The Prehistory of the Mind

THERE WAS a cultural explosion in the fourth and final act of our past. This happened in the time period 60,000–30,000 years ago, which marks the blurred start of the second scene of Act 4. The start of the act itself is marked by the entry of the final, and sole surviving, actor, H. sapiens sapiens at 100,000 years ago. This new actor appears immediately to have adopted certain forms of behaviour never previously seen in the play. Most notable are the making of bone artifacts in southern Africa, and the placing of parts of animals into human burials in the Near East – the only two areas of the world where 100,000-year-old H. sapiens sapiens fossils are known. But other than these glimpses of something new, the props of H. sapiens sapiens in the first scene of Act 4 are almost identical to those of the Early Humans. I will therefore refer to these first H. sapiens sapiens as Early Modern Humans. The cultural explosion only occurs after they have been on the stage for at least 40,000 years. And consequently it is the start of Scene 2, and not the first appearance of H. sapiens sapiens, which archaeologists denote as one of the major turning points in prehistory, referring to it in an ungainly phrase as the ‘Middle/Upper Palaeolithic transition’.

In this chapter I want to look at the behaviour of H. sapiens sapiens in the first two scenes of Act 4 – immediately before and after this transition – and ask how their minds were different from those of Early Humans. But I want to take the two scenes in reverse order, beginning with the dramatic cultural changes which happened after 60,000 years ago, notably the origin of art.

Now recall that by the start of Act 4 the cathedral of the modern mind is almost complete. The four chapels of technical, natural history, social and linguistic intelligence, the traces of which we saw when we looked at the modern mind in Chapter 3, are in place. But the walls of these chapels are solid; the chapels are closed to each other, trapping within them the thoughts and knowledge of each specialized intelligence – except for the flows between the chapels of linguistic and social intelligence. To constitute the modern mind, the thoughts and knowledge located in all these chapels must be allowed to flow freely around the cathedral – or perhaps within one ‘superchapel’ – harmonizing with each other to create ways of thought that could never have existed within one chapel alone.

Archaeologists have often described the Middle/Upper Palaeolithic transition as a cultural explosion. Recall from Chapter 2 that it is at, or soon after, this transition that Australia was colonized, that bone tools became wide spread (after having made their very first appearance in Act 4 Scene 1), and wall paintings were created. Scene 2 of Act 4 is a frenzy of activity, with more innovation than in the previous 6 million years of human evolution. As the start of this scene is so often described as a cultural explosion, it seems obvious to ask whether this noise is an explosion at all; perhaps it is the sound of doors and windows being inserted into the chapel walls, or even the noise of a ‘superchapel’ being constructed. In other words, the start of the final phase of our architectural history of the mind.

It is quite easy to think of the Middle/Upper Palaeolithic transition as a cultural explosion, or a big bang – the origins of the universe of human culture. Indeed a ‘big bang’ is the short hand description I will use in this chapter. Yet if we look a little more closely at the boundary between Scenes 1 and 2 we see that there is not so much a single big bang as a whole series of cultural sparks that occur at slightly different times in different parts of the world between 60,000 and 30,000 years ago. The colonization of Australia, for instance, seems to reflect a cultural spark which happened between 60,000 and 50,000 years ago, yet at this time all remained relatively quiet elsewhere in the world. In the Near East a cultural spark happened between 50,000 and 45,000 years ago when the Levallois technology was replaced by that of blade cores. The cultural spark in Europe seems not to have been until 40,000 years ago with the appearance of the first objects of art. Indeed, it is perhaps only after 30,000 years ago that we can be confident that the hectic pace of cultural change had begun in earnest throughout the globe. Some archaeologists go so far as to deny that there is such a thing as a major transition at all, and view the cultural changes as no more than the result of a long process of gradual change. They suggest that the new types of artifacts that appear in the archaeological record during Act 4 reflect better preservation and recovery rather than new forms of behaviour.¹ But I disagree.

As with the majority of archaeologists I believe something fundamental occurs at the Middle/Upper Palaeolithic transition, even if at slightly different times in different parts of the world. There have been several ideas previously put forward as to what this fundamental thing might be. These include notions about the ‘re-structuring of social relations’,² the appearance of economic specialization,³ a technological ‘invention’ similar to that which caused the transition to agriculture 30,000 years later,⁴ and the origin of language.⁵ I think that these are all wrong: either they are merely consequences rather than causes of the transition, or they fail to recognize the complexity of social and economic life of the Early Humans.

My explanation of the big bang of human culture is that this is when the final major re-design of the mind took place. It is when the doors and windows were inserted in the chapel walls, or perhaps when a new ‘superchapel’ was constructed. The modern mind might thus be represented as in Figure 17. With these new design features the specialized intelligences of the Early Human mind no longer had to work in isolation. Indeed I believe that during the last two decades of research the explanation for the Middle/Upper Palaeolithic transition has been found – not by archaeologists but by the cognitive scientists whose work we examined in Chapter 3.

Recall how Jerry Fodor finds the ‘passion for the analogical’ to be a central feature of the distinctly non-modular central processes of the mind and how Howard Gardner believes that in the modern mind multiple intelligences function ‘together smoothly, even seamlessly in order to execute complex human activities’. We saw how Paul Rozin concluded that the ‘hallmark for the evolution of intelligence … is that a capacity first appears in a narrow context and later becomes extended into other domains’ and Dan Sperber had reached a similar idea with his notion of a metarepresentational module, the evolution of which would create no less than a ‘cultural explosion’. Also recall the ideas of Annette Karmiloff-Smith regarding how the human mind ‘re-represents knowledge’, so that ‘knowledge thereby becomes applicable beyond the special-purpose goals for which it is normally used and representational links across different domains can be forged’, which is so similar to the notion of ‘mapping across knowledge systems’ as proposed by Susan Carey and Elizabeth Spelke, and the ideas of Margaret Boden regarding how creativity arises from the ‘transformation of conceptual spaces’.⁶

None of these cognitive scientists was writing about the Middle/Upper Palaeolithic transition. Nor were they necessarily writing about the same aspects of the modern mind: some were addressing child development while others were discussing cognitive evolution, or simply how we think as we go about our daily lives. But their ideas share a common theme: that in both development and evolution the human mind under goes (or has under gone) a transformation from being constituted by a series of relatively independent cognitive domains to one in which ideas, ways of thinking and knowledge flow freely between such domains. Although they did not know it, Gardner, Rozin, Boden and the others were providing the answer to the Middle/Upper Palaeolithic transition.

At least I think they were. It is the purpose of this chapter and the next to evaluate this proposition. I will begin by asking whether such developments can explain the new types of behaviour we see early on in Act 4, when people continued to live by hunting and gathering during the period we call the Upper Palaeolithic. In the Epilogue I will take us a little bit closer to the present day and lifestyles that are familiar to us by considering the origin of agriculture.

We must start with the event of Act 4 that at last brings some colour to the play: the appearance of art.

We cannot discuss the origin of art unless we agree what we are talking about. Art is another of those words pervading this book which defy easy definition, words like mind, language and intelligence. As with those words, the definition of art is culturally specific. Indeed many societieswho create splendid rock paintings do not have a word for art in their language.⁷ The communities of the Upper Palaeolithic are likely to have had a very different concept of art (if one at all) from that which is the most popular today: non-utilitarian objects to be placed on pedestals in galleries. Yet these prehistoric hunter-gatherers were producing artifacts which we regard as priceless today, and which are very readily placed on pedestals in our own galleries and museums. Let us for a moment consider the earliest pieces of art known to us, before generalizing about their essential qualities.

In the debris left from Act 3 a few pieces of scratched stone and bone have been found which some have claimed to be of symbolic significance, such as a bone from Bilzingsleben in Germany with incised parallel lines.⁸ I doubt if there is any justification for such claims, and these objects should be excluded from our admittedly ill-defined category of art. The majority can be explained as by-products of other activities, such as cutting plant material on a bone support – but there may be some exceptions to which I will return below.

Membership of the elite group of artifacts that we call ‘art’ must go to those which are either representational or provide evidence for being part of a symbolic code, such as by the repetition of the same motifs. The earliest phase of the Upper Palaeolithic provides us with examples of both.

In terms of representational art we can do no better than start with the ivory statuette from Hohlenstein-Stadel in southern Germany, some 30,000–33,000 years old (see Figure 18). This is a figure of a man with a lion’s head carved from the tusk of a mammoth, a remarkable combination of technical expertise and powerful imagery. It was found shattered in tiny pieces and meticulously restored to provide us with the earliest work of art known.⁹ Also from southern Germany at this time we have a series of animal figures carved in ivory including felines and herbivores such as mammoth, horse and bison. Some of these have incised markings on their bodies.¹⁰

Contemporary with this representational art, we find images which appear to be part of a symbolic code being created in south west France (see Figure 19). These are predominantly ‘V’-shaped signs engraved on to lime stone blocks in the caves of the Dordogne. Although they have been traditionally described as images of vulvas, archaeologists now discount the idea that they have any simple representational status. The critical feature is that the motifs which have the same form are repeatedly engraved.¹¹

19 (Right) Engraved symbols on a small boulder, 60 cm wide, from Abri Cellier, Dordogne, France, c. 30,000–25,000 years old. Images such as these are repeated in other sites in southwest France during this period, including Abri Blanchard, Abri de Castanet and La Ferrassie, as illustrated on the left.

Along with these pieces of art, the period between 40,000 and 30,000 years ago saw the first production of items for personal decoration such as beads, pendants and perforated animal teeth. At the site of La Souquette in south west France ivory beads were carved to mimic sea shells.¹² At the same time as, or soon after, these items were being produced the first caves in south west Europe were being painted with images of animals, signs and anthropomorphic figures, a tradition which would culminate in the paintings of Lascaux at around 17,000 years ago.¹³ Indeed some of the paintings in Chauvet Cave in the Ardèche region of France, a cave discovered as recently as 18 December 1994, have been dated as being 30,000 years old. The 300 or more paintings of animals in this cave – including rhinoceroses, lions, reindeer, horses and an owl – are quite remarkable. Many of them are highly naturalistic and demonstrate an impressive knowledge of animal anatomy and outstanding artistic skill. The cave is perhaps on a par with Lascaux, and certainly with Altamira in Spain, with regard to the spectacular nature of its art.¹⁴ Although this is the very first art known to humankind, there is nothing primitive about it.

While the production of art was most prolific in Europe, it was a world wide phenomenon by, or soon after, 30,000 years ago. In southern Africa the painted slabs from Apollo Cave are well dated to 27,500 years ago while wall engravings in Australia date back beyond 15,000 and perhaps to 40,000 years ago.¹⁵ Art remains rare, or even absent in several regions of the world until 20,000 years ago. But that is just 20,000 years after its first appearance in Europe – an almost insignificant amount of time when set against the more than 1.5 million years that Early Humans lived without art.

The variability in the intensity with which art was produced can be attributed to variation in economic and social organization, which in turn can be largely attributed to environmental conditions. The archaeological record shows us that Stone Age art is not a product of comfortable circumstances – when people have time on their hands; it was most often created when people were living in conditions of severe stress. The florescence of Palaeolithic art in Europe occurred at a time when environmental conditions were extremely harsh around the height of the last ice age.¹⁶ Yet there is unlikely to have been a human population living under more adaptive stress than the Neanderthals of western Europe. But they produced no art. They lacked the capacity to do so.

There can be little doubt that by 30,000 years ago this capacity was a universal attribute of the modern human mind. What does it entail? While the definition of a visual symbol is notoriously difficult, at least five properties are critical:

1. The form of the symbol may be arbitrary to its referent. This is one of the fundamental features of language, but also applies to visual symbols. For instance, the symbol ‘2’ does not look like two of anything.¹⁷

3. There may be considerable space/time displacement between the symbol and its referent. So, for example, I might draw a picture about something that happened long in the past, or what I imagine may happen some time in the future.

4. The specific meaning of a symbol may vary between individuals and indeed cultures. This often depends upon their knowledge and experience. A Nazi swastika has a different meaning to a young child, than to a Jew whose family was lost in the Holocaust. The swastika is in fact an ancient symbol, found in cultures as far apart as Mexico and Tibet.

5. The same symbol may tolerate some degree of variability, either deliberately or unintentionally imposed. For instance, we are able to read different people’s hand writing although the specific forms of the letters are variable.

These properties of visual symbols become particularly apparent when we consider the art created by recent hunter-gatherers, such as the Aboriginal communities of Australia. The last decade has seen a tremendous development in our understanding of this art.¹⁹ We now know that even the simplest of images, such as a circle, can have many different referents. Among the Walpiri of the Central Australian Desert, for example, a circle can represent an almost unlimited number of referents: campsites, fires, mountains, water holes, women’s breasts, eggs, fruit and other items. The intended meaning of the circle in any one composition can only be identified by the associated motifs. Such simple geometric motifs may have a wider range of possible meanings than complex naturalistic images²⁰ (see Box p. 158).

Naturalistic images, perhaps of animals or ancestral beings, can also have complex and multiple meanings. An Aboriginal child, lacking in knowledge about the Dreamtime (the mythical past/present), may initially interpret images in a literal fashion. To a child, images of fish, for instance, are about fishing which is an economically important activity for many Aboriginal groups. Such literal interpretations can be described as the ‘outside’ meanings of the art – they are learned in the context of daily life and are in the public domain. As the child matures and acquires knowledge about the ancestral world, the same image will be interpreted in a more metaphorical sense, often relating to the actions of the Ancestral Beings. There may be various levels of these, each requiring additional knowledge about the ancestral past, which may be restricted to certain classes of individuals. Consequently these are described as ‘inside’ meanings. For example, the child may gradually learn how fish are a potent symbol of spiritual transformation of both birth and death. They are good to paint not just because they are good to eat, but also because they are good to think. The metaphorical meanings of fish images, concerning birth and death, do not replace the literal interpretation concerning the practice of fishing, they are complementary. As a result, many images have different meanings to different people depending on their access to knowledge about the ancestral past.²¹

Complex meanings in simple designs of hunter-gatherer art

The complex and multiple meanings that may be found in the simplest geometric designs found in Palaeolithic art can be illustrated with an example from the art of the Australian Aborigines. The social anthropologist Howard Morphy has described how many of their paintings have a basic geometric template underlying the design. Each part of the template may encode a series of meanings. For instance, consider the image below which has two ‘loci’, (a) and (b).

At locus (a), the following meanings are encoded: ‘well’, ‘lake’, ‘vagina’. At locus (b) the meanings ‘digging stick’, ‘river’ and ‘penis’ are encoded. Consequently three different interpretations of this image would be a river flowing into a lake, a digging stick being used to dig a well, and a penis going into a vagina. All three of these are ‘correct’ interpretations, but each is appropriate in a different social context. Moreover, the interpretations may be connected within a single mythic sequence:

A kangaroo ancestor was digging a

well with a digging stick. When he

finished, a female wallaby bent down

to drink the fresh water, and the

kangaroo seized his opportunity to

have sexual intercourse with her. The

semen flowed out of her body and into

the waterhole. Today a river flows into

the lake at that place and the

kangaroo’s penis was transformed into

a digging stick which can be seen as a

great log beside the lake.

If such simple geometric designs can ‘encode’ such complex meanings, and by doing so express the transformational aspects of Ancestral Beings, one can only wonder at the types of meanings encoded in the geometric designs from the Palaeolithic period.

Whatever meaning is attributed to an image, that image is most likely to be displaced in time and space from the inspiration for the image. The water hole referred to by a circle may be far away, while the Ancestral Being has no clear location in either time or space.

We can find many of these features in the rock art tradition of other modern hunter-gatherers, such as the San of southern Africa.²² Indeed, we cannot doubt that the images created in the Upper Palaeolithic also had complex symbolic and multiple meanings involving those five properties listed above. Archaeologists are more likely to have success at reconstructing the ‘outside’ meanings of this art, rather than the ‘inside’ meanings which require access to the lost mythological world of the prehistoric mind – a world to which I will return at the end of this chapter when I consider the origins of religious ideas.

Having considered some of the properties of visual symbols, let us consider what mental attributes are involved in creating and reading them. There are at least three:

1. The making of a visual image involves the planning and execution of a preconceived mental template.

3. The attribution of meaning to a visual image not associated with its referent.

From what we established in the previous chapter – and as I will explain below – it is likely that Early Humans were competent in each of these cognitive processes. They are likely to have existed in as complex and as advanced a state as in the Modern Human mind. So why no art? The answer would appear to be that although they possessed these processes, they were found in different cognitive domains. They were inaccessible to each other and the origin of art only occurred following a marked increase in the connections between cognitive domains. So where in the Early Human mind were these processes located?

The making of marks on objects is something that happens unintentionally in the course of activities by many animals – marks such as hoofprints, scratches on trees and gnawmarks on bones. Some non-human animals also create marks intentionally: chimpanzees have created striking paintings in laboratories, although these do not appear to have symbolic meanings and are not created in the wild.²³ I would interpret such ‘artistic achievements’ in the same manner as the ‘linguistic’ achievements of chimpanzees – as the product of a generalized learning capacity. The earliest members of the Homo lineage we encountered in Chapter 6 were making marks with stone tools on bones in the process of butchery. We also have the series of artifacts made by Early Humans which have incised lines on them, such as a fossil nummulite from Tata in Hungary, which appears to have a line intentionally engraved perpendicular to a natural crack to make a cross and is thought to be 100,000 years old, and the marked bone from Bilzingsleben²⁴ in Germany (see Figure 20). Although it has yet to be demonstrated, I am sympathetic to the idea that some of these lines may have been intentionally created, and I will return to how they should be interpreted shortly. Similarly, the few pieces of red ochre from Early Human sites in southern Africa – no more than a dozen from the period prior to 100,000 years ago²⁵ – may suggest that archaic H. sapiens were marking their bodies. But there is no reason to believe that this is equivalent to the symbolic behaviour involved in producing objects of art. What we need to find in the mind of Early Humans is a capacity to intentionally create marks or objects of a preconceived form.

This can indeed be found – in the domain of technical intelligence. We have seen that Early Humans were regularly imposing form on to their stone artifacts. Handaxes and Levallois flakes required the extraction of objects of a preconceived form from nodules of stone. In view of such technical intelligence, the failure to make three-dimensional objects of art cannot reflect difficulties in conceiving of objects ‘within’ a block of stone or ivory, or the mental planning and manual dexterity to ‘extract’ them. The cognitive processes located in the domain of technical intelligence used for making stone artifacts appear to have been sufficient to produce a figurine from an ivory tusk. But they were not used for such ends.

With regard to the second of the three critical cognitive capacities for art, intentional communication, this was established in the previous chapter as a critical feature of Early Human social intelligence. Indeed Early Humans were probably as dependent on intentional communication as are modern humans today. Among the last of the Early Humans this capacity became manifest in spoken language; in the earlier ones it was probably restricted to vocalizations too simple to be described as language, as well as gesture. In Chapter 5 we saw that both monkeys and apes also engage in intentional communication, suggesting that this capacity has had a long evolutionary history: there can be little doubt that not only Early Humans, but also the common ancestor and the earliest Homo were engaging in frequent, intentional communication.

The third element of a capacity for art is an ability to attribute meaning to inanimate objects or marks displaced from their referents.²⁶ Can this ability be found within one of the cognitive domains of Early Humans? It certainly can: the capacity to attribute meaning to the unintentionally made tracks and trails of potential prey is a critical component of natural history intelligence. As I have argued in previous chapters, the ability to draw inferences from marks such as footprints most likely reaches back to when earliest Homo, or indeed australopithecines, began hunting and scavenging on the African savannah. These inferences often include the type, age, sex, state of health and current behaviour of the animal which made them.

The unintentionally made marks left by animals share a number of properties with the intentionally made ‘marks’ or symbols of Modern Humans, such as paintings on rock faces or drawings in the sand.²⁷ They are inanimate. They are both spatially and temporally displaced from the event which created them and that which they signify. Footprints, just like symbols, must be placed into an appropriate category if correct meaning is to be attributed. For instance, the hoofprint of a deer will vary depending upon whether it is made in mud, snow or grass, just as the drawing of a symbol will vary according to rock surface and the individual style of the artist. The marks left by animals will often be non-representational. While the hoofprint of a deer may look like the base of the hoof, it does not look like the event that is inferred from it, such as the passing of a male stag. Many marks have no visual resemblance to the animal which created them, such as the parallel lines left by the wriggling of a snake. And finally, the meaning of marks will vary according to the knowledge of the person viewing the mark, in a similar way that the meaning of symbols vary. For example a child may identify a hoofprint as coming from a deer, whereas a mature and skilled hunter may be able to infer that the deer is a pregnant female which passed two hours ago.

These points of similarity suggest that the same cognitive processes which are used to attribute meaning to marks unintentionally made by animals would be equally effective at attributing meaning to marks intentionally created by humans. But we have no evidence that they were used for such purposes before the arrival of Modern Humans.

The three cognitive processes critical to making art – mental conception of an image, intentional communication and the attribution of meaning – were all present in the Early Human mind. They were found in the domains of technical, social and natural history intelligence respectively. But the creation and use of visual symbols requires that they function ‘seamlessly and smoothly together’ (to quote Gardner). This would require ‘links across domains’ (to quote Karmiloff-Smith). And the result would be a ‘cultural explosion’ (to quote Sperber).

We do see a cultural explosion beginning 40,000 years ago in Europe as the first works of art were produced and I would suggest that this can be explained by new connections between the domains of technical, social and natural history intelligence. The three previously isolated cognitive processes were now functioning together, creating the new cognitive process which we call visual symbolism, or simply art (see Box p. 163).

If I had to choose just one feature of the earliest art to support this argument it would be that the very first images are of such technical skill and emotive power. No analogy can be drawn between the origins of art in evolutionary time and the development of artistic skills by a child. The latter consists of a gradual change from scribbles to representational images, and then a gradual improvement in the quality of those images. For some young artists one can then see a gradual understanding of how to use line and colour to convey not just a record of what is seen but one’s feelings for it. There is nothing gradual about the evolution of the capacity for art: the very first pieces that we find can be compared in quality with those produced by the great artists of the Renaissance. This is not to argue that the ice age artists themselves did not go through a process of learning; we can indeed find many images which appear to be those drawn by children or apprentice artists.²⁸ But the abilities to impose form, to communicate and to infer meaning from images, must have already been present in the Early Human mind – although there was no art. All that was needed was a connection between these cognitive processes which had evolved for other tasks to create the wonderful paintings in Chauvet Cave.

But before we leave the origins of art we must return to those scratched pieces of bone and ivory made by Early Humans, such as from Bilzingsleben and Tata. If – and it is a big if – these lines were intentionally made, how can they be accounted for? I suggest that they reflect the maximum amount of symbolic communication that can be achieved by relying on general intelligence alone. Early Humans may have been able to associate marks with meanings by using their capacities for associative learning alone. But relying on this would have severely constrained the complexity of the marks and meanings. There is a similarity between the simplicity of the toolmaking capacities of chimpanzees as compared with those of Early Humans, and the simplicity of Early Human intentional markings as compared with those of Modern Humans. Chimpanzees rely on general intelligence for toolmaking, just as Early Humans relied on general intelligence for ‘symbolic’ communication. As a result, chimpanzees and Early Humans appear to ‘underachieve’ in these activities in light of their accomplishments in behavioural domains for which they have specialized intelligences.

The new flow of knowledge and thought processes between cognitive domains of the modern mind can be readily seen not only in the existence of art, but also in its contents. Consider once again the image in Figure 18. This figure has a lion’s head and human body. We cannot prove, but equally cannot doubt, that it represents a being in the mythology of the Upper Palaeolithic groups of southern Germany. Whether it is an image of an animal that has taken on certain human attributes – reflecting anthropomorphic thinking – or a human who is descended from a lion – reflecting totemic thought – we do not know. But, which ever of these is correct (and the answer is probably both), the ability to conceive of such a being requires a fluidity between social and natural history intelligences.

Images like this pervade not only the art of Upper Palaeolithic groups, but that of almost all hunter-gatherer societies, and indeed those living by agriculture, trade and industry.²⁹ We have many spectacular examples from prehistory. In the art of the Upper Palaeolithic they include the ‘sorcerer’ from Trois-Frères – a painted figure that has an upright posture, legs and hands that look human, but the back and ears of a herbivore, the antlers of a reindeer, the tail of a horse, and a phallus positioned like that of a feline (see Figure 21) – as well as a bird-headed man from Lascaux and a female figurine from Grimaldi Cave paired back to back with a carnivore.³⁰ Indeed one of the paintings in the newly discovered Chauvet Cave, some of which are dated to 30,000 years old, is a figure with the head and torso of a bison and the legs of a human. Similarly the prehistoric hunter-gatherers who lived 7000 years ago in the forests of Europe after the ice had retreated made monumental carvings of fish/humans at the site of Lepenski Vir on the Danube.³¹ As I noted in Chapter 3, among the modern hunter-gatherers described by anthropologists, animals are frequently attributed with human-type minds.

21 The sorcerer from Trois-Frères, Ariège, France, as drawn by Henri Breuil. Height 75 cm.

Anthropomorphic thinking is something that pervades our own everyday lives. We indulge in anthropomorphic thinking in our relations with pets by attributing to them feelings, purposes and intentions. This may indeed be reasonable with regard to dogs and cats, but with a moment’s reflection it seems far-fetched with regard to pets such as gold fish. We seem unable to help anthropomorphizing animals – some claim that it is built into us by both nature and nurture – and while this gives us considerable pleasure, it is a problem that plagues the study of animal behaviour, for it is unlikely that animals really do have human-like minds.³² Anthropomorphism is a seamless integration between social and natural history intelligence (see Box p. 166). The very first pieces of Palaeolithic art indicate that it stretches back to the cultural explosion of 40,000 years ago. But I doubt if it goes back any further.

Totemism is the other side of the human/animal coin. Rather than attributing animals with human characteristics, it involves embedding human individuals and groups within the natural world, epitomized by tracing descent from a non-human species. The study of totemism – and attempts to define it – formed the core of social anthropology as it developed during the 19th century. Between 1910 and 1950 major works on totemism were produced by the pioneers of social anthropology including Frazer, Durkheim, Pitt-Rivers, Radcliffe-Brown and Malinowski. Such works provided the foundations for Levi-Strauss’ The Savage Mind. This in turn has been followed as from the 1970s by a renewed surge of interest in totemism.³³

In the light of this long history of study, it is not surprising that totemism has been defined and interpreted in a variety of ways. Lévi-Strauss’ position is perhaps the most widely known: animals are not just good to eat but also ‘good to think’. He viewed totemism as the practice of humanity brooding on itself and its place in nature. To his mind, the ‘study of natural species, provided non-literate and prescientific groups with a ready-to-hand means of conceptualizing relations between human groups’.³⁴

Whether or not this is a correct interpretation, we may simply note three features of totemism that are particularly relevant for an understanding of the evolution of the modern mind. First, when broadly defined, totemism is universal among human groups who live by a hunting-gathering lifestyle; second, it requires a cognitive fluidity between thinking about animals and people; and third, on the basis of archaeological evidence it is likely to have been pervasive in human society since the start of the Upper Palaeolithic. The evidence we can invoke here includes that of imagery in Palaeolithic art and that from burials, such as at the 7800-year-old cemetery at Oleneostrovski Mogilnik in Karelia where we find two clusters of graves, one associated with effigies of a snake, and the other with effigies of an elk.³⁵ In contrast, we have no reason to believe that Early Human society was structured on a totemic basis.

We must also note here that it is not just other living things which are thought of as possessing human qualities. Hunter-gatherers do not just live in a landscape of animals and plants, rocks, hills and caves. Their landscapes are socially constructed and full of meaning. Once again the Aboriginal communities of Australia provide a good example. The wells in their landscape are where ancestral beings dug in the ground, the trees where they had placed their digging sticks and the deposits of red ochre where they had bled.³⁶ John Pfeiffer has argued that the encompassing of the features of the landscape in a web of myths and stories is of great utility to the Aborigines, for it helps them to remember enormous quantities of geographic information.

Whether or not this is the case, when we look at a region such as that of south west France in which we find both a range of topographic features universally attributed with social and symbolic meanings by modern hunter-gatherers,³⁷ and caves and rockshelters covered with paintings, we can be in no doubt that the Upper Palaeolithic hunters were also living in a landscape full of symbolic meanings.

It is useful to recall here the words of Tim Ingold that I quoted in Chapter 3: ‘For them [modern hunter-gatherers] there are not two worlds of persons (society) and things (nature), but just one world – one environment – saturated with personal powers and embracing both human beings, the animals and plants on which they depend, and the landscape in which they live and move.’³⁸ The anthropomorphic images and painting of caves and rockshelters that begin after 40,000 years ago suggests that the earliest Upper Palaeolithic hunter-gatherers had a similar attitude to the social and natural worlds: they were one and the same. One consequence, of benefit to us today, is that they expressed this view within their art, creating some of the most powerful and beautiful images ever made. But this collapse of the cognitive barrier between the social and natural worlds also had significant consequences for their own behaviour, for it fundamentally changed their interaction with the natural world. It is to this that we must now turn.

The hunter-gatherers of the Upper Palaeolithic were hunting the same types of animals as the Early Humans. In Europe, for instance, reindeer, red deer, bison and horse continued as the mainstay of their economies, while in southern Africa animals such as eland, cape buffalo and seals remained the most important prey. What differed, however, is the manner in which these animals were hunted. Modern Humans appear to have been considerably more proficient at predicting game movements and planning complex hunting strategies.

This is readily apparent from Europe. Almost all the sites of Early Humans have a mix of animal species, suggesting that these were hunted as individuals on an opportunistic basis. The site of Combe Grenal in south west France is typical in this regard. Each occupation level usually contains a few individuals of each of the types of large game being hunted. As the climate grew colder, animals such as reindeer become more prevalent in the occupation deposits, while red deer increase during periods of relative warmth. The Neanderthals were simply hunting whatever animals were available – although as I indicated in the last chapter, we should certainly not minimize their achievement at exploiting such game.

The first Modern Humans in Europe hunted in a very different fashion. Although they continued to kill individual animals, or at most small groups, they began to specialize on specific animals at specific sites.³⁹ Consequently many sites are dominated by one species alone, very often reindeer. Indeed certain sites seem to have been selected for ambush hunting, indicating that Modern Humans were much better at predicting the movements of animals than Early Humans. This becomes very apparent when we look at hunting methods in the period c. 18,000 years ago, when the last ice age was at its peak. At about this time, Modern Humans shifted from hunting individual and small groups of animals to slaughtering mass herds of reindeer and red deer. These are likely to have been attacked at critical points on their annual migration routes when the animals were constrained in narrow valleys, or when crossing rivers.⁴⁰

The same contrast between Early and Modern Humans can be seen in other parts of the Old World. In northern Spain, for example, animals such as ibex began to be hunted for the first time. This is significant because, as the archaeologist Lawrence Straus has written, ibex hunting required ‘elaborate strategies, tactics, weapons and … logistical camps’. By ‘logistical camps’ he refers to sites specifically located for ibex hunting.⁴¹ Similarly on the Russian Plain, Olga Soffer has described how the first Upper Palaeolithic hunters were locating sites for exploiting specific animals at specific times of the year. She suggests that they were taking greater account of the seasonal and long-term fluctuations in animal numbers and behaviour patterns.⁴² The same can be seen in southern Africa. For instance, Richard Klein has suggested that a new awareness of the seasonal variation in seal numbers had arisen, and was being used to plan hunting trips to the coast. This replaced a more opportunistic pattern of hunting and scavenging.⁴³

In general, the Modern Humans of the Upper Palaeolithic appear to have had a significantly greater ability both to predict the movements of animals and to use that knowledge in their hunting strategies. How were they managing to do this? The answer lies in what has already been a major theme of this chapter: anthropomorphic thinking. This is universal among all modern hunters and its significance is that it can substantially improve prediction of an animal’s behaviour. Even though a deer or a horse may not think about its foraging and mobility patterns in the same way as Modern Humans, imagining that it does can act as an excellent predictor for where the animal will feed and the direction in which it may move.

This has been recognized in several studies of living hunter-gatherers, such as among the G/Wi and the !Kung of the Kalahari, the Valley Bisa of Zambia and the Nunamiut of the Canadian Arctic. Anthropomorphizing animals by attributing to them human personalities and characters provides as effective a predictor for their behaviour as viewing them with all the understanding of ecological knowledge possessed by Western scientists.⁴⁴ The anthropologist Mary Douglas sees the similarity in the categories used for understanding the natural and social worlds as primarily being of practical value in terms of understanding and predicting the ways of animals. She suggests that this is of far more importance than using the natural world for addressing profound metaphysical problems about the human condition, as proposed by Lévi-Strauss.⁴⁵

Anthropomorphic thinking, therefore, has clear utilitarian benefits. Yet the new powers of prediction would have been of limited value had Modern Humans not also been able to develop new types of hunting weapons. And we do indeed see a striking elaboration of technology at the start of the Upper Palaeolithic. In Europe, Modern Humans could make all those types of tools which Neanderthals, with their Swiss-army-knife mentality, could not even think about: tools which required an integration of technical and natural history intelligences.

For example, we see many new types of weapons made from bone and antler, notably harpoons and spearthrowers. Experimental studies using replica artifacts have shown that these were very effective at piercing animal hides and organs.⁴⁶ We see many new types of stone projectile points, and find associations between specific types of points and specific types of animals.⁴⁷ We can see evidence for complex, multi-component tools being made, such as in the presence of microliths – small blades of flint used as points and barbs. Lying at the heart of these new technological innovations was the switch to ‘blade technology’, which provided standardized ‘blanks’, each of which could be turned into part of a highly specialized tool (see Figure 22).

It is not simply the introduction of new tools at the start of the Upper Palaeolithic which is important. It is how these were then constantly modified and changed. Throughout the Upper Palaeolithic we can see the processes of innovation and experimentation at work, resulting in a constant stream of new hunting weapons appropriate to the prevailing environmental conditions and building on the knowledge of previous generations. As the environments became very harsh at the height of the last ice age 18,000 years ago, large points were manufactured, specialized for ensuring that large game would be despatched on the tundra. As the climate began to ameliorate, and a wider range of game became available, hunting technology became more diverse, with an emphasis on multicomponent tools.⁴⁸ Lawrence Straus has appositely described this as a Palaeolithic arms race.⁴⁹ Such behaviour, geared to maintaining if not maximizing hunting efficiency, is markedly different from the monotony of the hunting tools of Early Humans during the equally variable environments that they exploited. It could only have arisen owing to a new connection between natural history and technical intelligence.

The design of hunting weapons is perhaps the best example of this new type of thinking, but it also resulted in a wide range of other technological developments. For instance, by 18,000 years ago people in North Africa were using grinding stones for preparing plant material. Such artifacts required integrated thought about the characteristics of both stone and plant material.⁵⁰ The elaboration in the range of scraping and engraving tools used for such tasks as cleaning hides and carving bone required thought about the nature of animal products during the process of tool manufacture. And perhaps most impressive of all is the development of facilities for trapping animals, such as small game or fish, and the technology for storing food, whether it be reindeer meat during the Upper Palaeolithic or hazelnuts once forests had spread across Europe after the end of the last ice age 10,000 years ago.⁵¹ The design and use of all these involve an integration of natural history and technical knowledge, resulting in a constant innovation of new technology.

Many of the new bone and antler tools of the Upper Palaeolithic carried elaborate designs engraved on to their surfaces, or were even carved into animal figures themselves, such as the spearthrower from Mas d’Azil (see Figure 23). Indeed it is very difficult to draw any division between what is a piece of ‘art’ and what is a ‘tool’, and such artifacts epitomize the absence of any boundaries between different domains of activity. Many of the art objects can indeed be thought of as a brand new type of tool: a tool for storing information and for helping to retrieve information stored in the mind.

The simplest tools of this new type are pieces of bone with incised parallel lines. The most complex have many hundreds of marks made by a number of different tools, creating a complex pattern on the face of the artifact, such as on the Taï plaque from eastern France (see Figure 24).⁵² The interpretation of these has always been controversial. When first discovered they were described as ‘tailles de chasse’ – hunting tallies recording the number of animals killed. A range of other interpretations have since been made, for example that they record the number of people attending social gatherings and lunar calendars.⁵³

Detailed microscopic study of such artifacts by Alexander Marshack and Francesco D’Errico has confirmed that on several of them the marks come in such regular patterns that they appear to be a system of notation.⁵⁴ These artifacts are likely to have acted as a form of visual recording device, most probably about environmental events. They look very similar to notched and engraved artifacts made by modern hunter-gatherers which are known to have been mnemonic aids and recording devices, such as the calendar sticks made from ivory by the Yakut people of Siberia.⁵⁵

23 Antler spearthrower from Mas d’Azil, Ariège, France. This depicts an ibex that is either giving birth, or excreting a large turd on which two birds are perched. Total length 29.6 cm.

Like the engraved pieces of bone, cave paintings also appear to have been used to store information about the natural world, or at least facilitate its recall by acting as a mnemonic device. Indeed, these paintings have been described as the ‘tribal encyclopedia’ by John Pfeiffer.⁵⁶ I myself have suggested that much of the animal imagery within this art served to help recall information about the natural world stored within the mind.⁵⁷ For instance, I have argued that the manner in which many of the animals were painted makes direct reference to the ways in which information was acquired about their movements and behaviour. In some images, while the animals were painted in profile, their hooves were painted in plan, as if hoofprints were being depicted to facilitate the memorizing and recall of tracks seen while in the environment, or even the teaching of children. Similarly the choice of imagery itself was selective towards those animals which provide knowledge about forth coming environmental events. The bird imagery is particularly telling, dominated as it is by ducks and geese, which are likely to have been migratory. Modern hunters in glaciated environments keep a very close lookout for the annual arrival and departure of such birds, since such information gives a clue as to when the big freeze of the winter, or the spring thaw, will happen. Some of the most evocative images of this type are ivory carvings of geese in flight found at the Siberian site of Mal’ta, where the hunters had relied on mammoths for food but no doubt eagerly watched for the passing of migrating birds indicating the arrival of spring.⁵⁸

The way in which Upper Palaeolithic cave paintings may have functioned to help store information about the natural world is perhaps analogous to the way in which Wopkaimin hunter-horticulturalists of New Guinea use the bones from the animals they hunt. These bones are placed on the rear walls of their houses where they are described as ‘trophy arrays’. But they are carefully arranged to act as a mental map for the surrounding environment to facilitate the recall of information about that environment and animal behaviour. They thus play an important role in decision-making about use of resources and improving the predictions about animal location and behaviour.⁵⁹ There is clear patterning in the arrangement of animal figures in the cave paintings of the Upper Palaeolithic.⁶⁰ Michael and Anne Eastham have suggested that the paintings and engravings in the caves of the Ardèche region of France served as a model or a map for the specific terrain around the caves.⁶¹

In summary, although the specific roles that prehistoric artifacts may have played in the management of information about the natural world remain unclear, there can be little doubt that many of them served to store, transmit and retrieve information. Major benefits of this will have been enhanced abilities to track long-term change, to monitor seasonal fluctuations and to devise hunting plans. Many of the paintings, carvings and engravings of Modern Humans were tools with which to think about the natural world.

Beads, pendants and other items of personal decoration first appear at the start of the Upper Palaeolithic. They too arise from the new cognitive fluidity of the mind – an integration between technical and social intelligence. Such artifacts are initially found in abundance in occupation deposits of caves in south west France, and are particularly important during the very harsh climatic conditions at around 18,000 years ago.⁶² They are often found in burials, most dramatically on the 28,000-year-old burials at Sungir in Russia (see Box p. 175). Describing beads and pendants as ‘decoration’ risks belittling their importance. They would have functioned to send social messages, such as about one’s status, group affiliation and relationships with other individuals, just as they do in our own society today. And of course these messages need not have been ‘true’; beads and pendants provide new opportunities for deception in the kind of social tactics that we saw are prevalent even among chimpanzees. To have produced such artifacts required not only specialized social and technical intelligences – as possessed by Early Humans – but also an ability to integrate these.

It is likely that all types of artifacts, including those that might appear to be mundane tools for hunting or even processing animal hides, became imbued with social information at the start of the Upper Palaeolithic.⁶³ In effect the ‘goal posts’ of social behaviour were moved; whereas for Early Humans the domains of hunting, toolmaking and socializing were quite separate, these were now so integrated that it is impossible to characterize any single aspect of Modern Human behaviour as being located in just one of these domains. Indeed as Ernest Gellner stated: ‘the conflation and confusion of aims and criteria, is the normal and original condition of mankind’.⁶⁴

Many of the new behaviours I have been describing, such as the anthropomorphic images in the cave paintings and the burial of people with grave goods, suggest that these Upper Palaeolithic people were the first to have beliefs in supernatural beings and possibly an after life. We are indeed seeing here the first appearance of religious ideologies. This can be explained by the collapse of the barriers that had existed between the multiple intelligences of the Early Human mind.

Just as we did with art, we must first reach some agreement on quite what we mean by the notion of religion. While it is difficult to identify features universal to all religions, there are nevertheless a series of recurrent ideas. The importance of these has been stressed by the social anthropologist Pascal Boyer in his 1994 book The Naturalness of Religious Ideas. Boyer explains that a belief in non-physical beings is the most common feature of religions; it may indeed be universal. In fact, ever since the classic work of E.B. Tylor in 1871 on Primitive Cultures, the idea of non-physical beings has been taken for the very definition of religion itself. Boyer notes three other recurrent features of religious ideologies. The first is that in many societies it is assumed that a non-physical component of a person can survive after death and remain as a being with beliefs and desires. Second, it is very frequently assumed that certain people within a society are especially likely to receive direct inspiration or messages from supernatural agencies, such as gods or spirits. And third, it is also very widely assumed that performing certain rituals in an exact way can bring about change in the natural world.

Sending social information by material culture: the Sungir burials

The burials at Sungir, Russia, have been dated to 28,000 years old. They consist of the graves of a 60-year-old man, and a joint burial of a male and a female adolescent. Each of these individuals were decorated with thousands of ivory beads, which had probably been attached to clothing. The archaeologist Randall White has studied these graves and provides the following descriptions:

The man was adorned with 2936 beads and fragments arranged in strands found on all parts of his body including his head, which was apparently covered with a beaded cap that also bore several fox teeth. His forearms and biceps were each decorated with a series of polished mammoth-ivory bracelets (25 in all), some showing traces of black paint…. Around the man’s neck he wore a small flat schist pendant, painted red, but with a small black dot on one side….

The supposed small boy was covered with strands of beads – 4903 of them – that were roughly 2/3 the size of the man’s beads, although of exactly the same form. Unlike the man, however, he had around his waist – apparently the remains of a decorated belt – more than 250 canine teeth of the polar fox. On his chest was a carved ivory pendant in the form of an animal. At his throat was an ivory pin, apparently the closure of a cloak of some sort. Under his left shoulder was a large ivory sculpture of a mammoth. At his left side lay a medial segment of a highly polished, very robust human femur, the medullary cavity of which was packed with red ochre. At his right side … was a massive ivory lance, made from a straightened woolly mammoth tusk…. Near it is a carved ivory disc which sits upright in the soil.

The supposed girl had 5274 beads and fragments (also roughly 2/3 the size of the man’s beads) covering her body. She also wore a beaded cap and had an ivory pin at her throat, but her burial contains no fox teeth whatsoever. Nor does she have a pendant on her chest. However, placed at each of her sides there was a number of small ivory ‘lances’, more appropriate to her body size than that accompanying the boy. Also at her side are two pierced antler batons, one of them decorated with rows of drilled dots. Finally, she was accompanied by a series of three ivory disks with a central hole and lattice work, like that adjacent to the supposed boy’s burial.

(White 1993, 289–292)

If we look at the archaeological evidence from the start of the Upper Palaeolithic, we get hints that each of these features was present. Few can doubt that the painted caves, some of which were located deep underground, were the locus for ritual activities. Indeed the anthropomorphic images within this art, such as the sorcerer from the cave of Les Trois-Frères, are most easily interpreted as being either supernatural beings or shamans who communicated with them. As was most forcefully argued by the French prehistorian André Leroi-Gourhan, these painted caves are likely to reflect a mythological world with concepts as complex as those of the Dreamtime held by the Australian Aborigines.

In addition to the art we have the evidence from the burials. It is difficult to believe that such investment would have been made in burial ritual, as at Sungir, had there been no concept of death as a transition to a non-physical form. Indeed, since only a tiny fraction of the Upper Palaeolithic population seems to have been buried, it is likely that these people played a special religious role within their society.

Pascal Boyer has explored how the characteristics of supernatural beings as found in religious ideologies relate to the intuitive knowledge about the world genetically encoded in the human mind. In Chapter 3 I described three types of intuitive knowledge, that regarding psychology, biology and physics, and argued that these ‘kickstarted’ the formation of cognitive domains or multiple intelligences during child development. Boyer argues that a typical feature of supernatural beings is that they have characteristics which violate this intuitive knowledge.

For example, Boyer explains that the supernatural beings of religious ideologies commonly violate intuitive biological knowledge. While they may have bodies, they do not undergo the normal cycle of birth, maturation, reproduction, death and decay. Similarly, they may violate intuitive physics by being able to pass through solid objects (as with ghosts) or simply be invisible. Nevertheless, supernatural beings also have a tendency to conform to some intuitive knowledge; for instance, they are very frequently intentional beings who have beliefs and desires like normal human beings. The Ancestral Beings of the Australian Aborigines provide an excellent example of such entities which both violate and conform to intuitive knowledge of the world. On the one hand, they have very weird characteristics, such as existing in both the past and the present. On the other hand, in many of the stories they play tricks and engage in deception in a manner which is very human.⁶⁵ A more familiar example to many people will be the gods of Greek legends who have supernatural powers but also suffer jealousies and petty rivalries much like those of normal people.

Boyer argues that it is this combination of violation of, and conformity to, intuitive knowledge that characterizes supernatural beings in religious ideologies. The violations make them something different, but by conforming to some aspects of intuitive knowledge people are able to learn about them; if there was nothing about supernatural beings which conformed to intuitive knowledge of the world, the concept of them would simply be too difficult for the human mind to grasp.

An alternative way of viewing this feature of supernatural beings is as a mixing up of knowledge about different types of entities in the real world – knowledge which would have been ‘trapped’ in separate cognitive domains within the Early Human mind. For example, Early Humans would have known that rocks are not born and do not die like living things. And Early Humans would also have known that people have intentions and desires, while inert nodules of stone do not. Because they had isolated cognitive domains, there was no risk of the Early Human mind getting these entities mixed up, and arriving at a concept of an inert object that is neither born nor dies, but which nevertheless has intentions and desires. Such concepts, which Boyer argues are the essence of a supernatural being, could only arise in a cognitively fluid mind.

Boyer himself suggests that a combination of knowledge about different types of entities explains another recurrent feature of religious ideologies – the fact that some individuals are believed to have special powers of communication with supernatural beings. At the heart of this notion, Boyer argues, is the belief that some people have a different ‘essence’ from others in the group. I discussed the notion of essence in Chapter 3, where it was explained to be a critical feature of intuitive biology, a means by which even young children are able to classify animals into different species. Boyer explains the differentiation of people into different social roles, exemplified by that of shaman, as an introduction of the notion of essence into thought about the social world. In other words, it is a consequence of cognitive fluidity.

We cannot, of course, reconstruct the religious ideologies of the earliest Upper Palaeolithic societies. But we can be confident that religious ideologies as complex as those of modern hunter-gatherers came into existence at the time of the Middle/Upper Palaeolithic transition and have remained with us ever since. This appears to be another consequence of the cognitive fluidity that arose in the human mind, which resulted in art, new technology, and a transformation in the exploitation of the natural world and the means of social interaction.

The new cognitive fluidity transformed the human mind and all aspects of human behaviour (see Figure 25). It is not surprising that with new abilities to use materials such as bone and ivory for tools, and to use artifacts to store and transmit information, humans were able to colonize new areas of the world. At around 60,000 years ago a second major pulse of movement across the globe began, following that of the first Early Humans to leave Africa more than 1.5 million years ago. As Clive Gamble has described in his recent study of global colonization,⁶⁶ Australasia was colonized by extensive sea voyages, and then the North European Plain, the arid regions of Africa and the coniferous forests and tundra of the far north were colonized soon after 40,000 years ago. Early Humans may have temporarily entered these environments, but they did not remain on a long-term basis. Modern Humans not only colonized them but used them as stepping stones to the Americas and the Pacific islands.

The emergence of a cognitively fluid mentality provides the answer to the Middle/Upper Palaeolithic transition. But remember that this transition does not happen until half-way through Act 4. The start of that act is defined by the appearance of H. sapiens sapiens in the fossil record at 100,000 years ago. We must complete this chapter by asking how the minds of these Early Modern Humans – those who lived before the Middle/Upper Palaeolithic transition – were different from those of the Early Humans of Act 3 (who also continued into the first scene of Act 4), and the Modern Humans who lived after the Middle/Upper Palaeolithic transition, among whom we must include ourselves.

There is, I believe, a simple answer to this question. The Early Modern Humans seem to have been achieving some degree of integration between their specialized intelligences, but not gaining the full cognitive fluidity that arose after 60,000 years ago. Their minds were a half-way house between a Swiss-army-knife and a cognitively fluid mentality.

We can see this most clearly in the Near East, where we find the remains of Early Modern Humans in the caves of Skhūl and Qafzeh dating to between 100,000 and 80,000 years ago. While their stone tools are practically indistinguishable from those of the Neanderthals who used the cave of Tabūn before the Early Modern Humans arrived (c. 180,000–90,000 years ago), and Kebara after they left (63,000–48,000 years ago), the Early Modern Humans seem to have had two unique features to their behaviour.

The first is that they placed parts of animal carcasses within human graves. For instance, in the cave of Qafzeh a child was found buried with the skull and antlers of a deer. At Skhūl one of the burials contained a body which had been laid on its back, with the jaws of a wild boar placed within its hands.⁶⁷ These seem to imply ritualized burial activity, and a belief in religious ideologies. Recall that while Neanderthals did bury some individuals, there is no evidence for the intentional placing of items within the graves, or for any ritual activity associated with the act of burial.

The second contrast concerns the hunting of gazelle. This was the most important animal hunted by both Neanderthals and Modern Humans, and both appear to have used short thrusting spears with stone points. But their hunting patterns were quite different. The Early Modern Humans used their caves on a seasonal basis, and probably expended less physical energy in their hunting behaviour. In addition, they appear to have needed to repair their spears less frequently.⁶⁸ In other words, they were hunting with greater degrees of planning and more efficiently than the Neanderthals. This, in turn, is likely to reflect an enhanced ability at predicting the location and behaviour of their prey.

At first glance these two differences between the Early Modern Humans and the Neanderthals of the Near East appear unrelated. But there is in fact a very significant relationship: both derive from an integration of natural history and social intelligence in the minds of Early Modern Humans. As I argued earlier in this chapter, improvements in the ability to predict animal behaviour over what can be achieved with a natural history intelligence alone probably derive from anthropomorphic thinking, as is universal among living hunter-gatherers. I also discussed how concepts of religious belief arise from cognitive fluidity, particularly the integration of natural history and social intelligence. The placing of animal parts within the burials of Early Modern Humans implies that some associations were being made between people and animals, probably reflecting some form of totemic thought. It is significant, I think, that artifacts were not placed within the burials, which is common practice during the Upper Palaeolithic. This suggests that technical intelligence remained isolated within the Early Modern Human mind. This is indeed confirmed by the fact that in spite of their abilities at predicting the behaviour of gazelle, Modern Humans continued to use the same types of hunting weapons as the Neanderthals. They do not appear to have been designing more effective hunting weapons, which would have arisen if technical and natural history intelligence had been integrated, nor were they investing their stone tools with social information, as would have arisen if technical and social intelligence were integrated.

In summary, the minds of the Early Modern Humans of the Near East seem to be a half-way house between the Swiss-army-knife mentality of Early Humans and the cognitively fluid mentality of Modern Humans (see Figure 26).

We reach a similar conclusion when we consider the Early Modern Humans of South Africa. Their fossils, found in the caves of Klasies River Mouth and Border Cave, are less well preserved than those of the Near East, but date to the same time period of around 100,000 years ago. The South African specimens contain some archaic features and this region is likely to have been the original source of H. sapiens sapiens.⁶⁹

The long stratified sequence of archaeological deposits in Klasies River Mouth is of most interest.⁷⁰ It covers the period between around 140,000 years ago and 20,000 years ago. Towards the end of this sequence, at around 40,000 years ago, we see a change in stone technology from a predominantly flake to a blade production method, which denotes the Middle/Upper Palaeolithic transition – although in Africa this is referred to by archaeologists as the change from the Middle to Later Stone Age. Prior to this event, the stone tools in almost the whole of this sequence are very similar to those made by Early Humans elsewhere in the African continent during Act 3, even though those after 100,000 years ago appear to have been made by Early Modern Humans, the first H. sapiens sapiens.

However, the levels likely to correlate with the first appearance of Early Modern Humans are notable for a significant increase in the quantity of red ochre.⁷¹ Some of these pieces seem to have been used as crayons. The pieces of red ochre remain quite rare, less than 0.6 percent of the artifacts from any one layer, but are nevertheless at much higher frequencies than in sites associated with Early Humans. Indeed there are no pieces of red ochre known prior to 250,000 years ago, and only a dozen pieces before 100,000 years ago. Red ochre is also found at other sites in southern Africa after this date, and there have even been claims that it was mined at Lion Cavern in Swaziland. It remains unclear what the Early Modern Humans were doing with the ochre. As the anthropologists Chris Knight and Camilla Powers have argued, body painting is the most likely explanation, since there are no objects of art known in South Africa prior to 30,000 years ago, nor are there any beads or pendants.⁷²

A few other traces can be found of new types of behaviour by the Early Modern Humans in southern Africa. In Border Cave there appears to have been a burial of an infant within a grave dating to between 70,000 and 80,000 years ago. This is the only burial known from the Middle Stone Age of the region and it is notable for not only being that of an Early Modern Human, but for also containing a perforated Conus shell that had originated more than 80 kilometres away.⁷³ Another innovation – along side the more wide spread stone flake technology – was the introduction of small blades, made from higher quality stone and chipped into forms which would not be out of place in the Upper Palaeolithic of Europe. These blades look as if they were designed for multi-component tools.⁷⁴ A final type of novel behaviour is the working of bone. The most dramatic evidence comes from the sites at Katanda in Zaire, where bone harpoons with multiple barbs have been found. These are as complex as any bone artifact from the Upper Palaeolithic of Europe. They were made by grinding and are at least 90,000 years old – making them 60,000 years earlier than any other known examples. They are associated with typical Middle Stone Age stone artifacts.⁷⁵

If we are indeed dealing with a single type of human in southern Africa after 100,000 years ago, then the mentality of the Early Modern Humans appears to drift in and out of cognitive fluidity. It is as if the benefits of partial cognitive fluidity were not sufficient for this mental transformation to have been ‘fixed’ within the population. The minds of these Early Modern Humans seem like those of the Early Modern Humans of the Near East in showing some degree of cognitive fluidity, but one that did not match what arose after the start of the Upper Palaeolithic.

Nevertheless, this partial cognitive fluidity was to prove absolutely critical in giving Early Modern Humans the competitive edge as they spread from Africa and the Near East throughout the world between 100,000 and 30,000 years ago. The Early Modern Humans of the Near East are likely to be representatives of – or at least closely related to – the source population of H. sapiens sapiens that left Africa, spread into Asia and Europe and replaced all existing Early Humans.⁷⁶

The strongest evidence for this replacement scenario is the limited amount of genetic diversity among living humans today. Although there is considerable controversy as to how modern genetic variability should be interpreted, there is strong evidence that there has been a recent and severe ‘bottleneck’ in human evolution. In general, living Africans have a higher degree of genetic variability than people elsewhere in the world, suggesting that as the first H. sapiens sapiens left Africa there was a considerable loss of genetic variation. This implies that for a short period of time there was a very small breeding population. One recent estimate has suggested no more than six breeding individuals for 70 years, which would reflect an actual population size of around 50 individuals, or 500 individuals if this bottleneck lasted for 200 years.⁷⁷

If the Early Modern Humans of the Near East are indeed part of this source population, or closely related to them, then as they spread throughout the world, they took with them their partially cognitively fluid minds. This feature of their mentality was presumably encoded within their genes. It was their integration of natural history and social intelligence which enabled them to compete successfully with resident Early Human populations, pushing the latter into extinction – although the possibility of some hybridization remains. And consequently we find H. sapiens sapiens in China at 67,000 years ago, represented by the fossil skull from Liujang.⁷⁸

At slightly different times in different parts of the world the final step to a cognitively fluid mind was taken. This was the integration of technical intelligence with the already combined social and natural history intelligences. That all H. sapiens sapiens populations dispersed throughout the world took this final step – a case of parallel evolution – was perhaps inevitable. There was an evolutionary momentum to cognitive fluidity; once the process had begun it could not be halted. It appears that as soon as a set of adaptive pressures arose in each area, technical intelligence became part of the cognitively fluid mind, the final step on the path to modernity.

In this chapter I have argued that the events of Act 4 can be explained by the emergence of cognitive fluidity in the human mind. This process began with the very first appearance of H. sapiens sapiens and its culmination caused the cultural explosion that archaeologists call the Middle/Upper Palaeolithic transition. But, as in so much of science, answering one question merely raises another. How did it happen? How did the thoughts and knowledge escape from their respective chapels of the Early Human mind?