Back to Don's Maps

Early Hominids

Click on the photos to see an enlarged version

The major sites of discovery of Australopithecus africanus. Here in South Africa were unearthed the earliest known human fossils, about 600 000 years ago.
Photo: Lissner - Man, Myth and Magic

The major sites of discovery in Java.      Photo: Lissner - Man, Myth and Magic

Beijing man
Photo: Lissner - Man, Myth and Magic

Beijing man
Photo: New Scientist 17th April 2004

Beijing man graph
Photo: New Scientist 17th April 2004

Brute of Dragon Bone Hill

Far from being civilised, was Peking Man at heart a simple creature, shunning the cold and with only tenuous control over fire? A reconstruction by Noel T. Boaz and Russell L. Ciochon is forcing us to rethink the lifestyle of our famous ancestor

Anyone trying to imagine what life was like for our almost human ancestor, Homo erectus , will sooner or later be drawn to Dragon Bone Hill. The site, which lies 40 kilometres south-west of Beijing, was home to Peking Man. Here the fossilised bones of some fifty individuals have been found with the remains of thousands of other animals from many species. The picture researchers have built up is one of cave dwelling hominids who skillfully bagged an assortment of game, butchered it at home with simple stone tools, then cooked the meat over fires before eating it. They were reputedly the first people to master fire, using it for light and warmth in their ice age homes and to scare off predators, as well as for cooking.

This is a compelling reconstruction of the first hominid to possess many traits we consider to be human, and one of the most revered and tenaciously held hypotheses in palaeoanthropology. There is just one problem - our own research at Dragon Bone Hill has convinced us it is wrong.

After a decade investigating Peking Man, with colleagues Qinqi Xu and Jinyi Liu from the International Zhoukoudian Research Center in Beijing, we have concluded that these individuals were neither cave dwellers, nor hunters, but scavengers. In fact, they were much more akin to early African species of small-brained Homo than to our more recent ancestors or to the Neanderthals. Peking Man's forebears undoubtedly had what it took to migrate out of Africa and across Asia. Nonetheless, H. erectus was still in essence a tropical animal, shunning the cold and with a tenuous grasp on fire that probably terrified them almost as much as the other animals with whom they competed for food.

Dragon Bone Hill, or Longgushan, near the town of Zhoukoudian is a unique place, and now one of the UN's World Heritage sites. No other cave, with the possible exception of the recently discovered Atapuerca Cave in Spain, has yielded as many fossils of human ancestors. Palaeoanthropologists have wondered for decades why other sites like Zhoukoudian have not been discovered in the Far East. The answer seems to be that Dragon Bone Hill is a fortuitous accident: formed initially by river flow, the cave was then gradually uplifted by fault movements, while layers of sediment covered the bones within, protecting them from erosion. Its position, where the Western Hills, the Zhoukou river and the North China Plain meet, may also have made the site unusually popular with migrating ungulates and carnivores and hominids looking for shelter and water.

From the start, however, the hominids stole the show. Peking Man became an international celebrity in the late 1920s following the publication in Nature of a paper by Davidson Black, then professor of anatomy at Peking Union Medical College, describing two teeth discovered in the first quarrying operations at the site. Writing about another fossil tooth discovered shortly afterwards, Black coined a new genus and species name, Sinanthropus pekinensis or "Chinese man from Beijing" for its owner. Black predicted that the hominid would look like Piltdown Man, the fossil idol of the day. But when the first skull was unearthed in December 1929, it shocked everyone. The heavy brow ridges, thick-boned skull, and low, squat profile of the head were reminiscent of Java Man, a species discovered by legendary anthropologist Eugene Dubois in the 1890s who named it Pithecanthropus erectus . Piltdown of course was a fraudulent chimera, combining an artificially stained modern human skull and a broken orang-utan mandible with filed teeth. Today Peking Man and Java Man are both classified as H. erectus .

Black and other early researchers had little to go on when trying to estimate the age of their hominid. By comparing other fossilised bones from the site with similar animal remains from previously dated excavations they described the cave as broadly "middle Pleistocene". In recent years two new geochronological techniques have helped pin down its age. Palaeomagnetism can accurately date particular sedimentary layers that have been deposited at times when the Earth has undergone periodic reversals of the geomagnetic poles. There are 17 sedimentary layers at Zhoukoudian and layer 14 coincides with one such reversal 780,000 years ago dubbed the Brunhes-Matuyama boundary. This finding, together with results from a new dating technique called uranium-series isotope dating, which measures radioactive decay in calcium carbonate deposits such as flowstone, indicates that H. erectus occupied the cave intermittently from 670,000 to 410,000 years ago.

This makes Zhoukoudian a relatively late settlement for H. erectus in Asia. We know the species migrated out of Africa 1.9 to 2 million years ago, reaching Java at least 1.8 million years ago, according to recent re-dating of the most ancient remains there. This first great hominid expansion at the beginning of the Pleistocene marks a turning point in human evolution.

Climatic fluctuations during the Pleistocene, which may have been a spur for the move out of Africa (during times when Eurasia was warm and moist), could also have curtailed movement into northern parts of Asia when climate was colder. Levels of heavy oxygen-18 isotopes fall in sediments laid down during cold periods because they accumulate in glacier ice, leaving the lighter oxygen-16 isotopes temporarily more abundant in the environment. Variations in the ratios of oxygen-18 to oxygen-16 in rocks across China record fluctuations in the Pleistocene climate. Analysis of the sedimentary layers from Zhoukoudian suggests that hominids lived near the cave only when temperatures were balmy. We suspect this is no accident but indicates that these hominids were adapted to tropical climes and migrated south during the frigid periods of the Pleistocene.

But why would H. erectus need to leave Dragon Bone Hill when the cold weather set in? If they had fire and lived in a relatively cosy cave, could they not have hunkered down and stayed where they were during the glacial winter, just as the Neanderthals did later in Europe? Perhaps all was not what it seemed at Zhoukoudian.

Was the cave actually a hominid dwelling at all, we wondered? If so, you would expect the hominid fossils to be the most complete at the site and the most numerous. But that honour falls to a giant, lion-sized hyena, Pachycrocuta brevirostris . The remains of lions, sabretooths and wolves are also abundant, but hominid bones are fragmentary. And our detailed study of the fossilised remains of H. erectus , ungulates and other prey species shows a close similarity with modern bones gnawed by hyenas. So it looks as if the big carnivores, especially hyenas, were the number one residents. Hominids probably lived in more exposed places nearby, although stone tools discovered in the cave, cut-marks on bones and burnt horse skull bones indicate that they did occasionally visit of their own accord.

And what about the idea that Peking Man had mastered fire? Geologist Paul Goldberg's team at Boston University has found that what earlier researchers took to be charring on hominid remains are in fact microscopic layers of detritus, probably from decaying organic matter. The bones in question were found in a tightly circumscribed area of the cave, which has led us to believe that they are the remains of hominids cached in a stagnant pool by hyenas that apparently never came back to claim their prize.

Even the discovery of ash at Dragon Bone Hill cave is not conclusive proof that Peking Man regularly used fire. When geochemist Steve Weiner of the Weizmann Institute of Science, Israel, compared it with ash from Neanderthal caves, he found a crucial difference - no silica. Silica, which occurs in wood in structures known as phytoliths, survives burning. While the Neanderthals were obviously burning wood, the fires at Zhoukoudian seem to have been fuelled by dried guano, deposited on the cave floor by bats and owls that lived there. So, it looks as though Peking Man's grip on fire was tenuous - more like that of his ancestors in Africa, where patches of baked clay indicated single fires dating back 1.7 million years. Ephemeral fires may have allowed H. erectus to drive hyenas and other large carnivores from their kills, but this is far from the cosy hearth-side existence suggested in the traditional picture of life at Dragon Bone Hill.

Our re-analysis shatters the myth of a heroic early nimrod pitted against wild beasts and the elements, occasionally cannibalistic, managing to establish a cave home with primitive tools. Instead, it depicts a hungry scavenger skulking off to the cave, fire and sharp stones in hand, to steal away the half-chewed haunches of big carnivore kills, some of which by chance happen to be remains of its fellow hominids.

Further support for this idea comes from new research on the molecular evolution of tapeworms, by Eric Hoberg from the US Department of Agriculture, Maryland. According to Hoberg, the three species of tapeworms that infest humans are most closely related to species found in hyenas, large cats and dogs. Human tapeworms diverged from these relatives 780,000 to 1.7 million years ago. This suggests there must have been very close ecological relationships between hominids and their carnivorous contemporaries - much closer than would have arisen had they occasionally shared the same prey species. We take this as further evidence early humans habitually ingested meat scavenged from large carnivores.

Then there is the archaeological record. It speaks of the strangely static cultural life of H. erectus . Modern human culture changes so rapidly that almost anything we manufacture can be pinpointed to within a few years. Not so H. erectus culture, where things stayed the same for a million years in sites ranging across Africa and Eurasia. We believe this indicates these hominids were incapable of speaking as we do, a conclusion supported by the small size of the H. erectus brain, which at 750 to 1250 cubic centimetres, is just three-quarters the size of the modern human brain. It is only with the evolution of complex language that our ancestors would have been capable of the sorts of behaviours we think of as uniquely human.

But perhaps the biggest give-away that Peking Man was far less like us than has previously been suggested comes from the fossils themselves. The most distinctive anatomical difference between H. erectus and other hominids is the skull. The massively thick bony wall surrounding the brain - which has been likened to a tortoise carapace and a cycling helmet - has defied an adequate explanation, until now. Increasing brain size would have necessitated a larger skull, but not a thicker one. And the notion that a thick skull might have been required to support chewing muscles makes little sense since H. erectus has smaller teeth than earlier hominids. Searching for a plausible evolutionary explanation, we looked to other species and were struck by a similarity between a variety of thick-skulled animals, from the Cretaceous dinosaur Pachycephalosaurus to bighorn sheep. All these seem to have evolved thick bone for the same purpose - protection. Could this also explain the skull of H. erectus , we wondered? And if so, from what did these hominids need protection?

The answer, we believe, can be found on a number of skulls from Dragon Bone Hill. They show signs of trauma, and in particular the sorts of depression fractures that come from a sharp blow to the head. Our re-analysis of these fractures, originally identified by anatomist Franz Weidenreich on Zhoukoudian fossils in Beijing in the 1930s, have convinced us that, like some modern human populations, H. erectus regularly resorted to head-bashing to settle disputes.

This explanation is bound to disturb many people who want to believe that humanity has a basic adaptation for cooperation and sociality, but it makes more sense of the evidence than any other. Besides, we are not questioning fundamental human qualities, merely pointing out that the evolution of human behaviour may have been more complex than it initially seemed. If our work at Dragon Bone Hill has taught us anything, it is that even the most fervently held stories of human evolution sometimes need rewriting. H. erectus still has much to teach us about the evolution of human behaviour, and our work continues.

Them and us

The first Homo erectus started out in Africa about 2 million years ago, and spread to Eurasia shortly after that. So what happened to their descendants? There are two hotly contested theories. But our studies of Homo erectus have convinced us there is a third way.

These days, most palaeoanthropologists support the Out-of-Africa theory, which proposes that groups of hominids, or perhaps waves of immigrants, moved into Eurasia at discrete times, displacing resident populations and driving them to extinction. Our reading of the fossil record, so far as it goes, tends to agree with this scenario. So too does most mitochondrial DNA research. Mitochondrial genes from all human populations have been traced back to a common ancestral genotype with roots in Africa 100,000 to 200,000 years ago. But these findings have been widely taken to indicate total genetic replacement of pre-existing Eurasian hominid populations by Homo sapiens migrating out of Africa, a conclusion that we find unlikely. MtDNA is a rapidly evolving molecule and probably cannot measure evolutionary relatedness much beyond the level of the subspecies. Additionally, the studies refer to gene lineages, not necessarily lineages of organisms, and they cannot show that specific mtDNA genes were associated together in one woman who was ur-mother to us all. Nevertheless, this line of research has added a shadowy heroine to the narrative epic of human evolution: Mitochondrial Eve.

A less mainstream but equally tenaciously held view comes from the so-called multiregionalists. They maintain that resident Eurasian hominids were not all driven to extinction by the later waves of African incomers. Multiregionalists point to the anatomy of fossils to support interbreeding and genetic continuity from one time period to another within regions. But we believe the fossil record is too incomplete to support these claims. Instead, we argue, the evidence suggests relatively rapid regional evolutionary change at times, which is consistent with mass population movements. Nevertheless, multiregionalism is underpinned by sound population genetics and, we believe, is more reflective of the evolution of whole organisms than the abstract evolving gene lineages of the Eve theorists.

Our own theory starts from hominid ecology and population structure. H. erectus was a widespread species evolving in parallel with its large mammalian carnivorous contemporaries in many aspects of its adaptation, such as meaty diet and large home range. We know that during the Pleistocene, these carnivores - lions, bears, hyenas and dogs - dispersed throughout Eurasia as single widespread species. H. erectus may well have done the same. If so, groups within the species would have been territorial, but also interbred freely with neighbouring groups, spreading genes rapidly throughout the range. There would have been regional clusters of more closely related populations - subspecies, races or varieties - connected to adjacent populations by gene bridges called "clines".

Our "clinal replacement" model holds that H. erectus , with its carnivore-like adaptations, evolved across a broad front - genes rippling though clines as individuals found mates in neighbouring groups. Occasionally, a bigger immigrant wave might sweep over a region, replacing more than the average number of genes. But, it would not totally replace the original gene pool. Even in rare cases where immigrants displaced whole populations, some of the pre-existing genes would still remain in the region because the invaders would have shared genes with the population they replaced through their common ancestors. Ripples and waves of change could flow in different directions at different times, impelled by such vagaries as changing climate and inter-species competition.

Clinal replacement fits fossil and genetic evidence. It accounts for the broad, species-level continuity in anatomical change from H. erectus to more modern Homo , and can also explain discontinuities associated with a finer analysis of the fossils. Our theory provides a more plausible explanation of the population dynamics of evolving Pleistocene populations than does the total replacement notion central to Out of Africa. Yet, unlike multiregionalism, it can explain regional extinctions in human evolution and discontinuities in the Pleistocene fossil record.

Noel T. Boaz
Noel T. Boaz is professor of anatomy at Ross University School of Medicine, New Jersey

Russell L. Ciochon
Russell L. Ciochon is professor of anthropology at the University of Iowa. Their book, Dragon Bone Hill, an Ice-Age Saga of Homo Erectus is published this month in the UK by Oxford University Press

Back to Don's Maps