Discuss this article at First Drafts, Prospect's blog
Humans first emerged in Africa around 2.5m years ago. Over 160,000 years ago a new group-the first "anatomically modern" humans-arose in the lands of east Africa. Every human on earth today is descended from that group. Then, around 80,000 years ago, a splinter group of these new Africans journeyed out of Africa and their descendants spread out to the far reaches of the continents. This incredible journey across land, river and ocean can now be mapped and plotted in time, through a combination of archaeology, climate study and, most recently, the study of genes.
The earliest humans
Our human story really begins 7m years ago, when cool, dry weather devastated the habitat of numerous forest-dwelling African ape species and opened new pastures for those apes that could survive on the savannah. At some point soon afterwards, the first evolutionary steps were taken towards the two-legged, large-brained creature we now call Homo sapiens. The earliest known walking ape, which evolved on the savannah 4.5m years ago, had the same moderately large brain as chimps. But following an intensification of the dry cool phase the first humans-Homo rudolfensis and Homo habilis-appeared about 2.5m years ago. Diminutive, with brains initially not much larger than chimps, they none the less made stone tools and were joined rapidly by Homo ergaster, and then Homo erectus, also both toolmakers. The latter was the Model T for humans, lasting well over 1m years and spreading throughout Africa, Asia and Europe. These early human species and members of their vegetarian sister group Paranthropus (now extinct) were all characterised by a dramatic growth in brain size. The rapidity of that initial brain growth was never to be repeated.
Some new behaviour shared by all these new species must have been behind the selective pressure that began to favour individuals with larger brains. Whatever new behaviour, unique to these two groups of hominids, drove the growth of their brains, it arose long before evidence for complex culture. It seems to have given them a special advantage in this period of drought, since they replaced-in some cases violently-all other hominids apart from forest-dwelling great apes. The most obvious skill which would benefit from a large brain is the one which still clearly separates today's humans from all other living species: speech. Speech does not depend on diet or toolmaking, which explains how such rapid brain growth could have occurred in vegetarians and omnivores alike while, over the same period, technology apparently languished in both.
Just over 1m years ago, the brain of one human species-Homo rhodesiensis-had grown to within the range of volume of modern humans. By around 300,000 years ago, the climate-driven brain-growth machine had reached a plateau of size greater than that of today's humans. Since then, our brains and bodies have got smaller. Perhaps, as with cars, a law of diminishing evolutionary returns set in, making it no longer economical or feasible to build models with ever larger "engines." (Moreover, the head of the newborn baby might simply have become too large for maternal safety.) Except for changes in limb proportions, eyebrows and skull shape, the physical evolution of the human family had by now slowed to a snail's pace. The real physical and behavioural threshold of Homo sapiens may have been reached at this point. All the characteristic behaviour of modernity-the grinding and use of pigment, the making of fine stone points and blades, fishing and long distance trade-can be traced back to Africa within the past 300,000 years. The evidence emphasises the subsequent acceleration in human technology: first slow, then faster and faster. As more and more knowledge began to be transmitted orally and accumulated down the generations, cultural evolution began to leave genetic evolution far behind. Looked at another way, if cultural evolution really took over from genetic evolution 300,000 years ago, then the major differences between us and them are merely cultural-that implies that if archaic Homo sapiens individuals from 300,000 years ago were transplanted to modern society they could well have the intellectual potential to put a man on the moon.
First hinted at by Darwin, the "garden of Eden" or "out of Africa" theory began to gain ground less than 30 years ago based on skull comparisons. It suggests that all human species, including the archaic types, arose in Africa. Anatomically modern humans who emerged 160,000 years ago (as opposed to archaic Homo sapiens who emerged 300,000 years ago) replaced all the others first in Africa, then outside-in some cases this seems to have been achieved by wiping them out. But an older theory known as multiregionalism was influential until recently. This theory, held principally by fossil specialists, argued for multiple sources of the different human races, claiming that each present-day regional Homo sapiens type (or "race") evolved slowly from local Homo erectus colonies. Homo erectus, one of the first humans, had made it rapidly out of Africa to colonise the whole of Eurasia. According to multiregionalists, African Homo erectus gave rise to Africans, Neanderthals (Homo neanderthalensis) gave rise to Europeans, east Asians derived from Peking man (Asian Homo erectus) and Australians from Java man (a descendant of Asian Homo erectus). These two views of the modern human genesis have quite different timescales. The out of Africa theory describes a complete replacement of other humans by a close-knit family of pioneers within the past 200,000 years, while multiregionalism claims deep regional divisions between modern human races going back well over 1m years. As a result, multiregionalism has been accused of providing justification for racist views. While this may be unfair, it is surprising that the view has lasted so long. Part of the reason the argument still runs on is the paucity of the fossil record. Well-dated fossil specimens of the earliest anatomically modern humans, dating back 160,000 years, were unearthed only recently in Ethiopia. The specimens confirm the African origin of our own special version of Homo sapiens. Genetics, with its ability to build and date trees, has now vindicated the out of Africa theory and the multiregionalists have become a dwindling-albeit vocal-band regarded by some geneticists as akin to flat-earthers.
The Adam and Eve genes
The genes we carry in our cells are inherited from our parents and define us as individuals. Almost since this was first understood, over 100 years ago, geneticists have dreamt of using genetic markers to classify human races, show how they are related, and determine their origins. Genes are made up of a long string-like molecule, DNA, which carries a sequence of coded instructions to build and maintain our bodies. Over many generations, small harmless mutations build up in the DNA code. The new mutations are then passed down the generations where they act as markers to identify new branches of a family tree for that small stretch of DNA.
The trouble is that most of our DNA has the tendency to get shuffled up and spliced after fertilisation of the ovum at every generation; this blurs the family tree. Luckily, two small parts of our DNA do not suffer this shuffling process and are passed down intact through the generations. One is the Y chromosome, which is passed down only from father to son. The other is so-called mitochondrial DNA (mtDNA), which we all have in our cells, but is only transmitted by our mothers. Sixteen years ago the Hawaiian geneticist Rebecca Cann and her colleagues published the first mtDNA tree showing that all modern humans can be traced back to a single recent African female ancestor. The mtDNA finding was subsequently mirrored for the Y chromosome. The two genetic trees have revolutionised our view of our past. (These two small parts of the genome are a fraction of our genetic heritage. We may have thousands of other "common ancestors" corresponding with the 30,000 other genes in our genome.)
The fact that mutations occur at a constant (although random) rate made it possible to date not only the branches but the base of the tree-less than 200,000 years. This confirmed the garden of Eden theory that modern humans, arising in Africa, had only recently replaced all pre-existing human species throughout the world and sounded the death knell for the multiregional theory.
The Adam and Eve trees have also helped to answer the question of whether our ancestors interbred with older human species like Neanderthals; anatomically modern Cro-Magnon Europeans are known to have coexisted with Neanderthals for more than 10,000 years. MtDNA has now been sequenced from a number of well-preserved Neanderthal bones, revealing at least 18 mutational differences from our African genetic Eve. If there had been mixing of Cro-Magnon with Neanderthal mtDNA, their mtDNA could theoretically have continued into modern times. If Neanderthal mtDNA had survived in Europe such a great difference in sequence should have been detected among the thousands of modern people whose mtDNA has been sampled-but it has yet to be found. This absence does not, however, prove conclusively that interbreeding did not occur.
More crucially, the new genetic tools can confirm or reject controversial archaeological theories of migrations. Which exit route did we take out of Africa, how many exit points were there and where did we go next? The most important advance in answering these questions over the past five years has been the fine resolution of the tree by British geneticist Martin Richards and colleagues from Britain, New Zealand and Germany. This showed just one of the multiple African mtDNA branches peopling the whole of the rest of the world. Geneticists have confirmed a similar finding for several other gene trees, including the Y chromosome. This single branch pattern makes it extremely unlikely there was more than one successful exodus.
Where and when did we leave Africa?
Knowing where we left Africa may reveal not only when and where we went next but how each modern regional group is related to the others. Unfortunately, it is not quite as easy as that. Regional populations as in India, Europe and China are not the same as single branches of gene trees. The tree is in reality more like several strands of creeping ivy spreading and branching over the earth. One region will share strands of different genetic branches with neighbouring regions; but each region has its own unique new growth. It is the new twigs and leaves that grew on the older strands in different regions that tell us where people migrated to and where they went after that. Genetic branches, however, only have approximate dates, while climatic and archaeological events can be better dated. So the reconstruction of these ancient migrations has to be matched with archaeology, the dramatic effects of changing climate and natural geographic corridors and barriers.
There are only two routes out of sub-Saharan Africa to Asia: one up the Nile corridor through Egypt and the Suez to the Levant, and the other to the south, across the mouth of the Red sea and along the Arabian coast to modern-day Yemen and Oman. For most of the past 100,000 years, the Syrian and Arabian deserts separated south Asia from the Levant and Europe. So taking the northern route meant that emigrants could only go further north to Europe and the Caucasus. Taking the southern route meant continuing along the coast of the Indian ocean to India, the far east and Australia.
Archaeology and climate both favour the southern route (see map page 53). Australia was colonised at least 20,000 years before Europe. If there was only one exodus through the northern route, Europe should have been colonised earlier. My suspicion is that Europe was colonised late because the ancestors of west Eurasians had to settle somewhere in south Asia, like the Arabian gulf, until the climatic amelioration that began 50,000 years ago allowed them to make their way north to the east coast of the Mediterranean. From there, they could enter Europe. The genetic dating is consistent with this pattern.
There are other strands to the climatic argument. Between 50,000 and 80,000 years ago the world was much drier than today, sea levels were lower, and the short crossing of the mouth of the Red sea directly from sub-Saharan Africa would have been less hazardous than crossing the Sahara desert to get to north Africa. Evidence of systematic beachcombing, stretching back as far as 125,000 years, has recently been found near the mouth of the Red sea on the west coast. So a means of survival along the Arabian coast was definitely available to southern migrants. Increasing salinity of the Red sea as a result of falling sea levels, a shallow mouth and increased evaporative loss may have prompted the move about 80,000 years ago.
The evidence from the distribution of non-African genetic branches is the clincher for the southern route. We might reasonably suppose that our emigrant band from Africa would leave a trace of their early genetic branches at the start of their trail. There is no evidence for this through north Africa to the Levant or Europe. In these places, we see only derivative genetic branches dating from after 50,000 years ago, agreeing with the archaeological evidence for a later colonisation. Furthermore, one of the two earliest mtDNA branches outside Africa, the Asian M group, is virtually absent here. By contrast, when we look at India, the first major dispersal point along the southern route, we find all the earliest genetic branches outside Africa.
So a picture is emerging of a single exit by the southern route about 80,000 years ago and a rapid spread of beachcombers around the Indian ocean over land bridges through Indonesia to Bali. From there a few short island hops could have taken our migrants to Timor. The final stretch was more of a problem. Today there are 300 miles of sea between Timor and Australia. But 65-70,000 years ago a severe glaciation briefly locked up enough water to lower the sea level over 80 metres, taking the coast of Timor to within 100 miles of Australia. Archaeological evidence for the earliest occupation of Australia by modern humans suggests this must have been the only time they could have got across. To have reached Australia by that time, our ancestors would need to have left Africa at least 10,000 years earlier-80,000 years ago.
Other dates further back on the trail support the 80,000 year exodus claim. A famous, anatomically modern human skull from Liujiang in south China was recently redated to over 70,000 years ago. There is increasing consensus that certain pebble tools only appeared in southeast Asia with the arrival of modern humans over 70,000 years ago. One group of these tools was found in Kota Tampan in the Malay peninsula, encased in volcanic ash from the great Toba eruption. That volcanic eruption has been precisely dated to 74,000 years ago, indicating that modern humans had arrived in the Malay peninsula, halfway to Australia, by this time. Again, this makes the 80,000-year exit from Africa look very reasonable.
Later events, such as the peopling of the Americas and the drama of the last ice age 18,000 years ago, are also described and illuminated by the new genetic tools.
Why do humans look so different?
The implications of only one exodus are enormous. In particular, it means all non-African peoples-including Europeans, Indians, Chinese, Australians and native Americans-are related and recent descendants of that one small family band. This conclusion raises the question: "Why do they all look so different?" The answer is that most differences are only skin deep. Our acute sense of facial recognition tends to exaggerate anatomical differences in physiognomy. In evolutionary terms, changes of skin and hair colour from black to brown, blond or white are relatively minor events and can take place in any human population over a period as short as 20,000 years. Other animal species can also show enormous differences in skin and hair pigmentation. The black panther, for instance, is just a black or "melanistic" leopard, and several arctic animals change pelt colour from white to coloured with season.
Our skin and hair colour is the effect of melanin, a brown pigment. Melanin production is controlled by only a few genes. The normal or original state for modern humans was probably black because melanin protects against the harmful effects of the sun-lethal skin cancers in particular-in the tropics. Melanin genes can mutate in ways that reduce the amount of melanin our skin and hair follicles make. In northern Europe and northern Asia these mutations are beneficial and have been selected for, because too much melanin in a low-sun environment prevents production of vitamin D, leading to rickets, which can be fatal in infants. Over thousands of years, people with paler skin are naturally selected to survive better in the north. This is the most likely explanation for the pale skin of Europeans and northern Asians.
As for the rest of the physical differences between races, the bulk are probably the result of isolation and so-called "genetic drift." Variation within regional groups is often as great as that between them. Overall, the groups that have changed facial structure least since the exodus are western Eurasians (including Europeans), southern Indians, some Malaysian aborigines and New Guinean highlanders. Those that have changed most are perhaps the so-called Mongoloids (including native Americans), possibly as an adaptation to a harsh, cold Palaeolithic environment on the central Asian steppe, though the genetic basis of such differences has yet to be studied.
The new genetic story of our origins illuminates the trail of our wanderings from a single exodus. It also reveals the truth that the modern races of humanity represent a small, closely related cluster that expanded very recently to replace all the diverse species of humans that had walked and talked before us.
Find out more about Stephen Oppenheimer's work here, at the Bradshaw Foundation website
Discuss this article at First Drafts, Prospect's blog
