The Incredible Human Journey

by Alice Roberts

  ‘I honestly thought all the specimens had been lost,’ I said.

  ‘After the war, in the 1950s, we carried out some new excavations, and we got some new specimens,’ explained Wu.

  So I was holding in my hands fragments of a skull of an earlier human who had lived in China some one million years ago. It was quite a strange moment: there was something about the physicality of holding something I knew to be truly ancient. The vast depths of time through which these fossilised bones had passed, and then to end up in the present as some sort of magical talisman that could give us the power to know the past, made me feel almost giddy. A great range of dating techniques has been used to pin down the dates of the layers at Zhoukoudian. Most suggest that the hominin fossils come from layers dating to between 400,000 and 250,000 years ago. But the latest uranium series dates suggest that the fossils may be more than 400,000 years old, and perhaps as much as 800,000 years.5, 6 I laid the fragments back down on the red cloth. This was Chinese erectus, the real thing, with its massive brow and sloping forehead.

  Having got over the shock of seeing Peking Man, I turned my attention to the reconstructed casts – and the face of Chinese Homo erectus. The Weidenreich/Swan and Tattersall/Sawyer reconstructions were quite different looking. Professor Ian Tattersall had wanted to make a new reconstruction using more fragments from the face than had been used in the Weidenreich model. This had been possible, because, although the majority of original fossil fragments had been lost, casts had survived – and there were fragments of twelve separate skulls to work from. The original reconstruction had been based on just three pieces of skull: the skullcap, part of the right side of the mandible and a fragment of the left maxilla (the bone that bears the upper teeth and forms much of the cheek), and photographs and casts of this reconstruction didn’t make it obvious which parts were real and which were ‘restored’. Tattersall and Sawyer had wanted to keep artistic licence to a minimum, by using as many fragments of facial bones as possible in their reconstruction.3

  Professor Wu held up the Weidenreich/ Swan reconstruction and started to point out the features that he thought to be evidence for regional continuity. He indicated the flatness of the bridge of the nose, and the shape of the edges of the cheekbones. Then he pointed to these features on his own face and mine: it was true that he had a much less pronounced bridge of the nose than I had, and much wider, flatter cheekbones. But I couldn’t really see any similarity between Professor Wu’s face and that of Peking Man, even though it seemed that the Weidenreich/Swan reconstruction had somehow accentuated these Chinese-looking features. On the Tattersall/Sawyer skull, the nasal bones were more prominent, the face taller and narrower across the cheekbones, and the jaw more protruding. In general, this reconstruction looked less ‘Chinese’ and more like other erectus specimens around the world.3

  The shape of the front teeth was another feature that Wu considered as evidence of regional continuity. Weidenreich had proposed that shovel-shaped incisors were a regional trait shared by Chinese Homo erectus and many modern Chinese. But this tooth shape is also seen in African erectus and in Neanderthals, so, while it may be an archaic trait, it is not specifically East Asian. And the shovel shape seen in the teeth of some living Chinese people seems to be developmentally different from the shovelling of the archaic teeth, even though it looks superficially similar – in which case it can’t be used to argue for a connection at all.7

  Then we looked at the modern human specimens, including casts of two of the Upper Cave skulls. Their dates haven’t really been pinned down, and range between 10,000 and 30,000 years old, based on radiocarbon dating of animal bones in the same layer. Some archaeologists argue that the Upper Cave skulls might be burials that have been moved around by animals, and so seem older than they should be – preferring the lower estimate of around 10,000 years old.

  ‘There are many common features among them,’ said Wu, indicating the Upper Cave skulls and the Peking Man reconstruction. ‘I think it is most probable that the Upper Cave man are the descendants of Homo erectus man.’

  To me, the Upper Cave skulls looked undeniably modern, but they didn’t really look East Asian. Neither had the characteristic flattened nasal bones or tilted-out cheekbones. In fact they both looked quite similar to European modern humans: Cro-Magnon man. And the Peking Man skull looked completely different.

  Professor Wu had failed to convince me, by showing me the skulls, that there was any evidence here for regional continuity in China. But it’s not that easy. It was clear to me that interpretation of the characteristics of these skulls could be very subjective. Professor Wu, who had spent a lifetime studying these skulls, was utterly convinced by the features, which seemed so obvious to him.

  Chris Stringer, a prodigious analyser of skulls, has tackled the problem of quantifying differences and similarities between archaic and modern skulls. Rather than describing and counting up ‘archaic’ and ‘modern’ features in skulls, he prefers to take measurements and objectively compare differences in skull shape in that way. The regional continuity model as put forward by Weidenreich proposes not only that Homo erectus in China is a direct ancestor of modern humans, but also that later archaic fossils represent an intermediate or ‘Neanderthal’ stage between erectus and sapiens. Specimens that seem to fit into Weidenreich’s ‘in-betweener’ stage of human evolution in China include fossil skulls from Maba and Dali.2 The Maba cranium was discovered in 1958 in Guandong Province in southern China, while the Dali cranium was found in 1978, in Shaanxi Province. Uranium series dating of a cow tooth from the site gave a date of around 200,000 years ago for Dali, but it’s not clear how close the tooth was to the skull, so this date should be treated with some caution.5 Maba has been reported to be around 150,000 years old.

  Stringer included the Maba and Dali skulls in a study where he took skull measurements from a range of archaic and modern skulls from Africa, Europe and East Asia, and then mathematically compared the ‘shape distance’ between skulls to show how closely related they might be. He also looked at the three anatomically modern skulls from the Upper Cave at Zhoukoudian. The results showed that the archaic African skulls were the ‘best shape ancestors’ for modern human skulls, including those from East Asia, and that the Dali and Maba skulls were very different from modern skulls, and so were not convincing ‘in-betweeners’.8 It seems they might represent East Asian populations of Homo heidelbergensis, or even Neanderthals, that were later replaced when modern humans arrived.9

  It is clear that analysing skull shape is a very complicated business. Anatomical features don’t just vary from one population (or species) to another: they also vary within populations. Just to make things even more complicated, the variation within populations is often greater than that between them.2, 7 If you’re looking for differences between groups, you have to pick your features carefully. Another major problem is that we don’t yet understand how different anatomical features of the skull are connected or related to each other. Such connections could really skew the data one way or the other, whether we are just ‘counting up’ features or indeed taking measurements and trying to do something a bit more objective, as in Stringer’s analyses.

  These connections are a bit difficult to get your head around (again, forgive the pun), but just imagine that, for instance, eating a really crunchy diet from a young age could produce widespread effects on your skull. For the sake of argument, let’s say it would make the corners of your mandible more pronounced, give you a stronger browridge to resist the powerful forces of the chewing muscles, and maybe even affect the whole shape of your skull. (This isn’t entirely hypothetical: there’s good evidence to suggest that a switch to softer diets in the last thousand years is linked to smaller face size.10) Now, if I compared your skull with that of someone who ate much softer food, there would be several features that looked different, all of them linked to diet. If I compared your skull to an ancient fossil, from an early human who also ate hard food, then you might l
ook more like them than your soup-drinking counterpart would. I could count up the features and find at least three where your skull was similar to the fossil, but this wouldn’t mean that you were closely related, just that you ate a similarly tough diet.7

  As well as acquired characteristics which may be linked by some aspect of function, there are likely to be genetic connections between various skull features. There’s no way that each tiny feature of your skull is neatly controlled by a separate gene; one gene might affect a whole suite of features in different parts of your head. So again, just counting differences between skulls isn’t going to give you a real idea of how closely, genetically related two populations or species are.

  The genetics of morphology is hugely complicated, and an area with which researchers are really only just starting to get to grips. Genes don’t operate independently; they work as a team, with proteins sticking their oar in as well. Geneticists can look at whole genomes now, but it is like having a book in a foreign language (in this case, spelt ‘AGTCTGTTAATCCGG’ etc …), where we are only just beginning to understand what a few of the words mean. Some of them are about chemistry inside cells, but others dictate anatomy. Somehow, those gene-words create a conversation telling one fertilised cell to multiply and change and multiply and change until an adult human being is produced. Picking apart the complex tapestry of development, and finding out which genes are responsible for each motif, is a hugely exciting area of research in the twenty-first century.

  As many features of skull shape may be tied together by function or by genes, it means that huge, inclusive lists of features linking modern and archaic skulls (or not) are misconceived. It explains how it’s possible for such long-list-making studies to have been used on both sides of the argument, both for and against the regional continuity. This doesn’t mean we should give up on using morphology to help work out our evolutionary past, but we do have to be careful about how we go about it. And at the moment, while we’re still finding out how morphological features are related through function and genetics, all we can do is try not to use sets of features that seem to have a tendency to occur together.2

  Of course, while the relationship between genes, function and morphology is still being elucidated, the genes of living people offer us another powerful tool for reconstructing human lineages and migrations. I asked Professor Wu what he thought about genetic studies that suggested all modern humans had a recent African origin. He was quite sceptical about the potential for building evolutionary trees based on genes, and especially about the power of genetic studies to predict dates of divergence of lineages. ‘Different genetic studies even disagree on the age of the last common ancestor,’ he said. ‘Putting a date on it using the molecular clock assumes a constant mutation rate, which we can’t be sure about.’

  It’s true that different genetic studies have produced different predictions of age of a last common ancestor – but most agree that the date lies between 100,000 and 200,000 years ago.11

  It was clear that Professor Wu had much more faith in what he considered to be the hard – fossil – evidence, and he was absolutely sure that the fossils pointed to regional continuity. For him, as for Alan Thorne, Homo erectus and Homo sapiens were not even separate species, but subspecies. Wu preferred the labels Homo sapiens erectus and Homo sapiens sapiens. He argued that one form had gradually changed into the other over time, without speciation, and that the unity of the species across the world was maintained through gene flow between populations. Wu himself had proposed this theory, of ‘continuity with hybridisation’.

  But there does seem to be a significant temporal gap between archaic and modern human fossils in China. The most recent archaic fossils, from Xujiayo, date to around 100,000–125,000 years ago. The oldest well-dated modern human remains in China, including a mandible and limb bones, were discovered in Tianyuan Cave, about 6km away from the main site at Zhoukoudian. AMS radiocarbon dating placed these fossils at 39,000–42,000 years old.12 The next oldest modern human remains in the Far East are some leg bones from Yamashita-cho, Okinawa, dated to around 32,000 radiocarbon thousand years old (about 37,000 calendar years), and the Upper Cave skulls, at around 10,000 to (at a push) 30,000 years old. And, on balance, it seems that most investigators believe that the gap between archaic and modern Chinese fossils is not only temporal, but also morphological and genetic. Having seen the fossils and casts of Peking Man, I was not at all convinced that I had seen the ancestors of the Chinese.

  But there was something else in China that Wu believed supported his theory of regional continuity: stone tools. And here, I had to admit, he had a point. In Europe, the arrival of modern humans on the scene was marked by a clearly new ‘archaeological signature’, a sudden change in stone tool technology, with the appearance of the Upper Palaeolithic.

  In the East, modern humans seem to have been around for a long time before a distinct toolkit appears.

  An Archaeological Puzzle: Zhujiatun, China

  From around 1,000,000 to 30,000 years ago, the stone tools of East Asia are predominantly fairly crude, Oldowan-style pebble and flake tools. The archaeology of this part of the world is particularly strange, when viewed from a European perspective. The Acheulean, with its classic hand axes, doesn’t really happen, and neither does the Middle Palaeolithic.1, 2 People just seem to carry on using really basic pebble tools that even Neanderthals would have considered crude. In 1955, this moved American archaeologist Hallam Movius deprecatingly to call the East ‘a marginal area of cultural retardation’.3

  It’s not until 30,000 years ago, in the late Upper Pleistocene, that the Upper Palaeolithic appears in China, with more sophisticated tools like end-scrapers, burins, blades and microblades, as well as tools made from bone and antler. But that transition happened some 20,000 to 30,000 years after genetic estimates for the arrival of modern humans in East Asia, and a good 10,000 years after the first fossil evidence of modern humans in China.4 Before then, the tools the modern humans were making were no different from the tools made by earlier archaic humans in the East. It was this persistence of pebble tools – the so-called ‘chopper-chopping’ technology – that Wu interpreted as archaeological evidence for regional continuity. If all we had to look at were the stone tools, then regional continuity does seem like a reasonable explanation.

  But if the balance of evidence favours a recent African origin for modern humans, and if we take those dates from the genetic and fossil record, suggesting a colonisation of South-East and East Asia by (presumably ingenious and adaptable) modern humans between 40,000 and 60,000 years ago, then why were they putting so little thought into their tools? Were they really culturally retarded, or had Movius missed the real technology of Palaeolithic East Asia?

  Palaeolithic archaeology is a difficult area just because so little is left behind. As I had seen in Siberia, it was possible to erect a comfortable home, live in it, then move on without leaving any clue for future archaeologists to uncover. Archaeological clues in prehistory are rare and precious. As for tools, anything made out of something we would now consider as biodegradable – wood, other plant materials, animal skin – is by its very nature going to disappear from the archaeological record.

  Some archaeologists believe that the riddle of stone tools in East Asia could be explained by something that is both biodegradable and ubiquitous in the region. I met up with Australian archaeologist Jo Kamminga and we made our way to the small village of Zhujiatun in southern China, to find out more about this theory.

  Jo had armed himself with a selection of typical pebble and flake tools. I had a look at one of them.

  ‘This is a fairly crude tool,’ I observed.

  ‘Well, this is what we find in China, and South-East Asia, and in Australia as well. It’s not beautifully shaped, but it’s incredibly sharp,’ said Jo.

  ‘But at the end of the Palaeolithic in Europe, people are making really sophisticated stone tools – so what’s going on here?’

  ‘Well
, we’re in a different world here. Firstly, you don’t have the big cobbles of flint that are found in the chalk and limestone areas of Europe. The chert here comes in smaller nodules, and the material is not so nice to work. You can’t make those fine points. But the most important thing is that we have a different climate and vegetation here. You have a different range of raw materials available: not only stone, but also bamboo. It’s a very resilient material and it can be used to make the sorts of things you might be using stone for in other places.’

  So perhaps those most basic of pebble tools are just enough to make more sophisticated tools out of plant material, and, perhaps, out of that most prolific subfamily of grasses in East Asia: bamboo. Bamboo is used so widely in the East today that it is easy to imagine it would have been seized upon by the first colonisers. But there would be no trace of it left in the ground, only that of the stone tools which had been used to shape it. Although Movius had concentrated on pebbles from which flakes had been struck, leaving a sharp edge on a heavy tool, perhaps he had missed the point. Certainly, the flaked pebble could be used as a crude ‘chopper’, but the ‘waste flakes’ also had useful cutting edges.

  ‘Why would you go to all that trouble of making a sophisticated stone tool,’ asked Jo, rhetorically, ‘when you can just take a piece of bamboo, use that as a knife – and throw it away when you’ve finished? Because it’s everywhere.’

  He was right. Zhujiatun was surrounded by bamboo forest. From a distance, the hillsides looked feathery: the wind blew through the bamboo leaves as through a field of corn. Jo and I were going to try a bit of experimental archaeology. We used a large, very crudely sharpened cobble to bash at the base of a bamboo trunk. The bamboo was thick, about 15cm in diameter, and I prepared for some hard physical work to fell it. But after just a few minutes of bashing, the bamboo fell. With a little twisting, it ripped away from its base, and we had the raw material we needed to try making some ‘Palaeolithic’ bamboo tools.