Lecture Three: Shamanism 3.5

3.5 Prehistoric sources of meaning

It has been mentioned that shamanism is not only widespread worldwide, but can be dated back to prehistory. How is it possible to support such a claim? As mentioned, shamanic cultures do not leave behind impressive monuments or elaborated written witnesses. Most of their culture is transmitted orally and left to individuals to preserve. One of the reasons why several historians and archaeologists have argued that something like shamanism must have been common even in prehistorical times has to do with two related facts. On the one hand, there are some similarities between archaeological findings and ethnographic evidence from today’s cultures in which shamanism is still practiced. Based on these analogies, it is tempting to use evidence from contemporary small-scale cultures to interpret archaeological findings. On the other hand, and perhaps more importantly for the present discussion, there seems to be a conceptual need to assume that prehistorical humanity operated and regarded the world in a shamanic-like way, in order to fully understand some of the major transformations and evolutions that humanity itself underwent. As an appendix to the above discussion, two connected case studies can be briefly considered: the invention of agriculture at the beginning of the Neolithic era (ca. 12,000 years ago) and the explosion Upper Palaeolithic cave art in the Franco-Cantabrian region (ca. 44,000 years ago).

How did humans begin to domesticate nature, by raising cattle and cultivating fields? In his The Birth of the Gods and the Origins of Agriculture (original French edition 1994), Jacques Cauvin suggested that this advancement has to be explained by taking into account changes in the way human beings understood the world and their place in it. Cauvin focuses on what has been called by former generations of archaeologists ‘the Neolithic revolution,’ namely, the period around 10,000 years BCE where human groups living in the ‘fertile crescent’ in the Middle East began to practice agriculture and animal farming. Relatively quickly, these practices spread to other regions and transformed the way of life of human groups, who shifted from being mostly nomadic hunter-gatherers to a more sedentary life in a fixed territory, eventually leading to the foundation of the first urban cultures in Mesopotamia some five millennia later. Domesticating wild plants and animals was clearly a major step in the human quest for control and dominion over the natural world.

Cauvin’s main claim is that previous attempts at explaining this transformation have given excessive attention to material, environmental, and utilitarian aspects, while neglecting equally important transformations that took place in the way in which Neolithic human beings thought and conceived their own experience. According to Cauvin, archaeology has been strongly influenced by Marxist ideas about the fundamental role of economic structures, shaped by power and production relations among the individuals constituting a certain culture. This Marxist ideology would then consider any intellectual activity, including symbolic expressions, art, and religion, as part of the ‘superstructure’ of a culture. Cauvin concludes his book by writing:

The hard facts mentioned in the quote concern archaeological evidence that (i) domestication of plants and animals cannot be explained simply by an appeal to growing populations (and needs for food supply), nor to a reaction to environmental changes; (ii) various elements that led to domestication were already integrated in pre-Neolithic cultures, without them necessarily being associated with the practice of agriculture or animal farming. Based on these considerations, Cauvin argues that something other than sheer utilitarian needs must have propelled certain human groups to experiment with domestication, and he locates this factor in what he calls ‘the revolution of symbols.’ Cauvin focuses on the Khiamian period (occurring just before the Neolithic) and identifies in it the emergence of a new form of symbolism, compared with those of previous periods. From the Palaeolithic period onwards, human beings produced images, mostly of a relatively restricted group of animals, and less commonly fully human forms (as we shall discuss in a moment). In the Khiamian period, though, Cauvin sees a concentration of two key symbols: woman goddess and bull.

None of these symbols are entirely new. Female figurines were produced throughout the Palaeolithic, but according to Cauvin:

In his discussion, Cauvin goes on to argue that the emergence of this new symbolic and religious thought increasingly dramatized the way in which former cultures represented the divine. He detects a new connection between the divine and a sense of tragedy, death, suffering; and hence the urge for change in order to find salvation:

According to Cauvin, this revolution of symbols created the mental energy needed for early humans to attempt domestication, and for this to eventually spread in time and space. The development of Neolithic culture thus also shows a development of religious cults, with sanctuaries constructed on the ground (in contrast with the use of natural caves in the Palaeolithic), increasingly rich and complex burial rituals performed, and a growing production of religious images. All of this occurred, though, without compromising ‘the concept of an egalitarian structure for Neolithic societies’ (Cauvin 2000, 120).

By countering a more materialist approach that identifies the main drivers of human change in economic and utilitarian motives, Cauvin shows that something more is needed to explain such a profound transformation as the one that occurred in the Neolithic period. Human thought, and religious symbolism in particular, need not be the only factors taken into account seriously, but can help explain the different mind-set that led prehistoric human beings to radically change their relationship with natural resources and, in turn, their whole way of life. However, Cauvin is very cautious in elaborating on the sort of thinking proper of Neolithic people. When he ventures in this domain, he seems to resort to a rather standard (and arguably anachronistic) interpretation of symbols as mimetic representations of concepts. Female figures are thus interpreted as representing ‘mother,’ hence perhaps ‘security’ and ‘salvation,’ while the bull is a symbol of ‘virility.’^[1] More significantly, Cauvin stresses the emerence of some form of transcendence, in which divine symbols are meant to represent a reality that (based on Cauvin’s description) should be conceived as beyond the human realm, ‘external to man and elevated above him.’ This point seems necessary in order to create the dynamic element that Cauvin is looking for, as the propeller for an initiative of domesticating nature. His intuition seems to be that only by conceiving of something superior (and hence better, safer, more powerful, and so on), human beings could start thinking about how to improve their own material conditions. Thinking of the divine in a more vertical (if not transcendent) way would then offer a blueprint for imagining a better condition, and this would serve as a prelude to the human initiative of somehow actualizing or striving for improving the ordinary condition.

For present purposes, it should be asked: where did Neolithic people find their source of meaning, in Cauvin’s reconstruction? If female images and bulls are symbols of the divine, where is their meaning located in the spectrum of experience? Fertility, mother, birth, virility, fight, seem to be the sort of meanings that Cauvin attributes to Neolithic religious symbols, and these are all accessible in ordinary waking experience. Hence, religious symbols talk about non-religious experiences, perhaps amplifying and embellishing them, but substantially remaining in the same spectrum of meanings. In other words, Cauvin’s view entails a form of naturalism and realism in its interpretation of Neolithic religious symbolism.

To represent (or signify or create a symbol of) a domain of divine ‘above-ness’ (or ‘transcendence’), human beings need to have access to the meaning of ‘above-ness,’ namely, they need to have a way of experiencing something (some content) that will then be signified as ‘what is above.’ Perhaps the best way of thinking about this involves taking something ordinary and imagining the same content but in a much more powerful and amplified way. However, the invention of meaning clearly does not work in this way. If the imaginative increase somehow added to the ordinary content does not create a qualitative discontinuity, then a very powerful bull remains just a bull, the most maternal woman remains a woman. On the contrary, if imagination creates the discontinuity, then it is only by equivocation that one might keep referring to ‘mother’ or ‘virility’ as the meaning conveyed by religious symbols, since these symbols are meant to refer to something qualitatively different (a content experienced in a different realm, or in a different domain of experience) with respect to the ordinary meanings (like ‘mother’ or ‘virility’). Hence, Cauvin’s realism either undermines his case for the indispensable role played by religious thinking, or it needs to abandon its realist stance and seek the meaning of religious symbols somewhere else than in the ordinary way of life.

Cauvin rightly points out that the material and technical conditions needed for the practice of agriculture were to some extent already present prior to the Neolithic period proper, although they were not put to use. Rightly, again, he contends that the activation of those potentials for transformation required more than just material or utilitarian stimuli, they required a change of mind-set, what he calls a revolution of symbols. But symbols are products of human imagination, and their meaning is constructed, changed, and reshaped by how humans play with the power of their imagination. For present purposes, we do not need to decode prehistoric symbols and get clear on what exactly the change amounted to. It suffices to say that if a symbolic change seems necessary to account for a new mindset, this change must have originated in that domain of experience in which symbols are created and enacted, namely, in some sort of poietic practice. And this observation entails that those poietic practices must have been already well established at the time for them to undergo a profound reshaping. It is thus tempting to say that something akin shamanic poietic practices must have been established, and some change in their domain might have underpinned the revolution of symbols tackled by Cauvin.

In contemporary archaeology, David Lewis-Williams has been perhaps the boldest and most controversial advocate of the thesis according to which shamanic practices must have been operating in prehistoric time. More specifically, he used this hypothesis as a hermeneutic key to interpret significant monuments of prehistoric art. Lewis-Williams shares with Cauvin a dissatisfaction for the materialist approach, insofar as it has led to an overemphasis on utilitarian motivations as driving factors of human development. He also stresses that the earliest instances of human artmaking must be understood as the expression of religious experiences, which were so deeply felt that it urged humans to somehow fix them in outside artefacts and paintings. Lewis-Williams develops this interpretation by drawing explicitly from ethnographic studies on today’s shamanism and on (early) neuroscientific studies on altered states of consciousness. Combining these strands, he contends that since the Palaeolithic period, human artmaking has been an attempt at expressing experiences resulting from the induction of altered states of consciousness, then interpreted in religious and cosmological terms. It was this sort of experiences that fashioned the Palaeolithic worldview and must also have impacted social interactions (and possibly early forms of discrimination not based on age and sex, but on the ability of ‘seeing’ or experiencing those altered states). Lewis-Williams has extended this interpretation from the Palaeolithic to cover Neolithic art as well (Lewis-Williams and Pearce 2005) discussed by Cauvin.

The genre of Palaeolithic art discussed by Lewis-Williams (2002) mostly concerns cave paintings in France and Spain, which can be dated between 45,000 and 35,000 years ago. Cave panting is also associated with the first ritual burial decoration and the production of portable art (little carved objects). Lewis-Williams confidently asserts that

This point on language seems to be crucial for Lewis-Williams’ argument, since he associates language skills with the ability to entertain a higher form of consciousness, capable of acting upon memories and anticipations, and also of experimenting with altered states. This form of consciousness would be typical of homo sapiens, but perhaps lacking in Neanderthals.^[2] Lewis-Williams then argues that having access to this more sophisticated consciousness, early human beings had also to face its implications, and they had to come to term with them, or rather decide how to interpret them. The evolution of human society can then be seen (from such an early period) as a progressive negotiation between the ‘spectrum of consciousness’ and social settings in which individuals experiencing various conscious states (more or less altered) lived. There is an interplay between a ‘social contract’ and a ‘consciousness contract’ (Lewis-Williams and Pearce 2005) that determines what each society at each stage of its history accepts as normal and meaningful in terms of the range of experiences available to its members.

A key point in this reconstruction is the fact that both individual experience and how this experience is framed in the social context co-determine the construction of meaning:

While Cauvin stressed that even Neolithic societies remained fundamentally egalitarian (and significant social structures and hierarchies would not emerge until the age of the first urban cultures), Lewis-Williams sees social differentiations (hence inequality and even forms of discriminations) as intrinsic to human sociality almost since the beginning of it.

The core of this interpretation consists in rejecting any form of direct realism or mimetics between an outside reality and image-making. In order to explain, for instance, why Palaeolithic people chose to focus only on a relatively restricted number of animals in their paintings, it is necessary to understand what these animals meant for those people. However, ‘people did not invent two-dimensional images of things in their material environment. On the contrary, a notion of images and the vocabulary of motifs were part of their experience before they made parietal or portable images’ (Lewis-Williams 2002, 185). Images and motifs (the interpretation goes) were provided by the induction of altered states of consciousness, which (in Lewis-Williams’ account) show recurrent patterns and structures, albeit variously inflected based on different cultural and historical circumstances.

Palaeolithic caves were used to seek and induce altered states of consciousness (or trance-like states, to use the terminology introduced above), which then served as basis for artmaking in the same caves; images thus became prompts for resurrecting and re-experiencing those altered states. Moreover, access to caves (and hence to altered states) was socially regulated and constrained, and the source of meaning was controlled and asymmetric. Not everybody had the same opportunity (or even the right) to have the same experiences. Artmaking was thus not only socially cohesive (insofar as it provided an embodied repertoire of meaningful signs), but also divisive and discriminative.

Lewis-Williams’ account entails that meaning does not come from outside material experience, but from the autogenous power of imagination, its visions and its emotions, which urge humans to express and socialize them. The divine is presented as somehow ‘above’ humans in order to express this difference between ordinary daily states and altered states. But the experience of the divine is not just a sheer fancy. It is as real as any other experience, albeit it happens in a mode of experience different from the one usually encountered in daily life. However, once this mode of experience has been explored and taken as the source of meaning, it sheds its light on all the other aspects of experience. Cave walls become ‘membranes’ that separate human seers from the spiritual realm, and stones are not just inert materials but speaking entities. As happens in shamanic cultures, religious meanings forged in the domain of altered states determine the cosmology that is then used to interpret all other aspects of experience, which progressively become nothing but signs of those meanings. Poietic practices bring forth a whole world of meaning together with those who inhabit them.

Lewis-Williams’s account is provocative. While its early reception was positive, it has more recently, received sharp criticism. The most complete account of this controversy is available in Paul Bahn’s Prehistoric Rock Art: Polemics and Progress (2010, especially chapters 3 and 4). The main objection put forward is that Lewis-Williams’ model is at best an unfalsifiable hypothesis, and at worse wishful thinking based on an ad hoc selection of data. For instance, the notion of ‘shamanism’ seems either too broad (and thus explanatorily insufficient), or too focused on the case study of circumpolar Asian shamanic cultures, and thus unsuitable for the interpretation of other cultures around the world, especially Palaeolithic art. Lewis-Williams used a specific neuroscientific model to analyse altered states of consciousness and derive fixed experiential patterns to be uncovered in artmaking. But this model was based on studies conducted with hallucinatory drugs (LSD), which have unique effects, and which are unavailable in many cultures and (most likely) in prehistoric times. Hence, the kinds of patterns discovered by Lewis-Williams would not be something ‘wired in the human brain,’ but heavily shaped by the use of specific psychotropic substances. What seems at stake in these (and other) criticisms is a concern with the sort of ‘one-fits-all’ explanation that Lewis-Williams’s account suggests.

Critics of Lewis-Williams favour an approach that involves accurate description of different cases considered in their own right, without jumping to undue generalizations, and an acceptance that most information about prehistoric times will be unrecoverable for us. Yet, one might question whether this line of criticism strays too far from the essentials of Lewis-Williams’s approach that have been briefly presented above.

Derek Hodgson, one of Lewis-Williams’ critics, has presented a significantly different account. He introduces a different neurological model based on the visual mechanisms that allow the brain to recognize visual patterns and geometric forms in order to articulate and make sense of experience. Concluding his discussion, Hodgson states:

Altered states and shamanism are dismissed as inadequate. A neurological explanation is retained, but of a much different kind, since Hodgson’s account is motivated predominantly by evolutionary and utilitarian factors. Prehistoric human beings privileged certain shapes (either certain animals or geometrical patterns) and even reproduced them in their art, because doing so constituted a certain adaptive advantage. This would entail that caves were perhaps more akin to ‘schools’ than to ‘sanctuaries’ (although, it might be added, the difference between the two is notoriously difficult to trace).

But what animal is so badly adapted to its environment and so slow-minded that it needs to ‘take notes’ (even invent painting) in order to discriminate ‘the threatening from the useful and benign’? If this were the case with humans, they would probably have gone extinct long ago. Even granting that cave art might have also had utilitarian benefits (since art usually tends to have multiple reasons and ways of employment), seeking the origin of art in these utilitarian factors seems to run into the same problems already exposed by Cauvin. In fact, more recent critiques of Lewis-Williams tend to incorporate a more serious consideration of other rationales.

Bahn himself, for instance, accepts that

Myths are not far from what Cauvin called ‘symbolism.’ Why are myths relevant? Because of religious values associated with them. Why do religious values matter to people (past and present)? Because of the emotional overtones of religious experiences, because of the meaningfulness that they possess.

In a more recent paper, Patricia Helvenston and Derek Hodgson (‘The Neuropsychology of Animism,’ 2010) proposed replacing ‘shamanism’ with ‘animism,’ and then offered a neuropsychological model in which ‘animism’ would be connected with neural mechanisms through which human beings project life onto non-human objects (e.g., their having a ‘spirit’ or enacting a form of agency) and interpret them as somehow partaking in the same ‘spiritual world’ as human do. This model is based on the mechanisms that underpin visual perception and claims to more accurately explain artmaking, although the authors make absolutely clear that the actual attribution of meaning to images is deeply shaped by emotions. Helvenston and Hodgson note that human-forms are attributed to other objects most likely in conditions of uncertainty. Animism itself might have a ground in neurophysiology and neuropsychology, but it is actually seen as a meaningful option because it allows individuals to cope with uncertainty by decoding their experience while keeping fear and anxiety at bay. When the authors contrast animism with shamanism, they focus on the relatively more structured core of beliefs that is found in relatively late versions of the latter:

What is at stake here is mostly a matter of definition. In short, the authors propose using the term ‘animism’ to refer the older and less-systematic set of beliefs concerned with the idea that all sorts of non-human entities were endowed with ‘spirit,’ while reserving ‘shamanism’ only for a more specific, structured and localized (both in space and time) phenomenon, which should not be overly generalized and used as a paradigm for ‘animism.’ In this perspective, shamanism is a form of animism, but animism is not reducible to shamanism.

This being said, the authors do not deny that animism is associated with ‘primordial ecstatic experiences.’ In fact, without such experiences, it would be impossible to explain why the projection of spiritual forces upon non-human entities should have been taken seriously by individuals, instead of being dismissed as a perceptual mistake. When the stick looks broken in water, one might think that one’s eyes are tricked, or that something supernatural is going on. Why opt for one interpretation rather than another? It depends on the source of meaning. ‘Primordial ecstatic experiences’ (or trance-like states, to come back to Rouget’s more precise terminology, or ‘poietic practices’ to use the expression introduced in this Lecture) seem necessary to make sense of the fact that ‘primordial people’ acted in ways that clearly suggest they were deriving meaning from a specific mode of experiencing reality, endowed with a cogency and emotional vividness probably different from that of more ordinary waking experience. Having put the source of meaning there, they interpreted other aspects of their experience from that point of view. However, can we know what ‘primordial ecstatic experiences’ might have been like? If ‘ecstatic experiences’ were completely lost, the expression could not make any sense. But this is not the case, since ethnographic evidence shows that they are still practiced (in a rich variety of ways) by a large number of different cultures around the world, and historical records suggest the same is true of the past.

Margaret Bullen, in another critique of Lewis-Williams’ neurological model, concludes:

Here, again, the ability of prehistorical people to access trance is not questioned but granted. Image makers are described as ‘clever people capturing their dream images.’ Dream images are usually taken as a (mild) instance of altered states of consciousness. Hence, cave art comes from altered states. The issue for such critics seems to be the implicit determinism lurking in Lewis-Williams neuroscientific model. In turn, their aim seems to be to defuse the potential threat that it might represent for human (artistic) agency. This is surely a point worth stressing. But this concern leads us away from what is perhaps the most important philosophical point in Lewis-Williams’s account: the centrality of a directly felt poietic experience in prompting Upper Paleolithic people to express that experience in images. It was this sort of experience that established a sign-meaning distinction. With this distinction in place, trance-like states become the source of meaning that is then expressed through the manipulation and interpretation (more or less advanced) of natural objects like cave walls, which are then transformed in signs referring to that meaning.

We might swap shamanism for animism or mythology, but the fundamental sort of poietic experience cannot be removed without jeopardizing the intelligibility of the signs left in prehistoric art. Perhaps ‘shamanism’ is too broad or fuzzy a category and perhaps ‘animism’ is more precise and accurate. Lewis-Williams offered an outdated and partly biased neuroscientific model directly based on altered states of consciousness. But also Helvenston and Hodgson provide a neuropsychological model for the ‘animist projection’ that the human brain makes onto various entities that must be complemented by some account of the emotional charges and structures that would induce individuals to take this projection seriously, instead of dismissing it as a mistake. In any case, as we discussed in Lecture Two, neuroscience can help us appreciate the enabling conditions for first-person experience, but it cannot be used to reduce the latter to an epiphenomenon of brain states.

Shamanism, however constructed, is a sufficiently clear and well-instantiated phenomenon that allows us to look more closely at how particular domains of experience are used as a source of meaning and then provide a basis for creating signs expressing those meanings. It also shows how this process is embedded in a social substratum, which is made not only of material needs and power-relations, but also of a diverse and complex set of beliefs and emotional structures. Moreover, poietic practices are explicitly enacted for a purpose, namely, that of mastering uncertainty, in whatever way it is manifested and felt by both individuals and their community. In this process of mastery, the goal is to restore a form of harmony between different centres of agency that are involved in human affairs and upon which human wellbeing depends. But poietic practices do more, because in the enaction of a world of spirits and agents, and in the narration of how they can interact, clash, fight, or be pacified, they help one  to construct a distinctive form of selfhood. On our spectrum, this form of self is further away from the more naturalistic accounts we discussed in the previous lectures, insofar as it relies on a form of weak embodiment that denies a one-to-one correspondence between agency and physical individual bodies. As we saw, this form of selfhood has its own specific approach to the problem of uncertainty (domestication), its own techniques (poietic practices), its own heroes (shamans), and its own problems (the precarity of harmony).

What if we move even farther away from the naturalist ways of conceiving of the self and pursue even further that idea of ‘above-ness’ or even ‘transcendence’ evoked by Cauvin? This is the topic for the next Lecture.