Out of Eden: The Peopling of the World

by Oppenheimer, Stephen

Kennewick Man’s teeth were Sundadont, suggesting Southeast Asian origins, and his eye sockets were also unlike those of Europeans or modern Native Americans. Intensive studies of combinations of his skull features revealed that the extant modern populations he most resembled were the Ainu and South Pacific peoples, including Polynesians.51 Most striking of all was how unlike any peoples of the past 5,000 years he was. A lot of other things were learnt about Kennewick Man. He had suffered many injuries and bone fractures, and had enjoyed a diet rich in marine protein, such as salmon.

  Radiocarbon dating confirmed the age of the skeleton to be 8,400 years.52 The skeleton was then impounded by the US Army Corps of Engineers and became the object of a legal tug-of-war between different advocacy groups. In spite of this, several DNA labs attempted to extract DNA from the bones. For technical reasons, and contamination by modern DNA, they were unsuccessful, which is regrettable.

  A successful DNA extraction and analysis could have provided us with any of several different results, most of which might not have answered everyone’s questions. Mitochondrial DNA, although it is useful for tracing movements of peoples, is but a small fragment of our genetic heritage. It does not tell us what we should look like. But it can often tell us where our maternal ancestors came from. For example, if in the unlikely possibility that uncontaminated Kennewick mtDNA had shown him to belong to one of the typical European clans and to have close type matches in modern Europeans, that would have been some evidence of direct European input into North America. Most other possible outcomes are more likely to have been generally confirmatory of a Native American maternal origin, but this would not have proved that there had been no European colonists over 9,000 years ago. The lack of numbers of successful uncontaminated extractions is one of the many problems of studying ancient DNA; if uncontaminated DNA can be extracted, it is still only a single sample which tells us little about the rest of the population or their history. For these reasons, large-scale mtDNA studies of modern populations are generally much more informative than one-off ancient DNA investigations.

  As if to tickle up the link between Kennewick Man and Europeans, his location on the Columbia River in Washington State near the American north-west coast put him near one of the two regions where the X group had been found, in particular near the Yakima tribe.

  The Solutrean hypothesis

  The idea of Europeans and European technology crossing the Atlantic in prehistoric times is not new, and was around long before Kennewick Man was found. One of the best known of such theories is called the Solutrean hypothesis (see Figure 7.1). Denis Stanford, an anthropologist at the Smithsonian Institution, has been particularly identified with it.53 The story goes that, at the height of the last ice age or maybe a little later, some hunters from the south-west European refuge area bordering France and Spain (see Chapter 6) took to their boats. They made it all the way from the European Atlantic coast, round the north Atlantic, to the east coast of America, where they became the Clovis people. The culture in the southwestern European refuge around the time of the LGM is known as Solutrean, and is famous for its beautifully worked bifacial points.

  Solutrean points are argued to be a rare invention and similar to Clovis points or their precursors such as those found at Cactus Hill.

  In support of the Solutrean theory is the habit, present in cultures both sides of the Atlantic, of driving herds of game over cliffs. This is hardly strong support since there is evidence for this practice even among pre-modern humans in China. It is also claimed that the highest concentration of the oldest Clovis points, which is said to be in the south-east USA, points to an Atlantic landing of immigrants. There certainly are problems with this theory, the most important of which are practical. The Solutrean cultures came before Clovis – and, as Vance Haynes so clearly demonstrated, Clovis did not appear until well after the LGM. If makers of Solutrean points had ventured round the northern Atlantic rim at the LGM, they would have to have skirted a coast that was ice-bound all the way round to New York.54

  Unfortunately for such wishful theories of Europeans having an early stake the Americas, as opposed to the view of them as much more recently responsible for the rape of the beautiful New World, the genetic story has no supporting evidence to offer. Rather the opposite, according to Atlanta geneticist Michael Brown, of the Emory University School of Medicine in Atlanta. Confirming that the American X line was indeed a single fifth American cluster, Brown and colleagues estimated the age of arrival of the founder at 23,000–36,000 years ago, far too early for the Solutrean hypothesis or any other post-glacial entry. X was easily as ancient in America as were any of the other four founders. Five daughter X clusters had spread over North America into the different regions of the Great Lakes and West Coast both around the 50th parallel at around the same time.55

  There is no doubt that the American and European X lines have a common X ancestor, but they may have split from this ancestor as long ago as 30,000 years. If there had been a recent but pre-Columbian European admixture in the Americas, it should reveal itself in Native Americans with typical European X subgroups or other commoner modern European lines, since X is quite rare in Europe. They are not to be found in America. The American X is therefore most likely to have arrived from Asia, the same way as the four other mtDNA founders (see Figure 7.3).

  The power of mitochondrial DNA to trace such ancient relationships was demonstrated during the making of The Real Eve, the Discovery documentary film based on this book. The producers decided to take a small number of DNA samples at random from volunteer Americans of different ethnicity. The mtDNA was extracted and analysed by Martin Richards in his lab at Huddersfield, West Yorkshire. Towards the end of the film is a shot of the participants being told their results (with full consent). By rare coincidence, two of the them, an American of Greek ancestry and a Native American of the Cree tribe (see Plate 25), both belonged to the X mtDNA clan. For a moment they stared at each other across the lab, at first silently and in amazement, then finally able to express their emotion at the realization of an ancient link between them.

  How many migrant groups, and where from?

  The real genetic puzzle about the origins of X in the Americas is its rarity in Asia and the linked question of where in Eurasia it originally came from. That is part of the larger problem of deciding whether the Americas were populated by single or multiple immigrations, which must be addressed first. So, before we look at how X got to both Europe and the Americas, we must look again at what the evidence can tell us about how many migrations there might have been to the Americas. But haven’t I already resolved that, by presenting the genetic evidence of all five lines entering America before the ice age, the so-called ‘one-wave and re-expansion’ hypothesis? Not so. The uncertainty associated with genetic ages is usually in terms of thousands of years. And while it is possible to use genetics to argue against Clovis-first, we cannot use genetics the other way round to show that that all five founders entered the Americas at exactly the same time.

  Looking through the genetic research on the peopling of the Americas, we can see a clear tendency, as the 1990s progressed, to reductionism, with four migrations turning to three, three to two, and finally to a single colonization of the Americas followed by a re-expansion from the same founder types. This minimalist approach is not mirrored in a rationalization of the total number of maternal American founder lines, which started at four over ten years ago and is now at least five.

  By simple analogy, the five mtDNA strands that colonized the Americas are five times as many as the single African mtDNA line that colonized the non-African world; so either there was a large founding group with no genetic drift, or there were multiple entries. This does not give strong support to a single band colonizing the Americas, for the following reasons. In Chapter 2, I argued that the single African exodus follows logically from the single paternal and maternal African lines that parented the rest of the world. To get down to just one line, the founding group must have g
one through a period of isolation and drift. This unitary argument clearly cannot be applied to the Americas. Although it is possible that the Americas were colonized by a single large hunter-gatherer group, this cannot be inferred from any of the genetic evidence, including and not forgetting the Y chromosome. The weakness in the evidence has been tacitly acknowledged by geneticists Anne Stone and Alan Stoneking, who have recently argued for a ‘single’ entry. Three lines before they postulate a ‘single wave of people’, they state disarmingly that, ‘Most likely, many small groups of people wandered across Beringia as they followed game . . .’56

  But does it matter whether the single wave was composed of one or many small groups? I think it does, because one band implies, logically, there was only one geographical source. Several genetics labs have taken the idea of a single American entry one step backwards into Asia by looking for a single Asian geographical source of Native Americans. So convinced are they that there must be one origin in Asia that they look for Asian regions that hold all five of the American founder lines.57 The idea that there can be one part of Asia that had all the right gene lines and the right Palaeolithic technology is not only naive but, as we shall see, does not fit other observations.

  In any case, the genetic story told by Peter Forster and others is not of an Asian homeland. They tell us rather of a homeland in the now partly submerged Beringia, an environmental refuge which straddled the two great continents and was continuously occupied through the ice age. It was Beringia, then, that received incoming migrations from, presumably, several parts of Asia, and acted as a staging post for onward migration. The climatic conditions leading up to the LGM cut Beringia off ecologically, first from Asia (by polar desert) and then from America (by ice). Before the Americas were colonized, and before the ice sheets finally blocked the route onwards into Canada, the maternal lines in the Beringian settlements had drifted down to the five that eventually did enter the Americas. Then, during the ice age and after the Americas had been colonized, Beringians dwindled further down to just one of the five mtDNA founder lines, Group A.

  We shall come back to the possible Asian and other sources of those five lines. But first, what other evidence is there that could tell us whether the migrations to Beringia from Asia was single or multiple? We have just been looking at one clear piece of evidence that tells us that the first Americans were anything but – if I may be excused for using the expression – clones from a single Asian tribe. That piece of evidence is Kennewick Man. He physically breaks the mould with his non-Mongoloid, South Pacific, and Ainu characteristics. Obviously, we can never know his skin colour or whether his hair was dark, curly, straight, or whatever, but he was clearly different from the majority of modern Native Americans, and he was not the only early North American to be so different.

  Spirit Cave Man, at around 9,400 years, is another famous ancient American, this time from the Great Salt Lake (see Figure 7.1). He was also different from any modern population, including Native Americans; uniquely in his case, his body had mummified and his black hair could be seen for a while before it bleached red in the light. The nearest source groups suggested for him were Ainu, Polynesians and Australians. Three other atypical Palaeo-Indians from the northern States include Pelican Rapids, or Minnesota Woman (7,800 years old and found in 1938), Browns Valley Man (also Minnesota and 8,900 years old), and Buhl Woman from Idaho (10,800 years old). We should remember, however, that not all North American Palaeo-Indians of that vintage were different from modern Native Americans. Wizards Beach Man, at 9,200–9,500 years, did resemble modern Native Americans.58

  South America has its own claims for extremely old dates of colonization, and clearly was not going to keep quiet on the issue of ‘unusual’ palaeo-ancestors. Although they have been known about for some time, these skeletal reminders of the southern past came to public attention only quite recently. As with Kennewick Man, this happened through media speculation over a rather dubious interpretation of an interesting discovery. In August 1999 a BBC documentary reported that Brazilian human evolutionist Walter Neves had been studying so-called Negroid people from extinct tribes in South America. The centrepiece was the skull of a twenty-two-year-old Brazilian woman born over 11,500 years ago, nicknamed Luzia. Initially discarded and donated to the National Museum of Brazil by a French archaeologist excavating in Minas Gerais, Brazil, in the 1970s, her skull had features which suggested an Australian, Melanesian, or even African origin. In 1999 the forensic artist Richard Naeve, from Manchester University, reconstructed Luzia and fleshed her out.59 Reborn, she looked anything but modern Native American (see Plate 27). Judge for yourself – we all have a claim to be innate experts in such recognition. Negroid, Australian, Melanesian, Liujiang possibly, but to me she looks just like one of the 3,000-year-old great carved Olmec heads of Central America (see Plate 28).

  Is Luzia a remarkable link to an unknown past, or is she just one of a kind? Human variation is such that one swallow does not make a migration. More numbers are needed. A time-depth study of several ancient skulls from Serra da Capivara in north-east Brazil indicated to Neves that there was a change in skull shape around 9,000 years ago from robust-featured skulls to more Mongoloid and modern types. He speculated on an earlier migration of robust non-Mongoloid people, similar to those I referred to in Chapters 5 and 6 as remnants of the beachcombing route. The beachcombers represented the least physically changed types outside Africa. Australian, Melanesian, Ainu, and Polynesians (and even Europeans) are all likely to be nearer to the original beachcombers than to the Mongoloid peoples of East Asia and America (see Chapters 4 and 5). So if there is a possibility that these non-Mongoloid types formed part of the migrations to America, appropriate comparisons have to be made. However, the best representatives of the Asian beachcombers of over 12,000 years ago are more likely to be the rather robust East Asian skulls of that vintage found in China, than today’s admixed relict groups. Neves did actually find links between the Palaeo-Indians and Palaeolithic skulls in northern China as well as South Pacific peoples.60

  Unlike the makers of the 1999 BBC documentary Tracking the First Americans, Neves apparently did not favour the theory that these Luzia-like people had sailed to South America across the South Pacific from Australia. In his view, these robust types were more likely to have taken the same route as everyone else – either through Beringia or along its coast. He also felt that they had been replaced or wiped out by a later influx of Mongoloid types in a separate migration into the Americas.

  Are there any remnants of those beachcombers left in modern American populations? Well, there may have been until very recently. The first European explorers to reach the southern tip of the Americas noticed numerous fires burning on the land, hence the name given to that extreme region – Tierra del Fuego, the ‘Land of Fire’. The people who kept these fires burning to counter the bitingly cold wind were hunter-gatherers entirely different biologically and culturally from their neighbours and from all other known Native American groups. In the maze of waterways to the south-west, around Cape Horn, were the so-called ‘canoe indians’.

  Slightly farther north were the so-called ‘foot indians’. These included the Tehuelche of Patagonia and the Selknam of northeastern Tierra del Fuego. Large, robust hunter-gatherers, their burning campfires were seen by Ferdinand Magellan and his crew as they sailed through what is now called the Magellan Strait. Many skulls of these people are in the possession of museums – the English ranchers who subsequently occupied the area placed a bounty on Selknam and Tehuelche heads when the indigenous peoples turned to sheep-raiding after their ‘free-range guanaco’ lands were taken over as sheep stations. There may only be a couple of descendants of these tribes still alive today (see Plate 23).

  Cambridge-based Brazilian biological anthropologist Marta Lahr included a group of skulls from the historic period (twenty-nine, including Selknam and Tehuelches) in her classic study of human cranial variation.61 This group shared several features with Australia
ns, in particular their degree of robusticity. However, apart from robusticity and its associated features, the Selknam and Tehuelches did not share much else specifically with the Australian skull morphology (e.g. non-metric traits) – and were, in particular, much larger – indicating most likely simply a shared retention of a Pleistocene trait if not a more distant relationship. Lahr was of the opinion that these hunter-gatherers were unlike other modern Native Americans and East Asians, and more like Southeast Asians and South Pacific Islanders. She thought that these differences between Amerindians most likely indicated that there had been separate immigrant groups to America.

  Recently, Walter Neves, of the University of Sao Paulo, and Joseph Powell, University of New Mexico, two of the foremost specialists in this field, joined forces in a major comparative analysis to see whether they could make sense of the accumulating evidence for Palaeo-Indians being different from, and more variable than, modern Native Americans.62 Their findings confirmed that if there were multiple founding migrations (the alternative would have to be local differentiation), Palaeo-Indians would have been more like the undifferentiated Indo-Pacific beachcomber type, while modern Native Americans are more like Northeast Asians, and also have affinities with Europeans. Furthermore, they argued, North American Palaeo-Indians were different from South American Palaeo-Indians in that the former strongly resemble Polynesians, while the latter are more like early Australians. They did not accept the latter as evidence of a direct migration across the South Pacific. Instead, like Lahr, they argued that both Australians and Southern Palaeo-Indians derived from a common ancestral source on the Asian mainland.

  Such a picture of multiple separate morphological affiliations between different American groups (ancient and recent) and different Asian/Oceanic types is consistent with at least two, and possibly three or more, separate migrations from different Asian regions, with the Mongoloid type eventually expanding largely to replace Palaeo-Indian types.63 We can definitely fit this in with the emerging genetic consensus of Beringia as the homeland, acting as a staging post for inward migrations of different peoples from various parts of East Asia.