Signs of Consciousness:

Speculations on the Psychology of Paleolithic Graphics

J. A. Cheyne


Part I:

Graphics in an Iconic Culture

Part II:

Peirception and Consciousness

The Search for Positive Identification

Part III:

The Role of Technological Developments in the Late Paleolithic


Part I:

Paleolithic Graphic Images:

Based the evidence accumulated over the last two hundred years, we can conclude that graphic activity appears to have begun sometime prior to 33,000 BP during the last Ice Age (Würm), in the Upper Paleolithic Period (Aurignacian through Magdelenian in Europe) and reached its apogee during the Solutrean-Magdelanian cultures (22,00-10,500). The evidence consists of images carved, scratched, etched, pecked out, and/or painted on stone, bone, ivory, horn, and antler. Portable, or mobiliary graphics appears to be somewhat more widespread than parietal graphics (cave art) found in rock shelters ("abri"), shallow recesses, and occasionally, deep within caves, especially in the Dordogne and Ariège regions in France and in Cantabrian Spain. All such graphic activity is exclusively, or virtually exclusively, associated with modern Homo sapiens at sites in Africa, Australia, Middle East, and Eastern and Western Europe. It is the earliest evidence of humans, or any creature, creating traces that are not the incidental residue of instrumental activity, such as tool or shelter making, butchering, or simply the result of walking in soft volcanic ash. The beginning of graphic activity appears to coincide with the historical "vanishing point" of language and hence it is tempting to suggest that these graphics may offer clues about early, or even the origins of, representational activity. It is scarcely surprising then that these images have been the subject of intense scrutiny and speculation since their discovery.

The approach taken in this argument is to avoid postulating the possession of capacities by the graphists for which there is little or questionable evidence other than thick interpretation of the images themselves. Rather than assuming cognitive symbolic abilities and cultural traditions that might plausibly have found "expression" in graphics I invert this argument and conjecture that symbolic activities and cultural traditions, myths, and rituals were, in part, a result of the production of images. These images, I argue, preceded and produced the very notion of representation itself. This argument I characterize as a multiple factor theory comprising a componential theory of perception, behavioral strategies for positive identification of ambiguous images, a specialized form of sequential and hierarchical control of fine-motor hand-eye coordination emerging from developments in tool construction that led to a "mastering of the trace," and issues entailed in the sharing and transmission of technology.


Overview of "Signs of Consciousness" in Paleopsychology

[Bull-Lascaux]Some time between 30,000 and 40,000 thousand years ago modern humans began creating the first deliberate "signs" ever created on earth - perhaps in the universe. These signs were something other than simple traces of motor activities (e.g. footprints) or the results and residues of instrumental labor (e.g., tool marks). They were marks left as signs carrying meaning for those who left them and, subsequently, for we who find them now. These signs appeared in the Late Paleolithic in relative abundance, rather suddenly, in great variety, and often reflecting great skill. A remarkable number of signs and techniques for producing appeared almost simultaneously. These include crude scratches and colored blotches, complex enigmatic signs and, above all, figurative art of surpassing beauty. They are placed on stone, bone, antler, horn, ivory, clay in small, portable (mobiliary) sculpture and etching and, most spectacularly, as murals (Parietal Art) on the walls of rock shelters and deep in caves. How did those early humans manage this and why did they bother? What was so different about these people whose own recent ancestors were either unable or disinclined to create such signs?

It seems a safe assumption that the paintings of bison, horses, aurochs, and other large animals on the walls and ceilings of deep caves were not drawn from life, but from imagination. And what is imagination?

  1. Perceiving and Imagining: Take a moment to conjure up a serious image of a cow in your imagination. Now ask yourself about the color. Perhaps it seemed your cow had a particular color before, or perhaps only after, the question of color was raised. Now ask yourself if your imagined cow had an udder, and what color it was that? Finally, ask yourself what color the hooves were. As you proceed through such questions you may become less sure about whether or not you had thought to put in the features and colors. However, it is also likely that you imagined yourself "checking" your imaged figure and having little problem seeing the feature "after the fact," as it were. Visual imagining appears to consist of moving from feature to feature, and sometimes zooming in on particular features such as feet or horns. Subjects in studies by Kosslyn (1980) study report scanning images from point to point and the time taken to report completion of the task is linearly related to inferred "distance" scanned. Perhaps the best-known studies of this sort are those of Shepard (Shepard and Metzler, 1971) on mental rotation that may the strongest evidence that imagery relies directly on low-level perceptual processes. A vivid example of parallelism of perception and imagery is a case of acquired total color-blindness in which the stricken individual could neither see nor imagine color is provided by Sacks (1995). Hence imagining objects in the mind seems not dissimilar from inspecting an object in the world and may draw on exactly the same underlying neurological processes. A useful model, for present purposes, of such process is provided by Biederman (1987), Figure 1: A model of human object perception. In the same way that we extract features, parse, or visually break down objects, determine components and their relations to identify objects in the external world, so also do we seem to be able to isolate significant features (by parsing at points of discontinuity), zoom in on details, move around the image (or move it around in our imaginary space). Our imagined visions very likely depend upon structures and processes that support object perception, structures and processes that have somehow become rather independent of external objects. When, how, and why did this happen?
  2. Hominids: Australopithecus in Geological Time and Homo in Archaeological Time: In the next figure, Figure 2: Time Chart, I present two timelines: a geological timeline and an archaeological (cultural) timeline. Discussions of Hominid evolution typically begin with the Australopithecines. I have placed these in geological time because, although they share our upright stance, their pongid (ape-like) brain structure and the absence of convincing indications of culture or technology separates them from the members of the genus Homo. Homo habilis, Homo erectus, and Homo sapiens are placed in cultural time because of their brain structure and their association with cultural artifacts. For a more detailed description of various fossil hominids see Foley.
  3. The next figure charts the evolution of lithic technology and brain size, as well as a few significant structural changes. It is clear from Figure 3: Evolution of Brains and Technology that cranial capacity and, by inference, brain sizes, have increased dramatically since Homo habilis. Indeed the cranial capacity (an indirect measure of brain size) of Homo habilis does not much exceed that of some species of Australopithecus. However, close examination of the interior surface of the fossil crania reveals a human structure, or at least a human surface pattern. Two features that appear in Homo habilis and differentiate them from Australopithecines are the displacement of the lunate sulcus (a pongid or ape-like feature) by development of the POT (the area around the junction of the Parietal, Occipital, and Temporal lobes - also referred to, in the so-called dominant hemisphere, as Wernicke's area) and portions of the frontal lobe known (in the dominant hemisphere) as Broca's area. (See Wilkins and Wakefield, 1994, for a review.)

Why did these changes take place? There are certainly some good concrete arguments against such changes. There are, for example, numerous constraints on big brains and big heads. Brains are metabolically costly and large heads are incompatible with biomechanics of upright stance, most particularly because of the birthing problem created by big heads passing through a birth canal the structure of which is very much constrained by the upright stance. Moreover, arguments in favor of large brains tend to be rather vague, arguing that more brain power is somehow an unqualified good. The assumption appears to be that more neural cells mean better problem solving and memory. This argument seems reasonable (but see Cosmides and Tooby, 1997) as far as it goes but it does not go far enough. If increased problem solving and memory inevitably lead to greater fitness one would think that we would have many more competitors in the big brain business. One way to begin to overcome such vagueness is to consider not simply the overall increase in size but also the specific structural changes and their possibly equally specific functions in generating (yet again) very specific behavioral capacities. One strategy for overcoming vagueness is to consider the biological fitness-conferring uses to which these skills and capacities were put.

Some Structural Changes and their Functions: The POT appears to be an area in which information from each of three lobes (e.g., somatosensory-visual-auditory) may become integrated into amodal circuits. Broca's area appears to be important for controlling motor sequencing, especially for speech, in modern humans. Broca's area is connected to the POT by the arcuate fasciculis. Damage to Broca's area has long been known to produce production deficits in language. These considerations might be taken to suggest that Homo habilis was possessed of language and that the subsequent evolution of these areas and the consequent growth in brain size were a result of selection pressures on language development and communication (see Dunbar, 1992), for an interesting example of this line of reasoning). The issue of language capacities of pre-modern hominids is extremely controversial. Direct evidence one way or the other is completely lacking. There are, however, alternate interpretations of the evidence that are preferred by many researchers (such as Wilkins and Wakefield). This alternative focuses on the increasing skills required for cultural activities such as tool-making and tool use. The development of successive lithic techniques led to an ever expanding set of niches for tool using cultures to occupy (See Schick and Toth, 1993, for an interesting discussion of analogies between hominid tools and features of animal anatomy and the niches they open up.). These changes in Lithic (Stone) Technology provide some evidence compatible with increasing sensory integration potentially afforded by developments of the POT and of the coordination of these with frontal lobe development supporting more intricate motor sequencing. Consider some of the characteristics of Paleolithic ( = Old Stone Age) stone tools and their changing character over a period of 2.4 million years of so.

Throughout the Paleolithic, over time, stone tools provide evidence of increasing:

  • selection of materials
  • useful cutting edge,
  • symmetry,
  • investment of effort,
  • diversity and number of techniques,
  • number of steps,
  • complexity of stages, and
  • diversity and number of different tool types.

In addition, by the Late Paleolithic, tool making was very hierarchically organized, employing specialized stone knapping techniques such as: pressure flaking, indirect percussion, and soft hammer techniques that require considerable perceptual, judgmental, and planning skills as well as fine motor coordination. Moreover, stone tools were also being used to make secondary tools further increasing the complexity and temporal demands of forward planning. There is a fairly regular development of such techniques throughout the Paleolithic as there is of the brain. As noted above these developments are also correlated with potential niche exploitation.

Examples of many of these changes may be seen in Figure 5: Lithic Technology, in which some examples of the stone artifacts from each of the four classifications in Figure 3 above. A rough sense of the development of lithic technology from the following highlights. Oldowan tools are made from large pebbles with a small number of blows producing a small amount of rough cutting edge from a relatively large amount of material. Acheulian tools, by comparison, have more cutting edge and are more extensively worked. Mousterian tools are made from more carefully selected material, from flakes struck from a core, and further refined (retouched) by pressure flaking. Aurignacian - Magdelanian stone tools are generally made from numerous slender blanks removed from carefully prepared cores by difficult to master techniques such as soft hammer or indirect percussion techniques. These blanks were then sorted and reworked to provide a great variety of cutting tools, some of which are subsequently employed to create secondary tools. All of the forgoing developments imply increasing visualization, manipulation of internal images, and bringing of fine motor control under the guidance of these images. This implies access to, and integration of, information from somatosensory (Parietal), visual (Occipital), and perhaps auditory (Temporal) modalities.

In contrast to non-human primates, the successive forms of the genus Homo have left evidence of increasing ability to voluntarily shape and modify instrumental actions. These activities would have required increasing amodal modeling integrated in POT areas and organized by Broca's Area. This would be functional for visual-spatial-manual coordination not only in tool making but also for using tools, for such activities as: aimed throwing (e.g., Calvin, 1992), searching and digging tubers, and, certainly by the late Paleolithic, sewing. Also likely important for all of this is the ability of autocueing (Donald, 1991), that is, the voluntary retrieval of stored memories. Working with these memories would require the development of cognitive devices such as the visual-spatial sketchpad and the articulatory loop (Baddeley, 1986) for manipulating images and possibly sharing them. In summary, the interaction of brain development and the evolutionary pressures for amodal modeling, intersensory integration, and fine motor coordination linked to centres for amodal modeling have produced human cognition.

By the Late Paleolithic, it is these cognitive skills and capacities that would have prepared Homo Sapiens to transcend internal cognitive mechanisms and invent external working memory. Such external memory devices might be viewed as external visual-spatial-manual sketchpads and generally known as pictures, ideographs, and phonetic symbols. For a detailed consideration of the relative advantages of such exograms (As Merlin Donald calls them) over engrams (internal memory structures) see Donald's Discussion at the end of his précis (or a local version of his summary Table). These advantages have to do with things such as the flexibility, capacity, and permanence of the records. Other advantages of working with exograms include the ability to employ external senses, such as vision and audition in the analysis of imaginative productions that otherwise have no access to external senses. (This is, after all, a primary distinction between imagination and perception.) The first ever of these externalized imaginative constructions appear as small portable artifacts of material such as stone, bone, and ivory and on the walls and ceilings of caves and rock shelters in the Late Paleolithic.

Paleographics (Stone Age "Art"): Prior to 33,000 B.P. (Before Present) we see the beginnings of graphic activity. I will use the term "graphic activity" to designate any activity that results in the production of visual signs in any medium. This will incorporate what is generally referred to as "art", such as pictures or figurines, as well as the production of non-figurative marks that are typically designated as signs and (often mistakenly, I believe, when applied to Paleolithic marks) symbols. In Europe, this activity began in the Aurignacian and peaked in the Magdelenian (See Figure 2: Time Chart. These are exclusively associated with Homo sapiens with the possible exception of a few very ambiguous cases. There exists evidence of graphic activity, associated with modern humans, at sites in Africa, Australia, Middle East, Eastern and Western Europe.

There are two very general classes of graphic activity:

  1. Mobiliary statuary and graphics in stone, bone, ivory, horn, antler, clay.
  2. Parietal graphics in Rock shelters and Caves.

Parietal graphics consist largely of megafauna (large animals: mainly horses, bison, aurochs (wild cattle), mammoths, various species of deer, and goats), a few birds and smaller mammals, enigmatic signs (rectilinear shapes, wedges ("claviforms"), tectiforms (like a roof), dots, lines, strands ("spaghetti"), hand prints. Human figures are rare (except for the so-called "Venus" figurines) and in contrast to some of the animal images, almost always very crudely rendered. Why did Paleolithic people make these graphics? There are several questions implied. What motivated them to bother? Why did they occur when they did and not earlier? What capacities were required? It is on this final question that the present essay is focused.

First, I think they are guided by our manner of perception, which is feature-based. As hominids came to do more internal processing on the visual-manual sketchpad they came to work, I will argue, relatively independently, on certain parts of the images, namely, the most distinctive features. This assumes a capacity for hierarchical organization and nested sequencing – a capacity that was increasingly necessary as tool making and related technologies became increasing complex. Features also became available for new combinations, necessarily so for tool planning, making, and, increasingly, for the training of apprentices in these increasingly demanding tasks. Working for long periods of time on only one component of a tool, or article of clothing or decoration required that one understand the connection of the part to the whole. Parts, or components, of things would come to stand for whole objects as iconic signs.

I will now provide a few samples to try to convince you that Paleographic figurative images are compositions of components that are iconic and occasionally indexical signs by pointing out some stylistic features that suggest that the figurative graphics of the Late Paleolithic are constructions based on basic perceptual elements. The images are not vague abstract symbolic images. The images in question are concrete constructions that emphasize the distinctive features for the identification of the subject matter of the images. I will focus mainly, though not exclusively, on the most common images -- large prey animals. These are understandably objects of intense interest to large game hunters such as the people who created the images.

  1. Use of Occluding contours: It has been pointed out by perceptual psychologists that a basic element of Paleolithic graphics is the occluding contour (e.g., Kennedy, 1976). This is essentially the outline or silhouette of the figure. Many Paleolithic images are simply outline figures. Even most of the more elaborate and detailed figures are based on an initial outline etched and/or painted on the surface. For the animals represented in the images the occluding outline of the animal, especially the dorsal contour, in profile, is the most distinctive feature for identification of the animal, especially at a distance. This outline figure, when parsed at points of discontinuity (See Figure 1), yields the major body components. One of the most significant occluding contours for identifying the large mammals portrayed is the cervico-dorsal line. This feature is almost universally emphasized in the drawings and sometimes the only component drawn (e.g., horse-Lascaux ).
  2. Use of canonical orientations by component: As noted in the previous section the animal figures were invariably portrayed in profile, this being the most informative orientation for the major occluding bounds. However, for certain very particular distinctive features, such as horns and hooves, this is not the case. Horns and antlers are best discriminated in 3/4 perspective and hooves when viewed either from above, dorsally or from below, ventrally. Excellent evidence was provided for the independence of feature orientation when the creators of these images combined the profile presentation of the major occluding bounds of the torso and the head with 3/4 and dorsal/ventral presentation of horns, antlers, and hooves, giving the impression that these features are "twisted". This feature of Paleolithic style may be seen in the following images of "twisted" feet. (The two images in the upper right-hand corner are of horses' hooves, the rest are those of aurochs or deer), "twisted horns and "twisted" antlers. That these are not simply a few isolated examples is illustrated in Figure 6. The "twisted" feet technique is a very common and frequently noted one employed at a variety of sites. Interestingly, the hooves of the images may actually be "indexical signs" in as much as they resemble hoof prints as much as they do hooves. (An indexical sign is one that stand for something by being associated with it, e.g., smoke for fire, a wound for a knife, etc. – See Part II)
  3. Change of scale by component: Paleolithic images are often characterized by large and sometimes puzzling changes of scale between components (i.e., at regions of concavity segmenting head, neck, torso, tail, legs, and feet). It appears as though the graphist "zoomed" in on components after parsing and failed to match the scales of the components, so that heads, for example, are often far too small or too large for torsos although the individual components are often beautifully and appropriately rendered. A striking demonstration of the relative independence of critical distinctive components may be seen here. At some sites, such as Lascaux, the mismatched scale between components sometimes became conventional as in the case of the famous ponies with voluminous bodies and small heads. (e.g., horse-Lascaux)
  4. Exaggeration of Distinctive Features: Distinctive features such as horns, antlers, and the distinctive shape of the cervico-dorsal lines of Bison and mammoths are frequently exaggerated. Sometimes the exaggerated features are more complex relations. Torso Convergence -- the narrowing of the torso from the shoulders to the loins -- is very marked for the bison but not the horse. Analysis of the parietal graphics of bison and horses reveals that this difference is significantly exaggerated. The difference in aspect ration between bison and horses is greater in Paleolithic drawings of these animals than in photographs of the descendants of these animals. The same finding obtains for body aspect ratio. Bison are much bulkier animals than horses and this is reflected in their aspect ratio (body length/body depth - shoulder blade to sternum).
  5. Isolation of Distinctive Features: At many sites one finds evidence that certain distinctive features are portrayed by themselves. The cervico-dorsal line has already been mentioned. Often only a clearly identifiable cervico-dorsal line of an animal, such as a horse or mammoth, appears with no evidence that a complete figure was ever added or intended to be added. Sometimes only the head, antlers, and the cervico-dorsal line of a deer are portrayed, or the head and cervico-dorsal line of a horse. Perhaps one of the most striking cases of parts standing for whole figures may be found in the portrayal of female figures in both mobiliary and parietal graphics. The so-called Venus figures. In the very great majority of these figures the distinctive features of breasts, buttocks, bellies and vulvas are emphasized and apparently greatly exaggerated, while the extremities, head, arms hands, legs and feet are very much diminished or missing. Some of the images are composed only of breasts, bellies, buttocks, and vulvas (or, more precisely, as Sharon Winn has pointed out to me, the pubic triangle). Finally, there are many wall carvings of bellies and vulvas, and of just the vulvas, or pubic triangle with or without an indication of the vulva. In the latter case we may be viewing one of the earliest of the pure iconic signs. One distinctive feature stands iconically for the object of which it is a part. Perhaps this is one route by which our ancestors the rich repertoire of signs that characterize human culture wherever it is found.

In Summary, there is a style to Paleolithic graphics that is consistent with the view that they represent the construction of perceptual images that stand individually and collectively as signs of the constructions of which they are components. These images are very different from 1) the Neolithic graphics that followed them, 2) the art of cultures newly introduced to figurative drawing, and 3) children's drawing in modern cultures. All of these later graphic forms tend to be less perceptually based. There is, for example, in all of these subsequent cases, less concern for the details of the shape of the cervico-dorsal line. For later cultures and groups, a straight line often appears to suffice, at least for basic representation. "Here is a line. Call it a back." Often only a very few very stylized features are used semi-iconically, almost symbolically, to stand for the object. In such cases, the image has ceased to be the center of attention. It seems now to serve a supplementary function, perhaps as an aid, a mere illustration, to accompany the telling of a story. The ultimate concern seems more to simply convey the notion that the subject involves a cow (or more likely in Neolithic graphics - a person). Yet it is not the perceptual image of the cow that is important anymore. It is something the cow stands for or something that is happening to the cow. It is being hunted, or tended, or slaughtered, or worshipped. Similarly, the people, much more commonly portrayed in Neolithic than in Paleolithic graphics, are busy hunting, running or dancing. Neolithic and graphics seem to have become increasingly concerned with telling or reminding the viewer of events, stories, legends, and myths. There is no clear evidence, on the other hand, that the earlier Paleolithic graphics were concerned with stories, myths, or anything event-like at all. Paleolithic art appears to have been concerned with the externalized perceptual image and with its components, especially those that were distinctive and diagnostic. They may have learned much from their externalized imagery. Importantly, they may have been able to make more explicit (i.e., conscious) just what differentiated one animal from another. Perhaps because these images were so revelatory they were viewed with much interest and awe, perhaps even something akin worshipful admiration. Perhaps the images gradually came to be associated with something like the religious-magical sensibility the 19th Century prehistorians imagined them to have been. However, in the present view the graphics were not merely an expression of such feelings but something instrumental in the creation of such feelings. It has been conjectured that the large prey animals portrayed in the Paleolithic graphics, being so central to the lives of the graphists, were simply "good to think about." A great discovery of these earlier graphists was that the externalization of their visual-spatial imagery helped them systematically to think about things that were "good to think about."

Graphics in an Iconic Culture: Summary

Central Questions

There are three questions to which any discussion of early graphics must address itself. The question that has been most frequently asked is that of motivation. Why did these early humans engage in such activity? Motivation has usually been sought within the graphists in the form of some emerging capacities that "expressed" themselves in the graphic productions. A second question that continues to be actively addressed might be characterized as the "How?" question. As late as 1982 Leroi Gourhan could write, "If plenty remains to be discovered about the 'why?' of the symbols, almost everything remains to be said about the 'how?'" The "how?" question, in turn, conceals at least two sets of very different but interrelated questions. One set of questions concerns the techniques of etching, carving, applying pigments, etc. (Vandiver, 1983; Lechtman, 1977; Lorblanchet, 1980, 1991). The second set of questions concerns the perceptual, cognitive and action systems enabling the activity (Davis, 1986, 1987; Halverson, 1987, 1992; Kennedy, 1976).

A critical question that cuts across the previous set of questions and constitutes a challenge for any understanding of the emergence of graphics is that of timing. Why and how did graphic activity arise just when it did and not earlier or later? Failure to address this last question directly tends to lead one into the clever hominid fallacy" (Davidson & Noble, 1989). This is a version of the "and then the magic happened" gambit in which it is implicitly assumed that, at some arbitrary point humans, or their ancestors simply became clever enough to put it all together. Hence in this paper, I address these questions and attempt to show that some progress towards providing a frame for answers may be achieved by a considering a specific historical confluence of perception, action, technology, and communication.

The graphics in question include frequent images of megafauna including horses, bison, mammoths, wooly rhinoceros, reindeer, red deer, and ibex. The range of quality, as judged by current standards, is considerable. Many of the animal images seem extremely crude and incomplete while others strike us as exquisitely beautiful and evocative. There are also numerous "enigmatic" signs: rectilinear forms and characteristic shapes such as the tectiforms or claviforms, dots (solitary or in clusters or series, lines, either rectilinear or wavy series sometimes referred to as "spaghetti". There are also fairly commonly found positive and negative images of handprints and even images of handprints carved in the cave walls. Human images are also found but are rare, often incomplete, and generally crudely rendered.

While there are both temporal and geographical variations there are a number of distinctive features that justify characterizing the graphics as having a particular style. Any theory about the development of graphic activity must address these stylistic features. It is an underlying assumption in the present effort that these characteristics provide significant clues about the how, the why, and the timing of the development of graphic activity. First, outline figures predominate. When animals or other objects are depicted often only the occluding bounds of the object are depicted. Even among solid chromatic images occluding bounds are outlined by scratching, engraving, and application of contrastive coloring. Charcoal sketches have been found under subsequent applications of pigments at Niaux (Clottes, Menu, & Walter, 1990a,b). This characteristic of outlining and the simplicity of the many of the images produce an effect of a cartoon-like economy. The cartoon-like quality is enhanced by another common feature, the use of caricature or exaggeration of distinctive features of the objects depicted.

Paleolithic images of animals are all but invariably depicted in profile, at least the major components such as head, neck, torso, and legs are in profile. However, these profiles also frequently have a "twisted" perspective in which horns, antlers, tusks, feet, and sometimes ears are presented in perspectives that deviate from over-all profile presentation in varying degrees. In both mobiliary and parietal graphics the images and designs often incorporate, modify, or accommodate natural contours, breaks, and edges in the supporting medium. Graphics, whether animals or symbols, appear to be executed as discrete, stand-alone, images and they are so thoroughly componential that even the parts may frequently stand alone. With rare and questionable exceptions there is a notable absence of context, either environmental or narrative. These images do not provide direct evidence that their creators were members of a mythic culture or that the images were involved in accounts of, or rituals concerning, hunting, initiation, fertility or derived symbolic activity. (They do, however, provide considerable evidence for an interest in certain details of animal anatomy.) That they may have provided, in part, the ground from which all of these and more emerged is a possibility that will be entertained at the end of this paper. Certainly graphic activity in the Neolithic in Spain and Africa provide a dramatic contrast in which there is considerable evidence of interest in depicting events and perhaps even some impatience with anatomical details. This contrast suggests the culture(s) of the Late Paleolithic were obviously not Literate but neither were they Oral in the sense of Ong (1966; Goody, 1977, 1986. 1987) but one that, based on the evidence and argument presented below, may be called Iconic (e.g., Donald, 1991).

Part II

[pce]Psychology, Culture, and Evolution.