The Mind in the Cave — the Cave in the

Mind: Altered Consciousness

in the Upper Paleolithic
David J. Lewis-Williams
Rock Art Research Unit
Department of Archaeology
University of the Witwatersrand
Johannesburg 2050
Südafrika
[email protected]

and

Jean Clottes
Conservateur Ge'ne'ral du Patrimoine
11, ruedu Fourcat
09000 Foix
France

Abstract
This brief overview argues that the evidence of the images themselves, as well
as their contexts, suggests that some Franco-Cantabrian Upper Paleolithic cave art
was, at least in part, intimately associated with various shamanic practices. Universal
features of altered states of consciousness and the deep caves combined to create
notions of a subterranean spirit-world that became, amongst other ritual areas, the
location of vision quests.

Introduction
In western Europe, the Upper Paleolithic began about 35,000 years ago with the
appearance there — elsewhere it was earlier — of anatomically modern human
beings, somewhat smugly self-styled Homo sapiens sapiens. The period ended in
Europe and the Middle East roughly 10,000 years ago. It was then that people in
western Europe and other parts of the world began to abandon a hunting and
gathering life-style and to domesticate plants and animals (see, for example, Mellars
and Stringer 1989) .
It seems fair to say that the Upper Paleolithic of western Europe is best known
for its remarkable efflorescence of art, or, as some writers wish, less prejudicially and
probably rightly, to call it, "image-making" (for an overview of the art, see Bahn and
Vertut 1988). Recently, Upper Paleolithic cave art, painted and engraved deep
underground in dark chambers and nooks, has received renewed attention with the
announcement in the world media of two startling new sites—Cosquer, the partially
flooded cave on the French Mediterranean coast (Clottes and Courtin 1996), and

Aruhropobgy of Consciousness 9(1): 13-21. Gipyright© 1998 American Anthropological Association

13
14 Anthropology of Consciousness [9( 1 )1

Chauvet in the Ardeche region of France (Chauvet et al. 1996). Although it is with
cave, or parietal, art that this paper is concerned, there are equally striking pieces of
mobile art, exquisitely carved in bone, antler and ivory. Mobile art is found in
deposits in rock shelters, some of which are at the entrances to deep caves in which
there is wall art, as well as in the depths of the caves.
Since its discovery, cave art has been a topic of intense speculation and debate.
Why did people penetrate the caves to make images of horses, bison, aurochs, woolly
mammoths, and so forth? What functions did the images and the caves serve? What
did the images "mean"? Was it for the purpose of hunting magic? Or was the art
simply art pour Part?

Method and Theory

Behind these questions lie more general methodological and theoretical issues:
is it, in principle, possible to know what people who lived so long ago believed ? Their
myths, rituals and beliefs have not been preserved; we have only their enigmatic
images. If it is difficult to know what living people believe about their own art, can
we ever find out what Upper Paleolithic people believed about theirs?
Faced with these fundamental epistemological questions, researchers divide
into two camps. On the one hand, there are professional pessimists. They claim that,
in principle, we can never know, and they make their living by attempting to
demolish the work of others. They point out, rightly, that all the nineteenth and
twentieth century hunter-gatherers were (or are) not living Upper Paleolithic fossils
who can readily answer our questions. On the other hand, there are researchers who
believe that we can find out something — no one argues for everything — about
Upper Paleolithic beliefs and art. But it will not be easy. Even the most optimistic
researchers recognize that the theoretical and methodological problems need to be
addressed directly and explicitly. Unless those problems are solved, hypotheses will
remain unassessable guesses; anyone's guess will have to remain as good as anyone
else's.
In the last decade a new approach to the problem has been essayed. It is posited
on the generally uncontested proposition that, because Upper Paleolithic people
were Homo sapiens sapiens, they had the same nervous system as everyone in the world
today, whether residual hunter-gatherers or industrialists. Here, it is argued, is a link
with the Upper Paleolithic, not the only link but at least a starting point: there is
a neurological bridge between us and that remote period (Lewis-Williams and
Dowson 1988; Lewis-Williams 1991).

Altered States of Consciousness

The human nervous system generates consciousness, an extremely difficult state
to define. It also generates altered states of consciousness that are easier (though not
much) to define, even if only in relation to an intuitively understood "normal
consciousness." Upper Paleolithic people must have experitnced not only "normal
consciousness" but also altered consciousness because altered states are wired into
the human nervous system and, moreover, are induced by a wide range of factors that
include the ingestion of psychotropic drugs, audio-driving, hyperventilation, sensory
and social deprivation, pain, intense concentration, and certain pathological
March 1998 Altered Consciousness in the Upper Palaeolithic

conditions. Add "dreaming" to this list, and the experience by at least some Upper
Paleolithic people of altered states becomes indisputable. Altered states of
consciousness are part of being human, part of a "package deal" (for a review of
research on altered states see Siegel and West 1975). What Upper Paleolithic people
made of their altered states is another question altogether.
The ways in which altered states are experienced and interpreted are not "given"
or universal. In understanding this point it is useful to think of consciousness as a
spectrum. At one end is "normal" or "alert" consciousness. This grades into day-
dreaming, deep reveries, dreaming, "light" trance states, and, at the far end, "deep"
trances in which subjects are not aware of their surroundings at all, but are part of a
fully hallucinatory realm with its own rules of causality and transformation. That is
the way that many Westerners think of it. But the spectrum is divided up by each
society or subculture in its own way. What passes for madness in one community may
be esteemed as divine revelation in another. What is a vision to some people is, to
others, hallucination. The definitions of variously distinguished altered states are
therefore socially situated. But there is more to it. The definition of altered states
is also implicated in the negotiation of social statuses and political power. Visions
of the future may earn a person admiration in some societies, but they will hamper
rather than facilitate election to Congress. Because altered states are part of being
human, all people have to come to terms with them in one way or another.
So too, it must have been during the Upper Paleolithic. Were those people
hyper-rationalists who dismissed all altered states as aberrations? Unlikely. Or were
they like all known hunter-gatherers (and, of course, others as well) who place high
value on certain precisely defined altered states? Indeed, the ubiquity of cross-
culturally very similar altered states among hunter-gatherers points to the high
antiquity of the form of ritualized altered states that we call shamanism.

The Shamanic Cosmos

We now consider two features of altered states of consciousness that contribute
to these cross-cultural similarities (Lewis-Williams 1997).
First, as people go into altered states, they often experience sensations of
attenuation, rising up and flying. As images appear before them, they believe that
they are entering a spiritual realm set in or above the sky. The sensation of flight,
naturally enough, suggests transformation into a bird, and, with changes in perspective,
they look down onto the level of daily life. Birds are, of course, closely associated with
shamans in many cultures.
Secondly, as people move towards the "far" end of the spectrum, they experience
and are drawn into a vortex. On the sides of this vortex there is sometimes a lattice,
in the segments of which appear the first iconic images (Siegel 1977). Feelings of
constriction, difficulty in breathing, and of being drawn into the vortex often suggest
entrance into a tunnel that leads underground. At the other end of the tunnel is a
new realm inhabited by its own beings, spirits, animals, and monsters. All this is
wired into the human nervous system. There is a cave in the mind. In shamanic
societies these experiences lead to belief in a chthonic realm, an underworld that
shamans have the power to visit.
The shamanic cosmos is thus tiered. At its simplest, there are three tiers: the
16 Anthropology of Consciousness [9(D]

level of daily life, a realm above, and

another below. The realm above and
the realm below impinge on daily life,
and it is the shamans who, by
travelling the axis mundi (often
thought of as a tree or a hole in the
ground) are able to mediate the
cosmos.
During the Upper Paleolithic,
we argue, the limestone caves of
western Europe were regarded as
topographical equivalents to the
psychic experience of the vortex and
a nether world. The caves were the
entrails of the underworld, and their
surfaces — walls, ceilings and floors
—were but a thin membrane between
those who ventured in and the beings
and spirit-animals of the underworld.
This is the context of west European
cave art, a context created by
Figure J. A horse's head painted on a flint interaction between universal
nodule. The rest of the animal seems to be neuropsychological experiences and
behind the rock wall. Rouffignac, Dordogne, topographically situated caves.
France. When people o( the Upper
Paleolithic embellished these ca\es
with paintings and engravings of animals, signs and, less commonly, apparently
human figures, they were at times exploiting certain defined altered states to
construct, in each cave, a particular, .socially and historically situated underworld
(Lewis-Williams 1995). Otherwise inexplicable features of Upper Paleolithic cave
art are explained and unified by this explanation (Lewis-Williams 1QQ7, Clottes and
Lewis-Williams 1996).

Membranous Rock
Many images incorporate features ot the rock surface on which they were placed.
Sometimes a small nodule became an animal's eye, sometimes a natural swell of the
rock face was taken to delineate the chest or shoulder ot an animal; sometimes the
edgeot ashclfbecameadorsal line. To these naturalfeatures,the artists added lines,
thereby transforming the given into the created. Frequently, these images appear to
be coming out of the rock wall. At Rouffignac, for instance, i horse's head is painted
on the .side of a protruding flint nodule. The n.st ot the horse is apparently behind
the rock face (Figure 1).
These and other features of the art suggest that people were searching for
animals, by touch as much a.s by si^ht, in the convolutions ot the underworld; not
"real" animals, but spirit-animals that could become their 'guides' and animal-
helpers. The caves were thus like the entrails of the underworld, and the flours walls,
March 1998 Altered Consciousness in the Upper Palaeolithic 17

and ceilings a thin "membrane

between the people who ventured
in and the spirit world behind the
surfaces. Shamans sought to draw
animals through this permeable
membrane.
This interpretation is
strengthened by a common char-
acteristic of certain hallucinations,
again created by the wiring of the
human nervous system and
therefore universal. Hallucin-
ations are often projected onto
surfaces such as walls or ceilings.
Western subjects liken this
experience to a slide or film show: Figure 2. A painting of a vertical bison, from Niaux
the visions "float" and move on cave in the Ari&ge region of France. The dorsal line
the "screen" (Siegel 1977). Given is created by shadow.
the sensory deprivation afforded
by the caves, and leaving aside for
the moment all the other inducing factors, it is virtually certain that at least some
Upper Paleolithic people must have hallucinated in them and, moreover, that some
of those visions would have been projected onto the surfaces around them. Then,
in an attempt to fix, to capture, those visions, we suggest that people searched the
surfaces and added marks to re-create their mental images. The pictures that they
thus made were not 'pictures" in the usual sense of the word, nor were they
representations of something else — their visions or, even less likely, "real' animals.
Rather, they were visions, fixed forever.
Another type of Upper Paleolithic image is immediately clarified. When one's
lamp (not the electric light now unfortunately installed in some caves) is in a specific
position the shadows cast across the rock sometimes represent, to the expectant eye,
part of an animal, say, the dorsal line of a bison, as in the Niaux cave in the Ariege
region in France (Figure 2; see Clones 1995:165, Figure 180). Then only a few deft
strokes were needed to add the head, legs and belly. If the light is moved, the animal
disappears back through the 'membrane. The person has thus mastered the animal:
he or she can make it come and go at will. On the other hand, the animal has
mastered the person, for, if the person wishes the animal to remain fully visible, he
or she must maintain a certain posture, relax and allow the light to move, and the
spirit-animal has gone.
The techniques we have so far described were, we argue, used to coax spint-
animals from behind the membrane. Other practices move, ,is it were, in the opposite
direction. In Enlene, one of the three Volp caves, for instance, hundreds of very small
pieces of bone were thrust into cracks (Begouen et al. 1996). They are at various
levels from the floor and slope in all possible directions. They could not have had
some pragmatic function. In other caves — Les Trois Freres, another of the Volp
caves, is an example — animal teeth and stone artifacts were placed in crevices
 Anthropology of Consciousness

(Begouen and Clottes 1981). The teeth and the artifacts must have been taken
underground with the express purpose of placing them in the niches. In all these
cases, people were sending things through to the underworld. Exactly what they were
thereby achieving, we do not at present know.

A Neuropsychological Model
So far we have dealt only with selected aspects of representational images. But
there is also a large non-representational, or geometric, component of Upper
Paleolithic art. These meanders, dots, zigzags and parallel lines have sometimes been
considered a separate, ifparallel.symbolling system. Indeed, one ofthe most puzzling
features of hunter-gatherer rock art worldwide is the intimate relationship between
representational and geometric images. The answer to this enigma is to be found in
the wiring ofthe human nervous system and the way in which that system goes deeper
and deeper into altered states. What has become known as the neuropsychological
model distinguishes three stages of altered consciousness (Lewis-Williams and
Dowson 1988).
As people begin to move along the spectrum from alert consciousness to light
altered consciousness, they sometimes experience geometric visual percepts that
include zigzags, undulating lines, bright dots in clouds or lines, meandering lines, sets
of parallel lines and so forth (Kliiver 1966; Eichmeier and Hofer 1974). This is Stage
One. The geometric percepts of this stage aTe variously known as "form constants,"
"phosphenes, and "entoptic phenomena." Because they are wired into the human
nervous system, they are experienced universally. All people, no matter what their
cultural backgrounds, have the potential to see them. They are induced by the
generating factors that we have already listed and also by migraine, the scotomata d
which are well known to many sufferers. Again, we must point out that it is the form,
or structure, that is universal. Selection from the full range o( forms and the
ascription of meaning to the selected forms is entirely cultural For instance, the
Tukano of South America take multiple undulating lines of dots to represent the
Milky Way, the L: >al of .shamanic flight (Reichel-Dolmatoff 1978). In other societies,
such dots may mean someching altogether different.
In Stage Two of altered consciousness, people try to make sense of entoptic
phenomena (Horowitz 1975). They do so in culturally specific ways and in
accordance with their emotional states. For instance, Westerners might interpret a
bright dot as a bomb if they are fearful or a cup or water if they arc thirsty. Christians
in a mood of religious expectation may take the bright dot to represent a chalice held
by the Light ofthe World.
Although first experienced in an early sta^e of altered consciousness, entoptic
phenomena persist into Stage Three — deep trance (bieyel 1977). Here they are
combined with iconic images d people, animals and monsters. They are therefore
an integral part ofthe spirit world, no matter what their specific Culturally controlled
meanings may be. Here, we argue, is the answer to the enigma of at least some
geometric hunter-gather rock arts and, what concerns as most at the moment, to the
combination of geometric and representational motifs in Upper Paleolithic art.

Figure' 3. The 'sorcerer' from Lea Trois Freres caie in France.

(Illustration by Donald Sutherland, after Breuil )
March 1998 Altered Consciousness in the Upper Palaeolithic 19

Together, these two kinds oi

motifs form a unified whole.
They are all part of spiritual
experience, and their juxt-
aposition on the membranous
walls of the deep caves is no
longer surprising.
Also in Stage Three,
people feel themselves to be
blended with their imagery.
Often they feel themselves to
be transformed into animals,
partially or completely In
shamanic thought, the shaman
shares the power of his or her
spirit-animal guide. In animal
form, too, shamans believe that
they can travel beyond their
human bodies.
Such images are depicted
in Upper Paleolithic art; they
are the so-called theri-
anthropes. One of the earliest
pieces of Upper Paleolithic Figure 4 A painting depicting a part-human,
carving comes from the part-bison figure. Chauvet caie, at Vallon
Aurignacian period (approx- Pont-d'Arc, France.
imately 35,000-28,000 years
ago). It is of a man with a feline, probably a lion, head (Hahn 1971). Painted and
engraved therianthropes are also knoun from the tinal period ofthe Upper Paleolithic,
the Magdalenian (approximately 16,000-11,000 years ago). The sorcerer" from Les
Trois Freres (Figure 3) is perhaps the best known Upper I aleolithic therianthrope
(for a discussion o( this figure see Ucko and Rosenfeld 1967). In the recently
discovered Chauvet cave there is another fine example: it has the head of a bison,
but the lowerpartof its body is human (Figure 4; see Chauvet et al. 1996: Figure 93).
Therianthropes are not common in Upper Paleolithic art; they are more
common in some other shamanic arts But they are a persistent feature that is
evidence for the intimate relationship between shaman and spirit-anim.il

Altered States and Social Control

So far we have considered the parts played by altered consciousness and
associated shamanic cosmology in the production or Upper Paleolithic art. We also
pointed out that the ways in which altered states and visions are defined are always
socially situated. Now having noted key aspects ot the subterranean images, we can
consider the social role o( the caves themselves.
There is always a struggle to O mtrot access to altered states and to restrict the
type of imagery that may be considered appropriate to different classes of people
20 Anthropology of Consciousness [9(1)]
1—

(Lewis-Williams 1995). This kind of social differentiation was almost certainly

projected onto the topography of the caves. Some caves have spacious chambers that
were richly embellished with large, imposing images. Elsewhere in the same caves
there are often small diverticules into which only one or two people can squeeze, yet
here there are also depictions, frequently made with just a few confident strokes. In
Lascaux, for instance, the comparatively large Hall of the Bulls is richly embellished.
Much deeper in, at the very back of the cave, the narrow Chamber of the Felines has
only sketched outlines. Clearly, different activities were taking place in these
different parts of the caves (Lewis-Williams 1997).
The large images of the embellished chambers were probably communally
produced. Groups of people probably co-operated differentially to make large
quantities of paint, to build scaffolds (as in the Axial Gallery in Lascaux; Leroi-
Gourhan and Allain 1979), and to make the large images. Moreover, it seems highly
probable that communal rituals were performed in these spaces. Each space was, in
effect, a constructed segment of the underworld. Perhaps some of these rituals were
preparatory to individual experiences to be sought by the few who ventured farther
into the bowels of the underworld in search of spirit-animals. In solitary contemplation,
experiencing sensory deprivation, searching with hands, sight and mind, some
people found the spirit-animals for which they were questing and, then or later, fixed
those visions and so acquired shamanic power and animal guides.
The caves thus may have been socially differentiated not only by topography but
also by images. Certain people may have been allowed into certain parts only. As
the spectrum of consciousness was divided up and defined, so too were the caves
socially differentiated. As people moved through the caves they established their
personae or challenged the statuses of others. Both cave and art became instruments
of social differentiation and contestation. Here, in the very origin of religion, were
the seeds of its social divisiveness and domination, hierarchies and cruelties.
Consciousness has always been a site of struggle.

Conclusion
By drawing on the results of neuropsychological research on altered states of
consciousness and on those aspects of shamanism that are probably wired into the
human nervous system, we have outlined an explanation of Upper Paleolithic cave
art that brings its diverse features together into a coherent pattern. In a cave, the
mind fills the space with spirit-animals and beings. But there is also a cave in the
mind, created there by the wiring of the human nervous system. When the
topographical cave and the cave in the mind came together in the Upper Paleolithic,
new dimensions were created and a stunning art was born.

Notes
Acknowledgments: We are grateful to colleagues who commented on drafts of this paper, G. Blundell
and R. McLean. We thank M. Jean Plassard, the owner of the Rouffignac cave,/or the photograph from
which Figure 1 was made. Figures 1,2, and 4 were prepared by R. McLean. Figure 3 was drawn by Donald
Sutherland. The San Heritage Centre is funded by the Human Sciences Research Council and the
University of the Witwaterstrand. Neither institute is responsible for the views herein expressed.
March 1998 Altered Consciousness in the Upper Palaeolithic 21

References
Bahn, Paul G., and Jean Vertut
1988 Images of the Ice Age. London: Winward.
Begouen, Robert, and Jean Clones
1981 Apports Mobiliers dans les Cavernes du Volp (Enlene, Les Trois-Freres, Le Tuc
d'Audoubert) Altamira Symposium. Pp. 157-188. Madrid: Ministerio de Cultura.
Begouen, Robert, Jean Clottes, Jean-Pierre Giraud, and Francois Rouzaud
1996 Os Plantes et Peintures Rupestres dans la Caverne d'Enlene. In Pyrenees PnShistoriques.
Arts et Sorters. Henri Delporte and Jean Clottes, eds. Pp. 283-306. Paris: Ed. du C.T.H.S.
Chauvet, Jean-Marie, Eliette Brunel-Deschamps, and Christian Hillaire
1996 Dawn of Art: The Chauvet Cave: The Oldest Known Paintings in the World, with an
Epilogue by Jean Clottes. New York: Harry N. Abrams.
Clottes, Jean
1995 Les Cavernes de Niaux: Art Preiiistorique en Ariege. Paris: Le Seuil.
Clottes, Jean, and Jean Courtin
1996 The Cave beneath the Sea: Paleolithic Images at Cosquer. New York: Harry N. Abrams.
Clottes, Jean, and Lewis-Williams, David J.
1996 Les Chamanes de la Preliistoire: Transe et Magie dans les Grottes Ornees. Paris: Le
Seuil.
Eichmeier, Joseph, and Oskar Hofer
1974 Mescal and Mechanisms of Hallucinations. Chicago: University of Chicago Press.
Hahn, Joachim
1971 La Statuette Masculine de la Grotte du Hohlenstein-Stadel (Wurtemberg).
L'Anthropologie 75:233-244.
Horowitz, Mardi J.
1975 Hallucinations: An Information-Processing Approach. In Ronald K. Siegel and Louis J.
West, eds. Hallucinations: Behavior, Experience, and Theory. Pp. 163-195. New York:
Wiley.
Kliiver, Heinrich
1966 Mescal and Mechanisms of Hallucinations. Chicago: University Press.
Leroi-Gourhan, Arlerte, and Jacques Allain
1979 Lascaux Inconnu. Paris: Centre National de la Recherche Scientifique.
Lewi^Williams, David J.
1991 Wrestling with Analogy: A Methodological Dilemma in Upper Palaeolithic Art Research.
Proceedings of the Prehistoric Society 57(1):149-162.
1995 Modelling the Production and Consumption of Rock Art, South African Archaeological
Bulletin 50:143-154.
1997 Harnessing the Brain: Vision and Shamanism in Upper Palaeolithic Western Europe. In
Beyond Art: Pleistocene Image and Symbol. Margaret W. Conkey, Olga Softer, and Deborah
Stratmann, eds. Memoirs of the California Academy of Sciences, No. 23:321-342.
Lewis-Williams, David J., and Thomas A. Dowson
1988 Signs of All Times: Entoptic Phenomena in Upper Palaeolithic Art. Current
Anthropology 29:201-245.
Mellan, Paul, and Chris Stringer, eds.
1989 The Human Revolution: Behavioural and Biological Perspectives on the Origins of
Modern Humans. Edinburgh: Edinburgh University Press.
Reichel-Dolmatoff, Geraldo
1978 Beyond the Milky Way: Hallucinatory Imagery of the Tukano Indians. Los Angeles:
UCLA Latin America Center.
Siegel, Ronald, K., and Louis J. West, eds.
1975 Hallucinations: Behaviour, Experience, and Theory. New York: John Wiley and Sons.
Siegel, Ronald, K.
1977 Hallucinations. Scientific American 237: 132-140.
Ucko, Peter J , and Andre* Rosenfeld
1967 Palaeolithic Cave Art. London: WeidenfeId and Nicholson.