The Origin of Humankind

by Richard Leakey

  These days, many anthropologists favor a recent, rapid origin of language—principally because of the abrupt change in behavior seen in the Upper Paleolithic Revolution. Randall White, a New York University archeologist, argued in a provocative scientific paper almost a decade ago that evidence of various forms of human activity earlier than 100,000 years ago implies “a total absence of anything that modern humans would recognize as language.” Anatomically modern humans had evolved by this time, he concedes, but they had not yet “invented” language in a cultural context. This would come much later: “By 35,000 years ago, these populations . . . had mastered language and culture as we presently know them.”

  White lists seven areas of archeological evidence that, in his view, point to dramatic enhancement of language abilities coincident with the Upper Paleolithic. First, the deliberate burial of the dead, which almost certainly began in Neanderthal times but became refined, with the inclusion of grave goods, only in the Upper Paleolithic. Second, artistic expression, which included image making and bodily adornment, begins only with the Upper Paleolithic. Third, in the Upper Paleolithic there is a sudden acceleration in the pace of technological innovation and cultural change. Fourth, for the first time regional differences in culture arise—an expression and product of social boundaries. Fifth, the evidence of long-distance contacts, in the form of the trading of exotic objects, becomes strong at this time. Sixth, living sites significantly increase in size, and language would have been necessary for such a degree of planning and coordination. Seventh, technology moves from the predominant use of stone to include other raw materials, such as bone, antler, and clay, indicating a complexity of manipulation of the physical environment which is unimaginable in the absence of language.

  White and other anthropologists, including Lewis Binford and Richard Klein, are persuaded that this cluster of “firsts” in human activity is underlain by the appearance of complex, fully modern spoken language. Binford, as I noted in an earlier chapter, sees no evidence of planning and little facility for predicting and organizing future events and activities among premodern humans. The great step forward was language—"language and, specifically, symboling, which makes abstraction possible,” he argues. “I don’t see any medium through which such a rapid change could occur other than a fundamentally good, biologically based communication system.” Klein, in essential agreement with this proposition, sees evidence, in archeological sites in southern Africa, of an abrupt and relatively recent increase in hunting skills. This is a consequence, he says, of the origin of the modern human mind, including language facility.

  Although the view that language was a relatively rapid development coincident with the emergence of modern humans is widely supported, it does not completely dominate anthropological thinking. Dean Falk, whose studies of the evolution of the human brain I referred to in chapter 3, defends the proposition that language developed early. “If hominids weren’t using and refining language, I would like to know what they were doing with their autocatalytically increasing brains,” she wrote recently. Terrence Deacon, a neurologist at Belmont Hospital in Belmont, Massachusetts, takes a similar view, but based on studies of modern brains, not fossil ones: “Language competence evolved over a long period (at least 2 million years) of continuous selection determined by brain-language interaction,” he notes in a 1989 article in the journal Human Evolution. Deacon has compared the differences in neuronal connectivity between the ape brain and the human brain. He points out that the brain structures and circuits that were altered the most in the course of human brain evolution reflect the unusual computational demands of spoken language.

  Words do not fossilize, so how can anthropologists settle this argument? The indirect evidence—the artifacts our ancestors made and the changes in their anatomy—seems to tell different stories about our evolutionary history. We will start by examining the anatomical evidence, including brain architecture and the structure of the vocal apparatus. Then we’ll look at technological sophistication and artistic expression—aspects of behavior that constitute the archeological record.

  We’ve already seen that the expansion of the human brain began more than 2 million years ago with the origin of the genus Homo and continued steadily. By half a million years ago, the average brain size among Homo erectus was 1100 cubic centimeters, which is close to the modern average. After the initial 50 percent jump from australopithecine to Homo, there were no further sudden large increases in the size of the prehistoric human brain. Although the significance of absolute brain size is a subject of controversy among psychologists, the tripling that occurred in human prehistory surely reflects enhanced cognitive capacities. If brain size is also related to language capabilities, then the history of brain-size increase during the past 2 million years or so suggests a gradual development of our ancestors’ language skills. Terrence Deacon’s comparison of the anatomy of ape and human brains suggests that this is a reasonable proposition.

  The eminent neurobiologist Harry Jerison, of the University of California at Los Angeles, points to language as the engine of human brain growth, dismissing the notion that manipulative skills provided the evolutionary pressure for bigger brains, as embodied in the hypothesis of Man the Toolmaker. “It seems to me to be an inadequate explanation, not least because toolmaking can be accomplished with very little brain tissue,” he stated in a major lecture at the American Museum of Natural History in 1991. “The production of simple, useful speech, on the other hand, requires a substantial amount of brain tissue.”

  The brain architecture that underlies language is much more complex than was once thought. There appear to be many language-related areas, scattered throughout several regions of the human brain. If such centers could be identified in our ancestors, we would be in a good position to decide the language issue. However, the anatomical evidence of the brains of extinct humans is restricted to surface contours; fossil brains yield no clue to internal structure. Fortunately, one brain feature related in some fashion both to language and to the use of tools is visible on the surface of the brain. This is Broca’s area, a raised lump located near the left temple (in most people). If we could find evidence of Broca’s area in fossil human brains, this would be a signal, albeit an uncertain one, of emerging language ability.

  A second possible signal is the difference in size between the left and right sides of the brain in modern humans. In most people, the left hemisphere is larger than the right—a result, in part, of the packaging of language-related machinery there. Also associated with this asymmetry is the phenomenon of handedness in humans. Ninety percent of the human population is right-handed; right-handedness and a capacity for language may therefore be correlated with a larger left brain.

  Ralph Holloway examined the shape of the brain of skull 1470, a fine example of Homo habilis found east of Lake Turkana in 1972 and determined to be almost 2 million years old (see figure 2.2). He detected not only the presence of Broca’s area, impressed on the inner surface of the cranium, but also a slight asymmetry in the left-right configuration of the brain, an indication that Homo habilis communicated with more than the pant-hoot-grunt repertoire of modern chimpanzees. In a paper in the journal Human Neurobiology, he noted that while it was impossible to prove when or how language began, it was likely that its origins extended “far back into the paleontological past.” Although Holloway has suggested that this evolutionary trajectory might have begun with the australo-pithecines, I disagree. All the discussion of hominid evolution so far in this book points to a major change in hominid adaptation when the genus Homo appeared. I suspect, therefore, that only with the evolution of Homo habilis did some form of spoken language begin. Like Bickerton, I suspect that this was a protolanguage of sorts, simple in content and structure, but a means of communication beyond that of apes and of australopithecines.

  The extraordinarily careful and innovative experimental toolmaking of Nicholas Toth, discussed in chapter 2, has buttressed this view that brain asymme
try was present in early humans. His duplication of their stone flakes demonstrated that the practitioners of the Oldowan industry were predominantly right-handed, and therefore would have had a slightly larger left brain. “Brain lateral-ization occurred in the earliest toolmakers, as evidenced by their toolmaking behavior,” Toth has observed. “This is probably a good indication that a capacity for language was already emerging, too.”

  I am persuaded by the evidence from fossil brains that language started to evolve with the first appearance of the genus Homo. At the very least, there is nothing in this evidence that argues against an early appearance of language. But what of the vocal apparatuses: the larynx, the pharynx, the tongue, and the lips? This represents the second major source of anatomical information (see figure 7.1).

  Humans are able to make a wide range of sounds because the larynx is situated low in the throat, thus creating a large sound -chamber, the pharynx, above the vocal cords. According to innovative work by Jeffrey Laitman, of Mount Sinai Hospital Medical School in New York, Philip Lieberman of Brown University, and Edmund Crelin of Yale, the expanded pharynx is the key to producing fully articulate speech. These researchers conducted considerable research on the anatomy of the vocal tract both in living creatures and in human fossils. It is very different. In all mammals except humans the larynx is high in the throat, which allows the animal to breathe and drink at the same time. As a corollary, the small pharyngeal cavity limits the range of sounds that can be produced. Most mammals therefore depend on the shape of the oral cavity and lips to modify the sounds produced in the larynx. Although the low position of the larynx allows humans to produce a greater range of sounds, it also means that we cannot drink and breathe simultaneously. We exhibit the dubious liability for choking.

  Human babies are born with the larynx high in the throat, like typical mammals, and can simultaneously breathe and drink, as they must during nursing. After about eighteen months, the larynx begins to migrate down the throat, reaching the adult position when the child is about fourteen years old. The researchers realized that if they could determine the position of the larynx in the throats of human ancestral species, they could deduce something about the species’ capacity for vocalization and language. This presented a challenge, because the vocal apparatus is constructed from soft tissues—cartilage, muscle, and flesh—which do not fossilize. Nevertheless, ancient skulls do contain a vital clue. It resides in the shape of the bottom of the skull, or basicranium. In the basic mammalian pattern, the bottom of the cranium is essentially flat. In humans, however, it is distinctly arched. The shape of the basicranium in a fossil human species should therefore indicate how well it was able to articulate sounds.

  FIGURE 7.1

  The vocal tract. The chimpanzee, left, like all mammals, has a vocal tract in which the larynx is high in the throat, a configuration that allows breathing and swallowing at the same time, but limits the range of sounds that can be produced in the pharyngeal space. Humans are unique in having a larynx low in the throat. As a result, humans cannot breathe and swallow at the same time without choking, but they can produce a greatly expanded range of sounds. All human species earlier than Homo erectus had a larynx in the chimpanzee position. (Courtesy of J. Laitman, P. Gannon, and H. Thomas.)

  In a survey of human fossils, Laitman discovered that the basicrania of the australopithecines were essentially flat. In this, as in so many other biological characteristics, they were apelike, and like apes their vocal communication must have been limited. Australopithecines would have been unable to produce some of the universal vowel sounds that characterize human speech patterns. “The earliest time in the fossil record that you find a fully flexed basicranium is about 300,000 to 400,000 years ago, in what people call archaic Homo sapiens,” concluded Lait-man. Does this mean that archaic sapiens species, who appeared before the evolution of anatomically modern humans, had a fully modern language? This seems unlikely.

  The change in the shape of the basicranium is to be seen in the earliest-known Homo erectus specimen, skull 3733 from northern Kenya, dating from almost 2 million years ago. According to this analysis, this Homo erectus individual would have had the ability to produce certain vowels, such as in boot, father, and feet Laitman calculates that the position of the larynx in early Homo erectus would have been equivalent to that in a modern six-year-old. Unfortunately, nothing can be said of Homo habilis, because none of the habilis crania discovered so far has an intact basicranium. My guess is that when we do find an intact cranium of the very earliest Homo, we will see the beginnings of the flexion in the base. A rudimentary capacity for spoken language surely began with the origin of Homo.

  Within this evolutionary sequence we see an apparent paradox. Judging by their basicrania, the Neanderthals had poorer verbal skills than other archaic sapiens that lived several hundred thousand years earlier. Basicranial flexion in Neanderthals was less advanced even than in Homo erectus. Did the Neanderthals regress, becoming less articulate than their ancestors? (Indeed, some anthropologists have suggested that the Neanderthals’ extinction may have been related to inferior language abilities.) An evolutionary regression of this sort seems unlikely; there are virtually no examples of it in nature. More likely, the answer lies in the anatomy of the Neanderthal face and cranium. As an apparent adaptation to cold climates, the Neanderthal midface protrudes to an extraordinary degree, resulting in large nasal passages, in which frigid air can be warmed and moisture in exhaled breath can condense. This configuration may have affected the shape of the basicranium without diminishing the species’ language capacity in a significant way. Anthropologists continue to debate this point.

  Overall, then, the anatomical evidence indicates an early evolution of language, followed by gradual improvement of linguistic skills. However, the archeological evidence for tool technology and artistic expression for the most part tells a different story.

  Although, as I’ve said, language does not fossilize, the products of human hands can, in principle, give some insight into language. When we talk about artistic expression, as we did in the previous chapter, we are aware of modern human minds at work, and that implies a modern level of language. Can stone tools also furnish an understanding of the language capacities of the toolmakers?

  This was the task Glynn Isaac faced when he was asked to present a paper on the origin and nature of language at the New York Academy of Sciences in 1976. He looked at the complexity of stone-tool industries from their beginning more than 2 million years ago to the Upper Paleolithic Revolution 35,000 years ago. He was not interested as much in the tasks that people performed with the tools as in the order the toolmakers imposed on their implements. Imposition of order is a human obsession; it is a form of behavior that demands a sophisticated spoken language for its fullest elaboration. Without language, the arbitrariness of human-imposed order would be impossible.

  The archeological record shows that the imposition of order is slow to emerge in human prehistory—glacially so. We saw in chapter 2that Oldowan tools, which date from 2.5 million years ago to about 1.4 million years ago, are opportunistic in nature. Toolmakers apparently were concerned mostly with producing sharp flakes without regard to shape. The so-called core tools, such as scrapers, choppers, and discoids, were by-products of this process. Even the implements in Acheulean tool assemblages, which followed the Oldowan and lasted until about 250,000 years ago, display imposition of form only minimally. The teardrop-shaped handaxe was probably produced according to some form of mental template, but most of the other items in the assemblage were Oldowanlike in many ways; moreover, only about a dozen tool forms were in the Acheulean kit. From about 250,000 years ago, archaic sapiens individuals, including Neanderthals, made tools from prepared flakes, and these assemblages, including the Mousterian, comprised perhaps sixty identifiable tool types. But the types remained unchanged for more than 200,000 years—a technological stasis that seems to deny the workings of the fully human mind.

  Only when
the Upper Paleolithic cultures burst onto the scene 35,000 years ago did innovation and arbitrary order become pervasive. Not only were new and finer tool types produced but the tool types that characterized Upper Paleolithic assemblages changed on a time scale of millennia rather than hundreds of millennia. Isaac interpreted this pattern of technological diversity and change as implying the gradual emergence of some form of spoken language. The Upper Paleolithic Revolution, he suggested, signaled a major punctuation in that evolutionary trajectory. Most archeologists agree generally with this interpretation, although there are differences of opinion over what degree of spoken language earlier toolmakers had—if any.

  Unlike Nicholas Toth, Thomas Wynn, of the University of Colorado, believes that Oldowan culture in its general features was apelike, not human. “Nowhere in this picture need we posit elements such as language,” he notes, in a jointly authored article in the journal Man, in 1989. The manufacture of these simple tools required little cognitive capacity, he argues, and therefore was not human in any way. Wynn does concede, however, that there is “something humanlike” in the making of Acheulean handaxes: “Artifacts such as these indicate that the shape of the final product was a concern of the knapper and that we can use this intention as a tiny window into the mind of Homo erectus.” Wynn describes the cognitive capacity of Homo erectus, based on the intellectual demands of Acheulean tool production, as equivalent to that of a seven-year-old modern human. Seven-year-old children have considerable linguistic skills, including reference and grammar, and are close to the point where they can converse without recourse to pointing and gesturing. In this connection it is interesting to recall that Jeffrey Laitman judged the language capacity of Homo erectus to be equivalent to that of a six-year-old modern human, based on the shape of the basicranium.