Out of Eden: The Peopling of the World

by Oppenheimer, Stephen

  The spectrum of Nasreen types in India is also different from further west. Although her granddaughter Europa types are present in 13 per cent of all Indians surveyed, these are nearly all accounted for by two Europa sub-branches, U7 and U2i (the Indian version of the U2 clan), along with a scattering of other Europa clans also found in the West, with virtually no root Nasreen types. This suggests that although U7 and U2i are both very ancient South Asian clans, they entered from further west at the time of the first colonization of Europe from South Asia 50,000 years ago – well after the initial out-of-Africa trek. There is also no evidence which would place the origin of Nasreen in India proper, although a scattering of her first-generation daughters, who have West Eurasian counterparts, is found there.38

  The real surprise among Nasreen’s clan representatives in India is her daughter Rohani. If we recall, Rohani is Nasreen’s most prolific daughter, being mother to most Westerners including Europa, not to mention two Far Eastern daughters with very large families (see Fig 4.2). In India, Rohani makes even that fecundity look like family planning. Estonian geneticist Toomas Kivisild has identified numerous other Rohani daughter branches, apparently all originating in India and none of them shared any other region. So rich are these new branches that he has been able to date their expansion to around 73,000 years ago.39

  Nowhere else, west or east, do we find such deeply branched diversity in the Rohani genetic line. On this basis alone there is a strong case for identifying South Asia as Rohani’s birthplace. It would seem more logical for Rohani and her European daughters to have been born in the Gulf than in India; but there is no genetic evidence as yet for that. Given the short distances and great time depth, however, either scenario is possible. One thing seems extremely unlikely, and that is that Rohani was born in the Levant or Europe. No Rohani roots are found anywhere in those regions to support that possibility, thus again slamming the door shut on any early northern migration out of sub-Saharan Africa.

  What is perhaps most interesting about the unique Indian flowerings of the Manju and Rohani clans is a hint that they represent a local recovery from the Toba disaster which occurred 74,000 years ago, after the out-of-Africa trail began. A devastated India could have been recolonized from the west by Rohani types and from the east more by Manju types. Possible support for this picture comes from the recent study by Kivisild and colleagues of two tribal populations in the south-eastern state of Andhra Pradesh.40 One of these populations, the Australoid Chenchu hunter-gatherers, are almost entirely of the Manju clan and hold most of the major Manju branches characteristic of and unique to India. The other group, the non-Australoid Koyas, have a similarly rich assortment of Indian type Manju branches (60 per cent of all lines), but have 31 per cent uniquely Indian Rohani types. The Chenchu and Koya tribal groups thus hold an ancient library of Indian Manju and Rohani genetic lines which are ancestral to, and include, much of the maternal genetic diversity that is present in the rest of the Indian subcontinent. Neither of these two groups holds any West Eurasian Nasreen types. The presence of Rohani types in the Koyas but not in the Australoid Chenchus might fit with some component of a recolonization from the Western side of the Indian subcontinent. As evidence of their ancient and independent development, and in spite of their clearly Indian genetic roots and locality, there were no shared maternal genetic types (i.e. no exact matches) between the two tribal groups thus indicating deep antiquity.

  The first Asian men: three roots

  Can we trace a similar route and history along the coast of Arabia and through Pakistan into India for the founder Out-of-Africa Adam? It would be easier if there were any representatives of Out-of-Africa Adam left, but like Out-of-Africa Eve, Adam’s root line is effectively extinct outside Africa and is best represented by his three immediate sons, Cain, Abel, and Seth, the three male lines which encompass all non-African peoples (see Chapter 3).41

  One of these three root branches, C (or RPS4Y), which I have called Cain after Adam’s first-born, is an exclusively non-African line and is found at generally low rates throughout South and East Asia, Oceania, and the Americas. Cain is present at an even rate of around 5 per cent throughout India, including among the Proto-Australoid Chenchus. Seth’s line, F (or M89), is nearly exclusively non-African, with the exception of several root types which, like a branch of the maternal Manju line (M1), appear to have found their way back into North Africa, in particular Morocco. Seth is by far the commonest of the three branches and is found throughout the non-African world (see Figure 4.5), with high rates in Australia and New Guinea, demonstrating clearly that he joined the other two genetic sons of Out-of-Africa Adam on that initial epic beachcombing trip. He accounts for 95–98 per cent of Indian male lines and has deep splits in his branches there.42

  Three of the commonest Seth branches in India are found hardly anywhere else in the world (except in nearby Central Asia, just to the north) and thus echo the pattern of the local flowering of Indian maternal lines. Two of these are offspring of Group K, or Krishna, Seth’s most prolific son (see Figure 4.4), while the third is a first-degree branch from Seth that has been found not only in nearby parts of Central Asia to the north but, significantly, further east in Cambodia. All these three types feature strongly in tribal populations of South India, especially the Australoid Chenchu. The Koya tribespeople in addition feature a rare root Seth type in a quarter of their population. This type is largely restricted to the Indian subcontinent and neighbouring parts of Central Asia, thus supporting the idea that India was the first major port of call on the southern route for Seth.43

  In another view of out-of-Africa, derived from Y-chromosome data, the initial beachcombing party featured only Cain and Abel; Seth came much later, through North Africa and the Levant, directly to Central Asia and thence to India and Southeast Asia. As should be clear from this chapter and elsewhere in the book, I disagree with this view, arguing for a number of reasons – including the tight geographical and genetic coincidence of the founder lines – that a single southern route best explains the findings. As far as Seth in Central Asia is concerned, the commonest Indian Y-chromosome type holds the clue to the question of whether the numerous Y lines shared between Central Asia and India had their roots in India or vice versa. This type is M17, which I described in Chapter 3 as a major European player that had come in from Central Asia. M17 has an overall Indian frequency of 27 per cent but reaches 47 per cent in the Punjab. M17 achieves its highest diversity in Iran, India, and Pakistan, much higher than in Central Asia or Europe. This points to South Asia as its original homeland. The clincher is to find M17 at both high rates and high diversity among the tribal populations of South India, including the Australoid Chenchu and Yadhava. If M17 had been intrusive rather than indigenous among such tribal populations, the diversity would have been low.44

  Figure 4.4 The beachcomber Y-chromosome tree. Shows Y branches found along the Indo-Pacific coast, Oceania and the Antipodes. In each, genetic continuity of spread can be seen from South Asia to Japan and to the south-west Pacific.

  The YAP controversy: African or Asian?

  The third primary line, often called YAP, I call Abel after Adam’s second-born. This line has a small, deep, early Asian branch found in a patchy distribution nearly exclusively in the Far East (of which more below). The other, western YAP, branch split again and mainly peopled northern parts of Africa and the Middle East, although he spread in small numbers into Pakistan, India, and Central Asia. Some derived YAP lines are commonest back in Africa, while others are characteristic of the Middle East.45

  There is a biblical analogy here: in the Book of Genesis, Shem and Ham were two brothers who peopled, respectively, the Middle East and northern Africa. Sadly for our story, the original root male Abel line from before the east–west YAP+ split does not seem to have survived, only his two sons. All he left us is the enigma of his birthplace: Ethiopia or Arabia? This question – ‘African or Asian?’ – persists into the name ‘Hamito-Semitic’ used fo
r the language family spoken by of most of YAP’s descendants. Whether we use that older, biblically inspired family name or the more modern term ‘Afro-Asiatic’, the dispute over whether the language family’s origins are African or Asian remains. Given the young age of most large language families, any such spread, whether west–east or east–west across the Red Sea, is likely to have happened within the past 12,000 years or so.

  One American geneticist, Mike Hammer, has insisted that YAP was born outside Africa and one of his two sons then returned, while Peter Underhill’s group insist upon an African origin.46 I am inclined to side with Mike Hammer and argue that YAP evolved in the first out-of-Africa colony, wherever that was, and then went back into Africa, but the ambiguity of the origins of the Afro-Asiatic language family perhaps carries more than an analogy. The non-African genetic lines that have crept back into Ethiopia and North Africa47 may well have done so after the Last Glacial Maximum as peoples spread back to re-occupy green lands that were former desert. Quite how soon after the last glaciation the re-expansion occurred is an interesting question. Some linguists and archaeologists see the enlargement of the world’s major language families as having accompanied the Neolithic agricultural revolution of the last 5,000–7,000 years.48 The genetic dates, however, suggest a re-expansion during the Mesolithic (i.e. 12,500 years ago), in other words much earlier than the Neolithic.49

  A puzzling aspect of the Abel trail is the big gap in his distribution between West Eurasia and the Far East and, notably, his complete absence from India (Figures 4.4 and 4.5). That he was on the beachcombing trail is evident from the presence of Asian YAP in the Andaman Islands, Cambodia, and Japan (see Chapter 5). Recent evidence shows Asian YAP at rates of 3–6 per cent alongside western YAP at similar frequencies farther north from the beach trail in Central Asia (Uzbekistan). This mirrors the picture for Cain, who although not completely absent in India, has much lower rates there than in Central Asia to the north. Given that, although rare, Asian YAP is present beside the trail and farther along it, it is possible his absence in India – and indeed Cain’s low rates in South Asia – could also be explained by the devastating effect of the Toba blast on the Indian subcontinent, as suggested by the geology and the maternal genetic story (see above). Toba could have created a genetic bottleneck in India which was followed by a predominant local recovery of the Seth line at the expense of his two beachcombing brothers, Cain and Abel.50

  Figure 4.5 Distribution of the three male founders Abel, Cain and Seth in ethnic groups of Eurasia. Seth predominates overall, but reaches c. 100% in India possibly resulting from the Toba explosion. The latter is consistent with near local extinction following Toba with recovery only of Seth on the east coast. Extreme drift to Seth also occurs in north-west Asia. High rates of Cain in north-east Asia and Australia and of Abel in Tibet, the Andamans and Japan may result from drift.

  The three Asian male founder lines: beachcombers all

  With our three primary out-of-Africa male lines, can we ask the same questions as with nuclear and mitochondrial markers about the earliest route of spread out of Africa? I think so. Dating of Y branches is in its infancy, but we can get a view of the geographical sequence of branches if we look for the geographical sites of ancestors and descendants. This is the phylogeographic approach. Taking Cain first, at 60 per cent of all Y-chromosome types in Australia he is the dominant line. In the Eastern Indonesian islands of the Moluccas and Nusa Tenggara, which the first beachcombers had to island-hop to get to Australia, we find the least changed or root Cain type, from which the unique Australian type is derived. The only other place in the world where this root Cain type is found is India, where it achieves a low rate throughout, including among the Australoid tribal groups. On the neighbouring island of New Guinea we find a brother group to the Australian Cain type, C2. When we look for other early colonies we see first that Cain is absent from West Eurasia but increasingly common to the east and north of India. A breakdown of Cain frequency shows that he appears most commonly on the East Asian Pacific coastline and Southeast Asia, and much less frequently inland in Central continental, North, and South Asia. The Asian types mainly belong to one derived clan which later moved over to America as a founder, although there is some greater diversity in Japan.51

  The East Asian coastal distribution of Cain and his offsprings’ strong regional differences in Southeast Asia and Oceania (see Figure 4.5) suggest that he was numerically the main male founder along the initial coastal beachcombing route round Asia, although clearly not the only one since his other two brothers also made it round the coast to East Asia and Japan. Only one of the two remaining founding male types got as far as Australia as well as Southeast Asia and round the East Asian coast – Seth. This is shown by the presence of Seth at low frequencies in his ancestral or root form in all these areas. Australia has two main local ancestral Y types: one comes from Cain (see above), and is the commonest at over 60 per cent; the other is from Seth’s Asian genetic son Krishna, at around 30 per cent. This 30 per cent rate for Krishna persists, with variation, more or less throughout Southeast Asia and up the Pacific Rim to Korea. Seth’s ancestral line, however, persists at a couple of per cent throughout Southeast Asia and Australia and also on the Pacific coast in Korea. These patterns strongly suggest that the first beach-combing route not only went to New Guinea and Australia, but continued round the coast to China, Japan, and Korea. The third root Y line, the Asian YAP, got as far as Cambodia but apparently did not cross to Australia, instead continuing round the Indo-Pacific coast to appear at low rates in Southeast Asia and at higher rates in Tibet, southern China, and Japan.52

  This picture of the first male beachcombers mirrors that of the maternal mtDNA lines and tells us much, albeit indirectly, about the identity and ages of the first male founder lines. A prediction of our southern route is that somewhere near the beginning of the trail, and before the parting of the ways, we should find these three oldest non-African lines all together. Southern Arabia and the Gulf region are still under study, but when we look for the three male founder lines in Pakistan and India as a whole, there they all are. We can find root and branch for Cain, Seth, and Seth’s genetic sons and grandsons. We can even find several unique YAP types from a YAP branch normally only found in Africa and West Eurasia.53

  Nowhere else outside Africa do we find such a diversity of deep Y roots and branches except, to a much lesser extent, in Central and North Asia. This picture of Central Asia as another transition zone between East and West is borne out in the rich mixture of European and Asian maternal mtDNA lines also found in that region,54 suggesting that one of the primary splits after the arrival in India was to travel north up the Indus to Central Asia. This early inland route is one of the main topics of the next chapter.

  In summary, the South Asian region, the first homeland of that single, successful southern exodus, shows the presence of the genetic roots of that expansion not only in the so-called aboriginal peoples around the Indian Ocean, but among the bulk of the modern populations. Among these roots we can detect genetic base camps for the most westerly of the subsequent pioneer treks inland to the vast Eurasian continent. These treks set off, after a pause, for Europe, the Caucasus, and Central Asia. It seems that the vanguard of the beachcombing trail retained a surprising proportion of the original genetic diversity left in the out-of-Africa group and moved rather faster round the shores of the Indian Ocean. So fast, in fact, that they travelled right round to Indonesia and on into Near Oceania, arriving in Australia long before their first cousins made it to Europe.

  The exact chronological relationship of the exodus and the subsequent arrival in Southeast Asia to the massive Toba volcanic explosion of 74,000 years ago is critical. First, Toba is one of the most accurately and precisely dated events of the Palaeolithic, and its ashfall acts as a time datum for the whole of southern Asia. Second, the effects of Toba’s direct ashfall followed by the inevitable ‘nuclear winter’ would have been disastrous for any l
ife in its path, and pretty bad farther afield. The presence of tools thought to be made by modern humans found with Toba ash in the Malay Peninsula suggests that the beachcombing vanguard had arrived in the Far East before the eruption. Triangulation of this anchor date with other pieces of evidence supports this scenario. Other clues include new dates for the Liujiang skull, luminescence dates from Australia, the date of the lowest sea level enabling passage to Australia at 65,000 years ago, genetic dates of the expansion of the L3 group at 83,000 years ago, and the onset of significant salinization of the Red Sea dated to 85,000 years ago. The best evidence for early modern humans in Asia should come from real fossils and their dated context. Such work is in process at the site of Liang Bua in Flores.

  Now, if Toba really did blow its top after India was first colonized, we would expect a mass extinction event on the Indian Peninsula which affected the eastern side more than the west. This is certainly one interpretation of the paradox of the Indian genetic picture, in which the genetic trail of the beachcombers can be detected, but the bulk of Indian subgroups of Manju and Rohani are unique to the subcontinent, especially among the tribes of the south-east. This is what we would expect for a recovery from a great disaster. The oldest of these local lines have been dated to around 73,000 years ago.