The Origin of Humankind

by Richard Leakey

  The first early human fossil in East Africa was found by Mary Leakey, in August 1959. After almost three decades of searching the sediments of Olduvai Gorge, she was rewarded with the sight of millstone molars, like those of the robust australopithecine species in South Africa. The Olduvai individual was, however, even more robust than its South African cousin. Louis Leakey, who, with Mary, had taken part in the long search, named it Zinjanthropus boisei: the generic name means “East African man” and boisei referred to Charles Boise, who supported my father and mother in their work at Olduvai Gorge and elsewhere. In the first application of modern geological dating to anthropology, Zinj, as the individual became known, was determined to have lived 1.75 million years ago. Zinj’s name was eventually changed to Australopithecus boisei, on the assumption that it was an East African version, or geographical variant, of Australopithecus robustus.

  FIGURE 2.1

  Australopithecine cousins. The principal difference between Australopithecus robustus (and boisei) and africanus is in the chewing mechanisms, which include the structure of the jaws, cheekbones, and associated sites for muscle attachments. The robustus species was adapted to a diet that contained tough plant foods, requiring heavy mastication. (Courtesy of A. Walker and R. E. F. Leakey/Scientific American, 1978, all rights reserved.)

  The names are not particularly important in themselves. What is important is that we are seeing several human species with the same fundamental adaptation, that of bipedalism, a small brain, and relatively large cheek teeth. This was what I saw in the cranium I found resting on a dry streambed on my first expedition to the eastern shore of Lake Turkana, in 1969.

  We know from the size of various bones of the skeleton that the males of the australopithecine species were much bigger than the females. They stood more than 5 feet tall, while their mates barely achieved 4 feet. The males must have weighed almost twice as much as the females, a difference of the sort that we see today in some species of savanna baboons. It’s a fair guess, therefore, that the social organization of australopithecines was similar to that of baboons, with dominant males competing for access to mature females, as noted in the previous chapter.

  The story of human prehistory became a little more complicated a year after the discovery of Zinj, when my older brother, Jonathan, found a piece of the cranium of another type of hominid, again at Olduvai Gorge. The relative thinness of the cranium indicated that this individual was of slighter build than any of the known australopithecine species. It had smaller cheek teeth and, most significant of all, its brain was almost 50 percent larger. My father concluded that although the australopithecines were part of human ancestry, this new specimen represented the lineage that eventually gave rise to modern humans. Amid an uproar of objections from his professional colleagues, he decided to name it Homo habilis, making it the first early member of the genus to be identified. (The name Homo habilis, which means “handy man,” was suggested to him by Raymond Dart, and it refers to the supposition that the species were toolmakers.)

  The uproar was based on esoteric considerations in many ways; it erupted in part because in order to assign the appellation Homo to the new fossil, Louis had to modify the accepted definition of the genus. Until that time, the standard definition, proposed by the British anthropologist Sir Arthur Keith, stated that the brain capacity of the genus Homo should equal or exceed 750 cubic centimeters, a figure that was intermediate between that of modern humans and apes; it had become known as the cerebral Rubicon. Despite the fact that the newly discovered fossil from Olduvai Gorge had a brain capacity of only 650 cubic centimeters, Louis judged it to be Homo because of its more humanlike (that is, less robust) cranium. He therefore proposed shifting the cerebral Rubicon to 600 cubic centimeters, thereby admitting the new Olduvai hominid to the genus Homo. This tactic surely raised the emotional level of the vigorous debate that ensued. Ultimately, however, the new definition was accepted. (It later developed that 650 cubic centimeters is rather small for the average adult brain size in Homo habilis; 800 cubic centimeters is a closer figure.)

  Scientific names aside, the important point here is that the pattern of evolution beginning to emerge from these findings was of two basic types of early human. One type had a small brain and large cheek teeth (the various australopithecine species); the second type had an enlarged brain and smaller cheek teeth (Homo) (see figure 2.2). Both types were bipedal apes, but something extraordinary had clearly happened in the evolution of Homo. We will explore this “something” more fully in the next chapter. In any case, anthropologists’ understanding of the shape of the family tree at this point in human history—that is, at around 2 million years ago—was rather simple.

  FIGURE 2.2

  Early Homo. This fossil, known by its museum acquisition number of 1470, was found in Kenya in 1972. It lived almost 2 million years ago and is the most complete early specimen of Homo habilis; it shows significant brain expansion and reduction in tooth size, compared with the australopithecines. (Courtesy of A. Walker and R. E. F. Leakey/Scientific American, 1978, all rights reserved.)

  The tree bore two main branches: the australopithecine species, all of which became extinct by 1 million years ago, and Homo, which eventually led to people like us.

  Biologists who have studied the fossil record know that when a new species evolves with a novel adaptation, there is often a burgeoning of descendant species over the next few million years expressing various themes on that initial adaptation—a burgeoning known as adaptive radiation. The Cambridge University anthropologist Robert Foley has calculated that if the evolutionary history of the bipedal apes followed the usual pattern of adaptive radiation, at least sixteen species should have existed between the group’s origin 7 million years ago and today. The shape of the family tree begins with a single trunk (the founding species), spreads out as new branches evolve through time, and then reduces in bushiness as species go extinct, leaving a single surviving branch—Homo sapiens. How does all this match up with what we know from the fossil record?

  For many years after the acceptance of Homo habilis, it was thought that 2 million years ago there were three australopithecine species and one species of Homo. We would expect the family tree to be heavily populated at this point in prehistory, so four coexisting species doesn’t sound like much. And in fact it has recently become apparent—through new discoveries and new thinking—that at least four australopithecines lived at that period, cheek by jowl with two or even three species of Homo. This picture is by no means settled, but if human species were like species of other large mammals (and there is no reason to think that they were not, at that point in our history), then such is what biologists would expect. The question is, What happened earlier than 2 million years ago? How many branches were there on the family tree, and what were they like?

  As noted, the fossil record quickly becomes sparse beyond 2 million years ago and blank further back much more than 4 million years ago. The earliest-known human fossils are all from East Africa. On the east side of Lake Turkana, we have found an arm bone, a wrist bone, jaw fragments, and teeth from around 4 million years ago; the American anthropologist Donald Johanson and his colleagues have recovered a leg bone of similar age from the Awash region of Ethiopia. These are slim pickings indeed upon which to re-create a picture of early human prehistory. There is, however, one exception to the sparse period, and that is a rich collection of fossils from the Hadar region of Ethiopia which are between 3 million and 3.9 million years old.

  In the mid-1970s, a joint French/American team, led by Maurice Taieb and Johanson, recovered hundreds of fascinating fossil bones, including a partial skeleton of one diminutive individual, who became known as Lucy (see figure 2.3). Lucy, who was a mature adult when she died, stood barely 3 feet tall and was extremely apelike in build, with long arms and short legs. Other fossils of individuals from the area indicated that not only were many of them bigger than Lucy, standing more than 5 feet tall, but also that they were more apelike in
certain respects—the size and shape of the teeth, the protrusion of the jaw—than the hominids that lived in South and East Africa a million years or so later. This is just what we would expect to find as we moved closer and closer to the time of human origin.

  When I first saw the Hadar fossils, it seemed to me that they represented two species, perhaps even more. I considered it likely that the diversity of species we see at 2 million years ago derived from a similar diversity a million years earlier, including species of Australopithecus and Homo. In their initial interpretation of the fossils, Taieb and Johanson supported this pattern of our evolution. However, Johanson and Tim White, of the University of California, Berkeley, conducted further analyses. In a paper published in the journal Science in January 1979, they suggested that the Hadar fossils did not represent several species of primitive human but instead were the bones of just one species, which Johanson named Australopithecus afarensis. The large range of body sizes, which earlier had been taken to indicate the presence of several species, was now accounted for simply as sexual dimorphism. All the known hominid species that arose later were descendants of this single species, they said. Many of my colleagues were surprised by this bold declaration, and it provoked strong debate for many years (see figure 2.4).

  FIGURE 2.3 (RIGHT)

  Lucy. This partial skeleton, known popularly as Lucy, was found in 1974 by Maurice Taieb and Donald Johanson and their colleagues, in Ethiopia. A female, Lucy stood at close to 3 feet tall. Males of her species were considerably taller. She lived a little more than 3 million years ago. (Courtesy of the Cleveland Museum of Natural History.)

  Although many anthropologists have since decided that Johanson and White’s scheme is probably correct, I believe that the scheme is wrong, for two reasons. First, the size difference and anatomical variation in the Hadar fossils as a whole is simply too great to represent a single species. Much more reasonable is the notion that the bones came from two species, or perhaps more. Yves Coppens, who was a member of the team that recovered the Hadar fossils, also holds this view. Second, the scheme makes no biological sense. If humans originated 7 million years ago, or even only 5 million years ago, it would be highly unusual for a single species at 3 million years ago to be the ancestor of all later species. This would not be the typical shape of an adaptive radiation, and unless there is good reason to suspect otherwise we must consider human history to have followed the typical pattern.

  The only way this issue will be resolved to everyone’s satisfaction is through the discovery and analysis of more fossils older than 3 million years, which seemed possible early in 1994. After a decade and a half of being unable, for political reasons, to return to the fossil-rich sites in the Hadar region, Johanson and his colleagues have made three expeditions since 1990. Their efforts have met with great success, being rewarded with the recovery of fifty-three fossil specimens, including the first complete cranium. The pattern seen previously from this time period—that of a great range of body sizes—is confirmed and even extended by the new finds. How is this fact to be interpreted? Is the issue of one species or more at the brink of resolution?

  Unfortunately it is not. Those who considered that the size range of the previously discovered fossils indicated a difference in stature between males and females viewed the new ones as supporting that position. Those of us who suspected that so broad a size range must indicate a difference between species, not a within-species difference, interpreted the new fossils as strengthening that view. The shape of the family tree earlier than 2 million years ago must therefore be regarded as an unresolved question.

  The discovery of the Lucy partial skeleton in 1974 seemed to offer the first glimpse of the degree of anatomical adaptation to bipedal locomotion in an early hominid. By definition, the first hominid species to have evolved, some 7 million years ago, would have been a bipedal ape of sorts. But until the Lucy skeleton came along, anthropologists had no tangible evidence of bipedalism in a human species older than about 2 million years. The bones of the pelvis, legs, and feet in Lucy’s skeleton were vital clues to this question.

  FIGURE 2.4

  Family trees. The existing fossil evidence is interpreted differently by different scholars, although the overall shape of the inferred evolutionary history is similar. Two versions are presented here, somewhat simplified. My preference is for B, in which specimens of the genus Homo are among the earliest known fossils; this would be ancestral to what we know as Homo habilis. The fossil record does not extend back as far as the origin of the human family—some 7 million years ago, as inferred from molecular genetic evidence.

  From the shape of the pelvis and the angle between the thighbone and knee, it is clear that Lucy and her fellows were adapted to some form of upright walking. These anatomical features were much more humanlike than apelike. In fact, Owen Lovejoy, who performed the initial anatomical studies on these bones, concluded that the species’ bipedal locomotion would have been indistinguishable from the way you and I walk. Not everyone agreed, however. For instance, in a major scientific paper in 1983 Jack Stern and Randall Susman, two anatomists at the State University of New York, Stony Brook, offered a different interpretation of Lucy’s anatomy: “It possesses a combination of traits entirely appropriate for an animal that had traveled well down the road toward full-time bipedality, but which retained structural features that enabled it to use the trees efficiently for feeding, sleeping or escape.”

  One of the crucial pieces of evidence that Stern and Susman adduced in favor of their conclusion was the structure of Lucy’s feet: the bones are somewhat curved, as is seen in apes but not in humans—an arrangement that would facilitate tree climbing. Lovejoy discounts this view and suggests that the curved foot bones are a mere evolutionary vestige of Lucy’s apelike past. These two opposing camps enthusiastically maintained their differences of opinion for more than a decade. Then, early in 1994, new evidence, including some from a most unexpected source, seemed to tip the balance.

  First, Johanson and his colleagues reported the discovery of two 3-million-year-old arm bones, an ulna and a humerus, that they attribute to Australopithecus afarensis. The individual had obviously been powerful, and its arm bones had some features similar to those seen in chimpanzees while others were different. Commenting on the discovery, Leslie Aiello, an anthropologist at University College, London, wrote in the journal Nature: “The mosaic morphology of the A. afarensis ulna, together with the heavily muscled and robust humerus, would be ideally suited to a creature which climbed in the trees but also walked on two legs when on the ground.” This description, which I support, clearly fits closely with the Susman camp and not the Love joy camp.

  Even stronger support for this view comes from the innovative use of computerized axial tomography (CAT scanning) to discern the details of the inner ear anatomy of these early humans. Part of the anatomy of the inner ear are three C-shaped tubes, the semicircular canals. Arranged mutually perpendicular to each other, with two of the canals oriented vertically, the structure plays a key role in the maintenance of body balance. At a meeting of anthropologists in April 1994, Fred Spoor, of the University of Liverpool, described the semicircular canals in humans and apes. The two vertical canals are significantly enlarged in humans compared with those in apes, a difference Spoor interprets as an adaptation to the extra demands of upright balance in a bipedal species. What of early human species?

  Spoor’s observations are truly startling. In all species of the genus Homo, the inner ear structure is indistinguishable from that of modern humans. Similarly, in all species of Australopithecus, the semicircular canals look like those of apes. Does this mean that the australopithecines moved about as apes do—that is, quadrupedally? The structure of the pelvis and lower limbs speaks against this conclusion. So does a remarkable discovery my mother made in 1976: a trail of very humanlike footprints made in a layer of volcanic ash some 3.75 million years ago. Nevertheless, if the structure of the inner ear is at all indicative of h
abitual posture and mode of locomotion, it suggests that the australopithecines were not just like you and me, as Love joy suggested and continues to suggest.

  In promoting his interpretation, Lovejoy seems to want to make hominids fully human from the beginning, a tendency among anthropologists that I discussed earlier in this chapter. But I see no problem with imagining that an ancestor of ours exhibited apelike behavior and that trees were important in their lives. We are bipedal apes, and it should not be surprising to see that fact reflected in the way our ancestors lived.

  At this point, I will switch from bones to stones, the most tangible evidence of our ancestors’ behavior. Chimpanzees are adept tool users, and use sticks to harvest termites, leaves as sponges, and stones to crack nuts. But—so far, at any rate—no chimpanzee in the wild has ever been seen to manufacture a stone tool. Humans began producing sharp-edged tools 2.5 million years ago by hitting two stones together, thus beginning a trail of technological activity that highlights human prehistory.

  The earliest tools were small flakes, made by striking one stone—usually a lava cobble—with another. The flakes measured about an inch long and were surprisingly sharp. Although simple in appearance, they were put to a variety of uses. We know this because Lawrence Keeley, of the University of Illinois, and Nicholas Toth, of Indiana University, microscopically analyzed a dozen such flakes from a 1.5-million-year-old campsite east of Lake Turkana, looking for signs of use. They found different kinds of abrasions on the flakes—marks indicating that some had been used to cut meat, some to cut wood, and others to cut soft plant material, like grass. When we find a scattering of stone flakes at such an archeological site, we have to be inventive to imagine the complexity of life that took place there, because the relics themselves are sparse: gone is the meat, the wood, and the grass. We can imagine a simple riverbank campsite, where a human family group butchered meat in the shade of a structure made from saplings and thatched with reeds, even though all we see today are the stone flakes.