Masters of the Planet

by Ian Tattersall

  A crude plot of average hominid brain sizes against time. After an initial flatlining, this plot appears to indicate a consistent enlargement of the hominid brain over the last two million years. But it is important to bear in mind that these brain volumes are averaged across an uncertain number of different lineages within the genus Homo, and that it is likely that what the plot actually reflects is the preferential success over this period of larger-brained hominid species, rather than steady increase within a single lineage. Illustration by Gisselle Garcia.

  A perfect example of the conflicted mindset that results from the tension between evidence and tradition came a few years ago, when a team working in the Turkana Basin described two new hominid fossils from the Ileret area to the east of the lake. One was a lightly built braincase (691 cc) some 1.55 million years old, that they allocated to Homo erectus even though it bears none of the morphological hallmarks of the Javan type specimen. The other was a piece of upper jaw, perhaps a hundred thousand years younger, that they assigned to Homo habilis. The researchers hailed these finds as evidence that at least two distinct lineages of the genus Homo had occupied the Turkana Basin at about the same time, emphasizing diversity among the hominids of the period. Yet their allocation of the braincase to Homo erectus could only have borne any conceivable logic in the context of the view that Homo erectus is the middle grade of a single, worldwide, variable, and gradually evolving hominid lineage—precisely the construct that they ostensibly wished to undermine.

  Of course, with modern humans presenting us with brains ranging from roughly 1,000 cc to 2,000 cc in volume, we can hardly use past brain size variation alone to reject the notion of a single variable human lineage that consistently gained in brain volume with time. But the huge variation we see in the morphology of the skulls those brains were contained in is at least highly suggestive. And if multiple hominid species were out there in the past—species whose brain size ranges and geological lifespans are, regrettably, unknown to us—then it is just as likely that the trend we see toward increasing brain size over the last two million years is due to the greater competitive success in the ecological arena of larger-brained hominid species, as that it is due to the reproductive success of larger-brained and thus smarter individuals.

  A scenario involving the consistent triumph of larger-brained species might be taken to suggest that the pressures favoring hominid brain expansion over time were essentially ecological, and thus external to the species themselves. Nonetheless, there is one important observation that suggests that members of the genus Homo have been consistently predisposed in some way toward brain size increase: brain enlargement has occurred independently in at least three lineages within the genus. The earliest Homo erectus in Java, dating from perhaps more than 1.5 million to a bit under one million years ago, have brains ranging from around 800 cc to a bit over 1,000 cc in volume. A later Javan group, poorly dated but maybe about a quarter of a million years old, comes in at 917 cc to 1035 cc; and the latest Javan Homo erectus group of all, perhaps no more than 40 thousand years old, varies from 1,013 cc to 1,251 cc. Similarly, Homo sapiens and Homo neanderthalensis diverged from a smaller-brained common ancestor well over half a million years ago, and independently gained their comparably sized large brains. Thus, a 600-thousand-year-old collection of Spanish fossils foreshadowing the Neanderthals had brains some 1,125 cc to 1,390 cc in volume, against the later Neanderthal average of 1,487 cc.

  Given that Homo erectus lived in tropical eastern Asia, the Neanderthals in Ice Age Europe, and the precursors of Homo sapiens in Africa, it is hard to see a common environmental thread in the trend toward bigger brains that all three lineages independently followed. Somehow, very early on in its evolution, the genus Homo must have acquired some underlying predisposition, biological or cultural, toward brain enlargement. Identifying that factor will be essential if we are ever to have a full account of how we became the extraordinary cognitive entity we are— even though, as we will see, while a large brain is clearly a necessary condition for our unique modern cognitive style, it is not a sufficient one.

  Still, predispositions notwithstanding, there is nothing inevitable about the enlarging brain in Homo. We were all forcibly reminded of this by the recent discovery of the remarkable “Hobbit” at Liang Bua Cave on the Indonesian island of Flores. The best specimen of this extraordinary hominid, technically known as Homo floresiensis, is the skeleton of a tiny individual, dubbed LB1. In life LB1 had stood not much more than three feet tall and, though bipedal, had possessed rather unusual body proportions. LB1 also has a head with a tiny vault that had contained a brain of maybe as little as 380 cc in volume—fractionally smaller even than Lucy’s, the smallest australopith brain known. What’s more, perhaps most bizarrely of all, the individual had lived only about 18 thousand years ago.

  Predictably, the announcement of the entirely unexpected Hobbit was greeted by enormous controversy. The scientists who described LB1 thought it might be a dwarfed descendant of a population of Homo erectus that had somehow contrived to reach Flores in the remote past. This is not in itself implausible: “island dwarfing” of mammals and reptiles is not unusual on smallish isolated landmasses such as Flores—and indeed, bones of a tiny elephant are found in the same cave deposits that yielded LB1. But there is little about its anatomy to suggest any close affinity to Homo erectus, and LB1’s brain is much smaller than you would expect to find as a result of normal processes of dwarfing, even from the typically modest-brained Homo erectus. Several different authorities have suggested, alternatively, that LB1 is simply the skeleton of a pathological modern human; but none of the suggested disease conditions fits the case well enough. As more is learned about this specimen, the more likely it appears that LB1 and its kind will in the end prove to be descendants of extremely early émigrés from Africa, preserving archaic features that may well eventually help us learn just what those émigrés were like. Meanwhile, independent of whether or not some degree of island dwarfism is indeed involved in this particular case, LB1 tells us that, larger patterns apart, time and brain enlargement need not necessarily be synonymous among members of the genus Homo—if, that is, it is appropriately assigned to our genus at all.

  EIGHT

  THE FIRST COSMOPOLITAN HOMINID

  The systematic picture among fossil hominids of the period around a million years ago remains rather unclear, because relevant fossils in the African center of innovation are few and far between and widely scattered. But it clarifies considerably about 600 thousand years ago, when we get the first indications of what was the world’s first Old World–wide hominid species, Homo heidelbergensis. This species is based on a mandible that was discovered back in 1908 in a gravel pit at Mauer, near the German city of Heidelberg, but that was only recently dated to 609 thousand years ago. On its own the Mauer jaw was a bit puzzling, but fortunately it matches well with jaws present in a fossil sample from Arago, a cave in the French Pyrenees that also preserves a face and associated vault bones dating to around 400 thousand years ago. Thus armed with much of a skull, we can be confident in also assigning to Homo heidelbergensis, or to something very much like it, a 600-thousand-year-old partial cranium from Bodo, in Ethiopia, and crania from Kabwe in Zambia, Petralona in Greece, and Dali and Jinniushan in China, plus other less complete specimens from Africa and elsewhere. Dating is rather poor for most of these fossils, but all seem to fall in a broad range between about 500 thousand and a bit less than 200 thousand years ago. Equally unfortunately, we can’t say much about the body structure of Homo heidelbergensis, since bones of the body skeleton are few and far between (except in the Chinese Jinniushan specimen, details of which are not yet fully available to the scientific community). Still, what we do know suggests a build along the basic modern body plan, if very robust and differing from us in various details that foreshadow those of the Neanderthal skeleton, which we’ll learn more about shortly.

  Given what we currently know, it seems most likely that Homo heide
lbergensis arose in Africa and then spread out of that continent, just as the first hominid émigrés had done before it. The details of its origin continue to elude us among the welter of hominid morphologies so tantalizingly suggested in the post-one-million-year record; but what is uncontestable is that, with the arrival of Homo heidelbergensis, we are entering new adaptive territory. In its cranial morphology this new species once more anticipates the future at least as much as it echoes the past, with robust but flatter faces with shorter tooth-rows than its predecessors have, tall eminences above its eyes, and capacious cranial vaults ranging in volume from 1,166 cc to 1,325 cc. Here we are comfortably within the modern brain size range, if still somewhat below today’s average.

  The partial cranium of Homo heidelbergensis from Bodo in Ethiopia. At 600 thousand years old this is one of the oldest known representatives of its species. Drawing by Don McGranaghan.

  All of the brain endocasts of this species that have so far been described are said to show expanded Broca’s areas; but beyond this the paleoneurologists who study them have been disappointingly mute, although they have generally been more impressed by the endocasts’ similarities to modern brains than by the differences they see. Still, while the endocasts are said to show cerebral asymmetries of the kind one finds between the left and right halves of the brain in modern people, the prefrontal areas of the cortex that lie directly above our eyes (though in Homo heidelbergensis they lie variably above and behind them) are usually characterized as broad and flat compared to those of Homo sapiens—exactly as you’d expect from looking at the skull that enclosed them.

  In modern humans the prefrontal cortex is vital to our complex cognition, governing such key areas of mental activity as decision making, the manifestation of social behaviors, and the expression of personality traits; and it seems reasonable to conclude that its role was broadly similar in Homo heidelbergensis. But exactly what the difference in external appearance between the prefrontal regions of the two species means in terms of their exact function is not at all clear, especially since we have no idea how this brain area was internally organized and connected to adjacent structures in Homo heidelbergensis. Equally, we really can’t say just what kind of mental edge the marginally greater brain size of Homo sapiens might by itself give us over the cognitive condition of Homo heidelbergensis. All of which means that, though we can legitimately conclude on the basis of brain size alone that Homo heidelbergensis was somehow “smarter” than its predecessors, there is no way to determine the details of this increase in intelligence by any form of direct observation. Once more, we have to turn to the indirect proxy indicators furnished us by the archaeological record.

  The oldest fossil specimens we have of Homo heidelbergensis either lack any archaeological context (Mauer) or have pretty archaic associations (Bodo). Indeed, though stone tools are quite common in the Ethiopian deposits from which the Bodo skull was recovered, they are of Oldowan type, and handaxes are conspicuously absent—a million years after their invention, and despite their occurrence in underlying (older) layers. Once again, we see no correlation between the appearance of a new kind of hominid and technological innovation. Interestingly, though, the Bodo specimen bears cut-marks made when the bone was fresh, as if the skull had been deliberately defleshed. What this means is uncertain, beyond the clear inference of hominid intervention. Cannibalism is unlikely, because the marks appear on the face and forehead, areas that would not have yielded much that was edible; but equally we should be careful not to read into these marks evidence of ritual in the sense in which it is familiar to us today.

  Fortunately, a number of archaeological sites in Europe fall within the timespan of Homo heidelbergensis, and help us flesh out the story of what hominids were up to at the time. One particularly interesting hominid occupation locality is at Terra Amata, in a suburb of Nice on the French Mediterranean coast. Some 380 thousand years ago the terrace on which Terra Amata lies was an ancient beach, to which a small group of hunters repeatedly returned (alas, without leaving any evidence of themselves). There they built a number of large shelters, as indicated by oval rings of large stones that had served to anchor rows of saplings that were embedded in the ground and brought together at the top. Whether or not these shelters were converted into true huts by covering them with hides is unknown, but seems likely. A break in the ring of stones defining the best-preserved of these structures indicates not only its entrance, but also a place through which smoke would have escaped from a fire built directly inside. This fire burned in a shallow scooped-out hearth, in which archaeologists found both blackened cobbles and the bones of animals whose meat had presumably been cooked. This hearth at Terra Amata is perhaps the oldest firm evidence we have for the domestication of fire following that extraordinary 800-thousand-year-old outlier in Israel, and it announces the advent of fire as a regular part of the documented human behavioral repertoire. For, from Terra Amata times on, hearths become an increasingly common feature of hominid occupation sites.

  Building structures and routinely using fire represent significant steps in the direction of modern behavioral patterns; yet at the same time, the numerous stone tools found at Terra Amata are remarkably unsophisticated. There are no bifacially shaped handaxes, and most of the tools are simply produced flakes. The crudeness of the tools may be accounted for by the fact that the local rock (cobbles of silicified limestone) did not provide good raw material for cutting tools, though it may be significant that the Terra Amata hominids had carried in some pieces of red and yellow pigment from distant sources, evidently attracted by their aesthetic qualities.

  Reconstruction of the earliest documented shelter, some 350 thousand years old, from Terra Amata in France. A shallow hearth has been scooped out just within the entrance indicated by a break in the ring of reinforcing stones. Concept by Henry de Lumley; drawing by Diana Salles.

  An equally unusual and interesting site is the locality of Schoeningen, lying in an anaerobic peat bog in northern Germany. This unusual environment has preserved the earliest evidence we have of wooden tools, as well as the remains of ten butchered horses and numerous other mammals—though sadly (for us), not those of the hominids who had hunted them. Dead wood is a fragile material that rarely lasts beyond a few decades, or centuries at best; and although it must have been used as a resource by hominids back into the deep recesses of time (australopiths very likely used digging sticks), the preservation of several long and carefully-crafted spears some 400 thousand years old is little short of miraculous. Indeed, before this find the earliest wooden implement was a yew spear tip, a mere 125 thousand years old, that had been found at the German site of Lehringen amid the ribs of a straight-tusked elephant that it had presumably been used to hunt.

  In a cool-temperate setting such as that at Schoeningen when the deposits there were formed, there is a fairly restricted variety of naturally occurring plant resources suitable for hominids to exploit. As a result, one might have suspected that a substantial meat component in the diet was necessary to sustain hominids at this latitude. But the sophistication of the spears nevertheless came as a surprise to archaeologists, many of whom believed in the mid-1990s, when the find was made, that any spears used by hominids that far back would have been of the handheld thrusting kind. Yet the Schoeningen spears clearly resembled throwing implements, six-plus feet long and shaped like Olympic javelins, with their weight concentrated toward the front. Made from spruce, their tips were finely shaped to a sharp point; and although it has been objected that a wooden point would likely have bounced off the thick hide of a large mammal unless the projectile had been thrown from very close range (which would have more or less eliminated its advantage relative to a manual thrusting spear), it is claimed that these spears were nonetheless designed to be thrown, implying the advent of sophisticated ambush hunting. And whatever the hominids’ exact technique, the ample remains of a range of butchered large mammals testifies to the hunting efficacy of the Schoeningen hominids.
/>
  The Schoeningen site yielded another “first” as well, and probably a more significant one. A major innovation in the technological history of hominids was the invention of the “composite” tool, made from more than one component. This can lead to a huge increase in efficiency, as anyone who has ever tried using a hammer without a handle will know. At Schoeningen, in addition to a variety of flint flakes, three worked sections of fir branch were found, several inches to a foot long, each of them notched at one end. It is believed that these branch sections had formed the handles of tools that were tipped with flint flakes glued or bound into the notches. There are no traces of potential glue on any of the Schoeningen stone tools; but we know from a bit later in time that the Neanderthals, for instance, used natural resins in this role.

  If the Schoeningen wood segments were indeed the handles of composite tools, then this improvement preceded the next major advance in stoneworking itself, at least in Europe, by 100 thousand years. Sometime between about 300 and 200 thousand years ago, “prepared-core” implements were introduced, representing a radically new approach to the shaping of stone: one that required a high quality, predictably fracturing stone such as flint or chert. The toolmaker elaborately shaped a stone core on both sides, using numerous carefully positioned strikes and usually employing a “soft” hammer of bone or antler, until a final blow would detach a finished tool. This would have a more or less continuous cutting edge nearly all the way around it. The core could be discarded, or another flake could be knocked from it; and the flake(s) could then, if desired, be modified into a specific shape, perhaps for scraping or cutting.