blank space - 0.1 K to archaeology to Greater Anthropologyto contents

Altered States: The Origin of Art in Entoptic Phenomena

J.D. Lewis-Williams and T.A. Dowson (1988) in their article 'The Signs of All Times' propose a neurobridge backwards in time to the Upper Palaeolithic by which we can gain insight into the nature of the origins of art. Our nervous system has not changed much in the past 100,000 years. We are still physically very much the hunter-gatherers we were prior to agrarianism. In the signs of Upper Palaeolithic art Lewis-Williams and Dowson see entoptic phenomena very similar to those produced by people in altered states of consciousness today. 'Entoptic' is derived from the Greek for 'within vision', that is, anywhere within the optic system between and including the eye itself and the cortex where signals from the optic nerve are interpreted (Lewis-Williams and Dowson, 1988). Lewis-Williams and Dowson further break these down into 'phosphenes' which can be produced by physical stimulation (such as the patterns seen when you close your eyes and apply gentle pressure to your eyelids), and 'form constants' which are produced beyond the eye in the cortex itself. It is the latter which Lewis-Williams and Dowson (1988) primarily focus on, though they do not exclude phosphenes, and refer to both under the general term 'entoptics'. They do, however, distinguish between entoptics and hallucinations. Entoptics are geometric patterns whose origins are in the nervous system itself, whereas hallucinations are iconic and culturally determined and may be experienced in all senses (aural, visual, tactile, olfactory and synesthetic) not just the visual. Hallucinations may arise out of entoptics as will be outlined shortly.

The six entoptic forms which Lewis-Williams and Dowson (1988) see in common with Upper Palaeolithic art and contemporary research into drug induced visions are the grid, parallel lines, dots, zigzag lines, nested catenary curves, and filigrees (thin meandering lines) (see diagram). Lewis-Williams and Dowson (1988) do not address the spiral form directly, stating that it needs to be addressed all on its own. These basic forms manifest according to six principles: replication, that is, the entoptic itself, fragmentation, or broken down, (example: a ladder like form is a fragment of a grid); integration, where two or more forms combine (example: a zigzag grid); superpositioning, where one form appears atop another; juxtapositioning, where forms appear next to one another; reduplication, or multiples; and rotation.

 Lewis-Williams and Dowson (1988) outline three stages in the process; first there are the entoptic images alone, then these begin to be elaborated into iconic forms, and the final stage is that of intensification of the iconic forms. In the second stage, entoptics are elaborated into iconic forms, and this elaboration is informed by cultural expectations as well as individual dispositions. In a shamanic society, such as Lewis-Williams and Dowson (1988) argue is the San, a shaman in training may be encouraged to 'guess' at what the entoptics represent, and some measure of control and manipulation of the entoptics into culturally prescribed forms is encouraged. In the final stage, intensification, the experience moves beyond that of simile (I see something like) to a perception of the experience itself as being real.

 While many things may contribute to altered states (psychoactive drugs, sensory deprivation, fatigue, intense concentration, migraine, hyperventilation, rhythmic movement, schizophrenia, brain damage, intense emotion, stress, food and water deprivation, withdrawal from alcohol, advanced syphilis, crystal gazing, fever, etc. (Lewis-Williams and Dowson, 1988; Asaad and Shapiro, 1986; Siegel, 1977)), Lewis-Williams and Dowson (1988) focus primarily on those induced by psychoactive drugs.

 Two papers which are cited by Lewis-Williams and Dowson (1988) bear special examination, those by Ronald K. Siegel (1977), and Ghazi Asaad & Bruce Shapiro (1986). Siegal begins his paper by outlining early research, especially that of Heinrich Kluver who, in 1926, noticed that hallucinations seemed to occur in two stages, the first being related to four types of geometric: the grid, described variously as lattice, filigree, honey comb, grating, fretwork and chessboard; cobwebs; tunnel, also associated with cone, vessel, funnel, alley; and spirals, the category deliberately unaddressed by Lewis-Williams and Dowson (1988) for reasons not entirely clear. The second stage was that of iconic images which Kluver interpreted as being drawn from memory. An interesting point is that there seemed to be constants of theme in the more elaborate iconic images, the most common being religious symbols and images, followed by images of small animals and human beings (Siegal, 1977). Given that religious iconography is culturally determined, it would seem to support Lewis-Williams and Dowson's assertion that this is a part of stage two in the process (Lewis-Williams and Dowson, 1988) and may explain a certain uniformity in subject matter in the more elaborate depictions. In the Upper Palaeolithic a cultural constant in hallucination may have been the horse, judging from its ubiquitousness in the art (Leroi-Gourhan, 1968).

 Siegel (1977) then goes on to describe an experiment of his own in which an experimental group and a control group were given an alleged drug (for the control group it was a placebo, for the experimental group it was THC, the psychoactive ingredient in marijuana), and then individuals from the groups laid down on a bed in a light and soundproof chamber. Their descriptions of what they were experiencing were recorded on tape. The control group experienced "amorphous" black and white forms including lines and curves which moved randomly in their visual fields. The experimental group typically described the experience as being like a motion picture or slide show. There were many geometric forms which would frequently combine, duplicate and superimpose. Some went on to experience complex imagery, scenes of people and objects (Siegel, 1977). Siegel goes on to describe a further experiment, a 'refinement' of the first, where subjects were trained how to respond so that they could keep up with the rapid progression of images, a problem in the first experiment. It will not be outlined here, since training subjects in what they may expect to see renders the results of the experiment virtually inadmissible. It's a process which went out with the trained introspection experiments of Wilhelm Wundt at the turn of the century (Matlin, 1994). The results are tainted by subjects' desire to please and by the influence of their trained expectations on their perceptions, and would be of special concern if indeed the shift from geometries to iconic imagery is as determined by learning as Lewis-Williams and Dowson (1988) suggest, and in regard to the cultural constants which I would suggest were present in the later stages of hallucination in Kluver's experiments (Siegel, 1977).

 Asaad and Shapiro's article (1986) is an overview of the psychiatric literature and has a reassuring quality found in good social science: qualifications, doubt, and caution. As qualified and tentative as many of their conclusions are, it would seem that the pattern of stages outlined by Lewis-Williams and Dowson (1988) and found in Kluver (Siegel, 1977) are indeed most strongly associated with the use of psychoactive drugs. Such hallucinations are predominantly visual and preceded by abstract forms, examples they cite being stars, lines, circles and flashes of light (Asaad and Shapiro, 1986). An example of their cautious tone which also includes support of a 'learned' element in elaboration ("psychological background") is as follows:

It is important to keep in mind that a certain drug may not produce the same hallucinatory effect every time. These effects may vary according to the person, dose, mood, social setting, and physical condition. Conversely, the same hallucinatory experience may be produced by a wide variety of different drugs. In any case, the nature of the material hallucinated is greatly influenced by the individual's psychological background. (Asaad and Shapiro, 1986)

These are only two of several articles from the disciplines of neuropsychology and psychiatry cited by Lewis-Williams and . Hopefully the evidence provided by the others is stronger than that in the Siegel (1977) article.

According to Lewis-Williams and Dowson (1988) the actual translation of experienced entoptics and hallucinations into art would have come about initially as people in altered states of consciousness literally 'traced' the images they were experiencing on the wall. They note that such images might have been well placed for this since often they are localized on walls and ceilings like projections, and they would not be creating something new, but "touching and marking what was already there" (Lewis-Williams and Dowson, 1988). Later, once they had gotten the idea, they would have been able to paint images recollected from when they were in an altered state of consciousness. Finally works could be created which were derived from earlier created works without the necessity of the artist ever having been in an altered state. In terms of this process, there is no difference between the entoptic and iconic images. They are each what was actually seen and marked, though at different stages or in different degrees of alteredness.

 Following Lewis-Williams and Dowson's (1988) article are some very critical responses to this theory. Paul G. Bahn notes that there are a few basic doodle shapes which would be conspicuous if they did not appear in Upper Palaeolithic art. Australian anthropologist John Cleg notes that if these images are the product of "neurological hardware" that they should occur in "ordinary people's ordinary pictures", and that the six entoptics represent a good set of basic forms, while the seven principles are "a good basic guide to variation. Graphic artists as well as musical composers sometimes work (or play) with variations on a theme." (Lewis-Williams and Dowson, 1988).

 Joan M. Vastokas, an anthropologist at Trent University in Ontario, also has criticisms related to art-making and questions the origin of art in the literal tracing of hallucinated images, especially since these images are not static, but dynamic, constantly moving, changing, shifting. Indeed, this is what caused Siegel (1977) to adopt the questionable strategy of training his subjects, so that they could report faster and keep up with the rapidly changing images. While supporting the concept of multidisciplinary exploration of this problem, Vastokas regrets that no actual artists have been consulted to provide insight into how it's done really (Lewis-Williams and Dowson, 1988). While information of this sort provided by artists would be analogical, it could prove useful, arguing along the same sort of lines as our nervous system not having changed since the Upper Palaeolithic: How different is pigment placing now from back then? Even the modern airbrush is foreshadowed in the blowing of pigment in Upper Palaeolithic art.

 A danger inherent in a multidisciplinary approach is the qualifications of members of one discipline to assess the quality of evidence provided by members of another discipline. An example of this may be Lewis-Williams and Dowson's (1988) citing of the Siegel (1977) article whose methodology would rate at best a C- in an Intro to psychological Research Methods course. With caution, however, it is a very good approach, perhaps even an essential one for certain questions, and coupled with review by professionals in the involved disciplines, is a strategy which should be more pursued.

 Colin Martindale, a psychologist at the University of Maine, cites the work of R. Kellog on children's art and wonders if perhaps the more abstract elements of parietal art are the work of children. Kellog theorizes that children's art is not 'abstract' because they are lousy artists, but rather that they are more concerned with abstract essences (what they 'know') than with realistic depiction (what they 'see'). Lewis-Williams (1991) thinks that child art doesn't answer several concerns - "the long association during the Upper Palaeolithic of entoptic with often remarkably 'realistic' depictions, combinations of entoptic and iconic elements predicted by the neuropsychological model, the occasional combination of different iconic images... the location of much of the art in remote galleries and inaccessible diverticules..." and that these are "better explained by the more extreme varieties of altered consciousness than by infantile perception." (Lewis-Williams, 1991)

 Clearly, the idea that parietal art is solely the production of children seems unlikely, but there is nothing to suggest that child art could not appear alongside "remarkably 'realistic' depictions" by adults, and the remote and "inaccessible" parts of caves might be more easily reached by a smaller person. We should put aside any contemporary ideas of it being wrong to let children run loose in caves, especially in view of the archaeological evidence for the presence of children in caves.

... practically all known footprints were made by young people... The long line of prints at Aldene suggest a group of youths running through the cave corridors; at Niaux and at Montespan, there are the tramplings of a few individuals at a point where the ceiling is low; in a puddle at Pech Merle, we have the footprints of a boy, perhaps accompanied by a woman, and the imprints of small heels in the puddle-shaped depression at Le Tuc d'Audoubert. Thus it is certain that children went inside the caves. Elsewhere the footprints are found at points well away from the normal route through the cave... (Leroi-Gourhan, 1968)

We also can't be certain that people in the Upper Palaeolithic would like us have discouraged children from painting on walls, and we can't be certain that there was a curator who insured that only sanctioned art appeared, and not any 'early works.' As for the entoptic elements "predicted by the neuropsychological model", Kellog has catalogued 20 basic forms in children's art, including the six 'entoptic' forms (cited by Martindale in Lewis-Williams and Dowson, 1988).

 Another concern is that of sequence in time. Leroi-Gourhan (1968) divides Upper Palaeolithic art up into a chronology of styles. The 'signs' fit into this chronology and change over the progression. This change is not problematic for the iconic images since in Lewis-Williams and Dowson's theory they are largely culturally determined derivatives from the fundamental forms, but according to their theory the entoptics have their origin in the nervous system itself - they are fundamental universals and should not change as a function of time and style.

 Richard Bradley (1989) believes he is supporting Lewis-Williams and Dowson with evidence from Neolithic tombs in southern Brittany. There is a clear stylistic progression over time, roughly divisible into two periods in one of which entoptic-like forms predominate. The problem is that it is the second period which is dominated by entoptics, which runs counter to what one would expect given the stages and the nature of the iconic to evolve out of the entoptic. At very least we would expect entoptics and iconics to coexist as they do in the Upper Palaeolithic art and in the art of the San which Lewis-Williams and Dowson use for analogical support.

 The Brittany evidence also raises the question of the direction of the progression (if indeed there is a progression between the abstract and the iconic). Perhaps these iconic images are actually 'signs' abstracted from more iconic representations. The direction would then be from the more literal to the more abstract, where the abstract is a kind of short hand and might be considered as an early forerunner of written language.

 Leroi-Gourhan said "At Lascaux I really believed they had come very close to an alphabet." (recalled by Brigitte Delluc and Gilles Delluc of the Musee de l'Homme, Paris, in Lewis-Williams and Dowson, 1988), and it is likely that early sign-alphabets would be indecipherable to all but those members of the group that developed them, the ones who knew what they stood for.

 Writing proceeds from pictures of visual events to symbols of phonetic events...

The closer writing is to the former, the more it is primarily a mnemonic device to release information which the reader already has. The protoliterate pictograms of Uruk, the iconography in the early depictions of gods, the glyphics of the Maya, the picture codices of the Aztecs... are all of this sort. The information they are meant to release in those who look upon them may be forever lost and the writing therefore forever untranslatable. (Jaynes, 1976)

Leroi-Gourhan also notes associations between signs and representational images, and it is tempting to think of an Upper Palaeolithic ABC book.

 The final criticism has to do with the question of the late appearance of art. Lewis-Williams and Dowson (1988) state that the ability to hallucinate is not exclusively the property of human beings, that it is something common to all higher mammals: "Indeed, Australopithicines probably hallucinated." (Lewis-Williams and Dowson, 1988).

 We've had a nervous system capable of hallucinating for millions of years, our hands have been free from the task of locomotion for millions of years, yet art doesn't appear until possibly 300,000 BP on the evidence of an engraved bone (Lewis-Williams and Dowson, 1988) and doesn't really blossom until the Upper Palaeolithic. Lewis-Williams and Dowson do not address this question adequately, leaving it until their closing paragraph where in a sentence they attribute it to changing social circumstances (Lewis-Williams and Dowson, 1988).

 The Swiss psychologist Carl Jung found that psychologically stressed clients would sometimes spontaneously draw mandalas (circular patterns with a clearly defined centre), and he considered the use of art important in psychotherapy. He didn't refer to it or consider it art, but 'active-imagination' in the case of conscious imagery, and 'amplification' in the case of dreams (Jung, 1959). Could art have been a collective response to a stress greater than had ever occurred since we became bipedal?

 It is something of a stretch to try and find the origin of art in art therapy, but the general direction of the thrust, the development of a new behaviour in response to stress, might suggest at least a direction in which to look. It is easier to question theories about the origin of art than to come up with much that is solid. The temptation is toward despair, but that may not be the only course, as suggested in Brian Hayden's article 'Alliances and Ritual Ecstasy: Human Responses to Resource Stress' (1987). Though Hayden is not dealing with the origin of art, he makes some general points which apply; "... archaeologists have also had to develop evolutionary models to explain the emergence of other basic kinds of behavior that characterize human beings..." and "... archaeologists have found it most useful to approach explanations of these widespread types of behavior in terms of evolutionary ecology and the selective survival advantages of such behavior."

Obviously, one of the drawbacks of the archaeological perspective is that many of the behavioral and cognitive details of rituals and other types of behavior are lost, so that only broad categories of behavior can generally be examined. However, this is not always a liability inasmuch as some of the most interesting questions that we can ask deal with why certain basic kinds of behavior characterize the human species. (Hayden, 1987)

He goes on to note in relation to the subject of his paper that effort, distribution and persistence of ritual and religion are such that there must be "clear and major selective advantages" and questions what they might be (thus setting up the rest of the paper).

 Without launching into what would amount to another essay, I will conclude with this question and with this proposed approach to the problem: what are the adaptive advantages to art? Is it a record of cultural memory that passes down the knowledge of past generations to future generations (education)? Does it support the creating of "bonds between people through an entity that transcends the importance and the identities of the individual participants" (Hayden, 1987) (religion)?

 Psychoactive drugs may well have played a part in the creating of art, as they have through the Romantic poets to the Beatniks of the 1960s, and beyond. Yet they are neither necessary nor sufficient to answer the question of the origin and persistence of art. In a time when art is perceived more or less as a luxury, it is difficult to see that it could ever have been 'essential'. But that is the premise to operate from in future exploration of this question.



Asaad, G., and B. Shapiro 1986. Hallucinations: Theoretical and clinical overview. American Journal of Psychiatry 143:1088-97

Bradley, R. 1989. Deaths and entrances: A contextual analysis of megalithic art. Current Anthropology 30:68-75

Hayden, B. 1987. Alliances and ritual ecstasy: Human responses to resource stress. Journal for the Scientific Study of Religion 26:81-91.

Jaynes, J. 1976. The origin of consciousness in the breakdown of the bicameral mind. Boston: Houghton Mifflin Company.

Jung, C.G. 1959. The archetypes and the collective unconscious. Princeton: Princeton University Press.

Leroi-Gourhan, A. 1968. Treasures of prehistoric art. Translated by N. Guterman. New York: Abrams.

Lewis-Williams, J.D. and T.A. Dowson 1988. The signs of all times: Entoptic Phenomena in Upper Palaeolithic Art. Current Anthropology 29:201-245

Lewis Williams, J.D. 1991. Wrestling with analogy: A methodological dilemma in Upper Palaeolithic art research. Proceedings of the Prehistoric Society 57:149-162

Matlin, M.W. 1994. Cognition. Harcourt Brace.

Siegel, R.K. 1977. Hallucinations. Scientific American 237:132-40

Thoughts or Comments (?)

© 1996 Eric Pettifor

to Archaeology contact Eric to Greater Anthropology to contents