On the origin of races: Fossil skulls and ancient remains are providing important clues about the evolution of human races. Interpreting them, however, is fraught with difficulty and dogma

	Homo Erectus Skulls

World events of last year paint a grim picture of the human species' obsession with race and ethnicity. From the turbulent fragmentation of what was Yugoslavia to the rise of neo-Nazism in Germany, the basic human need to define me and mine as opposed to you and yours seemingly overpowered all others. Controversial questions concerning race and cultural identity have also come to haunt the quieter world of palaeontology. At what stage in human evolution did the modern races evolve? And can ethnic behaviours and customs act as a selective force in human evolution, helping to shape the physical characteristics of the five or so genetic subgroups of humanity that we call races? Recent attempts to find clues in fossil and skeletal remains have triggered some fierce academic skirmishes.

The main battle centres on the attempts of a small band of researchers to prove that human races are hundreds of thousands of years older than conventional theories would have us believe. Milford Wolpoff of the University of Michigan and his colleagues maintain that the principal human races-Negroids, Caucasoids, Mongoloids, Australian aboriginal peoples and southern African Bushmen-began to evolve well before the appearance of anatomically modern humans, Homo sapiens. Contrary to mainstream thinking, races did not evolve as a result of modern humans leaving Africa to colonise the rest of the world some 100 000 to 200 000 years ago. Or so Wolpoff argues.

Another palaeontological hotspot is Australia, where scientists have been examining skeletal remains for evidence of racial variations in skull thickness. Already the research has carried Peter Brown and Graham Knuckey of the University of New England, in Armadale, New South Wales, into politically dangerous waters: the links between biological make-up and ethnic behaviour, in this case between skull thickness and the traditional Aboriginal custom of settling disputes with a ritual contest involving blows to the head.

Both lines of research rest on tiny differences in the structures of ancient skulls. And both touch on sensitive issues, inasmuch as scientific studies of racial differences are invariably vulnerable to exploitation by racial propagandists.

Ironically, in everyday life most of the divisions we recognise are ethnic-that is, concern learned behaviour and cultural traditions rather than biological factors. Yet in science it is race that is often the flashpoint. This much is clear from the numerous attempts, past and present, to prove that races differ in genetic potential for intelligence. A fierce debate has recently been played out on the correspondence pages of Nature, for example, over claims that brain size varies from race to race. History shows that most such claims crumble when factors such as nutrition, environment and economics are taken into account.

To anthropologists and palaeontologists, the question of when in prehistory races began to evolve is no less controversial. Wolpoff, Alan Thorne of the Australian National University and their colleagues would trace racial characteristics as far back as 2 million years ago, to the extinct human species Homo erectus. According to their so-called multiregional hypothesis (see 'The case against Eve', New Scientist, 22 June 1991), anatomically modern humans evolved from this more ancient form simultaneously in different parts of the world, and it was during this period of simultaneous evolution that the racial characteristics of Homo sapiens first emerged.

As evidence, the multiregion-alists point to peculiar anatomical features that seem to characterise humans in a particular geographic region from the time of Homo erectus to the present. Palaeontologists sometimes call these features skeletal racial markers. Yet using these markers to determine the race of a skull of unknown origin is a difficult task. Researchers must resort to complex statistical and quantitative procedures: there are no all-or-nothing traits. And even the best racial markers often prove misleading.

For example, shovel-shaped incisors (front teeth rimmed with enamel on the back surfaces to make them resemble shovel blades) are unusually common in modern Mongoloid peoples. The snag is that the trait also occurs in African Homo erectus specimens from 1.5 million years ago. So finding it in a fossil does not necessarily mean that the specimen is an ancestor of modern Mongoloids. As Alan Walker of Johns Hopkins University Medical School notes, the feature is probably common to the ancestors of all modern humans. Similar arguments have been made over other features, including the shape of cheekbones and the presence and extent of a keel, or longitudinal ridge, on the top of the skulls.

Recently, Wolpoff and his colleagues have taken their ideas a stage further. They are now proposing nothing less than the complete abolition of Homo erectus on the grounds that the species is insufficiently distinct from Homo sapiens. All fossil specimens of Homo erectus and archaic Homo sapiens (including Neanderthals), they maintain, should be reclassified into a single species, Homo sapiens, that is subdivided only into races.

This would make modern races very ancient indeed and would greatly expand the physical and genetic variability of Homo sapiens. According to the proposed new definition of Homo sapiens, the species' members would include hominids with brains as small as 850 cubic centimetres (formerly Homo erectus) and modern humans with brains of up to 2000 cubic centimetres. Wolpoff and Thorne base their abolitionist argument on a study of the three species of the genus Homo: habilis, the earliest species; erectus, the second in line; and sapiens, the modern species. While Homo habilis proved to be morphologically distinct from the other two, the researchers could not find any anatomical markers that consistently separated Homo erectus from Homo sapiens.

At first sight, some newly discovered fossils from Yunxian in China also seem to support the proposed burial of Homo erectus. Last year, Li Tianyuan from the Hubei Archaeological Institute and Dennis Etler of the University of California at Berkeley discovered two fossil skulls with mixed anatomical features. The construction of the brain case and the base of the skull seem to resemble those of Homo erectus, but the face and brain sizes are closer to those of Homo sapiens. Could these skulls be evidence that the two hominid forms are in fact members of the same species? That Homo erectus should be laid to rest?

Unfortunately the skulls are so badly crushed that nobody is quite sure how to interpret them. Their age is particularly uncertain. Tianyuan and Elter believe that the fossils may be up to 350 000 years old, but no absolute dating techniques-for example, the radiocarbon or potassium-argon methods-can be applied in this case, because so little organic material was present. Instead, the researchers' estimate of antiquity is based on fossil animals found near the skulls, a notoriously imprecise method. Moreover, even if the Yunxian fossils are as old as Tianyuan and Elter think, another quan-dary emerges: the skulls do not show some of the racial markers that would be expected in Asian specimens if the multiregional hypothesis is correct.

Scholars' schism

Certainly, the combined forces of Wolpoff, Thorne and the Yunxian skulls have failed to persuade many scholars, not least Philip Rightmire, author of the most authoritative account of the species, The Evolution of Homo erectus. 'Homo erectus in the broad sense- including specimens from Java, China and Africa-is a polytypic species,' Rightmire says, referring to the sort of regional differentiation that Wolpoff and Thorne identify as racial. 'But there is no problem diagnosing this species relative to living people,' he adds, citing distinguishing features such as cranial size and large teeth.

Rightmire disagrees profoundly with Wolpoff and Thorne's vision of an unbroken, unbranching human evolutionary lineage stretching from the earliest Pleistocene to the present. 'Lumping distinctive populations like Neanderthals and other archaic humans, such as those represented by skulls from Broken Hill (in Zimbabwe) and Petralona (in Greece), together with Homo erectus and suggesting that there were no important extinctions across the entire Old World during this period is not going to help us to explore the patterns of evolutionary change that ultimately produced populations like us,' Rightmire concludes.

One reason the scope for argument is so huge is the uncertainty surrounding many crucial traits of fossil skulls. Skull thickness is a prime example. A commonly cited characteristic of Homo erectus is an unusually thick vault bone. But just how good a marker for Homo erectus is this trait? It was the quest for an answer to this question that led Brown, and his student Knuckey, to study the vaults of Australian aboriginal skulls. While many previous studies had focused exclusively on European skeletons, these researchers also studied skulls from cemetery collections of southern Chinese (Mongoloid) peoples, and from the remains of Australian aboriginals (Australoids) found in cemeteries between a few hundred and a few thousand years old.

Difficulties in identifying meaningful points for measuring vault thickness soon became apparent. The relative positions of various anatomical features (bony ridges, protuberances, sinuses, and grooves for blood vessels) on the internal surfaces of skulls varied considerably from individual to individual. This meant that vault measurements sometimes covered both the vault bone and a bony ridge. 'This remains a possible source of error,' Brown cautions.

Even so, geographic variation among modern humans was obvious. Australian Aboriginals of both sexes had considerably thicker vault bones than the other samples, and the European specimens had the thinnest vault bones. Vault thickness showed no correlation with stature, head size or cranial capacity and was only mildly related to skeletal robustness. So why do the Australian skulls have such thick vaults?

Brown has a subtle explanation. The Australian skulls originate from a large region of southeastern and central Australia in which the customary method of settling disputes is a ritual contest involving blows to the head. Traditionally, two men resolve conflicts by standing face to face and striking each other in turn with their digging sticks, aiming at the head, until one concedes defeat. Ducking is considered cowardly, but defending oneself from a blow by throwing up an arm is acceptable. Women conduct a similar ritual, with each kneeling alternately to receive the blow; a victor is declared when her opponent can no longer rise.

Once, the ritual may have been very common within this geographic region. Knuckey conducted a larger and more thorough survey of 409 Aboriginal skeletons from South Australia. As many as 57.4 per cent of all these skulls, male and female, bore depressed fractures, and healed fractures of the lower forearm, used to parry blows, were just as common. The presence of similar fractures on Aboriginal skeletons more than 11 000 years old shows that the practice is of ancient origin.

What was the evolutionary effect of this head bashing? Since those with thicker cranial vaults would survive ritual combat better than those with thin vaults, it could take relatively few generations to produce a population-wide thickening of the cranial vault. Then, as long as wooden digging sticks were used, this traditional method of settling disputes would work for the greater good of the Aboriginal population. Tensions between individuals could be dissipated without usually causing deaths or provoking entire communities to go to war or seek expensive reparations.

Deadly ritual

The ritual continues to this day. However, it has recently become such a social problem that some Aboriginal leaders are attempting to stop the tradition. As in white Australian society, interpersonal or domestic violence among Aborigines is not always brought to the attention of the police, making it difficult to reliably estimate the prevalence of the custom. What little information there is, however, suggests that the tendency to replace wooden digging sticks with harder and heftier implements has led to severe injuries and even deaths. The injuries may also be worse now than 10 000 years ago, says Brown, because genes from white, thin-skulled races have entered the Aboriginal gene pool. Not everyone raised as ethnically Aboriginal today may have inherited thicker vault bones.

Since the vaults of modern Homo sapiens had shown such variability, Brown decided to measure specimens of Homo erectus from China and Java according to the same procedures. The extinct species had a thicker cranial vault than either the European or southern Chinese samples, but in the frontal region where head blows are delivered, the Homo erectus skulls were just as thick as those of Australian Homo sapiens. In other areas of the cranium, the Homo erectus skulls were thicker than the Australian specimens.

Brown draws three cautionary lessons from his observations. First, thickness may arise from different causes. The thickening of the Australian skulls involves an increase both in the dense tabular bone that forms a 'skin' on the outside and inside of the skull and in the intervening, spongy layer known as the diploe. Lest ritual combat be invoked as an explanation for the thicker vault of Homo erectus, too, Brown is quick to point out that these skulls have normal amounts of tabular bone and larger amounts of diploic bone only.

Secondly, the prevalent Eurocentrism of previous studies has obscured some of the subtlety of the physical patterns of regional difference in both Homo erectus and Homo sapiens. Race cannot be ignored or swept under the carpet, says Brown: comparing Asian Homo erectus to Caucasoid Homo sapiens is inappropriate and misleading. Finally, and most tellingly, these studies remind us that behavioural flexibility is a fundamental human attribute, and one which occasionally may influence our physical attributes. Behaviour and biology form a tangled web but there is no need to become ensnared in it.