Lone Survivors

by Chris Stringer

  Many years ago I remarked that the one thing you might have expected the Neanderthals to bequeath to the Cro-Magnons in any interbreeding was physical (in terms of body shape) and physiological adaptation to the cold, whereas the reality is that the early Cro-Magnons had completely contrasting linear physiques, while Europeans of today have poor cold tolerance compared to many other modern human groups—hardly what you would expect if their ancestors included Neanderthals. This evidence against interbreeding is seemingly duplicated when it comes to skin color, as we already saw, since modern Europeans evolved their own lighter pigmentation rather than borrowing that of the Neanderthals. But if genome comparisons show that there was interbreeding, why didn’t these apparently useful features transfer across?

  I think the answer may come from recognizing that the place everyone has focused on when thinking about interbreeding is Europe, at the time of the last Neanderthals. But by then they were a dying breed, few in number and with low diversity. If the interbreeding actually happened earlier, in a warmer region or a warmer period, maybe the Neanderthals involved were not light-skinned and cold-adapted European examples. In fact, the interbreeding might even have happened when people like those from Skhul-Qafzeh and Tabun were in the Middle East 120,000 years ago. If a thousand of those early moderns mixed with just fifty Neanderthals and then survived somewhere in Arabia or North Africa, could they have subsequently interbred with the Out of Africa emigrants 60,000 years later and passed on their hidden component of Neanderthal genes?

  However, there is yet another possibility. Ongoing research suggests that the 2 percent or so of Neanderthal DNA in French, Chinese, and New Guinean samples is not the same 2 percent, and modeling Neanderthal-modern interactions has led geneticists Mathias Currat and Laurent Excoffier to propose yet another scenario—that this 2 percent or so in the different populations is instead a measure of how ineffective hybridization was in separate interbreeding events. It thus indicates a natural limit on the process, caused by biological, social, or demographic factors.

  Given our possible role in their demise, should we reverse the process of extinction and attempt to clone a Neanderthal from its newly reconstructed genome? This is something I would have dismissed as pure science fiction only a few years ago, but with the staggering progress recently in genomics no one should rule out an attempt in the future. What I am sure about, though, is that it would be quite wrong to resurrect long-extinct species purely to satisfy our curiosity about them, especially if they were human. Neanderthals were the products of a unique evolutionary history in Eurasia that lasted for several hundred thousand years, but they are gone, along with the world in which they evolved, and we should let them rest in peace.

  The breakthroughs in reconstructing Neanderthal genomes were mirrored across Asia in equally remarkable work on what has become known as “Lineage X,” or the “Denisovans,” by David Reich and his colleagues. A fossil finger bone about 40,000 years old from Denisova Cave in Siberia, which could not be assigned to a particular human species, first yielded a surprising mtDNA sequence, neither Neanderthal nor modern. In fact its mtDNA was more different than that of Neanderthals and moderns, suggesting an origin time well over 500,000 years ago. This finding was backed up when a massive molar tooth from the same levels was shown to possess a very similar mtDNA pattern. But even more remarkably, autosomal DNA was then recovered from the finger bone, allowing parts of the whole Denisova genome to be reconstructed. Comparisons with the genomes of chimps, Neanderthals, and various modern humans produced even more surprises: Denisova was probably a derivative of the heidelbergensis lineage, but one that remained genetically closer to Neanderthals than to modern humans, perhaps indicating intermittent gene exchanges across central Asia where Neanderthals and Denisovans encountered each other. Enigmatic Asian fossils dated between 100,000 and 650,000 years ago, such as those found in Narmada in India and Yunxian, Dali, Jinniushan, and Maba in China, have been considered possible Asian derivatives of H. heidelbergensis, or relatives of Neanderthals, so they may be candidates for this ancient eastern lineage. In addition, there are fragmentary remains from Xujiayao in north China and Zhiren Cave in south China that are claimed to show both archaic and modern human features, dating from about 100,000 years ago. These fossils are too incomplete to determine their evolutionary status, but they also hint at additional complexity in the story of modern human evolution in China.

  But something else even more remarkable has still to be explained properly: the Denisovans are also related to one group of living humans—Melanesians—which may explain Jeffrey Long’s hunch, discussed earlier in this chapter, that they contain distinct archaic genes from the rest of us. The most plausible explanation for this is that Denisovans were present in southeast Asia as well as Siberia, and pre-Melanesian populations migrating through the region from Africa interbred with some of these Denisovans, picking up some 5 percent of their genes. That component might represent only twenty-five Denisovans mixing with five hundred pre-Melanesians, but it was sufficient to give the present-day inhabitants of places like New Guinea and Bougainville as much as 8 percent “archaic” genes—a small Neanderthal component they acquired first, probably in western Asia, and a larger Denisovan component they acquired later, on their way to Melanesia. As with the presence of small amounts of Neanderthal genes, there will now be considerable attention to what, if anything, those Denisovan genes might be doing in modern-day Melanesians. For example, could they have picked up useful defenses against some of the diseases endemic to southeast Asia? This is suggested from studies of the immune system by geneticist Abi-Rached and colleagues, who argue that some variants in the HLA (human leucocyte antigen) system of modern populations in Eurasia could have been picked up through interbreeding with Neanderthals and Denisovans. Attention will now turn to Australia, as this continent has not yet had full genome comparisons with Neanderthals and Denisovans, but its early colonization and subsequent isolation will make it an important test bed for further hybridization studies.

  With such successes in recovering ancient DNA from Neanderthals and Denisovans, unimaginable even fifteen years ago, it may seem strange that there have not been a plethora of papers comparing ancient DNA from archaic populations in many other regions. But the reality is that tropical and subtropical conditions with high temperatures or humidity, or both, severely affect ancient DNA preservation, which is most unfortunate in a case like Homo floresiensis, where authentic DNA could have rapidly resolved the fierce arguments about its status as an archaic species or a strange variant of modern humans. While there are hopes for other sites in northern Asia beyond Denisova, and perhaps also for high-altitude locations farther south, it is likely that many ancient human populations will never yield up their DNA for study, and we will be dependent on lateral thinking, such as the study of human parasites, whose DNA provides parallel but independent pathways to the study of our own DNA, and of fossil DNA history (discussed later in this chapter). Non-DNA biochemicals such as bone proteins may also offer better prospects of survival in hostile environments and may yet provide us with useful alternative windows into our evolutionary past. But what of early moderns such as the Cro-Magnons in Europe, who lived under similar conditions to the Neanderthals but even more recently—surely they would be prime candidates for ancient DNA studies? Many of them must certainly contain authentic ancient DNA sequences at least as well preserved as those of the Neanderthals, and indeed several such sequences have been published. However, in some cases there are lingering doubts about the possibility, even probability, of contamination by recent human DNA, which might be indistinguishable from the authentic material.

  Indeed, in 2001 there were claims for the recovery of ancient DNA from Australian fossils dated to between 10,000 and 40,000 years old. Ten out of twelve specimens tested from the Willandra Lakes and from Kow Swamp had apparently yielded human mtDNA sequences, and one of these, from the 40,000-year-old Mungo 3 burial, was claimed to form an outgroup co
mpared with all other fossil (Neanderthal) and recent human sequences. It was claimed that the distinctiveness of the Mungo 3 sequence undermined genetic support for a recent African origin, and instead was confirmation of a Multiregional or Assimilation model of modern human origins. (One of the authors of the study was the multiregionalist Alan Thorne.)

  However, critics like myself, Alan Cooper, and Matthew Collins soon pointed out that the claimed recovery rate for the Australian ancient DNA was quite exceptional compared with results from elsewhere, particularly considering that some of the specimens had been buried for millennia under desert sands in conditions of extreme summer heat. Moreover, in some cases the DNA had been recovered from scraps of bone left in storage boxes after the skeletons had, most unfortunately, been reburied or cremated at the behest of aboriginal communities. Furthermore, standard experimental protocols for ancient DNA work had not been employed, suggesting the possibility of contamination, and reanalysis of the published DNA data, using a larger number of recent Australian and African sequences for comparison, demonstrated that the Mungo 3 sequence did not, in fact, form an outgroup to recent human mtDNA; nor did it present any serious challenge to a recent African origin.

  Now, reanalysis of the samples using the latest techniques is being carried out, to establish the extent of recent contamination. The lessons learned will be applied to further studies, which are concentrating on distinguishing ancient from recent DNA segments by the signals of damage that they carry. Such work is likely, at last, to provide us with unequivocal samples of ancient DNA from Cro-Magnon and other early modern fossils.

  A somewhat surprising—even disturbing—insight into human evolution can also be obtained from the study of some of our fellow travelers: hair and body lice. Lice parasites, which live off human blood, got a mention earlier in research that tried to estimate when humans may have first regularly used the clothing and bedding in which some of them lived. We are the victims of two distinct forms of lice, Pthirus, the pubic louse, and Pediculus, the head and body louse, and there must be a complex history behind our hosting of these two distinct genera. Our head louse is most closely related to that of chimps, which fits with an evolutionary divergence of our species and theirs about 6 million years ago. Yet strangely, our pubic lice are most closely related to those of the gorilla, but with an apparent divergence of only about 3 million years. This suggests that our head louse diverged with us, as the human and chimp lines underwent their evolutionary separation, but the younger jump of our pubic louse from the less close gorilla lineage must have a different explanation: our ancient African ancestors had direct contact with ancestral gorillas—perhaps sexual, perhaps sociable, perhaps involving conflict or predation. The separate existence of pubic hair in our ancient African ancestors, which presented an opportunity for the transfer from gorillas about 3 million years ago, also implies that we had lost much of our intervening body hair by that time.

  The mtDNA of our head and clothing louse comes in three quite distinct lineages, unlike our own mtDNA. The most common group is worldwide in distribution today and shows evidence of an expansion about 100,000 years ago, which fits well with the expansion of modern humans within and then outside of Africa. The second lineage, most common in Europe, diverged from the first about a million years ago, and a rare third lineage found in just a few individuals in Africa and Asia had an even older divergence at around 2 million years. The geneticist David Reed explained that one way to account for the results is to argue that the root population of modern humans was large enough to host these distinct lineages for about 2 million years, but his calculations showed that this was very unlikely.

  Another possibility is that there was interaction between human populations that had been distinct for much longer than 200,000 years, which implies contact between modern and archaic humans such as Neanderthals or, as we have seen, the Denisovans. The kinds of interaction cannot be determined, but it could range from interbreeding to contact with bedding, through to aggressive confrontations or even cannibalism, where lice could have jumped from victims to the perpetrators. As an example of the latter, historical studies showed that the Torres Straits islanders, living between New Guinea and Australia, used to keep the heads of both their deceased relatives and their enemies. In the latter case, this sometimes involved eating parts of the face and eyes of the trophy head; such behavior in the past could have allowed the spread of parasites between distinct human populations and even species. If they were able to jump species before their host populations became extinct, these louse lineages literally gave themselves a new lease on life. And of course the transfer of pathogens could have gone in the reverse direction too, and it remains possible that infections of one kind or another brought out of Africa by modern humans contributed to the demise of archaic humans elsewhere.

  In this chapter we have discussed the data within our genomes and those of the Neanderthals and Denisovans: genomes that document the evolution of, and at least occasional contact between, these closely related lineages of humans. The evidence confirms that we have a predominantly recent African origin, but worldwide our species is not purely and entirely Out of Africa. Within that continent, our ultimate ancestors were few in number and probably lived in small pockets. Our earlier discussion of the development of behavioral modernity showed its patchy genesis across Africa—a genesis that I compared to brief episodes, like a candle flickering and then being extinguished. So what finally changed to keep that flame burning and then intensified, in order for our species to begin its seemingly inexorable rise to world domination? There are many ideas and theories, and I will start to explore them in the next chapter.

  8

  Making a Modern Human

  As I explained in the first chapter, when I began my doctoral research in 1970, the origin of modern humans was hardly recognized as a specific topic worthy of scientific study. The standard classification of humans had living people, the Neanderthals, and diverse remains from sites like Broken Hill in Africa and Solo in Java all classified as members of our species. With such different-looking fossil members within Homo sapiens, the origins of features like a chin, a small brow ridge, and a globular skull were, not surprisingly, lost in a morass of diversity. Moreover, with the predominance of Multiregional or Neanderthal Phase models, the origins of those features were apparently scattered among many different ancestors living right across the Old World, so modern human evolution was not so much an event as a tendency; we were merely the end result of continuing long-term trends in human evolution in features like increasing brain size and decreasing tooth and face size. For human behavior too, there was an emphasis on gradual evolutionary trends; for example, in France the “transition” from the Middle Paleolithic Mousterian to the Upper Paleolithic Gravettian via the Châtelperronian industry was seen as supporting a parallel local evolution from Neanderthals to Cro-Magnons.

  It looks very different forty years on. For most scientists, Africa has been established as the center for both our physical and our cultural origins. The evolution of “modern” Homo sapiens can be viewed as an important physical and biological event, backed up by both fossil and genetic evidence. Many researchers would also trace back to Africa the origins of the complex behavior apparent in the Upper Paleolithic figurines and painted caves of Europe. And yet, much as I am delighted with the way the subject of modern human origins has taken off to become one of the most dynamic areas of research in paleoanthropology, I am still puzzled by many aspects of the African origin of our species. When I look critically at what we do know and, more important, what we still don’t, I feel we are not yet close to a full understanding of those origins, as I hope to explore in these final chapters.

  In the 1980s the issue for people like me, Günter Bräuer, and Desmond Clark was to get people to take the idea of a recent African origin for modern humanity seriously at all, let alone discuss how that origin might have come about. In what was a real struggle against some very influential and at
times vitriolic opposition, I’m sure at times we oversimplified both our views and those of the multiregionists, and played down complexity in the data, in what became an increasingly polarized and sometimes bitter debate. At times also, as my views on a recent African origin developed, I favored the idea that our species evolved very rapidly in one small area—a sort of African “Garden of Eden.” But the general view has been that there was probably a relatively gradual evolutionary sequence in Africa from archaic humans (Homo heidelbergensis, sometimes also called Homo rhodesiensis) to our species, H. sapiens. Heidelbergensis fossils in both Africa and Europe are dated to about 500,000 years old, while, as we have seen, fossils representing our species have been found in Ethiopia at Omo Kibish and Herto, dating to between 160,000 and 195,000 years, with more fragmentary remains from Guomde in Kenya perhaps 250,000 years old. The assumption has been that a gradual accumulation of modern characteristics in Africa would have paralleled a comparable buildup of Neanderthal traits in Europe, from a similar heidelbergensis ancestor.

  What triggered the evolution of modern humans in Africa, and why this happened at all, is still uncertain. Did social or technological advances promote evolutionary change, or was geographic isolation following severe climatic change responsible? It is not yet even clear where the first “modern” population or populations lived, but the areas of eastern and southern Africa have vied for the title “Cradle of Modern Humanity.” The fossil record has highlighted Ethiopia and Kenya in East Africa as the most probable location for our origins, but this is also the region with the best fossil record for the period. In contrast, South Africa has a poorer fossil record but a much richer behavioral one for the Middle Stone Age, which is why some workers claim that region as the real focal point for modern human origins. Recent discoveries also shifted the focus farther north to Morocco, where reevaluation of previous finds and the discovery of new ones suggest that even northwest Africa cannot be excluded as a center for modern human origins. We must also remember that at least 50 percent of African regions that have stone tools from this period have not yielded a single fossil human relic to show us who was making the tools in question. So bearing these points in mind, I would like to take a fresh look at several aspects of our evolution that we already discussed, ranging from biology to behavior, to the role of climate in our evolution, in the hope of throwing further light on our mysterious African origins.