Institute For Ice Age Studies

This article originally appeated in French as Les achives du Paléolithique in La Recherche, July 1994

"...mnemonic processes are indivisible
from the material act of representation "

Like neurological memory, which endures despite the constant replacement of brain cells, human cultural traditions and practices endure for millennia despite the constant replacement of one generation of human organisms by another. What are the cultural processes of memory, and how far back in time can we trace them?

We tend to think of memories as if they were simply past events somehow stored in our brains. However, it is more accurate to think of neurological memory as an electro-chemical process of representation. In the same way that a voice heard on the telephone is not really the voice of the other party, but a sonic representation (as in re-presentation) of it, a memory is not really the event or object remembered but an electro-chemical re-presentation of certain qualities of that object or event.

This notion of representation is also useful in understanding cultural memory. A newspaper article, a family photograph, a soldier's uniform, a computer image, a spoken word, a handshake and a cave painting are all linked by the fact that they are representations of some experienced reality. This very process of representation, especially when it takes material form, is critical to the ability of human societies to store experience and to reproduce it in subsequent generations.

Connerton has recognized three categories of neurological memory:

1. Personal memory, by which individuals represent to themselves the details of their own biographies.
2. Cognitive Memory, by which individuals represent to themselves knowledge of their surroundings, including geography, vocabulary and other learned 'facts.'
3. Habit Memory, which is a memory of gestures and operations.
   In contemporary humans and higher primates, and undoubtedly among our ancestors of the past several million years, electro-chemical representations in the form of neurological memory are the basis for what we recognize as experience or trial-and-error. We avoid past 'mistakes' of all kinds by remembering them; or perhaps more correctly by remembering more successful practices in place of less successful ones. In sum, mental images of past and simulated future experiences are electro-chemical representations that condition future behavior and the ways in which experiences of the moment are assimilated to existing memories.

   But, if existing memories condition new experiences, the opposite is also true. The representation of an experienced past through neurological memory is not some kind of "objective' data-base with "files" retrievable at will. Memory is active, dynamic, ever-changing and heavily conditioned by new experiences. The perception of current experiences is flavored by a remembered past; and the remembered past is continually reformulated in light of new experiences. This reciprocal process is at the heart of what makes experience meaningful. And, as Rose (1992:91) has observed, "...brains do not work with information in the computer sense, but with meaning. And meaning is a historically and developmentally shaped process..."

   While neurological memory is a process of internally representing an experienced reality, over the course of hominid biocultural evolution two new media of representation developed which rendered memory external to the individual human organism. Leroi-Gourhan recognized more than thirty years ago that speech and gesture. were parallel expressions of the human brain, which rendered public for the first time the electro-chemical representations of individuals.

   Speech. Vocal utterances in the form of speech served as the basis for the system of representation that we know as language: speech utterances acting in concert with a number of other cognitive and neurological capabilities (e.g., metaphor, foresight, rhythm) of the enlarged hominid brain. The evolutionary significance of such linguistic acts is that they are explicitly social in that they facilitate the sharing of mental representations within a collectivity. This externalization and socialization of electro-chemical memories through their linguistic re-presentation means that experience is rendered public. It can be discussed, disputed, rejected, elaborated, compared and, most importantly, it can be communicated from one human generation to the next. Connerton has referred to this inter-generational communication of knowledge and experience as "acts of transfer" from the neurological memory of one generation to that of the next.

   Gesture. Leroi-Gourhan recognized that, if speech is a medium of representation involving the production of meaningful vocal utterances, it has a parallel and complementary counterpart in gesture. Gestures combined with tools constitute techniques. For Leroi-Gourhan, gesture was to technique what speech was to language. Technical expressions (movements, objects, forms, rhythms, textures, colors) were no less meaningful than spoken ones. For this reason, Leroi-Gourhan saw the origin 35,000 years ago of what he called "graphism" as the beginning point for a whole new range of representational possibilities, the culmination of which would be written language, perhaps the most powerful form of cultural memory.

   The use of gesture in creating material representations relies less on Connerton's biographical and cognitive memory than on his habit memory. Following in the footsteps of Mauss and Leroi-Gourhan, Connerton observes that "Any bodily practice, swimming or typing or dancing, requires for its proper execution a whole chain of interconnected acts," what Leroi-Gourhan called a "chaine opératoire." Bourdieu termed these learned operational sequences, which are highly conventionalized and culture-specific "habitus." These operational chains are the constituent elements of all material representations; and as we shall see, they and their constituent gestures are recoverable from the archaeological record.

   Most archaeologists believe that ancient tools such as handaxes, side scrapers, cleavers and choppers were forms that ancient hominids sought to achieve, and that over time they developed the neuro-motor capacity for more and more complicated chaines opératoires. These more complex chaines opératoires are a clear indication of greater and greater capacity for habitual memory representations. However, an evolutionary transformation occurred around 40,000 years ago that Leroi-Gourhan labeled the "origin of graphism." Humans melded gestures and tools into techniques for creating representations external to the individual organism. Mental representations were transformed into material ones; and these representations operated not in the vocal-auditory channel , but in the formal-visual and material-tactile channel. In the same way that speech led to shared vocal representations, gesture was used to construct shared material representations.

   The consequences of durable, material representations for human culture were enormous. Material objects last long after the act of their creation. They continue to carry meaning in the absence of the person who created them. They accumulate histories of their own, which enhance their power and meaning by linking them to ancestors . The materials from which they are constructed, and the knowledge and techniques employed in their production can be manipulated for profound visual and tactile effect.

   Material representations then, are parallel and complementary to linguistic representations. Perhaps what is more important, is their permanence relative to verbal representations, which makes them a primary locus for the storage of cultural knowledge. The same icons can continue to act as a kind of mnemonic around which myths, ceremonies, adaptive strategies, environmental knowledge, genealogical descent and complex social relations are recalled, discussed and operationalized. While, for example, mythology can exist without being materially represented, its depiction in what Leroi-Gourhan described as mythography, gives it credibility, durability, immediacy, and authenticity. At the same time however, cultural icons are sufficiently ambiguous to be adaptable to new social and historical contexts. Their meaning is never self-evident to those who are strangers to the immediate cultural and historical context. Like internal memory-representations, material representations, whether in the form of a biblical passage, a home movie, or a family heirloom, are given new meanings and nuances each time they are employed. For this reason archaeologists confronted with even the most explicit material representations such as the painted animals of Lascaux, are unable to read them. While we may recognize a horse painted on the wall, we are in no position to infer the range of meanings and metaphors that horses might have evoked for the Magdalenians of 17,000 years ago.

   My own research over the past decade has focused primarily on material forms of representation during the first few thousand years following the origins of graphism about 40,000 years ago. For me, the great innovation at this defining moment in human evolution was in the material rendering or objectivation of ideas, forms, emotions, social relations, etc. Representational objects thus became part of "the reality of everyday life," which according to Berger and Luckmann , "is not only filled with objectivations; it is only possible because of them." According to them, humans are "constantly surrounded by objects that 'proclaim' the subjective intentions" of their fellows." In other words, social relationships are codified in representational objects: a kind of material mnemonic.

   In particular, I have been concerned with the earliest traces of material representation that we recognize as personal adornment. While it might be obvious that a painted cave like Lascaux could have served as a reservoir of cultural memory, how does such a model apply to the several thousand objects that were pierced for suspension on the human body?

   First, it is fundamental to recognize with Andrew Strathern that personal adornment is one of the best examples of gesture and tool being combined to construct and define social categories in material form: "What people wear, and what they do to and with their bodies in general, forms an important part of the flow of information -- as age, sex and status, which are also defined in speech and in actions." Thus, material constructions (for example, who could mistake a gentleman in a dark suit, wearing a bowler hat and carrying an umbrella for anything but an English businessman?), which fix and reify a limited range of social identities, serve as powerful, externalized (and highly mnemonic) representations.

   But, if material representations of social identity exemplified by clothing and costume are key aspects of social memory, so are the very habitual practices by which they are constructed and recognized. Who, among us, gives a passing thought while getting dressed in the morning, to the various social selves that we are creating (e.g., wealthy New York businesswoman) through a highly codified set of garments and objects? And, moreover, who among us is conscious of the act of identifying such a social category of person without a single word ever being spoken?

   Such material construction of social identity and of cultural knowledge is first seen in the archaeological record of Europe some 40,000 years ago, among the first European members of Homo sapiens sapiens, the Cro-Magnons. The habitual operations involved in the material construction of social identity have left clear traces in the archaeological record in the form of raw material choice and procurement, technical production sequences and regional differences in representational systems related to the communicating of group and inter-group identity.

   I wish to illustrate by using two examples from the Upper Paleolithic, one in the domain of personal adornment, the other in the domain of parietal imagery.

   1. Sungir: The Material Construction of Social Identity at 28,000 bp

      About 28,000 years ago, the residents of the Russian site of Sungir produced thousands of personal ornaments and a number of ivory carvings in geometric and animal forms. Sungir is one of the oldest known cases in which ornaments are actually found on human skeletons. While inhabiting Sungir, at least five of the site's occupants perished. According to Russian physical anthropologists these consisted of a 60 year-old man, a 7 to 9 year-old girl, a 13 year-old boy, an unsexed (male?), headless adult and an adult female skull. From my reading of the evidence, the adult burial is clearly an older male. The sex of the adolescents remains ambiguous.

      In total, the three most intact burials were lavishly decorated with more than 13,000 painstakingly prepared ivory beads arranged in dozens of strands, perhaps attached to their clothing. Although it is almost certain that the three individuals buried intact at Sungir were members of the same social group, there are remarkable differences among them in details of body decoration and grave offerings. For example, the man's forearms and biceps were each decorated with a series of polished mammoth-ivory bracelets (25 in all), some showing traces of red and black paint. Around his neck, he wore a small, flat schist pendant painted red, but with a small black dot on one side. The boy had an animal pendant around his neck and a belt of more than 250 fox canine teeth. The young girl had no animal teeth, but was accompanied by a remarkable series of carved ivory disks. She was decorated with nearly as many beads as the man and boy together. Meanwhile, nearby, the fragmentary remains of as many as six other individuals seem to have been accorded no special treatment at all.

      Experiments reveal that each of the ivory beads at Sungir took more than an hour to fabricate. Hence, the man's beadwork took more than 3,000 hours, while that of each child took more than 5,000. Considering additional objects placed on and alongside the corpses, it is clear that each of the childrens' burials had substantially more labor invested in it than that of the man. Based on the differences in grave offerings and labor investment revealed among these burials, we might be justified in inferring that the social system represented at Sungir was a hierarchical one in which social position was inherited rather than achieved; suggesting that materially-marked, internally differentiated social systems arose prior to and independent from economic systems based on agricultural production.

      The chaine opératoire for manufacturing the dominant form of bead at Sungir was clearly a variation on the Aurignacian assembly-line approach to ivory bead manufacture found elsewhere in Europe. However, at Sungir the blanks were scored across the width of each face before the hole was drilled. This caused the beads to fall into a visually impressive interlocking pattern when strung. In other words, the desired esthetic effect was deeply and probably unconsciously (habitus) embedded in even the earliest stages of the chaine opératoire.

      These differences provide a fragmentary view of the complex array of inter-generationally durable and materially fixed social identities that constituted the social fabric within Early Upper Paleolithic northern Russia. On a broader scale, Sungir provides yet another example of inter-regional differences in ornament production techniques and cultural esthetics, supporting the hypothesis of the rapid emergence of a more complex social landscape early in the European Upper Paleolithic.

   2. The Mnemonic Dimension of Magdalenian Cave Art

      L'art pour l'art has long been discredited as an explanation for the remarkable record of parietal and portable art produced by Magdalenians from 18,000 to 11,000 years ago. Whether we accept the Reinach-Breuil view of Paleolithic art as hunting magic, Leroi-Gourhan's interpretation of it as mythogram or Mithen's recent interpretations that see it as a storehouse of practical knowledge about the environment, the commitment of enormous labor, technological innovation and creativity implies a fundamental role for Paleolithic art in social and material reproduction. The construction of a scaffold, the development of thermal treatment to obtain desired pigment colors and the use of binders and extenders in paint recipes support the idea that the Magdalenians of Lascaux saw it as essential to their survival. "Art," far from being merely a spare-time diversion served a number of critical social and technological functions in Cro-Magnon cultures of the last Ice Age.

The concept of chaine opératoire can be applied just as easily to the construction of meaningful cave compositions as to the construction of socially significant costume discussed above. Leroi-Gourhan recognized that the organization of imagery on cave walls was a process of cultural production that over and over again imposed cultural structure on natural space. The process of creating spatially organized compositions may be just as easily seen as the outcome of habitus as it is a mnemonic device for reinforcing and transferring from one generation to the next habitual modes of thought. And as Levi-Strauss long ago noted, "the animal world and that of plant life are not utilized merely because they are there, but because they suggest a mode of thought."

It has long been observed that animals and their biological and social characteristics have served as metaphors for human social distinctions. If animals are good to think it is largely because they serve as metaphors for comprehending the intangible complexities of human existence. Nisbet observes that, "Metaphor is our means of effecting instantaneous fusion of two separated realms of experience into one illuminating, iconic, encapsulating image. The frightening, sensually-charged environment of the deep underground is a perfect setting for imprinting lasting memories of such profound and complex cultural understandings.

In sum, material images, whether in the form of carefully constructed costumes on human bodies or arrangements of painted animals in mysterious underground settings, were powerful devices for the codification, storage and transfer from one generation to the next of complex bodies of cultural knowledge. Like neurological memories, material images were sufficiently structured, redundant and predictable to serve as a mnemonic infrastructure for Upper Paleolithic cultural systems. At the same time, they were sufficiently ambiguous as to be reworked, recontextualized and reinterpreted with each encounter, and in relation to changing historical and environmental circumstances of the late Ice Age.