Masters of the Planet

by Ian Tattersall

  Daniel Povinelli, a distinguished researcher of chimpanzee cognition, proposed a few years ago that a fundamental distinction between the ways in which chimpanzees and humans view the world is that, while humans form abstract views about other individuals and their motivations, “chimpanzees rely strictly upon observable features of others to forge their social concepts. . . . [They] . . . do not realize that there is more to others than their movements, facial expressions, and habits of behavior. They [do] not understand that other beings are repositories of private, internal experience.” It also implies that individual chimpanzees do not have such awareness of themselves, either. They experience the emotions and intuitions that arise in their own minds; and they may act on them, or suppress them, as the social situation demands or permits. But, just as in Povinelli’s words they “do not reason about what others think, believe and feel . . . because they do not form such concepts in the first place,” it seems legitimate to conclude that this exclusion also applies to self-reflection. Because, if individual chimpanzees lack the ability to perceive that others have internal lives, it is highly probable that they also lack equivalent insight into their own interior existences.

  Profound as it is, this cognitive difference between us and them may not always produce radically distinctive observable behaviors; and indeed the ways in which chimpanzees and humans behave sometimes appear strikingly similar. Still, we should be wary of overstating these similarities. The behavioral resemblances we perceive are conditioned by an enormously long shared evolutionary history, and by the resulting structural similarities. But, as Povinelli would point out, similar observable behaviors may also hide mental processes that differ greatly in form and complexity.

  So, for all the manifold talents that chimpanzees possess, that cognitive gulf still yawns. Among all those organisms that we can study in the world today, it appears that only modern human beings show “spiritual self-awareness” in William James’ sense; and even his “social self-awareness” appears to differ dramatically in quality between humans and nonhuman primates. Still, even though a lot of evolutionary water has flowed under the bridge on both sides since human beings shared an ancestor with any ape, most authorities find it reasonable to conclude that cognition of the kind we see among chimpanzees (and among other great apes as well) provides us with a reasonable approximation of the cognitive state from which our ancestors started some seven million years ago. To return to Povinelli’s words, one may reasonably assume that those ancestors were “intelligent, thinking creatures who deftly attend[ed] to and learn[ed] about the regularities that unfold[ed] in the world around them. But . . . they [did] not reason about unobservable things: they [had] no ideas about the ‘mind,’ no notion of ‘causation.’” In the human sense, they had as yet no idea of self. This is a very plausible characterization of our lineage’s cognitive starting point; but at the same time it more or less exhausts what can usefully be said on this subject, based on our existing knowledge of comparative cognition.

  The next question is, of course, to which of our known ancestors does Povinelli’s characterization apply? In all probability, if we were able to directly observe the very early hominids we met in chapter 1 we would find that Povinelli’s depiction fit them well enough; and we have no compelling reason for believing that it would not also have applied very broadly to the earliest Australopithecus afarensis. Still, if it was indeed A. afarensis (or something very like it) that introduced stone tool making, as the Dikika (and Bouri) evidence seems to suggest, then we have to reconcile the Povinelli viewpoint with the significantly advanced cognitive achievements of those first stone tool makers. For there is no doubt that the first hominids who made stone tools and used them for butchering carcasses were displaying evidence of an entirely new and radical way of interacting with the world around them; and there is no reason to believe that this innovation might not have had internalized effects too. The simplest and most plausible way of explaining this apparent discrepancy is to suggest that the cognitive potential to make stone tools was born in the large genetic alteration that must have been involved in the acquisition of the new and radically different bipedal body form; and that this potential lay dormant for some time before being expressed in the invention of stone tool making.

  This scenario is not as far-fetched as it might seem if we remember that, as I’ve already suggested, innovations are not acquired directly as adaptations, but must instead arise as exaptations that must be co-opted post hoc. Examples of dramatic physical exaptations making possible huge later behavioral leaps are no rarity in evolution: birds, for example, possessed feathers for years before recruiting them for flight. And the ancestors of the tetrapods, the four-footed land-dwellers, initially acquired their limbs in an entirely marine setting.

  We will return to this subject, because the one we’ve just discussed is far from the last spectacular example of exaptation as an indispensable agent of innovation in hominid evolution. But meanwhile, knowing that doesn’t help us much in understanding potential changes in early hominids’ perception of themselves and the world around them. Because although the toolmakers were displaying an insight into the potential to modify that world in ways that would eventually have literally earth-shattering consequences, we have no idea whatsoever how this new ability affected, or was reflected in, other areas of their behavior and experience. All we can say is that they were behaving in entirely unprecedented ways: ways that underwrote all the manifold changes that were to come, but into which we cannot yet read evidence of any of the other attributes that make us feel so different from all other creatures.

  FOUR

  AUSTRALOPITH VARIETY

  It is reasonable for many reasons to start our discussion of the australopiths with Australopithecus afarensis. For a start, this is widely believed to be the “stem” species from which later australopiths were derived. And so far it is by a clear margin the best-known of all the early hominid species, so it acts as the perfect foil for discussing virtually all the issues of early hominid lifestyle that the australopiths raise. But it should never be forgotten that A. afarensis was only one species among many in the early hominid spectrum over the period between about 3.8 and 1.4 million years ago. This places hominids very comfortably within the pattern of diversification that we see in all similarly successful mammal groups. One of the best-documented evolutionary phenomena is “adaptive radiation”—the rapid multiplication of species that occurs when an organism enters a new “adaptive zone” in the way the early bipeds did, and the descendants of a single hardy adventurer diversify to exploit all of the new possibilities available to them. This has happened over and over again, and there are few better examples of the phenomenon than what occurred after the first hominids committed themselves to the ground. Chimpanzees may do quite well in woodland settings; but they do so by employing their existing attributes in slightly new ways, rather than by making the kind of radical change that the early hominids made. This in turn helps to explain why the range expansion of chimpanzees into savanna environments has had no effect on their diversity—the savanna chimpanzees remain members of the same species as their forest-dwelling relatives. Early hominids, on the other hand, made a physical rather than a simply behavioral response to life (at least part time) on the ground; and this opened up a host of new opportunities to them which they exploited to the full.

  The cranium of a gracile australopith, Australopithecus africanus, from the South African site of Sterkfontein. The specimen is known as Sts 71, and it is probably around 2.6 million years old. Drawing by Don McGranaghan.

  The very first australopith fossil ever discovered was found in a lime mine on the high veld of South Africa in 1924, and in the late 1930s others began turning up not too far away at similar sites. Quickly it was realized that at least two radically different kinds of small-brained early hominid were represented among these fossils: a lightly built form known as Australopithecus africanus (“southern ape of Africa”), and a relative wit
h a more massively constructed skull called Paranthropus robustus (“robust near-man”). Both had small cranial vaults containing brains that, at best, were only slightly larger than those of modern apes in comparison to their assumed body size. However, while A. africanus had boasted a set of teeth that were proportioned very much like those of A. afarensis, the teeth of Paranthropus were very different. The incisors at the front were small, and the canines behind them were likewise reduced. But the premolars and molars were broad and flat, together forming an impressively specialized grinding machine. The teeth were set in a face that was relatively short from front to back (because those front teeth didn’t take up very much space) but that was massively built both to accommodate the huge molars and to absorb the stresses they generated. So large were the muscles that moved the jaw that a tall ridge (the “sagittal crest”) ran backward along the center of the braincase to allow additional attachment room for large temporal muscles—just as we see among male gorillas today.

  For want of postcranial bones, nobody knew quite how big-bodied P. robustus individuals would have been, but quickly a dichotomy was established between the “robust” Paranthropus types and the “gracile,” or slender, A. africanus. Neither did anybody know exactly how old these early hominids were, although from the accompanying faunas it was guessed that the graciles were generally older than the robusts. Now we have a pretty good idea that the South African graciles (recently augmented by the finding at a site called Malapa of a fabulous trove of fossils representing a new and in many ways advanced species, A. sediba) bracket the period between about 3.0 and 2.0 million years ago or so, and that their robust counterparts date from around 2.0 to 1.7 million years ago.

  At one gracile site, Sterkfontein, a slightly older find deserves special mention, if only because of the extraordinary circumstances of its discovery. The majority of hominid fossils from Sterkfontein date from about 2.5 million years ago or so, but in an underground section of the site some cave deposits are exposed that may be over three million years old. Rummaging around in a collection of bones dynamited decades ago from these older cavern deposits, the paleoanthropologist Ron Clarke came across parts of an ankle joint and foot that his practiced eye immediately recognized as hominid. Noticing that the break in the shin fragment that provided the upper part of the ankle joint looked fresh, he asked his colleagues Stephen Motsumi and Nkwane Molefe to go into the vast gloomy cavern and look for the counterpart whitish cross-section—smaller than a dollar coin—embedded in the gray cave wall, truly a search for a needle in a haystack. But miraculously, those eagleeyed searchers promptly located the ring of bone Clarke had predicted would be there. They then began the long labor of extracting the rest of the skeleton (dubbed “Little Foot”) from the rock-hard matrix in which it had been entombed for the last three million years—a process that isn’t yet complete as of the time of publication. Enough of the skeleton has been exposed, however, to reveal that it isn’t quite like the A. africanus fossils from the later layers at Sterkfontein, and that it may represent a species ancestral to a second and as-yet-unnamed species that also comes from those later deposits. At the single site of Sterkfontein, therefore, we are seeing evidence of unexpected diversity among the gracile australopiths alone.

  Just looking at the teeth of the robust and gracile australopiths, you would immediately imagine that they were eating totally different things. The teeth of Australopithecus africanus look much more generalized, like those of an opportunistic fruit-eater that also availed itself of everything else in sight—just as you would expect from a relative of the ancestral bipeds. The teeth of the robusts, on the other hand, look like the specialized grinding apparatus of a dedicated feeder on hard and maybe gritty vegetal materials, such as the roots, tubers, and other items typical of open environments. But studies of the way those teeth actually wore as a result of chewing showed that things may not have been quite so simple. Examination of the dental wear surfaces of both graciles and robusts, under very high magnification, showed that there had in fact been a great deal of overlap in what all the hominids had consumed overall, and that any significant dietary differences were probably confined to times of the year when environmental productivity was low. At such times, the different kinds of hominid may have resorted to different “fallback” foods: hard and brittle in the case of the robusts, tough but more yielding in the case of the graciles.

  The notion that robusts and graciles in South Africa had largely similar diets also emerges from studies of stable carbon isotopes. These showed a lot of variability among samples, but yielded basically identical patterns for both kinds of australopiths—with a strong C 4 signal in each case. Researchers believe that the C 4 indications result largely from the consumption of animals such as hyraxes, cane rats, and young C 4-plant-consuming antelopes—though they do not rule out the possibility that some of it was due to consumption of grasses, most probably in the form of their rhizomes (underground runners). Interestingly, although it’s known that South African climates and environments fluctuated quite considerably during the tenure of the australopiths, the differing carbon-isotope ratios observed do not correlate with time. Both kinds of australopiths thus seem to have maintained their generalist dietary proclivities even as the habitat changed around them.

  All this strongly supports the idea that the success of all or most of the early hominids was due to an opportunistic strategy: they all ate everything available, rather than restricting themselves to particular types of resource. Only under stress might they have resorted to significantly different foods. It is this ubiquitous increase in dietary breadth that seems to have distinguished the australopiths most strongly from chimpanzees, which as we’ve seen maintain a strong preference for forest-type products even when they range into the savanna. The generalist proclivity that we see among australopiths also has implications for the evolutionary origins of the diversity we see among them: perhaps we should look at their sometimes huge anatomical differences as being the result of the fixation of chance novelties, rather than as the result of fine-tuning adaptation over the eons.

  Crude stone tools are known from deposits at Sterkfontein dating close to two million years ago; and similar implements up to about 1.8 million years old are also known from a nearby robust australopith site called Swartkrans. This latter has also yielded some polished pieces of bone that bear scratches identical to those you would make today if you tried to dig up roots and tubers using similar improvised tools. Both sites have additionally produced rare fossils that have been attributed to our own genus Homo, and that were assumed to have been the remains of the tool-makers. Still, the new evidence from Ethiopia makes it much easier to entertain the more obvious explanation that both gracile and robust australopiths were making and using stone and other tools at least as far back as two million years ago. This fits nicely with the interpretation of hand bones from Swartkrans that are almost certainly those of Paranthropus and that show anatomical features compatible with high manipulative abilities. In known postcranial features the South African graciles (except Malapa) look very much like Lucy (we have very little evidence for the robusts). All in all, the emerging picture from South Africa is looking similar to the one we are deriving from the later part of the tenure of Australopithecus afarensis in Ethiopia.

  EAST AFRICA

  South Africa was the first region of the world to provide evidence of very early hominids. But from the early 1960s on, eastern Africa grabbed the limelight. In 1959 the legendary Louis and Mary Leakey announced the discovery at Tanzania’s Olduvai Gorge of the fossil cranium of a “hyper-robust” australopith that they termed Zinjanthropus (for the old Arab-ruled “Zinj” empire along the East African coast), or more fondly “Nutcracker Man,” on account of the flat and hugely expanded chewing teeth that totally dwarfed its tiny incisors and canines. Nowadays this specimen is assigned to the species Paranthropus boisei (named for a benefactor of the Leakeys’ research), and is dated to 1.8 million years ago. The Leake
ys had been finding primitive stone tools at Olduvai for years and, as a devotee of the “Man the Toolmaker” notion, Louis was already convinced that these crude implements must have been the work of an early member of the genus Homo. Having spent some three decades periodically prowling the Gorge under the blistering African sun in search of the toolmaker, the Leakeys were elated to find any hominid at all; but they were naturally a bit disappointed when the one that eventually turned up was not something they had any hope of squeezing into the genus Homo.

  Cranium of Olduvai Hominid 5, a.k.a. “Zinjanthropus,” a robust australopith of the species Paranthropus boisei, from Olduvai Gorge, Tanzania. 1.8 million years old. Drawing by Don McGranaghan.

  Still, they did not have to wait long to find what in their terms was a much better candidate for the honor of being the long-sought Olduvai Toolmaker. For in 1961, at about the same geological level in the Gorge as the Paranthropus, Louis reported a lower jaw of a much more gracile hominid. Various people at the time noted striking similarities between the teeth of this specimen and those of Australopithecus africanus; but Leakey was undeterred in his search for early Homo, and a few years later he and some colleagues made the Olduvai jaw the holotype of Homo habilis, “handy man,” named of course for its presumed manual skills. Thus began the paleoanthropological tradition of routinely assigning East African early gracile hominids not to the genus Australopithecus, but to our own genus Homo—a tradition that was only broken a decade and a half later, when the much earlier and more robust Australopithecus afarensis was first announced from Hadar and Laetoli. Those fifteen intervening years were very eventful ones in paleoanthropology.