz, W.
1986 Early Chinese Writing. World Archaeology 17: 420 - 36.
1994 The Origin and Development ofthe Chinese Writing System. American Oriental
Society, New Haven.
Boulding, K.
1956
Toward a General Theory of Growth. General Systems 1: 66 -75.
Bourdieu, P.
1977
Outline of a Theory of Practice. Cambridge University Press, Cambridge.
1978-79 Symbolic Power. Telos 38:77-85.
Bowersock, G. W.
The Dissolution of the Roman Empire. In The Collapse ofAncient States and
1988
Civilizations, edited by N. Yoffee and G. 1. Cowgill, pp. 165-75. University
of Arizona Press, Tucson.
Bowman, A. K.
1996 Egypt eifter the Pharaohs 332 BC-AD 642: From Alexander to the Arab Conquest.
British Museum Publications, London.
Bradbury, R. E.
1969
Patrimonialism and Gerontocracy in Benin Political Culture. In Man in
Africa, edited by M. Douglas and P. M. Kaberry, pp. 17-36. Tavistock,
London.
Brailsford, B.
1981 The Tattooed Land. A. H. and A. W. Reed, Wellington.
Brandes, M.
1979 Siegelabrollungen aus den Archaischen Bauschichten in Uruk-Warka. Freiburger
Altorientalisch Studien Band 3. Franz Steiner, Wiesbaden.
Braudel, F.
1973 The Mediterranean and the Mediterranean World in the Age of Philip II, vol. n.
Harper and Row, New York.
Bray, W.
1968
Everyday Life ofthe Aztecs. G. P. Putnam's Sons, New York.
1972
The City State in Central Mexico at the Time of the Spanish Conquest. Journal of Latin American Studies 4: 161-85.
1977a
1977b
360 I REFERENCES
Bright,
1984
Processes of Statemaking and popular Protest: An Introduction. In Statemaking and Social Movements: Essays in History and Theory, edited by C. Bright and
S. Harding, pp. 1-15. University of Michigan Press, Ann Arbor.
Brinkman,]. A.
1980 Kassiten. Reallexikon derAssyriologie 5 :464 -73.
Britan, G. M., and R. Cohen
1980 Hierarchy and Society: Anthropological Perspectives on Bureaucracy. Institute for the
Study of Human Issues, Philadelphia.
Brown, P., and A. Podolefsky
1976
Population Density, Agricultural Intensity, Land Tenure, and Group Size in the
New Guinea Highlands. Ethnology 15: 211-38.
Brumfiel, E. M.
1983 Aztec State Making: Ecology, Structure, and the Origin of the State. American
Anthropologist 85 :261- 84.
1992 Distinguished Lecture in Archeology: Breaking and Entering the Ecosystem-Gender, Class, and Faction Steal the Show. American Anthropologist 94:
551-67.
Brumfiel, E. M., and T. K. Earle
1987
Specialization, Exchange, and Complex Societies: An Introduction. In Specialization, Exchange, and Complex Societies, edited by E. M. Brumfiel and T. K.
Earle, pp. 1-9. Cambridge University Press, Cambridge.
Buck, P.
1934
Mangaian Society. Bulletin 122. Bemice P. Bishop Museum, Hawai'i.
Burger, R.
1992
Chavin. Thames and Hudson, London.
Burke, P.
1986
City-States. In States in History, edited by]. A. Hall, pp. 137-53. Basil Blackwell, Oxford.
Byland, B.
1980
Political and Economic Evolution in the Tamazulapan Valley, Mixteca Alia, Oaxaca,
Mexico: A Regional Approach. Ph.D. dissertation, Department ofAnthropology,
Pennsylvania State University, College Park.
Cabrera Castro, R.
1991
Secuencia arquitect6nica y cronol6gica de la Ciudadela. In Teotihuacan 19801982: nuevas interpretaciones, pp. 31- 60. Serie Arqueologia, INAH, Mexico,
D.E
Cabrera Castro, R., S. Sugiyama, and G. L. Cowgill
The Templo de Quetzalcoatl Project at Teotihuac:in: A Preliminary Report.
1991
Ancient Mesoamerica 2: 77-92.
Calnek, E. E.
1978
The City-State in the Basin of Mexico: Late Pre-Hispanic Period. In Urbanization in the Americasfrom its Beginnings to the Present, edited by R. P. Schaedel,
]. E. Hardoy, and N. S. Kinzer, pp. 463-70. Mouton, The Hague.
Calvet, Y.
1987
La phase 'Oueili de l'epoque d'Obeid. In Prehistoire de la Mesopotamie: la Mesopotamie prehistorique et l'exploration recente du djebel Hamrin, edited by].-L.
Huot, pp. 129-51. Editions du Centre National de la Recherche Scientifique,
Paris.
REFERENCES I 361
Campbell,]. K.
1964 Honour, Family, and Patronage: A Study cif Institutions and Moral Values in a Creek
Mountain Community. Clarendon Press, Oxford.
Campo,]. E.
1991
The Other Side cif Paradise: Explorations into the Religious Meanings cif Domestic
Space in Islam. University of South Carolina Press, Columbia.
Carlton, E.
1977
Ideology and Social Order. Routledge and Kegan Paul, London.
Cameiro, R. L.
1967
On the Relationship between Size of Population and Complexity of Social
Organization. Southwestern Journal cif Anthropology 23:234-43.
1969 The Measurement of Cultural Development in the Ancient Near East and
in Anglo-Saxon England. Transactions cif the New York Academy cif Sciences,
Series 11, vol. 31: 1013-23.
1970a A Theory of the Origin of the State. Science 169: 733-38.
1970b Scale Analysis, Evolutionary Sequences, and the Rating of Cultures. In A
Handbook cif Method in CulturalAnthropology, edited by R. Naroll and R. Cohen, pp. 843-71. Columbia University Press, New York.
1978 Political Expansion as an Expression of the Principle of Competitive Exclusion. In The Origins cif the State: The Anthropology cif Political Evolution, edited by
R. Cohen and E. R. Service, pp. 205-23. Institute for the Study of Human
Issues, Philadelphia.
1981 The Chiefdom: Precursor of the State. In The Transition to Statehood in theNew
World, edited by G. D. Jones and R. R. Kautz, pp. 37-79. Cambridge University Press, Cambridge.
1987 Further Reflections on Resource Concentration and Its Role in the Rise of
the State. In Studies in the Neolithic and Urban Revolution: The V Cordon Childe
Colloquium, Mexico, 1986, edited by L. Manzanilla, pp. 245-60. BAR International Series 349. British Archaeological Reports, Oxford.
1991 The Nature of the Chiefdom as Revealed by Evidence in the Cauca Valley of
Colombia. In Profiles in CulturalEvolution, edited by A. T. Rambo and K. Gillogly, pp. 167-90. Anthropological Papers 85. Museum of Anthropology,
University of Michigan, Ann Arbor.
Carrasco, P.
1971 Social Organization of Ancient Mexico. In Handbook cif Middle American Indians, vol. 10: Archaeology cif Northern Mesoamerica, PartI, edited by R. Wauchope,
pp. 349-75. University of Texas Press, Austin.
1974 Sucesi6n y alianzas matrimoniales en la dinastia teotihuacana. Estudios de Cultura Nahuatl11 :235-42.
1984 Royal Marriages in Ancient Mexico. In Explorations in Ethnohistory: Indians cif
Central Mexico in the Sixteenth Century, edited by H. R. Harvey and H.]. Prem,
pp. 41-81. University of New Mexico Press, Albuquerque.
Casal, ].-M.
1964
Fouilles d'Amri. 2 vols. Publications de la Commission des Fouilles Archaeologiques, Fouilles du Pakistan, Paris.
Cashdan, E. A.
1980 Egalitarianism among Hunters and Gatherers. American Anthropologist 82:
116-20.
362 I REFERENCES
Caso, A.
1935 Las exploraciones en Monte Alban, temporada 1934-1935. Publicacion 18. Instituto Panamericano de Geografia e Historia, Mexico.
1938 Exploraciones en Oaxaca, quinta y sexta temporadas 1936-1937. Publicacion 34.
Instituto Panamericano de Geografia e Historia, Mexico.
Castells, M.
1983
The City and the Grassroots: A Cross-Cultural Theory cif Urban Social Movements.
University of California Press, Berkeley.
Castle, E. W.
1992
Shipping and Trade in Ramesside Egypt. journal cif the Economic and Social History cif the Orient 35 :239-77.
Champion, T., C. Gamble, S. Shennan, and A. Whittle
1984 Prehistoric Europe. Academic Press, London.
Chandler, T., and G. Fox
1974 3000 Years cif Urban Growth. Academic Press, New York.
Chang, K. C.
1983 Art, Myth, and Ritual: The Path to Political Authority in Ancient China. Harvard
University Press, Cambridge.
Charlton, T. H.
1972 Post-Conquest Developments in the Teotihuacan Valley, Mexico. PartI: Excavations.
Report 5. Office of the State Archaeologist, University of Iowa, Iowa City.
1991
The Influence and Legacy of Teotihuacan on Regional Routes and Urban Planning. In Ancient Road Networks and Settlement Hierarchies in the New
World, edited bye. D. Trombold, pp. 186-97. Cambridge University Press,
Cambridge.
Charlton, T. H., and D. 1. Nichols
1997
The City-State Concept: Development and Applications. In The Archaeology
cif City-States: Cross-Cultural Approaches, edited by D. 1. Nichols and T. H.
Charlton, pp. 1-14. Smithsonian Institution Press, Washington, D.e.
Charpin, D.
1980
Remarques sur l'adrninistration paleobabylonienne sous les successeurs d'Hammurapi. journal cif the American Oriental Society 100: 461-71.
1986 Le clerge d'Ur au sieded'Hammurabi (XIXe-XVIIIe av.j.-C.). Droz, Geneva and
Paris.
Charvat, P.
1988
Archaeology and Social History: The Susa Sealings ea. 4000 - 2340 B.C. Paleorient 14(1): 57-64.
Chase, A. F., and D. Z. Chase
1987 Investigations at the Classic Maya City cif Caracol, Belize: 1985-1987. Monograph 3. Pre-Columbian Art Research Institute, San Francisco.
Chase, A. F., and P. M. Rice (editors)
1985
The Lawland Maya Postclassic. University of Texas Press, Austin.
Chase-Dunn, e.
1990
World-State Formation: Historical Processes and Emergent Necessity. Political
Geography Quarterly 9: 108 - 30.
Chazan, N.
The Early State in Africa: The Asante Case. In The Early State in African
1988
Perspective: Culture, Power and Division cif Labot, edited by S. N. Eisenstadt,
M. Abitbol, and N. Chazan, pp. 60-97. E. J. Brill, Leiden.
REFERENCES I 363
Childe, V. G.
1942 What Happened in History. Penguin, Harmondsworth.
1952 New Light on the Most Ancient East. Routledge and Kegan Paul, London.
Cieza de Leon, P.
1959 (1553) The Incas. Translated by H. de Onis, edited by V. W. Von Hagen. University of Oklahoma Press, Norman.
Civil, M.
1975
Lexicography. In Sumerological Studies in Honor of Thorkild jacobsen on His Seventieth Birthday,june 7, 1974, edited by S. Lieberman, pp. 123-57. University
of Chicago Press, Chicago.
1987
Ur III Bureaucracy: Quantitative Approaches. In The Organization cif Power:
Aspects cif Bureaucracy in the Ancient Near East, edited by M. Gibson and R. D.
Biggs, pp. 19-41. Studies in Ancient Oriental Civilization 46. Oriental Institute, University of Chicago, Chicago.
1992 Education (Mesopotamia). AnchorBibleDictionary Il: 301-5. Doubleday, New
York.
Civil, M., R. D. Briggs, H. G. Giiterbock, H. J. Nissen, and E. Reiner
1969
The Series lu=sa and Related Texts. Materials for the Sumerian Lexicon 12.
Pontificium Institutum Biblicum, Rome.
Claessen, H. J M.
1978a The Early State: A Structural Approach. In The Early State, edited by H. J. M.
Claessen and P. Skalnik, pp. 533-96. Mouton, The Hague.
1978b Comment. Current Anthropology 19:769.
1978c Early State in Tahiti. In The Early State, edited by H. J. M. Claessen and
P. Skalnik, pp. 441-68. Mouton, The Hague.
1984
The Internal Dynamics of the Early State. Current Anthropology 25:365-79.
Claessen, H. J. M., and P. Skalnik
1978a The Early State: Theories and Hypotheses. In The Early State, edited by H. J. M.
Claessen and P. Skalnik, pp. 3-29. Mouton, The Hague.
1978b The Early State: Models and Reality. In The Early State, edited by H. J. M.
Claessen and P. Skalnik, pp. 637-50. Mouton, The Hague.
Claessen, H. J. M., and P. Skalnik (editors)
1978
The Early State. Mouton, The Hague.
1981
The Study cif the State. Mouton, The Hague.
Claessen, H. J. M., and P. van de Velde (editors)
1987 Early State Dynamics. E. J. Brill, Leiden.
1991 Early State Economics. Transaction Publishers, New Brunswick, New Jersey.
Claessen, H. J. M., P. van de Velde, and M. E. Smith (editors)
1985 Development and Decline: The Evolution cif Sociopolitical Organization. Bergin and
Garvey, South Hadley, Massachusetts.
Clastres, P.
1977 Society against the State. Translated by R. Hurley. Urizen Press, New York.
Clay, A. T.
Amurru, the Home cif the Northern Semites: A Study Showing That the Religion and
1909
Culture cif Israel Are Not cif Babylonian Origin. The Sunday School Times,
Philadelphia.
Cobb, H. (editor)
1979 Cook's Voyages. British Museum, London.
REFERENCES I 365
364 I REFERENCES
Cobo, B.
1990 (1653) Inca Religion and Customs. Translated and edited by R. Hamilton. University of Texas Press, Austin.
Coe,M.D.
1992
Breaking the Maya Code. Thames and Hudson, New York.
Coe, M. D., and R. A. Diehl
1980
In the Land cif the Olmec, Vol. II: The People cif the River. University of Texas
Press, Austin.
Coe, W. R.
1965 Tikal: Ten Years of Study of a Maya Ruin in the Lowlands of Guatemala. Expedition 8: 5 -56.
1990
Excavations in the North Acropolis, North Terrace, and Great Plaza cif Tikal. Tikal
Report 14. Monograph 61. University Museum, University of Pennsylvania,
Philadelphia.
Cohen, R.
1978
State Origins: A Reappraisal. In The Early State, edited by H.]. M. Claessen
and P. Skalnik, pp. 31-75. Mouton, The Hague.
1981 Evolution, Fission, and the Early State. In The Study of the State, edited by
H.]. M. Claessen and P. Skalnik, pp. 87-115. Mouton, The Hague.
Cohen, R., and E. R. Service (editors)
1978
Origins cif the State: The Anthropology cif PoliticalEvolution. Institute for the Study
of Human Issues, Philadelphia.
Cohen, R., and]. D. Toland
1988
State Formation and PoliticalLegitimacy. Political Anthropology VI. Series editor,
M.]. Aronoff Transaction Books, New Brunswick, New Jersey.
Collins, R.
1981
Sociology Since Midcentury: Essays in Theory Cumulation. Academic Press, New
York.
Conklin, W.]., and M. E. Moseley
1988
The Patterns of Art and Power in the Early Intermediate Period. In Peruvian
Prehistory, edited by R. Keatinge, pp. 145-63. Cambridge University Press,
Cambridge.
Conrad, G. W.
1982
The Burial Platforms of Chan Chan: Some Social and Political Implications.
In Chan Chan: Andean Desert City, edited by M. E. Moseley and K. e. Day,
pp. 87-117. University of New Mexico Press, Albuquerque.
Cooper,]. S.
1973
Sumerian and Akkadian in Sumer and Akkad. In Approaches to the Study cif
the Ancient Near East: A Volume cif Studies Offered to IgnaceJ. Gelb, edited by
G. Buccellati, pp. 239-46. Orientalia 42.
1983
Reconstructing History from Ancient Inscriptions: The Lagasli-Umma Border Conflict. Sources from the Ancient Near East 2(1). Undena, Malibu, California.
1989
Writing. International Encyclopaedia cif Communications 4:321-31.
1993
Paradigm and Propaganda: The Dynasty of Akkade in the 21st Century. In
Akkad: The First World Empire, edited by M. Liverani, pp. 11-23. Sargon,
Padua.
Corcoran, P. E.
1979
PoliticallAnguage and Rhetoric. University of Texas Press, Austin.
Cordy, R. H.
1981
A Study cif Prehistoric Social Change. Academic Press, New York.
Cortes, H.
1986
Lettersfrom Mexico. Yale University Press, New Haven.
Costin, e. L., and T. K. Earle
1989
Status Distinction and Legitimation of Power as Reflected in Changing Patterns of Consumption in Late Prehispanic Peru. American Antiquity 54: 691714.
Cowgill,G.L.
1983 Rulership and the Ciudadela: Political Inferences from Teotihuacan Architecture. In Civilization in the Ancient Americas: Essays in Honor cif Gordon R. Willey, edited by R. M. Leventhal and A. L. Kolata, pp. 313-43. University of
New Mexico Press, Albuquerque, and Peabody Museum of Archaeology and
Ethnology, Harvard University, Cambridge.
Onward and Upward with Collapse. In The Collapse cif Ancient States and Civi1988
lizations, edited by N. Yoffee and G. L. Cowgill, pp. 244-76. University of
Arizona Press, Tucson.
1992a Social Differentiation at Teotihuacan. In Mesoamerican Elites: An Archaeological
Assessment, edited by D. Z. Chase and A. F. Chase, pp. 206-20. University of
Oklahoma Press, Norman.
1992b Toward a Political History of Teotihuacan, In Ideology and Pre-Columbian Civilizations, edited by A. A. Demarest and G. W. Conrad, pp. 87-114. School of
American Research Press, Santa Fe.
1993 Distinguished Lecture in Archeology: Beyond Criticizing New Archeology.
American Anthropologist 95: 551-73.
1997 State and Society at Teotihuacan, Mexico. Annual Review of Anthropology
26:129-61.
Crawford (Thompson), D.]'
1971
Kerkeosiris: An Egyptian Village in the Ptolemaic Period. Cambridge University
Press, Cambridge.
Crawford, H.
1991
Sumer and the Sumerians. Cambridge University Press, Cambridge.
Creel, H. G.
1970
The Origins cif Statecraft in China, Volume One: The Western Chou Empire. University of Chicago Press, Chicago.
Crumley, e. L.
1976 Toward a Locational Definition of State Systems of Settlement. American Anthropologist 78: 59-73.
1987 A Dialectical Critique of Hierarchy. In Power Relations and State Formation,
edited by T. e. Patterson and e. W. Gailey, pp. 155-69. Special publication
of the Archeology Section, American Anthropological Association, Washington, D.e.
1995 Heterarchy and the Analysis of Complex Societies. In Heterarchyand the Analysis cif Complex Societies, edited by R. Ehrenreich, e. L. Crumley, and]. E. Levy,
pp. 1-5. Archeological Papers 6. American Anthropological Association,
Washington, D.e.
Culbert, T. P.
1991 Polities in the Northeastern Peten, Guatemala. In ClassicMaya Political History,
edited by T. P. Culbert, pp. 128-46. School of American Research Advanced
Seminar Series, Cambridge University Press, New York.
366 I REFERENCES
Culbert, T. P. (editor)
1973
REFERENCES I 367
Demarest, A. A.
1978 Interregional Conflict and "Situational Ethics" in Classic Maya Warfare. In
Codex Wauchope: Feslschrift in Honor ofRobert Wauchope, edited by M. Giardino,
B. Edmonson, and W, Creamer, pp. 101-11. Human Mosaic, Tulane University, New Orleans.
1993 Violent Saga ofa Maya Kingdom. National Geographic Magazine 183(2): 95 - 111.
Demarest, A. A., and G. W. Conrad (editors)
1992
Ideologyand Pre-Columbian Civilizations. School ofAmerican Research, Santa Fe.
DeMarrais, E., L. J. Castillo, and T. Earle
1996
Ideology, Materialization, and Power Strategies. Current Anthropology 37(1):
15-31.
de Montmollin, O.
1989
368 I REFERENCES
REFERENeE S I 369
Easton, D.
1959
Political Anthropology. BiennialReview ofAnthropology 1959:210-62.
Edzard, D. 0., W. Farber, and E. Sollberger
1977
Die Orts- und Gewiissernamen der priiesargonischen und sargonischen Zeit. Repertoire Geographique des Textes Cuneiformes I. Beihefte zum Tiibinger Atlas
des Vorderen Orients B7(1). Ludwig Reichert, Wiesbaden.
Ehrenreich, R. M.
1995
Early Metalworking: A Heterarchical Analysis of Industrial Organization. In
Heterarchy and the Analysis of Complex Societies, edited by R. M. Ehrenreich,
e. L. Crumley, and]. E. Levy, pp. 33-40. Archeological Papers 6. American
Anthropological Association, Washington, D.e.
Ehrenreich, R. M., e. L. Crumley, and]. E. Levy (editors)
.
1995
Heterarchy and the Analysis ofComplex Societies. Archeological Papers 6. American Anthropological Association, Washington D.e.
Eisenstadt, S. N.
1965
Essays on Comparative Institutions. John Wiley and Sons, New York.
1969
The Political Systems ofEmpires: The Rise and Fall ofthe Historical Bureaucratic Societies. The Free Press, New York.
1977
Sociological Theory and an Analysis of the Dynamics of Civilizations and
Revolutions. Daedalus 106(4):59-78.
1980
Cultural Orientations, Institutional Entrepreneurs, and Social Change: Comparative Analysis of Traditional Civilizations. American Journal of Sociology
85:840-69.
1981
Cultural Traditions and Political Dynamics: The Origins and Modes of Ideological Politics. BritishJournal ofSociology 32: 155 - 81.
1986
The Axial Age Breakthroughs- Their Characteristics and Origins. In The
Origins and Diversity of Axial Age Civilizations, edited by S. N. Eisenstadt,
pp. 1-25. State University of New York Press, Albany.
1988
Beyond Collapse. In The Collapse of Ancient States and Civilizations, edited
by N. Yoffee and G. L. Cowgill, pp. 236-43. University of Arizona Press,
Tucson.
Eisenstadt, S. N. (editor)
1986
The Origins and Diversity ofAxial Age Civilizations. SUNY Series in Near Eastern Studies, State University of New York Press, Albany.
Ekholm, K.
1972
Power and Prestige: The Rise and Fall of the Kongo Kingdom. Scriv Service,
Uppsala.
Elam,]. M.
1989
Defensible and Fortified Sites. In Monte Alban's Hinterland, Part Il: The PrehispanicSettlement Patterns in Tlacolula, Et/a, and Ocotlan, the Valley ofOaxaca, Mexico, by S. A. Kowalewski, G. M. Feinman, L. Finsten, R. E. Blanton, and
L. M. Nicholas, pp. 385-407. Memoirs 23. Museum of Anthropology, University of Michigan, Ann Arbor.
el-Khouli, A. A.-R. H.
1978
Egyptian Stone Vessels, Predynastic Period to Dynasty Ill: Typology and Analysis.
3 vols. Philipp von Zabern, Mainz.
Elvin,M.
1973
The Pattern of the Chinese Past: A Social and Economic Interpretation. Stanford
University Press, Stanford.
REFERENCES I 371
370 I REFERENCES
Was There a Transcendental Breakthrough in China? In The Origins and Diversity ofAxial Age Civilizations, edited by S. N. Eisenstadt, pp. 325-59. State
University of New York Press, Albany.
El- Wailly, E, and B. Abu al-Soof
1965
The Excavations at Tell es-Sawwan: First Preliminary Report (1964). Sumer
21:17-32.
Ember, M.
1963
The Relationship between Economic and Political Development in Nonin. dustrialized Societies. Ethnology 2: 228 - 48.
Emery, W. B.
1949-58 Great Tombs of the First Dynasty, vols. l-Tll. Service des Antiquites de I'Egypte, Excavations at Sakkara, Government Press, Cairo (I), and Egypt Exploration Society, London (I1-III).
1961 Archaic Egypt. Viking Penguin Books, New York.
Engelbach, R.
1943
An Essay on the Advent of the Dynastic Race in Egypt and Its Consequences.
Annales du Service des Antiquites de l'Egypte 42: 193-221.
Engels, D. W.
1978 Alexander the Greatand the Logistics ofthe Macedonian Army. University of California Press, Berkeley.
Englund, R. K., and H. Nissen, with P. Damerow
1993 Die lexikalischen Listen der archaischen .Texte aus Uruk. Gebriider Mann, Berlin.
Engnell,l.
1943 Studies in Divine Kingship in the Ancient Near East. Almqvist and Wiksell, Uppsala.
Erickson, E. E.
1973
Life Cycle of Life Styles: Projecting the Course of Locai Evolutionary Sequences. Behavior Science Notes 8: 135-60.
1975
Growth Functions and Culture History: A Perspective on Classic Maya Cultural Development. Behavior Science Research 10:37-61.
Erman, A.
1923 Aegypten und aegyptisches Leben im Altertum. 2nd ed. Revised by H. Ranke.
J. e. B. Mohr (Paul Siebeck), Tiibingen. Originally published 1885.
Esson, G. M. (translator and editor)
1931 Discourses on Salt and Iron: A Debate on State Control ofCommerce and Industry in
Ancient China, chapters 1-28. E. J. Brill, Leiden.
Evans, Sir A.
1921-36 The Palace ofMinos at Knossos, vols. I-IV. Macmillan, London.
Evans, G.
1958
Ancient Mesopotamian Assemblies. Journal of the American Oriental Society
78: 1-11.
Evans, S. T., andJ. e. Berlo
1992
Teotihuacan: An Introduction. In Art, Ideology, and the City of Teotihuacan,
edited by J. e. Berlo, pp. 1-26. Dumbarton Oaks, Washington, D.e.
Eyre, e. J.
The Adoption Papyrus in Social Context. Journal ofEgyptian Archaeology 78:
1992
207-21.
Fairservis, W. A., Jr. .
1961 The Harappan Civilization: New Evidence and More Theory. Novitates 2055.
American Museum of Natural History, New York.
1986
The Origin, Character, and Decline ofan Early Civilization. American Museum
Novitates 2302: 9 - 48.
1971 The Roots ofAncient India: The Archaeology ofEarly Indian Civilization. Macrnillan, New York.
1986
Cattle and the Harappan Chiefdoms of the Indus Valley. Expedition 28(2):
43-50.
1992 The Harappan Civilization and Its Writing: A Mode/for the Decipherment of the Indus Script. Oxford and IBH, Delhi.
Fakhry, A.
1961 The Monuments ofSneferu at DahshurI1: The Valley Temple I: The Temple Reliefs.
Antiquities Department of Egypt, General Organization for Government
Printing Offices, Cairo.
Falkenhausen, L. von
1993
On the Historiographical Orientation of Chinese Archaeology. Antiquity 67:
839-49.
Falkenstein, A.
Zur Chronologie der sumerischen Literatur. Comptes Rendus du Rencontre As1951
syriologique International 2 : 12-28.
1953
Zur Chronologie der sumerischen Literatur: Die nachaltbabylonische Stufe.
Mitteilungen der Deutsihen Orient-Gesellschaft 85: 1-13.
1956 -57 Die neusumerischen Gerichtsurkunden. 3 vols. Bayerische Akademie der Wissenschaften, Philosophisch-Historische Klasse, Abhandlungen N.E 39, 40, 44.
Bayerische Akademie /C. H. Beck, Munich.
1963
Zu den Inschriften der Grabung in Uruk-Warka. Baghdader Mitteilungen 2:
1-82.
1974 The Sumerian Temple City. Introduction and translation by M. de J. Ellis.
Monographs on the Ancient Near East 1(1). Undena, Malibu, California.
Originally published 1951.
Fash, W. L.
1982
A Middle Formative Cemetery from Copan, Honduras. Paper presented at the
81st Annual Meeting of the American Anthropological Association, Washington, D.e.
1983 Maya State Formation: A Case Study and Its Implications. Ph.D. dissertation, Department of Anthropology, Harvard University, Cambridge.
1991 Scribes, Warriors, and Kings: The City of Copan and the Ancient Maya. Thames
and Hudson, London.
Febvre, L., and H.-J. Martin
1976 The Coming of the Book: The Impact of Printing, 1450-1800. Translated by
D. Gerard. NLB Books, London.
Feeley-Harnik, G.
1985
Issues in Divine Kingship. Annual Review ofAnthropology 14:273-313.
Feinman, G. M.
1986
The Emergence of Specialized Ceramic Production in Formative Oaxaca. In
Research in Economic Anthropology, Supplement 2: Economic Aspects of Prehispanic
Highland Mexico, edited by B. L. Isaac, pp. 347-73. JAI Press, Greenwich,
Connecticut.
1991
Demography, Surplus, and Inequality: Early Political Formations in Highland
Mesoamerica. In Chiefdoms: Power, Economy, and Ideology, edited by T. Earle,
pp. 229-62. Cambridge University Press, Cambridge.
1995
The Emergence of Inequality: A Focus on Strategies and Processes. In The
1967
372 I REFERENCES
REFERENCES I 373
REFERENCES I 375
374 I REFERENCES
1994
Preliterate Counting and Accounting in the Middle East. Orientalistische Literaturzeitung 89(5/6): 477 -502.
Fried, M. H.
1967
The Evolution of Political Society: An Essay in Political Anthropology. Random
House, New York.
Friedman, D.
1977 A Theory of the Size and Shape of Nations. Journal ofPolitical Economy 85:
59-77.
Friedman, ].
1979
System, Structure and Contradiction: The Evolution of "Asiatic" Social Formations.
National Museum of Denmark, Copenhagen.
Friedman,]., and M.]. Rowlands
1977
Notes Towards an Epigenetic Model of the Evolution of "Civilisation." In
The Evolution of Social Systems, edited by ]. Friedman and M. ]. Rowlands,
pp. 201-76. Duckworth, London.
Friedman, J., and M.]. Rowlands (editors)
1977
The Evolution ofSocial Systems. Duckworth, London.
Fry, R. E.
1990
Disjunctive Growth in the Maya Lowlands. In Precolumbian Population History
in the Maya Lawlands, edited by T. P. Culbert and D. S. Rice, pp. 285-300.
University of New Mexico Press, Albuquerque.
Gadd, c.j
1963 Two Sketches from the Life at Ur. Iraq 25: 177 - 88.
Gailey, C. W.
1987
Kinship to Kingship. University of Texas Press, Austin.
Garda Cook, A.
1994
Cantona. Salvat, Mexico.
Gardiner, A. H.
1941-52 The Wilbour Papyrus. 4 vols. (vo!. IV by R. O. Faulkner). Brooklyn Museum, Oxford University Press, London.
1951 A Protest against Unjustified Tax-Demands. Revue d'Egyptologie 6: 115 -33.
Garelli, P.
The Changing Facets of Conservative Mesopotamian Thought. Daedalus
1975
(Spring) :47-56. Proceedings ofthe American Academy of Arts and Sciences 104(2).
Garstang, ].
1904
Tombs of the Third Egyptian Dynasty at Reqsqnal: and Bet Khalldf Archibald
Constable, Westminster.
Gasse, A.
1988
Donnees nouvelles administratives et sacerdotales sur I' organisation du Domaine
d'Amon, XXe-XXIe dynasties. 2 vols. Bibliotheque d'Etude 104. Institut
Francais d'Archeologie Orientale, Cairo.
Geertz, C.
1980
Negara: The Theater State in Nineteenth Century Bali. Princeton University Press,
Princeton.
Gelb, I.].
1968 An Old Babylonian List of Amorites. Journal of the American Oriental Society
88:39-46.
1969
On the Alleged Temple and State Economies in Mesopotamia. In Studi in onore
di Edoardo Volterra 6:137-54. Giuffre, Milan.
376 I REFERENCES
REFERENCES I 377
Gregoire, ].-P.
1962
La province meridionale de l'etatde Lagash. Imprimerie Bourg-Bourger, Luxembourg.
Gregory, C. A.
1980
Gifts to Men and Gifts to Gods: Gift Exchange and Capital Accumulation in
Contemporary Papua. Man (n.s.) 24: 1-20.
1982
Gifts and Commodities. Academic Press, London.
Griffeth, R., and C. G. Thomas (editors)
1981
The City-State in Five Cultures. ABC-Clio, Santa Barbara.
Guaman Poma de Ayala, F.
1980 (1613) El primer nueva cor6nica y buengobierno. 3 vols. Edici6n critica de J. v.
Murra y R. Adorno. Siglo Veintiuno, Mexico.
Gumerman, G. G.
1986
The Role of Competition and Cooperation in the Evolution of Island Societies. In Island Societies: Archaeological Approaches to Evolution and Transjormation,
edited by P. V. Kirch, pp. 42-49. Cambridge University Press, New York.
Gunn,]., W.]. Folan, and H. Robichaux
1995
A Landscape Analysis of the Candelaria Watershed in Mexico: Insights into
Paleoclimates Affecting Upland Horticulture in the Southern Yucaran Peninsula Semi-Karst. Geoarchaeology 10:3-42.
Gupta, S. P.
1978
Origin of the Form of Harappa Culture: A New Proposition. Puratattva 8:
141-46.
Haas, ].
1982
The Evolution cif the Prehistoric State. Columbia University Press, New York.
Haggett, P.
1966
Locational Analysis in Human Geography. St. Martin's Press, New York.
Hall,]. A.
1989
Towards a Theory of Social Evolution: On State Systems and Ideological
Shells. In Domination and Resistance, edited by D. Miller, M. Rowlands, and
C. Tilley, pp. 96-107. Unwin Hyman, London.
Hallo, W. W.
1960 A Sumerian Amphictyony.journal of Cuneiform Studies 14:88-114.
1962
New Viewpoints on Sumerian Literature. Israel Exploration journal 12 : 13-26.
1963
On the Antiquity ofSumerian Literature. journal of the American Oriental Society 83: 167-76.
Hamayon, R. N.
1994
Shamanism in Siberia: From Partnership in Supernature to Counter-Power
in Society. In Shamanism, History, and the State, edited by N. Thomas and
C. Humphrey, pp. 76-89. University of Michigan Press, Ann Arbor.
Hammond, N.
1991
Inside the Black Box: Defining Maya Polity. In Classic Maya Political History,
edited by T. P. Culbert, pp. 253-84. School of American Research Advanced
Seminar Series, Cambridge University Press, Cambridge.
Hammond, N. G. L.
1986
A History cif Greece to 322 B. C. 3rd ed. Oxford University Press, Oxford.
Hardacre, H.
1993
The Impact of Fundamentalism on Women, the Family, and Interpersonal Relations. In Fundamentalisms and Society, Volume 2: Reclaiming the Sciences, the
Family, and Education, edited by M. E. Marty and R. S. Appleby, pp. 129-50.
University of Chicago Press, Chicago.
378 I REFERENCES
REFERENCES I 379
Hardin, K.
1993
The Aesthetics if Action: Continuity and Change in a West African Town. Smithsonian Institution Press, Washington, D.e.
Harpur, Y.
1987
Decoration in Egyptian Tombs of the Old Kingdom: Studies in Orientation and Scene
Content. KPI, London and New York.
Harris, R.
1964 The Naditu Woman. In Studies Presented to A. Leo Oppenheim, edited by]. A.
Brinkman, pp. 106-35. Oriental Institute, University of Chicago, Chicago.
1975
if an
1977
1970
Wisconsin, Madison.
Hoffman, M. A., B. Adarns, M. Berger, M. el Hadidi,]. Harlan, H. Harnroush, C. Lupton, ]. McArdle, W. McHugh, R. Alien, and M. Rogers
1982
The PredynasticofHierakonpolis: An Interim Report. Egyptian Studies Association
Publication 1. Cairo University Herbarium, Giza, and Department ofSociology and Anthropology, Western Illinois University, Macomb.
Hole, F., K. V. Flannery, and]. A. Neely
1969
Prehistoric Human Ecology if the Deh Luran Plain. Memoirs 1. Museum of Anthropology, University of Michigan, Ann Arbor.
Holt, E.
1962
The Strangest War. Putnam, London.
Homrnon, R.].
1986 Social Evolution and Ancient Hawai'i. In Island Societies: Archaeological Approaches to Evolution and Transformation, edited by P. V. Kirch, pp. 55-68. Cambridge University Press, New York.
Hornung, E.
1982
Conceptions if God in Ancient Egypt: The One and the Many. Translated by
]. Baines. Cornell University Press, Ithaca.
Houston, S. D.
1992 Weak States and Segmentary Structure: The Internal Organization of Classic
Maya Polities. Paper presented at the symposium "The Segmentary State and
the Classic Lowland Maya," Cleveland State University, Cleveland.
1993
Hieroglyphics and History at Dos Pi/as: Dynastic Politics if the Classic Maya. University of Texas Press, Austin.
380 I REFERENCES
Hsu, e.-Y.
1965 Ancient China in Transition: An Analysis of Social Mobility, 722-222 B.G. Stanford University Press, Stanford.
1986
Historical Conditions of the Emergence and Crystallization of the Confucian
System. In The Origins and Diversity ofAxial Age Civilizations, edited by S. N.
Eisenstadt, pp. 306-24. State University of New York Press, Albany.
1988
The Roles of the Literati and of Regionalism in the Fall of the Han Dynasty.
In The Collapse ofAncient States and Civilizations, edited by N. Yoffee and G. L.
Cowgill, pp. 176-95. University of Arizona Press, Tucson.
Hsu, c., Y., and K. M. Linduff
1988
Western Chou Civilization. Yale University Press, New Haven.
Hsu, F. L. K.
1949
Under the Ancestor's Shadow: Chinese Culture and Personality. Routledge and
Kegan Paul, London.
Humphreys, S. C.
1977 Evolution and History: Approaches to the Study of Structural Differentiation.
In The Evolution ofSocial Systems, edited by J. Friedman and M. J. Rowlands,
pp. 341-71. Duckworth, London.
1978 Anthropology and the Greeks. Routledge and Kegan Paul, London.
Hyslop,J.
1976 An Archaeological Investigation of the Lupaca Kingdom and Its Origins. Ph.D. dissertation, Department of Anthropology, Columbia University, New York.
1977
Chulpas of the Lupaca Zone of the Peruvian High Plateau. Journal ofField Archaeology 4: 149-70.
1984
The Inka Road System. Academic Press, San Francisco and New York.
1985 Inkawasi: The New Cuzco. BAR International Series 234. British Archaeological Reports, Oxford.
1990 Inka Settlement Planning. University of Texas Press, Austin.
Isbell, W. H., and G. F. McEwan (editors)
1991 Huari Administrative Structure: Prehistoric Monumental Architecture and State Government. Dumbarton Oaks, Washington, D.e.
Jacobsen, T.
1943
Primitive Democracy in Ancient Mesopotamia. Journal ofNear Eastern Studies
2: 159-72.
1957 Early Political Development in Mesopotamia. Zeitschrift fur Assyriologie 52:
91-140.
1976
The Treasures ofDarkness: A History ofMesopotamian Religion. Yale University
Press, New Haven.
Jacobson, J.
1979 Recent Developments in South Asian Prehistory and Protohistory. Annual Review ofAnthropology 8:467-502.
1986 The Harappan Civilization: An Early State. In Studies in the Archaeology ofIndia and Pakistan, edited by J.Jacobson, pp. 137-73. Oxford and IBH, and the
American Institute of Indian Studies, Delhi.
Jacquet-Gordon, H.
1962 Les noms des domaines [uneraires sous I' ancien empire egyptien. Bibliotheque
d'Etude 34. Institut Francais d'Archeologie Orientale, Cairo.
Jansen, M.
1983 Theoretical Aspects of Structural Analysis for Mohenjo-daro. In Reports on
Field Work Carried out at Mohenjo-daro, Pakistan, 1982-83, by the IsMEO-
REFERENCES I 381
382 I REFERENCES
1978
Information Sources and the Development of Decision-Making Organizations. In Social Archaeology: BeyondSubsistence and Dating, edited bye. L. Redman,]. Berman, E. V. Curtin, W. T. Langhorne,Jr., N. M. Versaggi, and]. C.
Wanser, pp. 87-112. Academic Press, New York.
1980
Spatial Organization of Early Uruk Settlement Systems. In L'Archeologie de
l'Iraq du debut de l'epoque Neolithique a 333 avant notre ere, pp. 233-63. Colloques Internationaux 580. Centre National de la Recherche Scientifique,
Paris.
1982
Organizational Structure and Scalar Stress. In Theory and Explanation in Archaeology: The Southampton Conference, edited by e. Renfrew, M. Rowlands,
and B. Segraves, pp. 389-421. Academic Press, New York.
1983 Decision-Making Organization and Pastoral Nomad Camp Size. Human Ecology 11:175-99.
1987
The Changing Organization of Uruk Administration on the Susiana Plain. In
The Archaeology ofWestern Iran: Settlement and Societyfrom Prehistory to the Islamic
Conquest, edited by F. Hole, pp. 107-39. Smithsonian Institution Press, Washington, D.e.
1988 - 89 Late Uruk in Greater Mesopotamia: Expansion or Collapse? Origini 14:
595-613.
Jones, e.
1991
Cycles of Growth at Tikal. In Classic Maya Political History, edited by T. P. Culbert, pp. 102-27. School of American Research Advanced Seminar Series,
Cambridge University Press, Cambridge.
Jones, T. B.
1981
Foreword. In The City-State in Five Cultures, edited by R. Griffeth and e. G.
Thomas, pp. ix-xii, ABC-Clio, Santa Barbara.
Joshi,]. P.
1986
Settlement Patterns in Third, Second and First Millennia in India-With Special Reference to Recent Discoveries in Punjab. In Rtambhara: Studies in Indology, edited by K. e. Varma, M. e. Bhartiya, L. B. Ram Anant, and T. Acharya,
pp. 134-39. Society for Indic Studies, Ghaziabad.
1990
Excavation at Surkotada 1971-72 and Exploration in Kutch. Memoirs 87. Archaeological Survey ofIndia, Delhi.
Julien, c.j,
1983
HatunqoIla: A View ofInca Rulefrom theLake Titicaca Region. Publications in Anthropology 15, University of California, Berkeley and Los Angeles.
Kahn,]. S.
1985 Peasant Ideologies in the Third World. Annual Review of Anthropology 14:
49-75.
Kaiser, W.
1987
Zum Friedhof der Naqadakultur von Minshat Abu Omar. Annales du Service
des Antiquites de I'Egypte 71: 119-25.
1990
Zur Entstehung des gesamtagyptischen Staates. Mitteilungen des Deutschen
Archaologischen Instituts, Abteilung Kairo 46: 287 -99.
Kamp, K. A., and N. Yoffee
1980 Ethnicity in Ancient Western Asia: Archaeological Assessments and Ethnoarchaeological Prospectives. Bulletin ofthe American Schools of Oriental Research
237:88-104.
Kanawati, N.
Governmental Reforms in Old Kingdom Egypt. Aris and Phillips, Warminster.
1980
REFERENCES I 383
Kantor, H.].
1992 The Relative Chronology of Egypt and Its Foreign Correlations before the
First Intermediate Period. In Chronologies in Old World Archaeology, 3rd ed.,
edited by R. Ehrich, pp. 3-21. University of Chicago Press, Chicago.
Kasarda,]. D.
1974 The Structural Implications of Social System Size: A Three-Level Analysis.
American Sociological Review 39: 19-28.
Katz, D.
1993
Gilgamesh and Akka. Styx, Groningen.
1995 Enmebaragesi King ofKish a Sister ofGilgameS? NouveIles Assyriologiques Breves
et Utilitaires 2(June) :24.
Kautz, R., and G. D. Jones (editors)
1981
The Transition to Statehood in the New World. Cambridge University Press,
Cambridge.
Keatinge, R. W.
1982 The Chimu Empire in a Regional Perspective: Cultural Antecedents and
Continuities. In Chan Chan: Andean Desert City, edited by M. E. Moseley and
K. e. Day, pp. 197-224. University of New Mexico Press, Albuquerque.
Kees, H.
1961 Ancient Egypt: A Cultural Topography. University of Chicago Press, Chicago.
Keightley, D. N.
1983 The Late Shang State: When, Where, and What? In The Origins of Chinese
Civilization, edited by D. N. Keightley, pp. 523-64. University of California
Press, Berkeley.
Kelly, R. e.
1993
Constructing Inequality: The Fabrication of a Hierarchy of Virtue among the Etoro.
University of Michigan Press, Ann Arbor.
Kemp, B.].
1972 Temple and Town in Ancient Egypt. In Man, Settlement and Urbanism, edited
by P.]. Ucko, R. Tringham, and G. W. Dimbleby, pp. 657-80. Duckworth,
London.
1977 The Early Development of Towns in Egypt. Antiquity 51: 185-200.
1981 The Character of the South Suburb at Tell el-Amarna. Mitteilungen der
Deutschen Orient-Gesellschaft 113:81-97.
1989 Ancient Egypt: Anatomy ofa Civilization. Routledge, London and New York.
Kennedy, K. A. R.
1981 Skeletal Biology: When Bones Tell Tales. Archaeology 34(1) : 17- 24, 51.
Kenoyer,]. M.
1991 The Indus Valley Tradition ofPakistan and Western India. Journal ofWorld Prehistory 5:331-85.
1992 Harappan Craft Specialization and the Question of Urban Segregation and
Stratification. The Eastern Anthropologist 45 (1-2) :39 - 54.
1994 The Harappan State, Was It or Wasn't It? In From Sumer to Meluhha: Contributions to the Archaeology ofSouth and West Asia in Memory of George F. Dales, Jr.,
edited by]. M. Kenoyer, pp. 71-80. Prehistory Press, Madison, Wisconsin.
Kent, S.
1993 Sharing in an Egalitarian Kalahari Community. Man (n.s.) 28 :479-514.
Kertzer, D. I.
1988 Ritual, Politics, and Power. Yale University Press, New Haven.
Khan, F. A.
1965 Excavations at Kot Diji. Pakistan Archaeology 2: 11-85.
384 I REFERENCES
REFERENCES I 385
386 I REFERENCES
Ninurta Mythology in the Babylonian Epic of Creation. In Keilschriftliche Literaturen, edited by K. Hecker and W. Sommerfeld, pp. 55-60. Dietrich
Reimer, Berlin.
Lambert, W. G., and A. R. Millard
1969 Atra-hasfs: The Babylonian Story if the Flood. Clarendon Press, Oxford.
Landa, D. de
1941 Landa's reladon de las cosas de Yucatan. Edited and annotated by A. M. Tozzer.
Papers of the Peabody Museum 18. Harvard University, Cambridge.
Landsberger, B.
1976
The Conceptual Autonomy if the Babylonian World. Translated by T. Jacobsen,
B. Foster, and H. von Siebenthal. Sources and Monographs on the Ancient
Near East 1(4). Undena, Malibu, California. (Originally published in 1926,
Die Eigenbegrifllichkeit der babylonischen Welt. Islamica 2: 355 -72.)
Lange, K., and M. Hirmer
1968 Egypt: Architecture, Sculpture, Painting. 4th ed. Phaidon, London.
Laporte, J. P., and V. Fialko
1990
New Perspectives on Old Problems: Dynastic References for the Early Classic
at Tikal. In Vision and Revision in Maya Studies, edited by F. S. Clancy and
P. D. Harrison, pp. 33 - 66. University of New Mexico Press, Albuquerque.
Larsen, C. E.
1983 Life and Land Use on the Bahrain Islands: The Geoarchaeology if an Ancient Society.
University of Chicago Press, Chicago.
Larsen, M. T.
1976
The Old Assyrian City-State and Its Colonies. Akademisk Forlag, Copenhagen.
1977
Partnerships in Old Assyrian Trade. Iraq 39: 119-45.
1982
Your Money or Your Life! A Portrait of an Assyrian Businessman. In Societies
and Languages if the Ancient NearEast: Studies in Honor if I. M. Diakonof], edited
by M. Dandamayev and J. N. Postgate, pp. 214-45. Aris and Phillips,
Warminster.
1987a Commercial Networks in the Ancient Near East. In Centre and Periphery in
the Ancient World, edited by M. Rowlands, M. T. Larsen, and K. Kristiansen,
pp. 47-56. Cambridge University Press, Cambridge.
1987b The Mesopotamian Lukewarm Mind: Reflections on Science, Divination and
Literacy. In Language, Literature, and History: Philological and Historical Studies
Presented to Erica Reiner, edited by F. Rochberg-Halton, pp. 203-25. American Oriental Series, vol. 67. American Oriental Society, New Haven.
1988
Introduction: Literacy and Social Complexity. In State and Society: The Emergence and Development if Social Hierarchy and Political Centralization, edited
by J. Gledhill, B. Bender, and M. T. Larsen, pp. 173-91. Unwin Hyman,
London.
1992
The Collapse of Civilizations: The Case of Mesopotamia. In Clashes of Cultures: Volume in Honour if Niels Steensgard, edited by J. Christian, V.Johansen,
E. Ladewig Petersen, and H. Stevnsborg, pp. 107-29. Odense University
Press, Odense.
Larsen, M. T. (editor)
1978 Power and Propaganda: A Symposium on Ancient Empires. Mesopotamia, vol. 7.
Akademisk Forlag, Copenhagen.
Laufer, B.
The Development of Ancestral Images in China. In Reader in Comparative
1965
Religion: An Anthropological Approach, edited by W. A. Lessa and E. Z. Vogt,
pp. 445 -50. Harper and Row, New York.
1986
Leach, E. R.
1954 Political Systems of Highland Burma: A Study of Kachin Social Structure. Beacon
Press, Boston.
Leahy, A.
The Libyan Period in Egypt: An Essay in Interpretation. Libyan Studies 16:
1985
51-65.
Le Brun, A.
1978
Le niveau 17B de 1'Acropole de Suse (campagne de 1972). Cahiers de la
Delegation Archeologique Franiaise en Iran 9: 57 -154. Association Paleorient,
Paris.
1980
Les 'ecuelles grossieres': etat de la question. In L'archeologie de l'Iraq: perspectives et limites de l'interpretation anthropologiques des documents, edited by M.- T.
Barrelet, pp. 59-70. Colloques Internationaux 580. Centre National de la
Recherche Scientifique, Paris.
1985
Le Niveau 18 de l'Acropole de Suse. Memoire d'argile, memoire du temps.
Paleorient 11(2) :31-36.
Le Brun, A., and F. Vallat
1978
L'origine de l' ecriture aSuse. Cahiers de la Delegation Archeologique Francaise en
Iran 9: 11-59. Association Paleorient, Paris.
Lehman, E. H.
1969
Toward a Macrosociology of Power. American Sociological Review 34: 453 - 65.
Lekson, S. H.
1985 Largest Settlement Size and the Interpretation of Socio-Political Complexity
at Chaco Canyon, New Mexico. Haliksa'i: UNM Contributions to Anthropology
4:68-75.
Lenski, G.
1966 Power and Privilege: A Theory if Social Stratification. McGraw Hill, New York.
Lewis, H. S.
1974a Leaders and Followers: Some Anthropological Perspectives. An AddisonWesley Module in Anthropology 50: 1-25.
1974b Neighbors, Friends, and Kinsmen: Principles of Social Organization Among
the Cushitic-Speaking Peoples of Ethiopia. Ethnology 13: 145 -57.
Lichtheim, M.
1973 Ancient Egyptian Literature: A Book if Readings I: The Old and Middle Kingdoms.
University of California Press, Berkeley.
Lieberman, S. J.
1980
Of Clay Pebbles, Hollow Clay Balls, and Writing: A Sumerian View. American Journal if Archaeology 84: 339-58.
Limet, H.
1971 Les legendes des sceaux cassites. Palais des Academies, Brussels.
Linne, S.
1942 Mexican Highland Cultures: Archaeological Researches at Teotihuacsn, Calpulalpan,
and Chalchicomula in 1934-35. Publication 7 (n.s.). Ethnographic Museum of
Sweden, Stockholm.
Liverani, M.
1990 Prestige and Interest: International Relations in the Near East ca. 1600-1100 B.C.
Studies in History of the Ancient Near East 1. Sargon, Padua.
1993
Model and Actualization: The Kings of Akkad in the Historical Tradition. In
Akkad: The First World Empire, edited by M. Liverani, pp. 41-68. Sargon,
Padua.
388 I REFERENCES
Liverani, M. (editor)
1993 Akkad: The First World Empire. Sargon, Padua.
Lloyd, P. e.
1965 The Political Structure of African Kingdoms: An Exploratory Model. In Political Systems and the Distribution cif Power, edited by M. Banton, pp. 63-112.
Tavistock, London.
Longyear,]. M.
1952
Copan Ceramics: A Study cif Southeastern Maya Pottery. Carnegie Institution of
Washington, vol. 597. Washington, D.e.
Lothrop, S. K.
1937
Code: An Archaeological Study cif Central Panama, Part I. Peabody Museum of
Ethnology and Archaeology, vol. 7. Harvard University, Cambridge.
Lowenthal, D.
1985
The PastIs a Foreign Country. Cambridge University Press, Cambridge.
Luttwak, E. N.
1976
The Crand Strategy cif the Roman Empirefrom the First Century A.D. to the Third.
]ohns Hopkins University Press, Baltimore.
Machinist, P. B.
1984/85 The Assyrians and Their Babylonian Problem: Some Reflections. Wissenscheiftskolleg zu Berlin jahrbuch 1984/1985: 353 - 64.
On Self-Consciousness in Mesopotamia. In The Origins and Diversity cif Axial
1986
Age Civilizations, edited by S. N. Eisenstadt, pp. 193-202, 511-18. SUNY
Series in Near Eastern Studies, State University of New York Press, Albany.
1992
Order and Disorder: Some Reflections on Mesopotamian Mythology. Manuscript on file, Department of Near Eastern Studies, University of Michigan,
Ann Arbor.
Mackay, E.]. H.
1937-38 Further Excavations at Mohenjo-daro, 2 vols. Government ofIndia, Delhi.
1943
Chanhu-Daro Excavations 1935-36. American Oriental Series, vol. 20. American Oriental Society, New Haven.
Mackey, e., and A. M. U. Klymyshyn
1990
The Southern Frontier of the Chimu Empire. In The Northern Dynasties:
Kingship and Statecraft in Chimor, edited by M. Moseley and A. Cordy-Collins,
pp. 195-226. Dumbarton Oaks, Washington, D.e.
Maekawa, K.
1987
Collective Labor Service in Girsu-Lagash: The Presargonic and Ur III Periods.
In Labor in the Ancient Near East, edited by M. Powell, pp. 49-71. American
Oriental Society, New Haven.
Maher, V.
1984
Possession and Dispossession: Maternity and Mortality in Morocco. In Interest
and Emotion: Essays on the Study cif Family and Kinship, edited by H. Medick
and D. W. Sabean, pp. 103-28. Cambridge University Press, Cambridge.
Maidman, M.
1984
Kassites among the Hurrians: A Case-Study from Nuzi. Bulletin cif the Society
for Mesopotamian Studies 8: 15 - 21.
Malik, S. e.
1968
Indian Civilization: The Formative Period. Indian Institute of Advanced Study,
Simla.
1979
Changing Perspectives of Archaeology and Interpreting Harappan Society.
In Essays in Indian Protohistory, edited by D. P. Agrawal and D. Chakrabarti,
pp. 187-204. B. R. Publishing, Delhi.
Mann,M.
1984 The Autonomous Power of the State. Europeanjournal cif Sociology 25: 185- 213.
The Sources of Social Power, Volume I: A History cif Powerfrom the Beginning to A.D.
1986
1760. Cambridge University Press, Cambridge.
1987 The Autonomous Power of the State: Its Origins, Mechanisms and Results. In
States in History, edited by]. A. Hall, pp. 109-36. Basil Blackwell, Oxford.
Manzanilla, L.
1993 Los conjuntos residenciales teotihuacanos. In Anatomla de un conjunto residencial
teotihuacano en Oztoyahualco, vol. I: las excauaciones, edited by L. Manzanilla,
pp. 31-46. Instituto de Investigaciones Antropol6gicas, Universidad Aut6noma de Mexico, Mexico, D.F.
Manzanilla, L. (editor)
1993 Anatomla de un conjunto residencial teotihuacano en Oztoyahualco. 2 vols. Instituto
de Investigaciones Antropol6gicas, Universidad Aut6noma de Mexico, Mexico, D.F.
Marcus, ].
1973 Territorial Organization of the Lowland Classic Maya. Science 180: 911-16.
1974 An Epigraphic Approach to the Territorial Organization cif the Lowland Classic
Maya. Ph.D. dissertation, Department of Anthropology, Harvard University,
Cambridge.
1976a Emblem and State in the Classic Maya Lowlands: An Epigraphic Approach to Territorial Organization. Dumbarton Oaks, Washington, D.e.
1976b Iconography of Militarism at Monte Alban and Neighboring Sites in the Valley of Oaxaca. In The Origins cif Religious Art and Iconography in Preclassic
Mesoamerica, edited by H. B. Nicholson, pp. 123-39. UCLA Latin American
Center, Los Angeles.
1976c The Origins of Mesoamerican Writing. Annual Review of Anthropology 5:
35-67.
1978 Archaeology and Religion: A Comparison of the Zapotec and Maya. World
Archaeology 10:172-91.
1980 Zapotec Writing. Scientific American 242: 50 - 64.
1983a Lowland Maya Archaeology at the Crossroads. American Antiquity 48 :454 - 88.
1983b On the Nature of the Mesoamerican City. In Prehistoric Settlement Patterns:
Essays in Honor of Cordon R. WilIey, edited by E. Z. Vogt and R.Leventhal,
pp. 195-242. University of New Mexico Press, Albuquerque.
1983c Stone Monuments and Tomb Murals of Monte Alban IlIa. In The Cloud
People: Divergent Evolution cif the Zapotec and Mixtec Civilizations, edited by
K. V. Flannery and]. Marcus, pp. 137-43. Academic Press, New York.
1983d Changing Patterns of Stone Monuments after the Fall of Monte Alban,
A.D. 600-900. In The Cloud People: Divergent Evolution cif the Zapotec and Mixtee Civilizations, edited by K. V. Flannery and]. Marcus, pp. 191-97. Academic
Press, New York.
1983e Rethinking the Zapotec Urn. In The Cloud People: Divergent Evolution of
the Zapotec and Mixtec Civilizations, edited by K. V. Flannery and]. Marcus,
pp. 144-48. Academic Press, New York.
1984 Mesoamerican Territorial Boundaries: Reconstructions from Archaeology and
Hieroglyphic Writing. Archaeological Reviewfrom Cambridge 3(2) :48 - 62. Cambridge, England.
1987a The Inscriptions of Calakmul:Royal Marriage at a Maya City in Campeche, Mexico.
Technical Reports 21. Museum of Anthropology, University of Michigan,
Ann Arbor.
390 I REFERENCES
REFERENCES I 391
Martin-Pardey, E.
1976
Untersuchungen zur iigyptischen Provinzialverwaltung biszum Ende des Alten Reiches.
Hildesheimer Agyptologische Beitrage 1. Gerstenberg, Hildesheim.
Mathews, P.
1985 Maya Early Classic Monuments and Inscriptions. In A Consideration if theEarly
Classic Period in the Mdya Lowlands, edited by G. R. Wiliey and P. Mathews,
pp. 5-54. Institute for Mesoamerican Studies 10. State University of New
York at Albany.
Matthews, R.
1993 Cities, Seals, and Writing: Archaic Seal Impressions from Jemdet Nasr and Ur.
Gebriider Mann, Berlin.
Mayr, E.
1982 The Growth if Biological Thought. Harvard University Press, Cambridge.
McCulloch, W. S.
1945 A Heterarchy of Values Determined by the Topology of Neural Nets. Bulletin
if Mathematical Biophysics 7 :89 - 93.
McDowell, A. G.
1992a Agricultural Activity by the Workmen of Deir el-Medina. Journal if Egyptian
Archaeology 78: 195-206.
1992b Awareness of the Past in Deir el-Medina. In Village Voices: Proceedings oJ
the Symposium "Texts from Deir el-Medina and Their Interpretation," Leiden,
May 31-June 1, 1991, edited by R.-J. Dernaree and A. Egberts, pp. 95-109.
Centre of Non-Western Studies, Leiden.
McEwan, G. F.
1985 Excavaciones en Pikillacta: un sitio wari. DiaIogo Andino 4: 89-136. Arica,
Chile.
1991
Investigation at the Pikillacta Site: A Provincial Huari Center in the Valley of
Cuzco. In Huari Administrative Structure: Prehistoric MonumentalArchitecture and
State Government, edited by W. H. Isbell and G. F. McEwan, pp. 93-119.
Dumbarton Oaks, Washington, D.e.
McGuire, R.
1983 Breaking Down Cultural Complexity: Inequality and Heterogeneity. Advances
in Archaeological Method and Theory, vol. 6, edited by M. B. Schiffer, pp. 91142. Academic Press, New York.
McGuire, R., and D. Saitta
1996 Although They Have Petty Captains, They Obey Them Badly: The Dialectics
of Prehispanic Western Pueblo Social Organization. American Antiquity 61:
197-216.
Meadow, R. H. (editor)
1991 Harappa Excavations 1986-1990: A Multidisciplinary Approach to Third Millennium Urbanism. Prehistory Press, Madison, Wisconsin.
Meiliassoux, e.
1978
"The Economy" in Agricultural Self-Sustaining Societies: A Preliminary
Analysis. In Relations if Production: Marxist Approaches to Economic Anthropology,
edited by D. Seddon, pp. 127-57. Frank Cass, London.
Mendelssohn, K.
1974 The Riddle if the Pyramids. Thames and Hudson, London.
Michailidou, A.
1989 Knossos: A Complete Guide to the Palace if Minos. Ekdotike Athenon, Athens.
REFERENCES I 393
392 I REFERENCES
Michalowski, P.
1983
History as Charter: Some Observations on the Sumerian King List. Journal of
the American OrientalSociety 103:237-48.
1987
Charisma and Control: On Continuity and Change in Early Mesopotamian
Bureaucratic Systems. In The Organization ofPower: Aspects ofBureaucracies in
the Ancient Near East, edited by M. Gibson and R. D. Biggs, pp. 55-67. Oriental Institute, University of Chicago, Chicago.
1990 Early Mesopotamian Communicative Systems: Art, Literature, and Writing. In Investigating Artistic Environments in the Ancient Near East, edited by
A. Gunter, pp. 56-69. Smithsonian Institution Press, Washington, D.e.
1993a Memory and Deed: The Historiography of the Political Expansion of the
Akkad State. In Akkad: The First World Empire, edited by M. Liverani, pp. 6990. Sargon, Padua.
1993b Tokenism. American Anthropologist 95: 996-99.
1993c On the Early Toponymy of Sumer: A Contribution to the Study of Early
Mesopotamian Writing. In Kinattiitu sa darati: Rafael Kutscher Memorial Volume,
edited by A. Rainey, pp. 199-233. Institute of Archaeology, Tel Aviv.
1994 Early Literacy Revisited. In Literacy: Interdisciplinary Conversations, edited by
D. Keller-Cohen, pp. 49-70. Hampton Press, Cresskill, New Jersey.
Midant-Reynes, B.
1992
Prehistoire de l'Egypte, des premiers hommes aux premiers pharaons. Armand Colin,
Paris.
Miller, D.
1985
Ideology and the Harappan Civilization. Journal ofAnthropological Archaeology
4:34-71.
1989
The Limits of Dominance. In Domination and Resistance, edited by D. Miller,
M. Rowlands, and e. Tilley, pp. 61-79. Unwin Hyman, London.
Miller, M. E.
1986
The Murals ofBonampak. Princeton University Press, Princeton.
Miller, N.
The Use of Dung as Fuel: An Ethnographic Example and an Archaeological
1984
Application. Paleorient 10(2): 71-79.
Miller, W. B.
1955
Two Concepts of Authority. American Anthropologist 57 :271-89.
Millett, P.
1989
Patronage and Its Avoidance in Classical Athens. In Patronage in Ancient Society, edited by A. Wallace-Hadrill, pp. 15-47. Roudedge and Kegan Paul,
London.
Millon, R.
1973
Urbanization at Teotihuacan, Mexico, vol. I, The Teotihuacan Map, Part I, Text.
University of Texas Press, Austin.
1976
Social Relations in Ancient Teotihuacan. In The Valley ofMexico: Studiesin PreHispanic Ecology and Society, edited by E. R. Wolf, pp. 205-48. School of
American Research Advanced Seminar Series, University of New Mexico
Press, Albuquerque.
Where Do They All Come From? The Provenance ofthe Wagner Murals from
1988
Teotihuacan. In Feathered Serpents and Flowering Trees: Reconstructing the Murals
ofTeotihuacan, edited by K. Berrin, pp. 78-113. The Fine Arts Museum, San
Francisco.
Tcotihuacan Studies: From 1950 to 1990 and Beyond. In Art, Ideology, and the
City of Teotihuacan, edited by J. e. Berlo, pp. 339-429. Dumbarton Oaks,
Washington, D.e.
1993
The Place Where Time Began: An Archaeologist's Interpretation of What
Happened in 'Teotihuacan History. In Teotihuacan, edited by K. Berrin and
E. Pasztory, pp. 17-43. Thames and Hudson, New York.
Millon, R., and S. Sugiyarna
1991
Concentraci6n de pinturas murales en el Conjunto Arquitect6nico Grande, al
este de la Plaza de la Luna. In Teotihuacan 1980-1982: nuevas interpretaciones,
pp. 211-31. Serie Arqueologia, INAH, Mexico.
Miroschedji, P. de
Un four de po tier du IVe millenaire sur le tell de l'Apadana. Cahiers de la
1986
Delegation Archeologique Franiaise en Iran 7: 13- 46. Association Paleorient, Paris.
Moore, B.,Jr.
1978
Injustice: The Social Bases ofObedience and Revolt. M. E. Sharpe, White Plains,
New York.
Moore, S. F. (editor)
1993
Moralizing States and the Ethnography ofthe Present. Monograph Series 5. American Ethnological Society, Washington, D.e.
Moorey, P. R. S.
1978
Kish Excavations 1923-1933. Oxford University Press, Oxford.
1985
Materials andManufacture in Ancient Mesopotamia: The Evidence ofArchaeology and
Art. Metals and Metalwork, Glazed Materials and Glass. BAR International Series 237. British Archaeological Reports, Oxford.
1994
Ancient Mesopotamian Materials and Industries: The Archaeological Evidence.
Clarendon Press, Oxford.
Morelos Garcia, N.
1991 Esculturas y arquitectura en el Conjunto Arquitect6nico. In Teotihuaoin 19801982: nuevas interpretaciones, pp. 193 - 201. Serie Arqueologia, INAH, Mexico.
Morenz, S.
1969
Prestige- Wirtschaft im altenAgypten. Sitzungsberichte der Bayerischen Akademie
der Wissenschaften, Philosophisch-Historische Klasse fascicle 4. Bayerische
Akademie/C, H. Beck, Munich.
Morley, S. G.
1920
The Inscriptions at Copan. Carnegie Institution ofWashington, vol. 219. Washington, D.e.
Morris, e.
1967
Storage in Tawantinsuyu. Ph.D. dissertation, Department of Anthropology,
University of Chicago, Chicago.
1972
State Setdements in Tawantinsuyu: A Strategy of Compulsory Urbanism. In
Contemporary Archaeology: A Guide to Theory and Contributions, edited by M. P.
Leone, pp. 393-401. Southern Illinois University Press, Carbondale.
1974
Reconstructing Patterns of Non-Agricultural Production in the Inca Economy: Archaeology and Ethnohistory in Institutional Analysis. In Reconstructing
Complex Societies, edited by e. B. Moore, pp. 49-68. Supplement to the Bulletin of the American Schools of Oriental Research 20. Cambridge, Massachusetts.
1981
Tecnologia y organizaci6n inca del almacenamiento de viveres en la sierra. In
La tecnologfa en el mundo andino, vol. 1, edited by H. Lechtman and A. M. Soldi,
pp. 327-75. Universidad Nacional Aut6noma de Mexico, Mexico, D.F.
1992
1982
REFERENCES I 395
Mudar, K.
1988 The Effects of Context on Bone Assemblages: Examples from the Uruk Period in Southwestern Iran. Paleorient 14: 151-68.
Mughal, M. R.
1982 Recent Archaeological Research in the Cholistan Desert. In Harappan Civilization: A Contemporary Perspective, edited by G. L. Possehl, pp. 85-95. Oxford
and IBH, and the American Institute of Indian Studies, Delhi.
1997
Ancient Cholistan: Archaeology and Architecture. Firozsons, Lahore.
Miiller-Wollermann, R.
1986
Krisenfaktoren im iigyptischen Staat des ausgehenden Alten Reiches. PhD. dissertation, University ofTiibingen, Tiibingen,
Murra,J. V.
1960 Rite and Crop in the Inca State. In Culture in History: Essays in Honor cif Paul
Radin, edited by S. Diamond, pp. 393-407. Columbia University Press, New
York.
1962a Cloth and Its Function in the Inca State. American Anthropologist 64:710-28.
1962b An Archaeological "Restudy" of an Andean Ethnohistorical Account. American Antiquity 28: 1-4.
1964 Una apreciacion etno16gica de la visita. In Visita hecha a la provincia de Chucuito
por Garci Diez de San Miguel en 1567, documentos regionales para la etnologia y la
etnohistoria andina, vol. 1, pp. 421-42. Casa de la Cultura, Lima.
1980
The Economic Organization cif the Inka State. JAI Press, Greenwich, Connecticut. (PhD. dissertation, Department of Anthropology, University of Chicago,
1955.)
1986 The Expansion of the Inka State: Armies, War, and Rebellions. In Anthropological History cif Andean Polities, edited by J. v. Murra, N. Wachtel, and
J. Revel, pp. 49-58. Cambridge University Press, Cambridge and London,
and Editions de la Maison de Sciences de l'Homme, Paris (originally in French
in Annales [ESCj, vol. 33, nos. 5-6, Paris, 1978).
Naroll, R.
1956 A Preliminary Index of Social Development. American Anthropologist 58: 687715.
1967 Imperial Cycles and World Order. Peace Research Society Papers 7: 83-101.
Neely, J. A., and H. Wright
1994
Early Settlement and Irrigation on the Deh Luran Plain. Technical Reports 26.
Museum of Anthropology, University of Michigan, Ann Arbor.
Netting, R. M.
1990 Population, Permanent Agriculture, and Politics: Unpacking the Evolutionary
Portmanteau. In The Evolution of Political Systems: Sociopolitics in Small-Scale
Sedentary Societies, edited by S. Upham, pp. 21-61. School of American Research Advanced Seminar Series, Cambridge University Press, Cambridge.
Nicholas, L. M.
1989 Land Use in Prehispanic Oaxaca. In Monte Alban's Hinterland, Part 11: The Prehispanic Settlement Patterns in Tlacolula, Et/a, and Ucotlan, the Valley of Oaxaca,
Mexico, by S. A. Kowalewski, G. M. Feinman, L. Finsten, R. E. Blanton, and
L. M. Nicholas, pp. 449-505. Memoirs 23. Museum of Anthropology, University of Michigan, Ann Arbor.
Nissen, H. J.
1970 Grabung in den Quadraten K/L XII in Uruk-Warka. Baghdader Mitteilungen
5: 102-91.
396 I REFERENCES
1982
REFERENCES I 397
Oliver, D. L.
1974 Ancient Tahitian Society, Vols. 1-3. University Press of Hawai'i, Honolulu.
Oppenheim, A. L.
1975
The Position of the Intellectual in Mesopotamian Society. Daedalus (Spring):
37- 46. Proceedings of the American Academy ofArts and Sciences 104(2).
1977 AncientMesopotamia: Portrait of a Dead Civilization. 2nd ed. Edited by E. Reiner.
University of Chicago Press, Chicago. (Originally published in 1964.)
Otterbein, K.
1968
Internal War: A Cross-Cultural Study. American Anthropologist 70: 277- 89.
1973
The Anthropology of War. In Handbook of Social and Cultural Anthropology,
edited by]. Honigmann, pp. 923-58. Rand-McNally, Chicago.
Paddock,].
1983 Lord 5 Flower's Family: Rulers of Zaachila and Cuilapan. Publications in Anthropology 29. Vanderbilt University, Nashville.
Parkinson, R. B.
1998
The Dream and the Knot: Contextualizing Middle Kingdom Literature. In
Diftning Egyptian Literature: Proceedings of the Symposium "Ancient Egyptian
Literature, History and Forms," Los Angeles, March 24-26, 1995, edited by
G. Moers. Lingua Aegyptia Studia Monographica 2. Gottingen, forthcoming.
Parpola, S.
1983
Assyrian Library Records. Journal ofNear Eastern Studies 42: 1-30.
Parsons,]. R.
1971 Prehistoric Settlement Patterns of the Texcoco Region, Mexico. Memoirs 5. Museum
of Anthropology, University of Michigan, Ann Arbor.
Parsons, T.
1960
Max Weber. American Sociological Review 25: 750 -52.
1967 Social Theory and Modern Society. The Free Press, New York.
Pasternak, B.
1976 Introduction to Kinship and Social Organization. Prentice-Hall, Englewood Cliffs,
New Jersey.
Pasztory, E.
1988
A Reinterpretation of Teotihuacan and Its Mural Painting Tradition. In Feathered Serpents and Flowering Trees, edited by K. Berrin, pp. 45-77. The Fine Arts
Museum, San Francisco.
1992
Abstraction and the Rise of a Utopian State at Teotihuacan. In Art, Ideology,
and the City of Teotihuacan, edited by]. e. Berlo, pp. 281-320. Dumbarton
Oaks, Washington, D.e.
1997 Teotihuaoin: An Experiment in Living. University ofOklahorna Press, Norman.
Pattee, H. H. (editor)
1973 Hierarchy Theory: The Challenge of Complex Systems. George Braziller, New
York.
Patterson, T. C.
1991 The Inka Empire: The Formation and Disintegration of a Pre-Capitalist State. Berg,
Herndon, Virginia.
Patterson, T. C.; and e. W. Gailey (editors)
1987 Power Relations and State Formation. Special publication of the Archeology Section, American Anthropological Association, Washington, D.e.
Pendergast, D. M.
1985
Lamanai, Belize: An Updated View. In The LowlandMaya Postclassic, edited by
A. F. Chase and P. M. Rice, pp. 91-103. University of Texas Press, Austin.
398 I REFERENCES
1986
Stability Through Change: Lamanai, Belize from the 9th to the 17th Century.
In Late Lawland Maya Civilization: Classic to Postclassic, edited by J. A. Sabloff
and E. W. Andrews, pp. 223-49. School of American Research Advanced
Seminar Series, University of New Mexico Press, Albuquerque.
Peregrine, P. N.
1991
Some Political Aspects of Craft Specialization. World Archaeology 23: 1-11.
1992
Mississippian Evolution: A World-System Perspective. Prehistory Press, Madison,
Wisconsin.
Peregrine, P. N., and G. M. Feinman (editors)
1996
Pre-Columbian World Systems. Prehistory Press, Madison, Wisconsin.
Perkins, A. L.
1949
The Comparative Archaeology if Early Mesopotamia. Studies in Ancient Oriental
Civilization 25. Oriental Institute, University of Chicago, Chicago.
Petrie, W. M. F.
1900
Sequences in Prehistoric Remains. Journal if the Anthropological Institute if Great
Britain and Ireland 29:295-301.
1901
DiospoIis Parva: The Cemeteries of Abadiyeh and Hu, 1898-9. Memoir 20. The
Egypt Exploration Fund, London.
Pettinato, G.
1971
Das altorientaIische Menschenbild und die sumerischen und akkadischen Schopfungsmythen. Carl Winter, Universitatsverlag, Heidelberg.
Pierce, R. H.
1972
Three DemoticPapyri in the Museum: A Contribution to the Study if Contracts and
Their Instruments in Ptolemaic Egypt. Symbolae Osloenses, Supplement 24. Universitetsforlaget, Oslo.
Piggott, S.
1950
Prehistoric India to 1000 B.e. Penguin Books, Baltimore.
Pincemin, S.
1994
Entierro en el Palacio. Colecci6n: Arqueologia. Universidad Aut6noma de Campeche, Mexico.
Pittman, H.
1993
Pictures of an Administration: The Late Uruk Scribe at Work. In Between
the Rivers and Over the Mountains, edited by M. Frangipane, H. Hauptmann,
M. Liverani, P. Matthiae, and M. Mellink, pp. 235-246. Universita di Roma
"La Sapienza," Roma.
Pokora, T.
1978
China. In The Early State, edited by H. J. M. Claessen and P. Skalnik, pp. 191212. Mouton, The Hague.
Polanyi, K.
1957
The Economy as an Instituted Process. In Trade and Markets in the Early Empires: Economies in History and Theory, edited by K. Polanyi, C. M. Arensberg,
and H. W. Pearson, pp. 243-70. The Free Press, New York.
Pollock, S.
1992
Bureaucrats and Managers, Peasants and Pastoralists, Imperialists and Traders:
Research on the Uruk ~nd Jemdet Nasr Periods in Mesopotamia. Journal of
World Prehistory 6: 297 -336.
Pope, M., and S. Pollock
1995
Trade, Tool, and Tasks: A Study of Uruk Chipped Stone Industries. Research
in Economic Anthropology 16: 227 - 65. JAI Press, Greenwich, Connecticut.
REFERENCES I 399
Porada, E.
1965
The Relative Chronology of Mesopotamia. In Chronologies in Old World Archaeology, edited by R. Ehrich, pp. 133-200. University of Chicago Press,
Chicago.
1979
Some Thoughts on the Audience Reliefs of Persepolis. In Studies in Classical
Art and Archaeology: A Tribute to Petervon Blanckenhagen, edited by G. Kopcke
and M. Moore, pp. 37-44. J. J. Augustin, Locust Valley, New York.
Posener, G.
1965
Sur l' orientation et l' ordre des points cardinaux chez les Egyptiens. In Gottinger
Vortriige vom Agyptologischen Kolloquium der Akademie am 25. und 26. August
1964, edited by S. Schott, pp. 69 -78. Vandenhoeck and Ruprecht, Gottingen.
1975 L'[anachoresis] dans I'Egypte pharaonique. In Le mondegrec: hommages Ciaire
Preaux, edited by J. Bingen, G. Cambier, and G. Nachtergael, pp. 663-69.
Editions de l'Universite de Bruxelles, Brussels.
Posener-Krieger, P.
1976
Les archives du temple funeraire de Niferirkar~-Kakai' ~es papyrus d'Abousir). 2 vols.
Bibliotheque d'Etude 65. Institut Prancais d'Archeologie Orientale, Cairo.
Possehl, G. L.
1979
Ancient Cities if the Indus. Carolina Academic Press, Durham, North Carolina.
1986
Kulli: An Exploration of Ancient Civilization in South Asia. Carolina Academic
Press, Durham, North Carolina.
1990
Revolution in the Urban Revolution: The Emergence ofIndus Urbanization.
Annual Review if Anthropology 19:261-82.
1993
The Date of Indus Urbanization: A Proposed Chronology for the Pre-Urban
and Urban Harappan Phases. In South Asian Archaeology 1991, edited by A. J.
Gail and G. J. R. Mevissen, pp. 231- 49. Franz Steiner Verlag, Stuttgart.
1996
Indus Age: The Writing System. University of Pennsylvania Press, Philadelphia.
Possehl, G. L., and P. C. Rissman
1992 The Chronology ofPrehistoric India: From Earliest Times to the Iron Age. In
Chronologies in Old WorldArchaeology, 3rd ed., edited by R. Ehrich, pp. 465-90
(vol. 1) and 447-74 (vol. 2). University of Chicago Press, Chicago.
Postgate, J. N.
1992
Early Mesopotamia: Society and Economy at the Dawn if History. Routledge,
London.
1993 The Four "Neo-Assyrian" Tablets from Seh Hamad. State Archives of Assyria
Bulletin 7: 109-24.
Potter, T. W.
1980
Rome and Its Empire in the West. In The Cambridge Encyclopedia if Archaeology, edited by A. Sherratt, pp. 232-38. Crown, New York.
Pounds, N. J. G.
1969
The Urbanization of the Classical World. Annals of the Association if American
Geographers 59: 135-57.
Powell, M. A.
1981
Three Problems in the History of Cuneiform Writing: Origins, Direction of
Script, Literacy. Visible Language 15 :419 - 40.
1989-90 Masse und Gewichte. Reallexikon derAssyriologie 7: 457 -517.
Poyer, L.
1993 Egalitarianism in the Face of Hierarchy. Journal if Anthropological Research
49:111-33.
400 I REFERENCES
REFERENCES I 401
Rathje, W. 1.
1975
Last Tango in Mayapan: A Tentative Trajectory of Production-Distribution
Systems. In Ancient Civilization and Trade, edited by ]. Sabloff and e. c.
Lamberg-Karlovsky, pp. 409-48. School of American Research Advanced
Seminar Series, University of New Mexico Press, Albuquerque.
Ratnagar, S.
1981
Encounters: The Westerly Trade cif the Harappa Civilization. Oxford University
Press, Delhi.
1991
Enquiries into the Political Organization of Harappan Society. Ravish Publishers,
Pune.
Redding, R.
1979
The Fauna Remains from Tappeh Zabarjad. In Archaeological Investigations
in Northeastern Xuzestan, 1976, edited by H. T. Wright, pp. 91-93. Technical Reports 10. Museum of Anthropology, University of Michigan, Ann
Arbor.
The Animal Remains. In An Early Town on the Deh Luran Plain: Excavations at
1981
Tepe Farukhabad, edited by H. T. Wright, pp. 233-61. Memoirs 13. Museum
of Anthropology, University of Michigan, Ann Arbor.
Redfield, R.
1953
The Primitive World and Its Transjormations. Comell University Press, Ithaca.
Redfield, R., and M. Singer
The Cultural Role of Cities. Economic Development and Culture Change 3(1):
1954
53.
Redman, e. 1.
1978
The Rise cif Civilization: From Early Farmers to Urban Society in the Ancient Near
East. W. H. Freeman, San Francisco.
The Comparative Context of Social Complexity. In Chaco and Hohokam,
1991
edited by P. 1. Crown and W.]. Judge, pp. 251-81. School of American Research Press, Santa Fe.
Redmond, E. M.
1983
A fuego y sangre: Early Zapotec Imperialism in the Cuicatlan Canada, Oaxaca. Studies in Latin American Ethnohistory and Archaeology, vol. 1. Memoirs 16. Museum of Anthropology, University of Michigan, Ann Arbor.
Redmond, E. M., and C. S. Spencer
The Cacicazgo: An Indigenous Design. In Caciques and Their People: A Volume
1994
in Honor of Ronald Spores, edited by]. Marcus and]. F. Zeitlin, pp. 189-225.
Anthropological Papers 89. Museum of Anthropology, University of Michigan, Ann Arbor.
Reents-Budet, D.
1994
Painting the Maya Universe: Royal Ceramics of the Classic Period. Duke University
Press, Durham, North Carolina.
Reiner, E.
1961
The Etiological Myth of the "Seven Sages." Orientalia 30: 1-11.
Renfrew, C.
1972
The Emergence cif Civilisation: The Cyclades and the Aegeanin the Third Millennium
B. c. Methuen, London.
1974
Beyond a Subsistence Economy: The Evolution ofSocial Organization in Prehistoric Europe. In Reconstructing Complex Societies: An Archaeological Colloquium, edited by e. B. Moore, pp. 69-95. Supplement to the Bulletin of the
American Schools of Oriental Research 20. Cambridge, Massachusetts.
REFERENCES I 403
402 I REFERENCES
1975
zation. Studies in Ancient Oriental Civilization 48. Oriental Institute, University of Chicago, Chicago.
Rounds,].
1979
Lineage, Class, and Power in the Aztec State. American Ethnologist 6:73-86.
Rowe,]. H.
1946
Inca Culture at the Time ofthe Spanish Conquest. In Handbook ofSouth American Indians: The Andean Civilizations, edited by]. H. Steward, pp. 183-330.
Bureau of American Ethnology, vol. 2, Bulletin 143. Smithsonian Institution,
Washington, D.e.
1960
Cultural Unity and Diversification in Peruvian Archaeology. In. Men and Cultures: Selected Papers ofthe FiJth International Congress of Anthropological and Ethnological Sciences, edited by A. F. e. Wallace, pp. 627-31. University of
Pennsylvania Press, Philadelphia.
1962
Stages and Periods in Archaeological Interpretation. SouthwesternJournal of Anthropology 18(1):40-54.
1979 An Account of the Shrines of Ancient Cuzco. Nawpa Pacha 17: 1- 80.
Rowlands, M.
1985 Exclusionary Tactics in the Logic of Collective Dynamics. Critique ofAnthropology 5(2):47-69.
Rowlands, M., and].-P. Warnier
1988
Sorcery, Power and the Modern State in Cameroon. Man (n.s.) 23: 118-32.
Roys, R. L.
1943
The Indian Background ofColonial Yucatan. Publication 548. Carnegie Institution of Washington, Washington, D.e.
1957
TIle Political Geography ofthe Yucatan Maya. Publication 613. Carnegie Institution of Washington, Washington, D.e.
Rudolph, S. H.
1987
Presidential Address: State Formation in Asia- Prolegomenon to a Comparative Study. Journal of Asian Studies 46(4): 731-46.
Runciman, W. G.
1982
Origins of States: The Case of Archaic Greece. Comparative Studies in Society
and History 24: 351-77.
Russell,]. e.
1958 Late Ancient and Medieval Populations. Transactions of the American Philosophical Society48(3): 1-101.
Russmann, E. R.
1995
Two Heads of the Early Fourth Dynasty. In Kunst des Alten Reiches: Symposium im Deutschen Archiiologischen Institut Kairo am 29. und 30. Oktober 1991,
pp. 111-18. Deutsches Archaologisches Institut, Abteilung Kairo, Sonderschrift 28. Philipp von Zabern, Mainz.
SablotI,]. A., and E. W. Andrews, V (editors)
1986
Late Lowland Maya Civilization. School ofAmerican Research Advanced Seminar Series, University of New Mexico Press, Albuquerque.
SablotI, J. A., and]. S. Henderson (editors)
1993
Lowland Maya Civilization in the Eighth Century A.D. Dumbarton Oaks, Washington, D.e.
SablotI,]. A., and G. Tourtellot
1991
The Ancient Maya City ofSayil: The Mapping of a Puuc Region Center. Publication
60. Middle American Research Institute, Tulane University, New Orleans.
Safar, F., M. A. Mustafa, and S. Lloyd
1981 Eridu. Iraq Ministry of Culture and Information, Baghdad.
404 I REFERENCES
Sahlins, M. D.
1958
Social Stratification in Polynesia. University of Washington Press, Seatde.
The Segmentary Lineage: An Organization of Predatory Expansion. American
1961
Anthropologist 63: 322-45.
1970
Poor Man, Rich Man, Big-Man, Chief: Political Types in Melanesia and Polynesia. In Cultures ofthe Pacific, edited by T G. Harding and B. J. Wallace,
pp. 203-15. Free Press, New York.
1972
Stone Age Economics. Aldine, Chicago.
1983
Other Times, Other Customs: The Anthropology of History. American Anthropologist 85 :517 - 44.
Sahlins, M. D., and E. R. Service (editors)
1960
Evolution of Culture. University of Michigan Press, Ann Arbor.
Saitta, D. J.
1994
Class and Community in the Prehistoric Southwest. In The Ancient Southwestern Community: Models and Methods for the Study ofPrehistoric Social Organization, edited by W. H. Wills and R. D. Leonard, pp. 25-43. University of New
Mexico Press, Albuquerque.
Saitta, D. J., and A. S. Keene
1990
Politics and Surplus Flow in Prehistoric Communities. In The Evolution ofPolitical Systems: Sociopolitics in Small-Scale Sedentary Societies, edited by S. Upham,
pp. 203-24. Cambridge University Press, Cambridge.
Sajjidi, M.
1979
The Protoelamite Period on the Izeh Plain. In Archaeological Investigations in
Northeastern Xuzestan, 1976, edited by H. T Wright, pp. 93-98. Technical
Reports 10. Museum of Anthropology, University of Michigan, Ann Arbor.
Samaniego, L., E. Vergara, and H. Bischof
1985
New Evidence on Cerro Sechin, Casma Valley, Peru. In Early Ceremonial Architecture in theAndes, edited bye. B. Donnan, pp. 165-90. Dumbarton Oaks,
Washington, D.e.
Sanders, W. T
1965
The Cultural Ecology of the Teotihuacan Valley. Manuscript on file, Department of Sociology and Anthropology, Pennsylvania State University, University Park.
1973
The Significance of Pikillacta in Andean Culture History. Occasional Papers in
Anthropology 8: 380 - 428. Pennsylvania State University, University Park.
Chiefdom to State: Political Evolution at Kaminaljuyu, Guatemala. In Recon1974
structing Complex Societies: An Archaeological Colloquium, edited by C. B. Moore,
pp. 97-116. Supplement to the Bulletin of the American Schools of Oriental
Research, 20. Cambridge, Massachusetts.
1984
Pre-Industrial Demography and Social Evolution. In On the Evolution ofComplex Societies: Essays in HonorofHarryHoijer 1982, edited by T Earle, pp. 7-39.
Undena, Malibu, California.
Sanders, W. T, A. Kovar, T Charlton, and R. A. Diehl
1970
The Teotihuacan Valley Project FinalReport, vol. 1. The NaturalEnvironment, Contemporary Occupation, and 16th Century Population ofthe Valley. Occasional Papers in Anthropology 1. Pennsylvania State University, University Park.
Sanders, W. T, J. R. Parsons, and R. S. Sandey
1979
The Basin of Mexico: Ecological Processes in the Evolution of a Civilization. Academic Press, New York.
Sanders, W. T, and B. J. Price
1968
Mesoamerica: The Evolution ofa Civilization. Random House, New York.
REFERENCES I 405
406 I REFERENCES
Seaton, S. L.
1978
The Early State in Hawai'i. In The Early State, edited by H. Claessen and
P. Skalnik, pp. 269-88. Mouton, The Hague.
SeidJmayer, S. J.
1996
Town and State in the Early Old Kingdom: A View from Elephantine. In Aspects of EarlyEgypt, edited by A.J. Spencer, pp. 108-27. British Museum Press,
London.
Sejourne, L.
1966 Arquitectura y pintura en Teotihuacan, Siglo XXI, Mexico, D.F.
Service, E. R.
1962 Primitive Social Organization: An Evolutionary Perspective. Random House, New
York.
1963 Profiles in Ethnology. Harper and Row, New York.
1975
Origins ofthe State and Civilization. Norton, New York.
Sewell, W. H., Jr.
1992
A Theory of Structure: Duality, Agency, and Transformation. American journal
ofSociology 98: 1-29.
Shaffer, A.
1983
Gilgamesh, the Cedar Forest, and Mesopotamian History. journal of the American OrientalSociety 103: 307 -13.
Shaffer, J. G.
1982
Harappan Culture: A Reconsideration. In Harappan Civilization: A Contemporary Perspective, edited by G. L. Possehl, pp. 41-50. Oxford and IBH, and the
American Institute of Indian Studies, Delhi.
1992
The Indus Valley, Baluchistan and Helmand Traditions: Neolithic through
Bronze Age. In Chronologies in Old World Archaeology, 2 vols., 3rd ed., edited
by R. W. Ehrich, pp. 441-64 (vol. 1) and 425-46 (vol. 2). University of Chicago Press, Chicago.
1993
Reurbanization: The Eastern Punjab and Beyond. In Urban Form and Meaning
in South Asia, edited by H. Spodek and D. M. Srinivasan, pp. 53-67. Studies
in the History of Art 31. Center for Advanced Study in the Visual Arts Symposium Papers 15. National Gallery of Art, Washington, D.e.
Shaffer, J. G., and D. A. Lichtenstein
1989 Ethnicity and Change in the Indus Valley Cultural Tradition. In Old Problems and
New Perspectives in SouthAsian Archaeology, edited by J. M. Kenoyer, pp. 117-26.
Wisconsin Archaeological Reports 2. University of Wisconsin, Madison.
Shanks, M., and e. Tilley
1982
Ideology, Symbolic Power and Ritual Communication: A Reinterpretation of
Neolithic Mortuary Practices. In Symbolic and Structural Archaeology, edited by
1. Hodder, pp. 129-54. Cambridge University Press, Cambridge.
Sharer, R. J.
1978
Archaeology and History at Quirigua, Guatemala. journal ofFieldArchaeology
5:51-70.
1990
Quirigua: A Classic Maya Center and Its Sculpture. Carolina Academic Press,
Durham, North Carolina.
1991
Diversity and Continuity in Maya Civilization: Quirigua as a Case Study. In
Classic Maya Political History, edited by T. P. Culbert, pp. 180 -98. School of
American Research Advanced Seminar Series, Cambridge University Press,
Cambridge.
1995
The Beginning and End of the Copan Dynasty. Paper presented at the UCLA
Maya Weekend, October 28, Los Angeles.
REFERENCES I 407
REFERENCES 1409
408 I REFERENCES
Smith, H. S.
1972
Society and Setdement in Ancient Egypt. In Man, Settlement and Urbanism,
edited by P.]. Ucko, R. Tringham, and G. W. Dimbleby, pp. 705-19. Duckworth, London.
Smith, M. Estellie
1985
An Aspectual Analysis of Polity Formations. In Development and Decline: The
Evolution ofSociopolitical Organization, edited by H. ]. M. Claessen, P. van de
Velde, and M. E. Smith, pp. 97-125. Bergin and Garvey, South Hadley, Massachusetts.
Smith, Mary Elizabeth
Codex Selden: A Manuscript from the Valley of Nochixtlan? In The Cloud
1983
People: Divergent Evolution of the Zapotec and Mixtec Civilizations, edited by
K. V. Flannery and]. Marcus, pp. 248-55. Academic Press, New York.
1994
Why the Second Codex Selden Was Painted. In Caciques and Their People:
A Volume in Honor ofRonald Spores, edited by ]. Marcus and]. F. Zeidin,
pp. 111-41. Anthropological Papers 89. Museum of Anthropology, University of Michigan, Ann Arbor.
Smith, Michael E.
1987
Imperial Strategies in the Western Portion of the Aztec Empire. Paper presented at the 86th Annual Meeting of the American Anthropological Association, Chicago.
Rhythms of Change in Postclassic Central Mexico: Archaeology, Ethnohis1992
tory, and the Braudelian Model. In Archaeology, Annales, and Ethnohistory,
edited by A. B. Knapp, pp. 51-74. Cambridge University Press, Cambridge.
Smith, Michael E., and F. F. Berdan
1992
Archaeology and the Aztec Empire. World Archaeology 23 :353-67.
Smith, W. S.
A History of Egyptian Sculpture and Painting in the Old Kingdom. 2nd ed. Mu1949
seum of Fine Arts, Boston, and Oxford University Press, London.
Snodgrass, A. M.
The Dark Age of Greece. University of Edinburgh, Edinburgh.
1971
1977
Archaeology and the Rise ofthe Greek State. Cambridge University Press, Cambridge.
Archaic Greece: The Age ofExperiment. ]. M. Dent and Sons, London.
1980
1986
Interaction by Design: The Greek City-State. In Peer Polity Interaction and
Socio-Political Change, edited by C. Renfrew and]. F. Cherry, pp. 47-58. Cambridge University Press, Cambridge.
Sommerfeld, Walter
1987 -90 Marduk. Reallexikon derAssyriologie 7: 360 -70. Walter de Gruyter, Berlin.
Sorre, M.
The Concept of Genre de Vie. In Readings in Cultural Geography, edited by
1962
P. L. Wagner and M. W. Mikesell, pp. 399-415. University of Chicago Press,
Chicago.
Southall, A. W.
1952
Lineage Formation among the Luo. International Africa Institute, Memorandum
31. Oxford University Press, London.
Alur Society: A Study in Processes and Types ofDomination. Heffer, Cambridge.
1956
1965
A Critique of the Typology of States and Political Systems. In Political Systems
and the Distribution ofPower, edited by M. Banton, pp. 113-40. ASA Monograph 2. Tavistock, London.
The Segmentary State in Mrica and Asia. Comparative Studiesin Society and History 30:52-82.
1991
The Segmentary State: From the Imaginary to the Material Means ofProduction. In Early State Economics, edited by H. Claessen and P. van de Velde,
pp. 75-96. Transaction Publishers, New Brunswick, New Jersey.
Spencer, C. S.
The Cuicatlan Canada and Monte Alban: A Study ofPrimaryStateFormation. Aca1982
demic Press, New York.
1987
Rethinking the Chiefdom. In Chiefdoms in theAmericas, edited by R. D. Drennan and C. A. Uribe, pp. 369-90. University Press of America, Lanham,
Maryland.
1990
On the Tempo and Mode of State Formation: Neoevolutionism Reconsidered. Journal ofAnthropological Archaeology 9: 1-30.
1993
Human Agency, Biased Transmission, and the Cultural Evolution of Chiefly
Authority. Journal of Anthropological Archaeology 12: 41-74.
Spencer, H.
1885
The Principles ofSociology, vol. 1. 3rd ed. Appleton, New York.
The Evolution ofSociety. Edited and with an introduction by R. L. Carneiro.
1967
University of Chicago Press, Chicago.
Spores, R.
1967
The Mixtec Kings and Their People. University of Oklahoma Press, Norman.
1969
Setdement, Farming Technology, and Environment in the Nochixtlan Valley.
Science 166:557-69.
1972
An Archaeological Settlement Survey of the Nochixtlan Valley, Oaxaca. Publications
in Anthropology 1. Vanderbilt University, Nashville.
1974
Marital Alliance in the Political Integration ofMixtec Kingdoms. American Anthropologist 76 :297-311.
1983a Yucufiudahui. In The Cloud People: Divergent Evolution ofthe Zapotec and Mixtec
Civilizations, edited by K. V. Flannery and]. Marcus, pp. 155-58. Academic
Press, New York.
1983b Postclassic Setdement Patterns in the Nochixtlan Valley. In The Cloud People:
Divergent Evolution ofthe Zapotec and Mixtec Civilizations, edited by K. V. Flannery and]. Marcus, pp. 246-48. Academic Press, New York.
1983c Postclassic Mixtec Kingdoms: Ethnohistoric and Archaeological Evidence. In
The Cloud People: Divergent Evolution of the Zapotec and Mixtec Civilizations,
edited by K. V. F1annery and]. Marcus, pp. 255 - 60. Academic Press, New
York.
Spriggs, M.
1988
The Hawai'ian Transformation of Ancestral Polynesian Society: Conceptualizing Chiefly States. In State and Society, edited by]. Gledhill, B. Bender, and
M. T. Larsen, pp. 57-76. Unwin Hyman, London.
Stein, B.
The Segmentary State in South Indian History. In Realm and Region in Tradi1977
tional India, edited by R.]. Fox, pp. 3-51. Duke University Press, Durham,
North Carolina.
Stein, G.
Imported Ideologies and Local Identities: North Mesopotamia in the Fifth
1991
Millennium B.C. Paper presented at the 56th Annual Meeting of the Society
of American Archaeology, New Orleans.
1993 Ethnicity, Exchange, and Emulation: Mesopotamian-Anatolian Interaction at
1988
REFERENCES 1 411
410 I REFERENCES
Hacinebi, Turkey. Paper presented at the 58th Annual Meeting of the Society
of American Archaeology, St. Louis.
1994a Segmentary States and Organizational Variation in Early Complex Societies:
A Rural Perspective. In Archaeological Views from the Countryside: Village Communities in Early Complex Societies, edited by G. M. Schwartz and S. E. Falconer, pp. 10-18. Smithsonian Institution Press, Washington, D.e.
1994b Economy, Ritual, and Power in 'Ubaid Mesopotamia. In Chiefdoms and Early
States in the Near East: The Organizational Dynamics of Complexity, edited by
G. Stein and M. S. Rothman, pp. 35-46. Prehistory Press, Madison, Wisconsm.
Stein, G., R. Bernbeck, e. Coursey, A. McMahon, N. Miller, A. Misir, J. Nichols,
H. Pittman, S. Pollock, and H. Wright
1996
Uruk Colonies and Anatolian Communities: An Interim Report on the
1992-1993 Excavations at Hacinebi, Turkey. American Journal of Archaeology
100: 205-60.
Steinkeller, P.
1987
The Administrative and Economic Organization of the Ur III State: The Core
and the Periphery. In The Organization of Power:Aspects of Bureaucracy in the Ancient Near East, edited by M. Gibson and R. D. Biggs, pp. 19-41. Studies in
Ancient Civilization 46. Oriental Institute, University of Chicago, Chicago.
1989
Sale Documents of the Ur III Period. Freiburger Altorientalische Studien 17.
Franz Steiner, Wiesbaden and Stuttgart.
Stemper, D.
1989
The Persistence of Prehispanic Chiefdom Formations, Rio Daule, Coastal Ecuador.
Ph.D. dissertation, Department of Anthropology, University of Wisconsin,
Madison.
Steve, M.-J., and H. Gasche
1973
L'Acropole de Suse. Memoires de la Delegation Archeologique Francaise en Iran XLVI.
Paul Guethner, Paris.
1990
Le Tell de l' Apadana avant les Achemenides. Contribution i la topographie de
Suse. In Melanges Jean Perrot, edited by F. Vallat, pp. 15-60. Editions
Recherche sur les Civilisations, Paris.
Steward, J. H.
1949
Cultural Causality and Law: A Trial Formulation of the Development of
Early Civilizations. American Anthropologist 51 : 1-27.
1955
Theory of Culture Change: The Methodology ofMultilinear Evolution. University of
Illinois Press, Urbana.
Stol, M.
1982
State and Private Business in the Land of Larsa. Journal of Cuneiform Studies
33: 127-230.
Stone, E. e.
1981
Texts, Architecture, and Ethnographic Analogy. Iraq 43: 19-33.
1987
Nippur Neighborhoods. Studies in Ancient Oriental Civilization 44. Oriental Institute, University of Chicago, Chicago.
Strommenger, E.
1980
Habuba Kabira: Eine Stadt vor 5000Jahren. Phillip von Zabern, Mainz.
Strudwick, N.
1985
The Administration of Egypt in the Old Kingdom: The Highest Titles and Their
Holders. Studies in Egyptology. KPI, London..
Stuart, D.
1992
Hieroglyphs and Archaeology at Copan, Ancient Mesoamerica 3(1): 169-84.
1993
1995
Stuart, G.
1993
New Light on the Olmec. National Geographic Magazine 184(5): 88-114.
Stumer, 1. M.
1954
Population Centers of the Rimae Valley of Peru. American Antiquity
20: 130-48.
Suggs, R. e.
1960
The Island Civilizations of Polynesia. New American Library, New York.
Sugiyama, S.
1989 Burials Dedicated to the Old Temple of Quetzalcoatl at Teotihuacan, Mexico.
American Antiquity 54: 85 -106.
Siirenhagen, D.
The Dry-Farming Belt: The Uruk Period and Subsequent Developments. In
1986
The Origin of Cities in Dry Farming Syria and Mesopotamia in the Third Millennium
B.G., edited by H. Weiss, pp. 7-43. Four Quarters, Guilford, Connecticut.
Symonds, S. e., and R. Lunag6mez
1994
Settlement System and Population Development at San Lorenzo, Veracruz,
Mexico. Manuscript on file, American Museum of Natural History, New York.
Taagepera, R.
1968
Growth Curves of Empires. General Systems 13: 171-75.
1978a Size and Duration of Empires: Systematics of Size. Social Science Research
7:108-27.
1978b Size and Duration of Empires: Growth-Decline Curves, 3000 to 600 B.C. Social Science Research 7: 180 -96.
1979
Size and Duration of Empires: Growth-Decline Curves, 600 B.C. to 600 A.D.
Social Science History 3: 115 -38.
Tadmor, H.
Monarchy and the Elite in Assyria and Babylonia: The Question of Royal Ac1986
countability. In The Origins and Diversity of Axial Age Civilizations, edited by
S. N. Eisenstadt, pp. 203-24. SUNY Series in Near Eastern Studies, State
University of New York Press, Albany.
Tainter, J.
1988
The Collapse of Complex Societies. Cambridge University Press, London.
Tambiah, S. J.
1976
World Conqueror and World Renouncer: A Study of Buddhism and Polity in Thailand against a Historical Background. Cambridge Studies in Social Anthropology
15. Cambridge University Press, Cambridge.
1977 The Galactic Polity: The Structure ofTraditional Kingdoms in Southeast Asia.
Annals ofthe New York Academy ofSciences 293 :69-97.
Taylour, Lord W.
1983
The Mycenaeans. Revised ed. Thames and Hudson, London.
Thapar, B. K.
1973
New Traits of the Indus Civilization at Kalibangan: An Appraisal. In South
Asian Archaeology, edited by N. Hammond, pp. 85-104. Noyes Press, Park
Ridge, New Jersey.
Thomas, K.
1978
Religion and the Decline of Magic: Studies in Popular Beliefs in Sixteenth- and
REFERENCES I 413
412 I REFERENCES
Zur Chronologie der alten Culturen von lea. Journal de la Societe des Americanistes de Paris (n.s.) 10:341-67.
University of Chicago
1992
The Dictionary of the OrientalInstituteof the University of Chicago. Oriental Institute, University of Chicago, Chicago.
Upham, S.
1987
A Theoretical Consideration of Middle Range Societies. In Chiefdoms in the
Americas, edited by R. D. Drennan and C. A. Uribe, pp. 345-68. University
Press of America, Lanham, Maryland.
Valdes,]. A., and F. Fahsen
The Reigning Dynasty of Uaxactun during the Early Classic: The Rulers and
1995
the Ruled. Ancient Mesoamerica 6: 197 - 219.
Vallat, F.
1993
Les noms geographiques des sources suso-elamites. Repertoire Geographique des
Textes Cuneiformes XI. Beihefte zum Tubinger Atlas des Vorderen Orients,
Reihe B, No. 7. Reichert Verlag, Wiesbaden.
van de Mieroop, M.
1992
Society and Enterprise in Old Babylonian Ur. Berliner Beitrage zum Vorderen
Orient 12. Dietrich Reimer, Berlin.
van der Vliet, E. C. L.
1987
Tyranny and Democracy: The Evolution of Politics in Ancient Greece. In
Early State Dynamics, edited by H.]. M. Claessen and P. van de Velde, pp. 7090. E.]. Brill, Leiden.
van Driel, G.
1983
Seals and Sealings from Jebel Aruda, 1974-1978. Akkadica 33: 34-62.
van Zantwijk, R.
1985
The Aztec Arrangement: The Social History of Pre-Spanish Mexico. University of
Oklahoma Press, Norman.
van Zeist, W., and S. Bottema
1977
Palynological Investigations in Western Iran. Palaeohistoria 19: 19-95.
Vats, M. S.
1940
Excavations at Harappa. 2 vols. Government of India, Delhi.
Vayda, A.
1960
Maori Waifare. The Polynesian Society, Wellington.
1976
War in Ecological Perspective. Plenum Press, New York.
Veenhof K.
1972
Aspects of Old Assyrian Trade and Its Terminology. E.]. Brill, Leiden.
1980
Kanish, Karum. In Reallexikon derAssyriologie 5: 369-78.
Veyne, P.
1990
Bread and Circuses: Historical Sociology and Political Pluralism. Translated by
B. Pearce. Penguin, London.
Villar, M. del
1996
La indumentaria en los cautivos mayas del Clasico. Arqueologfa Mexicana 3(17) :
66-71.
Vincent,].
1986
System and Process, 1974-1985. Annual Review of Anthropology 15:99-119.
1990
Anthropology and Politics: Visions, Traditions, and Trends. University of Arizona
Press, Tucson.
Vogelzang, M., and H. Vanstiphout (editors)
1992
Mesopotamian EpicLiterature: OralorAural?Edwin Mellen, Lewiston, New York.
1913
414 I REFERENCES
Voorhies, B.
1989
A Model of the Pre-Aztec Political System of Soconusco. In Ancient Trade and
Tribute: Economies ofthe Soconusco Region ofMesoamerica, edited by B. Voorhies,
pp. 95-129. University of Utah Press, Salt Lake City.
Waddell, 1. A.
The Indo-Sumerian Seals Deciphered: Discovering Sumerians of Indus Valley as
1925
Phoenicians, Barats, Goths and Famous Vedic Aryans, 3100-2300 B.G. Luzac,
London.
Wallace, A. F. e.
1971
Administrative Forms of Social Organization. Addison-Wesley ModularPublications 9:1-12.
Wallerstein, 1.
A World-System Perspective on the Social Sciences. In The Capitalist WorldEconomy, pp. 152-64. Cambridge University Press, Cambridge.
1992
The Challenge of Maturity: Whither Social Science? Review 15: 1-7.
Walther, A.
1979
1917
Das altbabylonische Gerichtswesen. Leipziger Semitische Studien 6: 4 - 6. Leipzig.
Wattenmaker, P.
1990
On the Uruk Expansion. CurrentAnthropology 31: 67-68.
Weaver, M. P.
The Aztecs, Maya, and Their Predecessors. 3rd ed. Academic Press, New York.
1993
Weber, M.
REFERENCES I 415
Wei-Ming, T.
1986
The Structure and Function of the Confucian Intellectual in Ancient China.
In The Origins and Diversity of Axial Age Civilizations, edited by S. N. Eisenstadt, pp. 360-73. State University of New York Press, Albany.
Weiss, H., and T. e. Young, Jr.
1975
The Merchants of Susa. IranXIII: 1-17.
Wesson, R. G.
1978
State Systems: International Pluralism, Politics, and Culture. Free Press, New York.
Westenholz, A.
The Sargonic Period. In Circulation of Goods in Non-Palatial Context in the An1984
cientNear East, edited by A. Archi, pp. 17-30. Incunabula Grceca, Rome.
Wetterstrom, W.
1993
Foraging and Farming in Egypt: The Transition from Hunting and Gathering
to Horticulture in the Nile Valley. In The Archaeology of Africa: Food, Metals and
Towns, edited by T. Shaw, P. Sinclair, B. Andah, and A. Okpoko, pp. 165226. Roudedge, London.
Wheadey, P.
The Pivot if the FourQuarters: A Preliminary Enquiry into the Origins and Charac1971
ter of the Ancient Chinese City. Aldine, Chicago, and Edinburgh University
Press, Edinburgh.
Wheeler, R. E. M.
1947
Harappa 1946: The Defenses and Cemetery R-37. Ancient India 3:58-130.
1950 Five Thousand Years ofPakistan: An Archaeological Outline. Royal India and Pakistan Society, London.
1959 Early India and Pakistan: To Ashoka. Frederick A. Praeger, New York.
1968
The Indus Civilization. 3rd ed. Supplementary volume to the Cambridge Ancient History ofIndia. University of Cambridge Press, Cambridge.
Whitecotton, J.
1992
Culture and Exchange in Postclassic Oaxaca: A World-System Perspective. In
Resources, Power, and Interregional Interaction, edited by E. M. Schortman and
P. A. Urban, pp. 51-74. Plenum Press, New York.
Whiting, R. M.
1987
Old Babylonian Letters from Tell Asmar. Assyriological Studies 22. Oriental Institute, University of Chicago, Chicago.
Widmer, R. J., and R. Storey
1993
Social Organization and Household Structure of a Teotihuacan Apartment
Compound: S3W1: 33 of the Tlajinga Barrio. In Prehispanic Domestic Units in
Western Mesoamerica: Studies of the Household, Compound, and Residence, edited
by R. S. Sandey and K. G. Hirth, pp. 87-104. CRC Press, Boca Raton,
Florida.
Wilcke, e.
1973
Politische Opposition nach sumerischen Quellen. In La voix de l'opposition en
Mesopotamie, edited by A. Finet, pp. 37-65. Institut des Hautes Etudes de Belgique, Brussels.
Willey, G. R.
1974
The Classic Maya Hiatus: A Rehearsal for the Collapse? In Mesoamerican Archaeology: New Approaches, edited by N. Hammond, pp. 417-44. Duckworth,
London.
1991
Horizontal Integration and Regional Diversity: An Alternating Process in the
Rise of Civilization. American Antiquity 56: 197-215.
416 I REFERENCES
Williams, B. B.
1986
Excavations between Abu Simbel and the Sudan Frontier I: The A-Group Royal
Cemetery at Qustul: Cemetery L. Oriental Institute Nubian Expedition 3. Ori-
PrehispanicSettlement Patterns in the Lower Santa Valley, Peru: A Regional Perspective on the Origins and Development of Complex North Coast Society. Smithsonian
1968
Wolf, E. R.
1951 The Social Organization of Mecca and the Origins ofIslam. SouthwestJournal
of Anthropology 7: 329-56.
1982
Europe and the People without History. University of California Press, Berkeley.
REFERENCES I 417
418 I REFERENCES
sponses to Risk and Uncertainty, edited by P. Halstead and]. O'Shea, pp. 10613. Cambridge University Press, Cambridge.
Wrightson, K.
1982 English Society 1580-1680. Unwin Hyman, London.
Wrong, D. H.
1979 Power: Its Forms, Bases and Uses. Harper and Row, New York.
Yang, M. M.-H.
1989
The Gift Economy and State Power in China. Comparative Studies in Society and
History 31: 25-54.
Yoffee, N.
1977
The Economic Role of the Crown in the Old Babylonian Period. Bibliotheca
Mesopotamica 5. Undena, Malibu, California.
1982
Social History and Historical Method in the Late Old Babylonian Period. Journal ofthe American OrientalSociety 102: 347 - 53.
1988a Orienting Collapse. In The Collapse ofAncient Statesand Civilizations, edited by
N. Yoffee and G. L. Cowgill, pp. 1-19. University of Arizona Press, Tucson.
1988b The Collapse of Ancient Mesopotamian States and Civilizations. In The Collapse ofAncient States and Civilizations, edited by N. Yoffee and G. L. Cowgill,
pp. 44-68. University of Arizona Press, Tucson.
1991 Maya Elite Interaction: Through a Glass, Sideways. In Classic Maya Political History: Hieroglyphic and Archaeological Evidence, edited by T. P. Culbert,
pp. 285-310. School of American Research Advanced Seminar Series, Cambridge University Press, Cambridge.
1992
Reading between the Lines, Sideways, from a Distance: Anthropological Perspectives on Ancient Mesopotamian Economy and Society. Paper presented at
the conference on "Documenting Cultures: Written and Unwritten in Preindustrial Societies," Columbia University, New York.
1993a Mesopotamian Interaction Spheres. In Early States and the Evolution of
Mesopotamian Civilization: Soviet Excavations in the Sinjar Plain, Northern Iraq,
edited by N. Yoffee and].]. Clark, pp. 257-69. University of Arizona Press,
Tucson.
1993b Too Many Chiefs? or Safe Texts for the 90s. In Archaeological Theory: Who Sets
the Agenda? edited by N. Yoffee and A. Sherratt, pp. 60-78. Cambridge University Press, Cambridge.
1993c The Late Great Tradition in Ancient Mesopotamia. In The Tablet and the Scroll:
Near Eastern Studies in Honor of William W Hallo, edited by M. E. Cohen,
D. e. Snell, and D. B. Weisberg, pp. 300-308. CDL Press, Bethesda, Maryland.
1995a Conclusion: A Mass in Celebration of the Conference. In The Archaeology of
Society in the Holy Land, edited by T. E. Levy, pp. 542-48. Facts on File, New
York.
1995b Political Economy in Early Mesopotamian States. Annual Review ofAnthropology 24:281-311.
1997 The Obvious and the Chimerical: City-States in Archaeological Perspective.
In The Archaeology of City-States: Cross-Cultural Approaches, edited by D. L.
Nichols and T. H. Charlton, pp. 255-63. Smithsonian Institution Press,
Washington D.e.
Yoffee, N., and G. L. Cowgill (editors)
1988
The Collapse ofAncient States and Civilizations. University of Arizona Press,
Tucson.
REFERENCES 1419
Young, T. C.,]r.
1969 Excavations at Godin Tepe: First Progress Report. Royal Ontario Museum,
Toronto.
1986
Godin Tepe Period VI/V and Central Western Iran at the End of the Fourth
Millennium. In Gamdat Nasr: Period orRegional Style? edited by U. Finkbeiner
and W. Rollig, pp. 212-28. Beihefte zum Tiibinger Atlas des Vorderen Orients, Reihe B, 62. Reichert Verlag, Wiesbaden.
Zimansky, P.
1993 Review ofDenise Schmandt-Besserat, Before Writing. Journal ofField Archaeology 20: 513-17.
Zuidema, R. T.
1964
The Ceque System ofCuzco. International Archives of Ethnography, Supplement to vol. 50, Leiden.
Index
Bali,166
Becan, 325, 326, 328, 335
Blanton, R. E., 168
Blau, P. M., 167
Bloch, Maurice, 163, 165
bureaucracy, 148, 266, 303
Burke, P., 8-9
Buto, 86, 209, 217
Cacaxtla, 60, 72
Cajamarquilla, 76
Calakmul, 57; Dynamic Model of, 59,
61,62,63-64; as head of administrative hierarchy, 18, 19; palace at, 3334; royal tomb at, 51-53; warfare at,
347
Cantona, 60,72
Caracol, 65, 66, 317, 347
Carneiro, R. L., 68, 98, 99, 104, 107,
109, 266
CasaI,J-M., 271
Casma Valley, Peru, 47, 75
centralization, 136-37, 138, 207-8, 293,
308-9
.
Cerro Azul, 76
Cerro Sechin, 75
Chacchob, 325
Chalcatzingo, 56
Chan Chan, 76; ciudadelas at, 23-24, 25,
43, 54; royal tombs at, 50, 51
Chanhu-daro, 276
Chavin de Huantar, 75
Chichen Itza, 40, 41, 62, 334
chiefdoms, 21; competition between and
state formation, 92; differentiated from
states, 6-7, 104,265; diverse forms of,
279-80; origin of, 267
Childe, V. G., 157
Chimu, 23, 50, 61, 76, 77
China: corporate moral code in, 159-60,
164; decentralization in, 139, 160-61;
households in, 168-69; metallurgy in,
158-59
Chincha, 296-97, 304
Chingu, 60, 72, 73
Chogha Mish, 78, 184, 196
Chucuito, 298
Chunchucrnil, 325
Cieza de Le6n, P., 306
city-states: in Greece, 90-91; in
Mesopotamia, 80-81; not a stage leading to state, 92-93; validity of concept, 8-10, 74
civilization: applied to Indus, 261; contrasted with state, 254; cosmologies
and, 255, 258-59; high culture and,
235
Claessen, H.]. M., 97, 266-67, 284
Coe, M. D., 55-56
cognitive codes, 159, 162-63, 164, 167,
170
Cohen, R., 8
communication: meta-, 164-66; reflexive, 162-63, 164, 165; ritualization of,
165. See also writing
Confucianism, 168-69
Copan, 33,63,66-68,80, 321,334,349
corporate political economy: decentralization of ritual in, 160; in Mesoamerica, 149-50, 172; in Mesopotamia,
155; moral codes in, 159-60, 164; origin of, 171-72; reflexive communication in, 162-63; regulations in, 15659; semiautonomy in, 167, 169; writing and, 161-62
Cuicatlan, 69, 118, 121
INDEX 1423
Cuicuilco, 71
Cuzco, 298, 299
Dahshur, 245
Dakhla Oasis, 258
"dark ages," 99-100
decentralization, 139, 160-61; pseudo-,
167
Deh Luran Plain, 189-93, 196
Deir el-Medina, 230
Dholavira, 287
Diehl, R. A., 56
Djoser, 242-43
Dos Pilas, 59, 61, 64, 325, 331, 347
Dynamic Model: of Aegean, 89-91; of
Andes, 61, 74-76; of central Mexico,
60,71-74; of Egypt, 86-88; lengths of
cycles in, 93; of Maya, 59-60, 61-68;
of Mesopotamia, 60-61, 76-86; of
Zapotec, 60, 68-71
Eade, T K., 339, 342, 345
Easter Island, 345
Ebla,221
egalitarian behavior, 151-53
Egypt: artistic change in, 239; chronology
of, 202; collapse of civilization in,
256-57; conspicuous consumption in,
242-43; contrasted with
Mesopotamia, 204-13, 219, 232, 25253,255; Dynamic Model of, 86-88;
economy of, 228-32; elites in, 233;
exclusionary architectural strategies in,
160; exclusionary political economy in,
155; extent of, 204; high culture in,
242-45,246; kingship in, 205-7; local power in, 222-25; map of, 201;
order in, 212-13; political system of,
199-200,219-20; royal tombs in,
216-18,242,245; spiritual domination in, 160; state development in,
216-18; temples in, 255; urbanization
in, 209-10; writing in, 241-42, 244,
251
Eisenstadt, S. N., 101, 159
Ejutla, 69, 116-21
el-'Amama,231
Elephantine, 209
INDEX 1425
INDEX I 427
T aagepera, R., 98
Tahiti, 312, 329, 336, 338, 339
Tamazulapan, 118, 121
temples: in Egypt, 255; in Mesoamerica,
40-42; in Mesopotamia, 36, 37, 38; in
Peru, 42-43, 44
Teotihuacan: apartment compounds at,
168; contact with Maya, 67, 338; corporate political economy of, 172; Dynamic Model of, 60, 71-73; lack of
centralization at, 149; palaces at, 2731; temples and priests' residences at,
41-42; writing at, 161-62, 166
Tepe Gawra, 46, 214
Tepe Sharafabad, 179-81, 182
Tepe Zabarjad, 193-94
Texcoco, 73
Thapar, B. K., 270
Thebes, 222, 224, 230, 231
Tikal: Dynamic Model of, 59, 61, 62,
63-66; fortifications at, 324; population of, 320, 334; warfare by, 330, 333,
334,335,347
Tikopia, 345
Tiwanaku (Tiahuanaco), 74, 76
Tonga, 312, 332, 339,343, 346
Tonina,337
Tosi, M., 287
trade: ofIndus civilization, 288; and spatial extent of state, 110; in Susiana,
195-96
Trigger, B. G., 92
Tula, 60, 72, 73
36,338-39,340-46,348; as process,
314; ritual, 307, 338; status-rivalry,
310,340-41,345,348,349,350; war
leader, 220
Wari, 61, 76
Warka (Uruk), 199,214,215
wealth, 213, 253-54
Weber, M., 169
Wessex, 157
Wheeler, R. E. M., 274, 291
Williamson, R. W., 341-42
Wilson, D.]., 75
Wright, H. T., 137
writing, 54; control of, 236; corporate vs.
exclusionary political strategies and,
161-62; in Egypt, 241-42, 244,251;
in Mesopotamia, 162,215,246-51,
253
Xochicalco, 60, 72