Out of Eden: The Peopling of the World

by Oppenheimer, Stephen

  The two best examples of controversies that rumble on among anthropologists as a result of these imperfect tools are, first, the multiregional versus out-of-Africa theories of modern human origins, and second, whether or not there was interbreeding between Neanderthals and the first modern Europeans – the Cro-Magnons. Classical population genetics using simple protein markers gave as blurred a picture as did studies of skull shape. The unadulterated direct transmission of the Adam and Eve genes down the generations has changed all that.

  Clear genetic trees for both modern Y chromosomes and mtDNA point back to a recent common ancestor of all modern humans within the last 200,000 years and a migration out of Africa less than 100,000 years ago. This new line rather quickly replaced all preexisting human genetic lines, including the Neanderthals. Admittedly there could have been interbreeding between archaic and modern Homo sapiens, but there is no convincing evidence for this in our male and female gene lines. So, if there was miscegenation it would have been rather small-scale, and the ancient lines would have become extinct. Some have suggested that traces of such past interbreeding could reside undetected in our vast nuclear genome, but the key word is ‘undetected’. Because of their tendency to recombine and mix at each generation, it is difficult to draw unambiguous trees of nuclear genes, so there will always be questions.

  The first exodus to South Asia

  I argue in this book from the genetic evidence – with back-up from several landmark archaeological dates – that the first successful exodus was single, occurred 85,000 years ago, and took the southern route out of Africa, reaching the Malay Peninsula by 74,000 years ago and Australia by 65,000 years ago. The earlier exodus to the Levant had already died out 90,000 years ago. Although there remains the possibility of a later archaic human colonization of Australia from Eastern Indonesia, the earliest people there were modern humans, and so far there is no convincing evidence of archaic Adam and Eve gene lines in today’s aboriginal Australians.

  Explicit in all these predictions is the central role of South Asia (particularly India, Pakistan, and the Gulf) as the fount of all non-African dispersals. As far as the gene tree is concerned, the earliest branches of non-African gene lines are in South and Southeast Asia. The dates of first colonization of East and Central Asia by modern humans are problematic, because of uncertain skull dating and the evidence for a more recent Mongoloid replacement, but if the redating of the Liujiang skull at no less than 68,000 years is correct, South China could have been colonized at the same time as Southeast Asia. If modern humans had reached Southeast Asia before the great Toba volcanic explosion, the sharp genetic break between India and the Far East may be explained by the ash cloud that covered India around 74,000 years ago.

  Trekking north

  Europe and West Asia were colonized later, around or after 50,000 years ago, when the world warmed up, opening a land corridor from the Gulf region to the eastern Mediterranean countries.

  Archaeological evidence suggests that Central Asia was first colonized at the same time as West Eurasia, 40,000–50,000 years ago. A group of genetic lines still found in North and Central Asia support this and suggest a direct movement north from the Indian subcontinent round the western edge of the Himalayas. Some of these North and Central Asian lines eventually moved west into Europe and east to the Americas. Other Central Asian lines seem to have come ultimately from the early beachcombing Southeast Asian pool, presumably taking routes round to the east of the Himalayas. The homeland of the so-called Mongoloid physical type, if there is such a single complex, remains unknown and contentious, although associated dental evidence again points to the south. The simplest view of the genetic and geographic evidence is consistent with the idea that somewhere in east Central Asia (i.e. north of the Himalayas) was a region where certain physical trends which had already begun to appear in Southeast Asian peoples, as a result of drift, were exaggerated either by drift or under selection to cope with a cold climate.

  The last glaciation: refugees and pioneers

  Following Toba, the most dramatic recent climatic event to have affected modern humans the world over was the Last Glacial Maximum (LGM), 18,000 years ago. For Africa it meant another great desert and a population crash. For Europe, the best-studied region, it meant people retreating south to a few refuges in the Basque country, Italy, and the Balkans, and to the Ukraine in Eastern Europe. The genetic picture supports the Western and Central European population crash implied by the archaeology. In one conventional archaeological account, the remnant Palaeolithic hunter-gatherers were more or less replaced within the past 10,000 years by farmers from Anatolia and the Levant. In contrast, the genetics reveal that most modern European paternal and maternal gene lines derive from the Palaeolithic ancestral lines already in Europe which re-expanded from their refuges in the south before the Neolithic period.

  Much less is known of the effects of the LGM on Asia. Some suggest a near-total abandonment of the once vast Central Asian steppe, formerly inhabited by Upper Palaeolithic mammoth-hunters. The archaeology reveals that there was at least one refuge, in southern Siberia, which was inhabited right through the Big Freeze. The surviving genetic lines tell us that at least some people made it through the worst of the cold in those refuges.

  One thing we can be sure of is that, even if a few remained, others would have tried to escape the increasing cold of the steppe while they had the chance. Unlike Europe, where escape to the south was bounded by the Mediterranean and the Syrian Desert, Central and North Asia offered several escape routes to more temperate climes: west to Eastern Europe, north-east to Beringia and America, east to Japan and Korea, and south-east to south China and Southeast Asia. The last destination would have proved the most attractive, because unlike the others its habitable area was increasing, thanks to the falling sea level, not shrinking under the encroachment of the ice. In Chapter 6, I speculated on these pre-glacial diaspora in the light of available genetic and archaeological evidence.

  The peopling of the Americas remains a contentious field for American archaeologists. A die-hard old school hangs on to the post-glacial Clovis-first orthodoxy, while others are persuaded of a pre-glacial migration by more recent findings, and linguists are unable to supply live ammunition to either side of the academic battle. The genetics community, by contrast, seems mainly to agree that most, if not all, of the founding American lines entered North America before the LGM, between 22,000 and 30,000 years ago. The former date would fit well chronologically and genetically with the preglacial refugees referred to in Chapter 5. Not all American founding lines came from north Central Asia: at least one genetic match can be made with Japan and multiple sources seem more likely. The much-maligned discipline of measuring heads also supports the view that the first Americans were not all derived from a single Asian homeland, rather from a mixture of Mongoloid, Caucasoid and the more robust Pacific Rim types.

  As in Europe, there was re-expansion after the LGM – in Beringia and the far north of America – from the original Palaeolithic founder lines. This gave rise to the distinctive Arctic and Subarctic peoples found in Canada, on the west coast.

  How much of this do I expect the reader to believe?

  Presenting our genetic history before 10,000 years ago in fewer chapters than even Julian Barnes’ The History of the World in 10½ Chapters is an ambitious aim. Inevitably there are regions, such as east and north-east Asia, where the trail left by the bones, stones, and even genes wears thin or is confused by overlaying evidence. With so many new reconstructions of Palaeolithic prehistory being offered, I could be accused less of speculation than of semi-informed guesswork. I am in good company, however. Several multi-author reviews published in the last couple of years feature maps of the world festooned with big arrows all over the world map. Written by experts in genetics, archaeology, and palaeontology, their maps differ sufficiently from one another for human genetic history to be an open field.

  Flippancy aside, the trend toward multidiscipli
nary syntheses is essential, since no single discipline has the full story. In some parts of the reconstruction, especially with the dating, the synthesis relies on mutually supporting evidence from different disciplines joining together to give a logical whole that is not apparent from the parts. Clearly, there is a danger of incest and circularity since several stories can show coincidental but false congruence.

  As an alternative to defending my speculations (it is too late for that, after all) I should like to say something about the quality of the evidence – about where there is genetic consensus, where there is no consensus but I think the evidence is good, and where I know the evidence is shaky and expect to see raised eyebrows. Otherwise there is a strong chance that the good evidence I have presented will be tarred with the same brush as the indifferent.

  Out-of-Africa is secure. There are few geneticists now working in this field who doubt that recent expansions of modern humans from Africa replaced pre-existing archaic Homo sapiens and older human species throughout the rest of the world. The possibility of a few genetic traces of the previous non-African lines lingering on among today’s humans remains that – a theoretical possibility. The genetic logic of the single out-of-Africa replacement hypothesis is almost inescapable, though, surprisingly, it has yet to be generally accepted by all geneticists and archaeologists. The southern route out of Africa across the Red Sea has long been recognized by zoologists as a dry route for migrating mammals during the Pliocene, but it is only recently that it has been put forward as a possible parallel route among several for modern humans. The northern route for Europeans is still clung to.

  In this book I have offered a synthesis of genetic and other evidence. Everything points to a single southern exodus from Eritrea to the Yemen, and to all the non-African male and female gene lines having arisen from their respective single out-of-Africa founder lines in South Asia (or at least near the southern exit). I regard the genetic logic for this synthesis as a solid foundation, and I have based the rest of my reconstruction of the human diaspora upon it. Obviously, the ‘choice’ of starting point (mine or theirs) determined all the subsequent routes our ancestors and cousins took. Tracing the onward trails is only possible as a result of marked specificity in regional distribution of the genetic branches today. The geographic clarity of both male and female gene trees is a big departure from the fuzzy inter-regional picture shown by older genetic studies. The degree of segregation of lines into different countries and continents is in itself good evidence that once they got to their chosen new homes, the pioneers generally stayed put, at least until the Last Glacial maximum forced some of them to move. This conservative aspect of our genetic prehistory also provides a partial explanation for the fact that when we look at a person, we can usually tell, to the continent, where their immediate ancestors came from, and underlies differences that some of us still call ‘race’.

  Farther along the trail, north and east from India, the genetic reconstruction of the routes becomes progressively more speculative, with some routes clearer than others. The early route to Australia is easy. The early date of Australian colonization, if confirmed, also makes a coastal beachcombing adventure round the Indian Ocean via Southeast Asia virtually the only possibility. The deep genetic differences between New Guinea and Australia tend to indicate that there were at least two separate early colonizations of these outposts.

  The late colonization of Europe and the Levant from the Gulf, rather than through North Africa, is my idea and is based upon the structure of the genetic tree. It neatly explains why West Asia and Europe, alone among non-African regions, lack the Asian M super-clan. The late date of Levantine colonization is explained by a dry climate before 50,000 years ago preventing access between the Gulf and the Levant. A marked improvement in climate from 50,000 years ago opened the Fertile Crescent corridor finally allowing migration north-west to the Levant, the Caucasus and Europe.

  The rest of the peopling of Eurasia becomes increasingly speculative. The colonization of East Asia via the coastal beachcombing route, and of Central and North Asia by a pincer movement either side of the Himalayas, are simply the best geographical explanations of the available genetic data. The concept of the expanding East Asian coastal flatlands being swamped by refugees from the Asian high steppe at the time of the LGM is also my idea and is bound to raise eyebrows. But there are no clear alternative explanations of the present distributions of gene lines and physical types in East Asia, and the abrupt archaeological changes in Japan and Korea at the peak of the LGM are very suggestive of an intrusion of Central Asian Upper Palaeolithic cultures.

  As far as the peopling of the Americas is concerned, I choose to follow the convincing genetic consensus of an entry of all lines around or before the LGM, while putting some human behavioural perspective on the reasons for the archaeological hang-up on Clovis.

  A word on prehistoric dates

  Dates vary in their precision, with the method used to obtain them. Even carbon-dating has systematic errors. In certain circumstances dates calculated using the molecular clock from genetic lines provide very useful confirmation of archaeologically derived dates. The mitochondrial molecular clock is still in the process of being fine-tuned, however, and Y-chromosome dates are still very contentious. Furthermore, because the molecular clock depends on infrequent random mutations, it requires many measurements (i.e. human samples with different genetic types) to get anywhere near reliability. This means that many of the more recent archaeological dates using the radiocarbon method are much more precise or certain. However, there are consistent errors in radiocarbon dates of archaeological sites from the Last Glacial Maximum, and the method becomes seriously unreliable over 40,000 years ago. This 40,000-year ceiling does not affect the genetic clock.

  The only truly reliable ancient dates are those that can be lined up unambiguously against climate records, such as those more recent ones indicated by tree rings and the analogous layers found in cores from the Greenland ice cap spanning the whole of the past 100,000 years and beyond. These cores give accurate and detailed climate records. For example, we can now see the dates of the frequent warm-ups or interstadials that affected vegetation and the movement of human populations. In our story the most accurately dated, dramatic, and unambiguous event before the last ice age was the eruption of Toba in Sumatra 74,000 years ago; and Toba ash is also found in the Greenland ice-record and submarine cores in the Indian Ocean, allowing a precise date marker.

  I have stuck my neck out to place modern humans in Malaysia by this date on the basis of the Kota Tampan site where tools were found under a thick layer of volcanic ash from Toba. The key tools were indisputably artefacts, and the ash did come straight from the sky 74,000 years ago. But in spite of majority view that the Kota Tampan tools were the handiwork of modern humans, they could still theoretically have been made by other humans, since no bones have been found on-site which would confirm the identity of their makers. The only modern human remains of that antiquity found in the region are the now re-dated Liujiang skull and partial skeleton from southern China. The dating of the earliest Flores (Eastern Indonesia) occupation by modern humans remains to be published.

  I have several corroborating reasons for relying on such a shaky lithic connection. First of all, the logic of the low-water colonization of Australia 65,000 years ago fits; and second, increasing numbers of genetic dates outside Africa easily reach back to this time. The next available low-water slot for the colonization of Australia would have been around 50,000 years ago, but that does not fit the other evidence so well.

  The trouble is that it is difficult enough persuading a majority of Australian archaeologists that Australia was colonized by 60,000 years ago, let alone 65,000 years. The problem with suggesting any dates for modern humans outside Africa much beyond 45,000 years ago is again the radiocarbon ceiling. If one were to rely solely on radiocarbon dating, the whole human world would seem to have started just over 40,000 years ago. Only a few Australian date
s of greater antiquity (e.g. in the region of 60,000 years) have been obtained by other methods, and they have so many problems it is easy to rubbish them. We could take the view that the colonization date for Australia just happens to coincide with the radiocarbon ceiling, but that is likely to be wrong. The safest thing, of course, is to say the date of first colonization is unknown. In this speculative synthesis, however, I would rather go with an earlier date which fits Kota Tampan and low sea levels than one which is more precise and recent, but eventually turns out to be wrong.

  Toba is also regarded by some as having caused worldwide population extinctions as a result of the ‘nuclear winter’ that followed. I have taken this into account in my reconstruction. India bore the brunt of the massive ash fall, and may have suffered mass extinction, since the Toba plume spread north-west across the Indian Ocean from Sumatra. This may explain why most Indian M sub-groups are not shared elsewhere in Asia and the dates of the M clan’s re-expansion are paradoxically younger in India than elsewhere in Asia and Australasia.

  What of the future?

  In my view it is not quite as rosy as some would hope. We are like all other species in that our evolution and survival are both subject to the controlling and moulding influence of the changing environment, whether the changes result from the ice-age cycles or from our own prodigal exploitation of the planet’s resources. One of the lessons of the past is presented by the longest timescale, that of recurring glacial maxima. Eventually, it is certain, there will be another freeze-up, at which point our species will be stressed and may well hit another bottleneck. It is hard to overstate the bearing that ice-age cycles have had on our fate on the planet. Taking the long view, the effects of global warming could be little more than a blip on the way to the next glacial maximum.