Spencer Wells, Genetic Anthropologist, on the first Great Migrations

Peter Myers, September 4, 2008; update April 10, 2012.

Write to me at contact.html.

You are at http://mailstar.net/wells-genetics.html.

Spencer Wells is a Genetic Anthropologist. He is Director of the Genographic Project, operated by National Geographic to capture a "genetic snapshot" of humanity, before Globalization blurs our genetic trails. Geneticists in many countries - India, China etc - are participating.

In his book The Journey of Man, Wells shows that Europe's ancestry derives mainly from people in that continent around 30,000 years ago; not from early agriculturalists in the Middle East. But there was an invasion of Aryans from the steppes, which imposed the Indo-European languages on Europe and northern India.

Before that, Basque might have been more typical of European languages.

Spencer Wells discovered a genetic marker, M17, which is the signature of the Aryan invaders from the steppe into east & central Europe and northern India. See pp. 166-7 of The Journey of Man (2002) below.

More detail is provided from Spencer Wells' latest book, Deep Ancestry (2006).

UNESCO held a conference on the "First Great Migrations of Peoples" in Paris on June 19th 2008: http://www.firstgreatmigrations.org/.

(1) The Journey of Man: A Genetic Odyssey, by Spencer Wells
(2) Deep Ancestry: Inside the Genographic Project, by Spencer Wells

(1) The Journey of Man: A Genetic Odyssey, by Spencer Wells

The Journey of Man: A Genetic Odyssey

Spencer Wells.

ALLEN LANE an imprint of PENGUIN BOOKS

London 2002

{See the associated map at wells-Y-marker-map.jpg}

{p. 27} Inside each of our cells, what we think of as our genome - the complete DNA sequence that encodes all of the proteins made in our bodies, in addition to a lot of other DNA that has no known function - is really present in two copies. The DNA is packaged into neat, linear components known as chromosomes - we have twenty-three pairs of them. Chromosomes are found inside a cellular structure known as the nucleus. One of the main features of our genome is the astounding compartmentalization - like computer folders within folders within folders. In all there are 3,000,000,000 (3 billion) building blocks, known as nucleotides (which come in four flavours: A, C, G and T), in the human genome, and we need some way to get at all of the information it contains in a straightforward way.

{p. 28} The reason we have two copies of each chromosome is more compli- cated, but it comes down to sex. When a sperm fertilizes an egg, one of the main things that happens is that part of the father's genome and part of the mother's genome combine in a 50:50 ratio to form the new genome of the baby. Biologically speaking, one of the reasons for sex is that it generates new genomes every generation. The new combinations arise, not only at the moment of conception with the 50:50 mixing of the maternal and paternal genomes, but also prior to that, when the sperm and egg themselves are being formed. This pre-sexual mixing, known as genetic recombination, is possible because of the linear nature of the chromosomes - it is relatively easy to break both chromosomes in the middle and reattach them to their partners, forming new, chimeric chromosomes in the process. The reason why this occurs, as with the mixing of Mum's and Dad's DNA, is that it is probably a good thing, evolutionarily speaking, to generate diversity in each generation. If the environment changes, you'll be ready to react.

But wait, you might say, why are these broken and reattached chromosomes any different from the ones that existed before? They were supposed to be duplicates! The reason, quite simply, is that they aren't exact copies of each other - they differ from each other at many locations along their length. They are like duplicates of duplicates of duplicates of duplicates, made with a dodgy copying machine that introduces a small number of random errors every time the chromo- somes are copied. These errors are the mutations mentioned above, and the differences between each chromosome in a pair are the polymorphisms. Polymorphisms are found roughly every 1,000 nucleotides along the chromosome, and serve to distinguish the chromosomes from each other. So, when recombination occurs, the new chromosomes are different from the parental types.

The evolutionary effect of recombination is to break up sets of polymorphisms that are linked together on the same piece of DNA. Again, this diversity-generating mechanism is a good thing evolutionarily speaking, but it makes life very difficult for molecular biologists who want to read the history book in the human genome.

{p. 29} Recombination allows each polymorphism on a chromosome to behave independently from the others. Over time the polymorphisms are recombined many, many times, and after hundreds or thousands of generations, the pattern of polymorphisms that existed in the common ancestor of the chromosomes has been entirely lost. The descendant chromosomes have been completely shuffled, and no trace of the original deck remains. The reason this is bad for evolutionary studies is that, without being able to say something about the ancestor, we cannot apply Ockham's razor to the pattern of polymorphisms, and we therefore have no idea how many changes really distinguish the shuffled chromosomes. At the moment, all of our estimates of molecular clocks are based on the rate at which new polymorphisms appear through mutation. Recombination makes it look like there have been mutations when there haven't, and because of this it causes us to overestimate the time that has elapsed since the common ancestor.

One of the insights that Wilson and several other geneticists had in the early 1980s was that if we looked outside of the genome, at a small structure found elsewhere in the cell known as the mitochondrion, we might have a way of cheating the shuffle. Interestingly, the mitochondrion has its own genome - it is the only cellular structure other than the nucleus that does. This is because it is actually an evolutionary remnant from the days of the first complex cells, billions of years ago - the mitochondrion is what remains of an ancient bacterium which was swallowed by one of our single-celled ancestors. It later proved useful for generating energy inside the cell, and now serves as a streamlined sub-cellular power plant, albeit one that started life as a parasite. Fortunately, the mitochondrial genome is present in only one copy (like a bacterial genome), which means that it can't recombine. Bingo. It also turns out that, instead of having one polymorphism roughly every 1,000 nucleotides, it has one every 100 or so. To make evolutionary comparisons we want to have as many polymorphisms as possible, since each polymorphism increases our ability to distinguish between individuals. Think of it this way: if we were to look at only one polymorphism, with two different forms A and B, we would sort everyone into two groups, defined only by variant A or variant B. On the other hand, if we looked at ten polymorphisms with two variants each, we would have much better resolution, since the likelihood of

{p. 30} multiple individuals having exactly the same set of variants is much lower. In other words, the more polymorphisms we have, the better our chances of inferring a useful pattern of relationships among the people in the study. Since polymorphisms in mitochondrial DNA (mtDNA) are ten times more common than in the rest of our genome, it was a good place to look.

Rebecca Cann, as part of her PhD work in Wilson's laboratory, began to study the pattern of mtDNA variation in humans from around the world. The Berkeley group went to great lengths to collect samples of human placentas (an abundant source of mtDNA) from many different populations - Europeans, New Guineans, Native Americans and so on. The goal was to assess the pattern of variation for the entire human species, with the aim of inferring something about human origins. What they found was extraordinary.

Cann and her colleagues published their initial study of human mitochondrial diversity in 1987. It was the first time that human DNA polymorphism data had been analysed using parsimony methods to infer a common ancestor and estimate a date. In the abstract to the paper they state the main finding clearly and succinctly: 'All these mitochondrial DNAs stem from one woman who is postulated to have lived about 200,000 years ago, probably in Africa.' The discovery was big news, and this woman became known in the tabloids as Mitochondrial Eve - the mother of us all. In a rather surprising twist, though, she wasn't the only Eve in the garden - only the luckiest.

The analysis performed by Cann and her colleagues involved asking how the mtDNA sequences were related to each other. In their paper they assumed that if two mtDNA sequences shared a sequence variant at a polymorphic site (say, a C at a position where the sequences had either a C or a T), then they shared a common ancestor. By building up a network of the mtDNA sequences - 147 in all - they were able to infer the relationships between the individuals who had donated the samples. It was a tedious process, and involved a significant amount of time analysing the data on a computer. What their results showed were that the greatest divergence between mtDNA sequences was actually found among the Africans - showing that they had been diverging for longer. In other words, Africans are the oldest group on the planet - meaning that our species had originated there.

{p. 31} One of the features of the parsimony analysis used by Cann, Stoneking and Wilson to analyse their mtDNA sequence data is that it inevitably leads back to a single common ancestor at some point in the past. For any region of the genome that does not recombine - in this case, the mitochondrion - we can define a single ancestral mitochondrion from which all present-day mitochondria are descended. It is like looking at an expanding circle of ripples in a pond and inferring where the stone must have dropped - in the dead centre of the circle. The evolving mtDNA sequences, accumulating polymorphisms as they are passed from mother to daughter, are the expanding waves, and the ancestor is the point where the stone entered the water. By applying Zuckerkandl and Pauling's methods of analysis, we can 'see' the single ancestor that lived thousands of years ago, and which has mutated over time to produce all of the diverse forms that exist

{p. 32} today. Furthermore, if we know the rate at which mutations occur, and we know how many polymorphisms there are by taking a sample of human diversity from around the globe, then we can calculate how many years have elapsed from the point when the stone dropped - in other words, to the ancestor from whom all of the mutated descendants must have descended.

Crucially, though, the fact that a single ancestor gave rise to all of the diversity present today does not mean that this was the only person alive at the time - only that the descendant lineages of the other people alive at the same time died out. Imagine a Provencal village in the eighteenth century, with ten families living there. Each has its own special recipe for bouillabaisse, but it can only be passed on orally from mother to daughter. If the family has only sons, then the recipe is lost. Over time, we gradually reduce the number of starting recipes, because some families aren't lucky enough to have had girls. By the time we reach the present century we are left with only one surviving recipe - la bouillabaisse profonde. Why did this one survive? By chance - the other families simply didn't have girls at some point in the past, and their recipes blew away with the mistral. Looking at the village today, we might be a little disappointed at its lack of culinary diversity. How can they all eat the same fish soup?

Of course, in the real world, no one transmits a recipe from one generation to the next without modifying it slightly to fit her own tastes. An extra clove of garlic here, a bit more thyme there, and voila! - a bespoke variation on the matrimoine. Over time, these variations on a theme will produce their own diversity in the soup bowls - but the recipe extinction continues none the less. If we look at the bespoke village today we see a remarkable diversity of recipes - but they can still be traced back to a single common ancestor in the eighteenth century, thanks to Ock the Knife. This is the secret of Mitochondrial Eve.

The results from the 1987 study by Cann and her colleagues were followed up by a more detailed analysis a few years later, and both studies pointed out two important facts: that human mitochondrial diversity had been generated within the past 200,000 years, and that the stone had dropped in Africa. So, in a very short period of time - at least in evolutionary terms - humans had spread out of Africa to

{p. 33} populate the rest of the world. There were some technical objections to the statistical analysis in the papers, but more extensive recent studies of mitochondrial DNA have confirmed and extended the conclusions of the original analysis. We all have an African great-great ... grandmother who lived approximately 150,000 years ago.

{p. 42} The further apart the polymorphisms are, the more likely it is that they have been shuffled. And because shuffling obscures the historical signal, this means that most of our genome isn't terribly useful for tracing migrations.

There is one piece of DNA, though, that has recently proven to be an invaluable tool for inferring details about human history - providing us with far greater resolution than we ever thought possible about the paths followed by our ancestors during their wanderings. It is the male equivalent of mtDNA, in that it is only passed from father to son. For this reason, it defines a uniquely male lineage - a counterpart to the female line illuminated by studying mtDNA. It is the patrimoine in our Provencal village, and the details of lineage extinction and diversification that went on with the soup recipes also apply to this piece of DNA. It is known as the Y-chromosome.

Now wait a minute, you might be saying - what's going on with all of this maternal and paternal lineage gibberish? I thought that the whole idea of sex was to mix the mother's and father's genomes in a 50: 50 ratio to produce the child? Why do we have these oddities that break the rules? For the mitochondrial DNA the answer is easy - it is actually outside of what we think of as the human genome, an evolutionary remnant of a time when it was a parasitic bacterium living inside the earliest cells. The story for the Y is a bit more complicated.

One of the quirky features of sexual reproduction is that the chromo- somes that actually determine our sex - the so-called sex chromosomes - are exceptions to the 50: 50 sexual mixing rule. The double layout of our genomes, with two copies of each chromosome, fails us when we get to these chromosomes. This is because of the way in which sex is determined in most animals, through the presence of a mismatched sex chromosome. In the case of mammals, it is the male that is mismatched, with one X and one Y-chromosome. In females, the X-chromosome is present in two copies, like the other chromosomes, allowing normal recombination. In males, however, the Y only matches with the X in short regions at either end, which serve to align the sex chromosomes properly during cell division. The rest of the Y-chromosome, known as the non-recombining portion of the Y, is pretty much completely unrelated to the X. Thus it has no paired chromosome with

{p. 43} which it can recombine, and so it doesn't. It is passed unshuffled from one generation to the next, for ever - exactly like the mitochondrial genome.

The Y turns out to provide population geneticists with the most useful tool available for studying human diversity. Part of the reason for this is that, unlike mtDNA, a molecule roughly 16,000 nucleotide units long, the Y is huge - around 50 million nucleotides. It therefore has many, many sites at which mutations may have occurred in the past. As we saw in the last chapter, more polymorphic sites give us better resolution - if we only had Landsteiner's blood types to work with, everyone would be sorted into four categories: A, B, AB and O. To put it another way, the landscape of possible polymorphisms is simply much larger for the Y. And critically, because of its lack of recombination, we are able to infer the order in which the mutations occurred on the Y - just like mtDNA. Without this feature, we can't use Zuckerkandl and Pauling's methods to define lineages, and Ock the Knife can't help us with the ancestors.

{p. 44} As we have seen, we can only study human diversity by looking at differences - the language of population genetics is written in the polymorphisms that we all carry around with us. These differences define all of us as unique individuals - unless we have a twin, no other person in the world has an identical pattern of genetic polymorphisms. This is the insight behind a DNA 'fingerprint', used to identify criminals. Applied to the Y-chromosome, it allows us to trace a unique male lineage back in time, from son to father to grandfather, and so on. Taken to the extreme, it allows us to travel back in time from the DNA of any man alive today to our first male ancestor - Adam. But how does it link unrelated men to each other in regional patterns? Surely each man must trace his own unique Y-chromosome line back to Adam?

The answer is no, but the reason is a bit complicated. It's because we're not as unrelated as we think. Imagine the situation for the majority of our genome - the parts that don't come uniquely from our mother or our father. Since we inherit half of this DNA from each of our parents, the pattern of polymorphisms it contains can be used to infer paternity, since it connects us to both our mother and our father. If my DNA is shown in court to have a 50 per cent match with that of a child I've never met, it is likely that I will be paying for the support of that child for many years to come - the probability of a match

{p. 45} occurring by chance is infinitesimally small. So polymorphisms define us, and our parents, as part of a unique genealogical branch. No other group of people on earth has exactly the same story written in its DNA.

If we extend this further, and begin to think about our grandparents, and their grandparents, and so on, we lose some of the signal in each generation. I have a 50 per cent match with my father, but only a 25 per cent match with my grandfather, and only a 6 per cent match with his grandfather. This is because we acquire new ancestors in each generation as we go back in time, and they start to pile up pretty quickly. Each of my parents had two parents, and each of them had tvo parents, and so on. {p. 37} The earliest Homo Erectus fossils yet discovered date from around 1.8 million years ago, and they were all found in east Africa (the African)

{p. 38} variant of Homo erectus is sometimes given the name Homo ergaster). Recent discoveries in the medieval city of Dmanisi, in the former Soviet Republic of Georgia, show that they left Africa soon thereafter - perhaps reaching east Asia within 100,000 years. From this we can infer that all Homo erectus around the world last shared a common ancestor in Africa nearly million years ago. But according to the Berkeley mitochondrial data, Eve lived in Africa less than 200,000 years ago. How can we reconcile the two results?

It's all about timing

Let's step back for a moment and consider the case objectively. The evidence for an African Genesis of Homo erectus is circumstantial - we see evolutionary 'missing links' in Africa, either uniquely or first. These include an unbroken chain of ancestral hominids stretching back more than 5 million years to the recently discovered chimpanzee- like apes Ardipithecus. But is this evidence sufficient to conclude that Africa was also the birthplace of our species? ...

Rather, the conclusion from the mitochondrial data is that modern humans evolved very recently in Africa, and subsequently spread to populate the rest of the

{p. 39} globe, replacing our hominid cousins in the process. It's a ruthless business, and only the winners leave a genetic trace. Unfortunately, Homo erectus appears to have lost.

As we'll see, other genetic data corroborates the mitochondrial results, placing the root of the human family tree - our most recent common ancestor- in Africa within the past few hundred thousand years. Consistent with this result, all of the genetic data shows the greatest number of polymorphisms in Africa - there is simply far more variation in that continent than anywhere else. You are more likely to sample extremely divergent genetic lineages within a single African village than you are in whole of the rest of the world. The majority of the genetic polymorphisms found in our species are found uniquely in Africans - Europeans, Asians and Native Americans carry only a small sample of the extraordinary diversity that can be found in any African village.

Why does diversity indicate greater age? Thinking back to our hypothetical Provencal village, why do the bouillabaisse recipes change? Because in each generation, a daughter decides to modify her soup in a minor way. Over time, these small variations add up to an extraordinary amount of diversity in the village's kitchens. And - critically - the longer the village has been accumulating these changes, the more diverse it is. It is like a clock, ticking away in units of rosemary and thyme - the longer it has been ticking, the more differences we see. It is the same phenomenon Emile Zuckerkandl noted in his proteins - more time equals more change. So, when we see greater genetic diversity in a particular population, we can infer that the population is older - and this makes Africa the oldest of all.

{p. 56} While all African populations contain deeper evolutionary lineages than those found outside the continent, some populations retain traces of very ancient lineages indeed. These groups are found today in Ethiopia, Sudan and parts of eastern and southern Africa, and the genetic signal they contain is very good evidence that they are the remnants of one of the oldest human populations. The signals have been lost in other groups, but today these eastern and southern African groups still show a direct link back to the coalescence point - Adam.

The populations involved encompass the African Rift Valley, extending into south-western Africa, where people known as the San - formerly called Bushmen - have a very strong signal of the diversity that characterized the earliest human populations. They also speak one of the strangest languages on the planet, notable for its use of clicks as integrated parts of words - like the clicking sound we might make when we guide a horse, or imitate a dripping tap. ...

The pattern of deep genetic lineages within the San is also seen for mitochondrial DNA, and the convergence of these three independent

{p. 57} lines of evidence - Y, mtDNA and linguistic - strongly suggests that the San represent a direct link back to our earliest human ancestors. ...

One of the distinguishing features of the San people is their 'non- African' physical appearance. ... When most of us think of Africans, we tend to picture the typically Bantu features of central Africans ... The San are a much smaller people, with lighter skin, more tightly curled hair and a thicker layer of skin over the eyes - the so-called epicanthic fold that also characterizes people from east Asia. It is this latter feature which has led some researchers to suggest that the epicanthic fold is an ancestral characteristic of our species, and was simply lost in western Eurasian and Bantu populations.

{p. 58} It is unlikely that our African ancestors were the hairy, brutish troglodytes portrayed in museums - these are probably overly influenced by our perception of Neanderthals, who may have been pretty hairy and brutish. Rather, they are likely to have been fairly gracile and elegant, at least in comparison to Neanderthals. The simple reason is that the great mass of a Neanderthal, and the likely hairy exterior, is thought to have been an adaptation to the cold Eurasian climate. Because our earliest ancestors lived in the relatively warm climes of southern and eastern Africa, they would not have needed the warmth provided by a furry exterior....

Early humans probably had fairly dark skin. This is because of the nature of the environment where they lived - a sunny African savannah - where the protection against solar radiation afforded by dark skin would have been a distinct advantage. It is also because at least some of the mutations that produce light skin colour in Europeans and north-east Asians are derived from the ancestral, darker form of the gene (known as MCI R, or melanocortin receptor), which is virtually

{p. 59} the only form found in Africa today. Thus, it seems likely that Africans have retained a darker colour, rather than evolving it from a lighter form.

Our ancestors of 60,000 years ago were probably about the same height as you and I ...

So, the picture that emerges is of a dark-skinned (although perhaps not as dark as some Africans today), reasonably tall, thin person - perhaps with an epicanthic fold. Someone who wouldn't look that out of place today dressed in a suit and sitting opposite you on the train. Not surprising, I suppose, given that he only lived about 2,500 generations ago.

Out of the nest

Accepting the evidence at face value, the implication is that Adam lived in population groups directly ancestral to the modern San, in eastern and/or southern Africa, around 60,000 years ago. The date of the earliest modern human populations - the first of our species - remains to be assessed, and could be anywhere between 60,000 and several hundred thousand years ago. We simply lose the signal from our genes at that stage, as all of the genetic diversity present today coalesces to a single ancestor. What is clearly implied by the data, however, is that all modern human genetic diversity found around the world was in Africa around 60,000 years ago. The mtDNA and Y-chromosome give us the same dates for the earliest non-African genetic lineages, and it is now agreed by most geneticists that humans began to leave Africa around this time. There may have been occasional

{p. 60} forays into the Middle East prior to this, as suggested by 100,000-year- old human remains at sites such as Qafzeh and Skuhl in present-day Israel, but the Levant of 100,000-15O,000 years ago was essentially an extension of north-eastern Africa, and was probably part of the original range of early Homo sapiens. The real expansion was beyond the Mediterranean world, into the uncharted territory of Asia proper.

Here we run headlong into what the Australians might call 'a curly one'. According to the dated remains in Australia, humans were there, 15,000 km east of Africa by the shortest land route, at the same time we are all supposed to have been in Africa, 50-60,000 years ago.

{p. 66} Africa is the most equatorial continent on earth. The entirety of its landmass is found between latitudes of 38°N and 34°S, and 85 per cent of its land area is in the tropical zone between Cancer and Capricorn. Sea-level freezing temperatures are rare in Africa - uniquely among all the continents.

{p. 68} Recent research by Robert Walter, an American geophysicist based in Mexico, suggests that a large-scale drying up of the African continent at the onset of the last ice age resulted in modern humans favouring coastal environments. This is because savannahs are unusual places. They are closely related to tropical forests in the chain of climatic relationships, and the two zones are interchangeable depending on the level of rainfall. In general, the areas of tropical Africa with more than three months of low rainfall are savannah, while those with fewer than three are forest. If there are substantially longer dry periods, the environment grades into steppe, and ultimately into desert as moisture becomes extremely scarce. While these regions are all found in particular locations in present-day Africa, their past extent has fluctuated. What Walter's research suggests is that as Africa began to dry up, the savannahs of eastern Africa were replaced by steppe and desert, except in a narrow zone near the coast. It was in these coastal savannah environments that early humans would have congregated, exploiting food sources from the sea as well as those of the land animals living near by.

While the universality of this theory is uncertain, and it may turn out to be a minor sideline of human evolution, one thing is clear: there is incontrovertible proof that early humans were able to live off of the sea. Large middens, or garbage dumps, of shells from clams and oysters have been found in Eritrea, on the eastern Horn of Africa, dating from around 125,000 years ago. These middens also have human stone tools interspersed among them, showing that humans were living in the region and exploiting coastal resources. The presence of butchered remains of rhinoceros, elephant and other large mammals conjures up a prehistoric 'surf 'n' turf' feast reminiscent of the massive platters of steak and shellfish served in American restaurants. It seems that our

{p. 69} distant ancestors had quite well-developed palates, even in those days of apparent hardship.

One of the most exciting details to emerge from Walter's work is the fact that there appears to have been exchange with coastal dwellers thousands of kilometres away, who were exploiting the same types of resources in southern Africa. This is suggested by the similarities in tools found at the sites, coupled with their roughly contemporary dates. It seems that humans were able to migrate over long distances, relatively rapidly, by following the coast of eastern Africa.

Now for the big leap: if humans could migrate over long distances within a continent, using the same technologies and exploiting the same resources, why couldn't they do the same between continents? The coastal route would be a sort of prehistoric superhighway, allowing a high degree of mobility without requiring the complex adaptations to new environments that would be necessary on an inland route. The resources exploited in Eritrea would be pretty much the same as those in coastal Arabia, or western India, or south-east Asia, or - wait for it - Australia. And because of the ease of movement afforded by the coast, the line of sandy highway circumnavigating the continents, this would allow relatively rapid migration. No mountain ranges or great deserts to cross, no need to develop new toolkits or protective clothing, and no drastic fluctuations in food availability. Overall, the coastal route seems infinitely preferable to anything further inland. There were only a couple of sections of open water that would have required a boat to cross. Most likely these boats would have been rather simple - probably a few logs lashed together - but we have no direct evidence, because wood disintegrates very quickly. Nevertheless, they did make it across.

It is clearly plausible that the early presence of humans in Australia, almost immediately after they left Africa, can be accounted for by migration along this coastal route - beachcombing along the southern coast of Asia. There are two remaining pieces of the puzzle to be evaluated, though - rather critical ones, in fact. If one of the early wavcs of migration out of Africa followed a coastal route, is there a telltale genetic pattern? It depends on the way in which the migration occurred, and what the migrants did along the way. We might expect to see a band of particular genetic markers along the coast,

{p. 70} differentiated from the populations living further inland. Or perhaps the signals have been homogenized among descendants of the coastal dwellers and the land migrants. The only way to find out is to examine populations from along the route and see what the genetic pattern is. The second critical piece of evidence is to be found in the pattern of archaeological remains along the route - are they consistent with such a journey?...

The beauty of the genetic data is that it gives us a clear, stepwise progression out of Africa into Eurasia and the Americas. The diversity we find around the world is divided into discrete, although related, units, defined by markers - the descendants of ancient mutation events. By mapping these markers on to the map of the world, we can infer details of past migrations. Following the order in which the mutations occurred, and estimating the date and any demographic details (such as population crashes or expansions), we can gain an insight into the details of the journey. And the first piece of evidence comes from one man in particular, who had a rather important, random mutation on his Y-chromosome between 31,000 and 79,000 years ago. He has

{p. 71} been named, rather prosaically, M168. More evocatively, he could be seen as the Eurasian Adam - the great ... great-grandfather of every non-African man alive today. The journeys taken by his sons and grandsons defined the subsequent course of human history.

It is perhaps surprising that the clearest evidence for the route followed by our ancestors on their journey out of Africa comes from the Y-chromosome - surely men tend to 'sow their oats', causing the widespread dispersal of regional genetic signals? Oddly enough, no - and the rapid loss of ancient soup recipes on the male lineage (which we used to explain Adam's recent date) means that men living in a particular area tend to share a recent common ancestor, providing us with clear 'fingerprints' of particular geographic regions. What this means is that the Y gives us the clearest evidence for the journeys followed by early humans. It is literally a 'journey of man', but it is the best tool we have for inferring the details of the trip. It is obviously important to examine the female lineage to see if it follows the same pattern - to make sure the fish stays with the bicycle, so to speak - but the Y-chromosome does provide us with the cleanest distillation of human migrational history.

As we look more carefully at the arrangement of branches on the mitochondrial tree, we find that there is a similar pattern - all of the non-African mitochondrial branches descend from a particular branch of the tree trunk, implying that our M168 Adam was paired with an Eve. Thankfully, this Eurasian Eve lived around 50-60,000 years ago, suggesting that she and Eurasian Adam could have met. She is called by the (again) rather mundane name L3, and her daughters accompanied the sons of M168 on their journey to populate the world.

Based on the distribution of the descendants of M168 and L3 in Africa today, it is likely that they both lived in north-east Africa, in the region of present-day Ethiopia and Sudan. Like all men alive today, M168 shared deeper roots with his African cousins. His lineage is a major branch leading off the human family tree, with his descendant 'terminal branches' found in the DNA of all of today's Eurasians, but he connects them back through M168 to our species' African root. In our tree metaphor, each marker that we study defines a node on the tree - a point where a branch splits into two smaller branches. If we had no markers apart from Ml68 and L3, our trees would be fairly sparse,

{p. 72} comprising a root (Adam and Eve) and one split on the tree, defined by M168 or L3, on the branch leading out of Africa, and another branch remaining in Africa. Luckily, the tree is packed with dense foliage, defining a pattern of growth that traces the map of our journey.

Intriguingly, on both the mitochondrial and Y branches, there is another split, immediately after M168 and L3, dividing the Eurasian branch structure into distinct clusters - two in the case of mtDNA, and three in the case of the Y. For both the Y-chromosome and mtDNA, one cluster is more common than the other(s), accounting for around 60 per cent of the non-African branches (or lineages) in the case of mitochondrial DNA, and more than go per cent in the case of the Y. In other words, the majority of non-Africans alive today have mtDNA and Y-chromosomes belonging to the more numerous clusters - people living all over the world, in places as disparate as Europe, India and South America. The rarer lineages, though, are found only in Asia, Australasia and the Americas. It is these rare lineages that constitute the majority of the mitochondrial and Y types in the Australian Aborigines.

Our rare mitochondrial cluster is given the name M - like the head of MI6 in James Bond movies. In biblical terms, Eve begat L3, and L3 begat M. According to recent research by Lluis Quintana-Murci, a Catalan researcher working in Paris, the distribution of the M cluster is indicative of an early migration out of Africa, which proceeded along the coast of south Asia, ultimately reaching south-east Asia and Australia. M is virtually absent from the Middle East, and is not found at all in Europe, but it constitutes 20 per cent or more of the mitochondrial types in India, and close to 100 per cent of those in Australia. Quintana-Murci estimates its age to be 50-60,000 years, and from its distribution it seems that people who carried the M lineage never made it into the interior parts of the Middle East. The most

[p. 73} likely explanation is that the 'M people' left Africa very early on, carrying their distinctive genetic signature across the south of the continent along the coastal highway.

And what about the Y? Is there a male counterpart to our M mitochondrial lineage? Luckily, the answer is yes. Again assuming biblical style, Adam begat M168, and M168 begat M130. M130 appears to have accompanied mitochondrial M on her coastal journey, and the present-day distribution of his descendants provides us with an insight into the nature of the trip. Like the M mitochondrial lineages, M130 Y-chromosomes are limited to Asia and America, but the dynamics of lineage extinction that we see for the Y have left a much more striking pattern than the one seen for their mitochondrial counterparts. M130's descendants are virtually unknown west of the Caspian Sea, but they comprise a substantial proportion of the men living in Australasia. M130 is only found at low frequency in the Indian subcontinent - 5 per cent or less. But as we move further east,

{p. 74} the frequency increases: 10 per cent of Malaysian, 15 per cent of New Guinean and 60 per cent of Australian aboriginal men trace their ancestry directly to M130. There is a quirkily high frequency of M130 in north-east Asia, particularly in Mongolia and eastern Siberia, which suggests a later migration that we will revisit in Chapter 7. For the purposes of our Australian story, though, M130 provides us with a clear fingerprint of the coastal migration out of Africa.

One other piece of evidence suggests a direct link between Africa and Australasia - physical appearance. The dark skin of the Australians is reminiscent of that found in Africa - something that begs an explanation. Most of the people living in south-east Asia today would be classified as 'Mongoloid' peoples, implying a shared history with those living further north in China and Siberia. There are, however, isolated populations of so-called Negritos living throughout south-east Asia who closely resemble Africans. The most obvious examples are found in the Andaman Islands, under the jurisdiction of India but

{p. 75} actually 400 km off the west coast of Thailand. The largest tribal groups, known as the Onge and Jarawa, have many features that link them with the Bushmen and Pygmies of Africa, including short stature, dark skin, tightly curled hair and epicanthic folds. Other Negrito groups, such as the Semang of Malaysia and the Aeta of the Philippines, have mixed substantially with Mongoloid groups and have a more 'Asian' appearance. The Andamanese, probably because of their island home, have escaped much of the admixture seen on the mainland. Because of this they are thought to represent a relic of the pre-Mongoloid population of south-east Asia - 'living fossils', if you will. The suggestion made by many anthropologists, particularly Peter Bellwood of the Australian National University, is that the population of south-east Asia prior to 6,000 years ago was composed largely of groups of hunter-gatherers very similar to modern Negritos. Migrations from north-east Asia over the past few millennia have erased the evidence of these early south-east Asians, except in the case of small groups living deep in the jungles or - in the case of the Andamanese - on remote islands.

So, both the Y-chromosome and the mtDNA paint a clear picture of a coastal leap from Africa to south-east Asia, and onward to Australia. Taking the genetic dates as a guide, modern humans could have made this journey around the same time as the earliest archaeological evidence pointing to human occupation in Australia. DNA has given us a glimpse of the voyage, which almost certainly followed a coastal route via India. But is there any archaeological trace of this journey along the route?

A swim in Ceylon

This brings us back to the issue of the dating, particularly as applied to the Australian remains. No evidence for other hominids has ever been found in Australia - Homo erectus did not make it across the long stretches of open ocean that separated it from south-east Asia, despite living only a few hundred kilometres away in Java. Because Homo sapiens is the only hominid species that has ever been found in Australia, any evidence of human occupation sticks out like a

{p. 76} proverbial sore thumb. Stone tools unearthed in Arnhem Land could only have come from one source - us. And if the radiometric dates say that stone tools were present in Australia 50-6,0000 years ago, almost immediately after the genetic dates show us that our ancestors were still in Africa, this means that modern humans must have made use of a route that afforded extremely rapid movement. The coastal superhighway seems to be the most likely one.

As we have seen, though, there were other hominids living along the route followed by these beach dwellers. They also made stone tools, and these have been found throughout Eurasia. The easternmost extension of the range of Homo erectus was Java, and it is possible that they even survived until around 40-5,0000 years ago - long enough for the coastal migrants to have encountered them as they moved through the Indonesian archipelago. It is clear, though, that they must have become extinct almost immediately after the arrival of the Moderns, if not before. What is uncertain is whether we actively forced them out of - the picture - a genocidal scenario that we will explore in greater detail when we get to Europe later in the book.

{p. 77} Even before the current era of globalization, the world had its 'killer apps' that dominated everything else. In the case of the period we are talking about, 50-6,0000 years ago, the killer apps are grouped into a common cultural phenomenon known as the Late Stone Age, or more technically, the Upper Palaeolithic. The tools of the Upper Palaeolithic mark a radical departure from those that pre-date them, and are clear evidence for the presence of anatomically modern humans, as opposed to Homo erectus or Neanderthals, who remained trapped in a Middle Palaeolithic time-warp.

The details of the Middle to Upper Palaeolithic transition will be examined in the next chapter, but for the purpose of the story of our Australian coastal dwellers it is sufficient to say that the earliest Upper Palaeolithic tools mark the initial migration of modern humans into any geographic region. And that is why India is unusual, since there is actually very little evidence of the Upper Palaeolithic there. There is a general dearth of human remains from all periods leading up to the Upper Palaeolithic, but at least there are abundant tools from the earlier periods. The Upper Palaeolithic provides no telltale signs until very late in the day, and even then they show up in an unexpected place.

Fa Hien cave in Sri Lanka provides us with the earliest sign of the Upper Palaeolithic in the Indian subcontinent. The date, however, is a problem - the earliest clearly modern artefacts date from no earlier than 31,000 years ago. Nearby Batadomba Lena cave contains the earliest skeletal material from anatomically modern humans, also dating from around 30,000 years ago. The combination of age and location gives us two clues in our search for traces of the coastal migration. First, the Sri Lankan caves suggest that the earliest modern humans arrived in India from the south, rather than from the north via the more obvious inland route. This implies that they were living on the coast, consistent with the theory of an early coastal migration.

The second clue, which comes from the date, is that the Batadomba people could not have been the ancestors of the Australians, since they actually lived over 20,000 years after the earliest evidence for human settlement in Arnhem Land. Another curly one. It may turn out that archaeological layers below those already excavated will yield earlier

{p. 78} evidence for modern human presence, but for the time being it appears that Batadomba is too late to help us along on our voyage. In fact, late dates are found along the entirety of our coastal route to Oz. In Thailand, for instance, there is evidence for modern human occupation from about 37,000 years ago at Lang Rongrien cave - but not before. As we move closer to the scene of the crime, the dates get older - advanced, Upper Palaeolithic stone tools dating from 40,000 years ago have been found at Bobongara, on the Huon Peninsula of eastern New Guinea. This would have been the final stepping-stone on the journey, but there is still nothing approaching the 50-60,000-year-old dates in Australia. Thus, in spite of the genetic pattern tracing an early coastal route out of Africa, the archaeology appears to have failed us. Where is the evidence for our coastal route?

Unfortunately we don't know, but there is a likely hypothesis. Since almost all archaeological work today is carried out on land, we are probably missing the artefacts that are hidden underwater. 'Rubbish - surely Atlantis is a myth!' you might be saying. Well, yes and no. While the evidence for an entire civilization falling catastrophically into the sea is fairly sparse, what is unequivocal is that sea levels have indeed fluctuated substantially - if somewhat more gradually - over the past 100,000 years. Those of 50,000 years ago were around 100 metres lower than they are today, as large amounts of moisture were tied up in the expanding ice sheets of the northern hemisphere. This may not sound like much of a difference, but remember that we are not as interested in the depth as we are in the extent of the land that would have been exposed by these fluctuations. Since the continents typically have very shallow slopes as they fall off into the sea, a difference of 100 metres can make a huge difference in the amount of land exposed. For example, a drop in sea level of this magnitude would expose as much as 200 km of land off the west coast of India. Sri Lanka and India would have been connected by a land bridge, the Persian Gulf and the Gulf of Thailand would have been fertile river deltas, and Australia and New Guinea would have been two bulbous extremities of a single landmass. All in all, our entire coastal route would have been much different 50,000 years ago.

What the recent sea-level rise means is that, if our coastal voyagers were living primarily off resources provided by the sea, the places

{p. 79}where they chose to live would have been those that are now underwater. The Y-chromosome pattern in Eurasia shows that our M130 coastal marker is found predominantly in the southern and eastern parts of the continent. Furthermore, M130 chromosomes in the south appear to be older than those found further north, suggesting a later migration originating in the tropics. These results, coupled with a lack of archaeological evidence for modern human occupation until after 40,000 years ago, suggest that the early coastal migrants did not stray far from the sea. Adapted to a coastal lifestyle, the surfers do not appear to have made significant colonial forays into the turf. Knowing this, it would seem more appropriate for archaeologists in search of the first Indians to be wearing scuba gear rather than pith helmets. It is likely that the earliest Upper Palaeolithic tools in the subcontinent will be found underneath thousands of years' worth of sand and coral growth.

{p. 108} The first Upper Palaeolithic humans may have reached the Middle East during the relatively warm and moist conditions around 50,000 years ago, when the eastern Sahara was in retreat and a gateway opened along the Red Sea. Perhaps they migrated down the Nile to the Mediterranean, then spread eastward across the Sinai peninsula. Alternatively, early human populations may have moved across the strait of Bab al Mandab into southern Arabia, a short hop of 20 km or so. Once there, the relatively moist conditions along the coastal mountain range of western Arabia - which served to scoop moisture from the prevailing westerly winds coming off the Red Sea - may have created savannah-like hunting conditions for these Upper Palaeolithic people. Even today there is a narrow strip of steppe extending as far north as the city of Medina in Saudi Arabia, unique in the harsh environment that defines most of the Arabian peninsula. In the past, this tenuous steppe environment may have been joined with its ecological equivalent extending southward from the Gulf of Aqaba in Jordan, effectively opening a door to the interior of Eurasia.

William Calvin, a neurobiologist who has written extensively on climate and early human evolution, has compared the Sahara to a kind of hominid 'pump'. During wetter periods, the Sahara would have sustained human populations, perhaps focused around oases or rivers, or limited to zones that received moisture from prevailing winds. As the conditions turned drier, the Sahara would have returned to uninhabitable desert, forcing human emigration. Calvin suggests that the climatological downturn after 50,000 years ago may have been the impetus for the migration of Upper Palaeolithic humans out of northern Africa and into the Middle East.

However the earliest Upper Palaeolithic moderns reached the Levant, it is clear that the deteriorating climate after 45,000 years ago effectively locked them into their new home. The Sahara would have been at its driest between 40,000 and 20,000 years ago, and it is likely that any previously inhabitable areas there would have been engulfed by desert during this time. Modern humans were trapped in a new continent.

{p. 109} The genetic pattern bears this out, and provides the next clue on our journey. M89, the marker that occurred immediately after M168 on our main line into Eurasia, has been dated using the absolute method detailed above to around 40,000 years ago. Due to possible errors in the assumptions that go into the calculation, particularly in determining the rate at which new mutations occur, this estimate actually encompasses a range between 30,000 and 50,000 years, and it is likely (given the climatic data) that it appeared at the earlier end of this range, perhaps 45,000-50,000 years ago. This is because it serves to unite populations living in north-eastern Africa - Ethiopia and Sudan in particular - with the populations of the Levant. The shutting of the Saharan gate after these M89-bearing populations were allowed through is suggested by the low frequency in north-eastern Africa of Eurasian markers that occurred later on the M89 lineage. If Africa and the Levant had been part of a continuous range occupied by humans throughout the Upper Palaeolithic, we would expect to see a relatively homogeneous distribution of markers throughout. In fact, it seems that the emigration of populations bearing M89, which we can call a Middle Eastern marker, signified the last substantial Upper Palaeolithic exchange between sub-Saharan Africa and Eurasia. The world had been divided into African and Eurasian, and it was to be tens of thousands of years until significant exchange was to take place again.

The presence of M89 in both north-eastern Africa and the Middle East, and the age of the Upper Palaeolithic archaeological sites in the Levant, helps us to answer the question of whether Eurasia was settled in a single southern coastal emigration from Africa. M130 chromosomes are not found in Africa, suggesting that this coastal marker arose on an M168 chromosome en route to Australia. Conversely, M89 Y-chromosomes are not found in Australia or south-east Asia - but they appear at fairly high frequency in north-eastern Africa. The implication is that M89 appeared slightly later than M130 in a population that stayed behind in Africa after the coastal migrants left for Australia. It was these people, sans M130 chromosomes, who first colonized the Middle East. There is archaeological evidence for a modern human presence in the Levant from around 45,000 years ago, consistent with the arrival of modern humans from somewhere else. North-eastern Africa is the only nearby location with archaeological

{p. 110} sites dating from around the same time - and, crucially, the same genetlc markers we see in the Levant. Thus, the genetic and archaeological patterns tell us that there was a second migration from Africa into the Middle East.

Once our Upper Palaeolithic migrants had arrived in the Levant, the road into the heart of Eurasia was open. There was a continuous highway of steppe - not unlike African savannah in terms of its species composition - that stretched from the Gulf of Aqaba to northern Iran, and beyond into central Asia and Mongolia. The hurdle of the Sahara having been overcome, the subsequent dispersal of these fully modern humans would have been limited only by their own wanderlust. They had all of the intellectual building blocks that would enable them to conquer the continent, and the process began with gradual migrations along this Steppe Highway, the continental equivalent of the southern Coastal Highway.

At this time, game would have been plentiful. The large, grazing mammals of the steppe zone - particularly antelope and bovids, the ancestors of the domestic cow - would have been easy prey for early humans, and they gradually expanded their range as their numbers

{p. 111} grew. Moving northward and westward, some may have entered the Balkans early on - the first modern humans in Europe. The numbers would not have been great, though, since it was far easier to stay within the bounds of the steppe zone to which they had become so well adapted. The mountains and temperate forests of the Balkan peninsula would have seemed rather alien to early Upper Palaeolithic people, and the genetic data bears this out. Very few Europeans trace their ancestry directly to the Levant of 45,000 years ago, as attested to by the Y-chromosome results. Our canonical Levantine Upper Palaeolithic lineage, M89, is found at frequencies of only a few per cent in western Europe. It may have been these few Middle Eastern immigrants who introduced the earliest signs of the Upper Palaeolithic to Europe, a culture known as the Chattelperronian, but they did not leave a lasting trace. The true conquest of Europe, and the demise of the Mousterian, would have to wait for a later wave of immigration - people with a few more ingredients in their genetic soup.

Eastward ho!

The main body of Upper Palaeolithic people began to disperse eastward. As with other early human migrations, it almost certainly wasn't a conscious effort to move from one place to another. Rather, it seems that the continuous belt of steppe stretching across Eurasia provided an easy means of dispersal, gradually following game further and further afield. It was during this time that another marker appeared on the M89 lineage, given the name M9. It was the descendants of M9, a man born perhaps 40,000 years ago on the plains of Iran or southern central Asia, who were to expand their range to the ends of the earth over the next 30,000 years. We will call the people carrying M9 the Eurasian clan.

As the steppe hunters migrated eastward, carrying Eurasian lineages into the interior of the continent, they encountered the most significant geographical bollards so far. These were the great mountain ranges that define the southern central Asian highlands - the Hindu Kush running west to east, the Himalayas running north-west to south-east and the Tien Shan running south-west to north-east. The three ranges

{p. 112} meet in the centre, at the so-called Pamir Knot in present-day Tajikistan, and each radiates off like a spoke in a wheel.

The first humans to see them must have been absolutely awe-inspired. Although they had encountered the Zagros range in western Iran, it was a permeable barrier, with numerous valleys and low passes that would have allowed easy movement. The Zagros themselves actually would have been part of the geographic range of the prey species hunted by Upper Palaeolithic people, with the herds migrating into higher pastures during the summer and descending to the surrounding plains in the winter. The high mountains of central Asia were a different beast altogether. Each of the ranges has peaks that soar to 5,000 metres or higher (in the case of the Tien Shan and Himalayas, over 7,000 metres), and the radiating high-altitude ridges would have been formidable barriers to movement. Remember that the world was in the grip of the last ice age, and temperatures would have been even more extreme than today. It was because of these mountains that our Eurasian migrants would have been split into two groups - one moving to the north of the Hindu Kush, the other to the south, into Pakistan

{p. 113} and the Indian subcontinent. How do we know this? The Y-chromoome again traces the route.

Those who headed north, toward central Asia, had additional mutations on their Eurasian lineage that we will trace below. The Upper Palaeolithic people who headed south, though, had an unrelated mutation on their Y-chromosome known as M20. It is not found a appreciable frequencies outside of India - perhaps 1-2 per cent in some Middle Eastern populations. In the subcontinent, though, around 50 per cent of the men in southern India have M20. This suggests that it marks the earliest significant settlement of India, forming a uniquely Indian genetic substratum - which we can call the Indian clan - that pre-dates later migrations from the north. The ancestors of the Indian clan, who moved into southern India around 30,000 years ago, would have encountered the earlier coastal migrants still living there. From the genetic pattern, it seems likely that any admixture with them was not reciprocal: as we saw in Chapter 4, mitochondrial DNA retains strong evidence of the coastal migrants in the form of haplogroup M, while the Y-chromosome primarily shows evidence of later migrants from the north. Thinking back to the scenario we imagined for the birth of the Upper Palaeolithic in Africa, this is the pattern we woul expect to see if the invaders took wives from the coastal population, but the coastal men were largely driven away, killed, or simply not given the chance to reproduce. The result would be the widespread introduction of M mtDNA lineages into the Indian population, while the Coastal Y-chromosome lineages would not be nearly as common - precisely the pattern we see. Today, the frequency of the Coastal marker is only around 5 per cent in southern India, and it falls in frequency as we move northward. This pattern suggests that the contribution from the coastal populations was minimal, at least on the male side. The contrast between the two types of data gives us a glimpse the behaviour of these first Indians, and hints at a cultural pattern we will explore in more detail in Chapter 8.

The migrating Eurasian masses were not only shunted down in India, of course - some of them also migrated to the north of the Hindu Kush, into the heart of central Asia. The Tien Shan would have been an even more formidable barrier than the Hindu Kush, keeping the Upper Palaeolithic hunters out of western China. It is around this

{p. 114} time that another mutation occurred on the Eurasian lineage. It was known as M45, and it will help us to trace two very important later migrations. Using absolute dating methods, we can infer that the M45 mutation occurred approximately 35,000 years ago in central Asia. Today, M45 is found only in central Asians and those who trace their ancestry to this region - thus, it defines a central Asian clan. Descendants of the central Asian clan occur only sporadically in the Middle East and East Asia, and at somewhat higher frequency in India, where the clan appears to have migrated much later (as revealed by the presence of additional mutations). The 'ancestral' form - the deepest split in the genealogy of Y-chromosomes from the central Asian clan - is found only in central Asia. This allows us to pinpoint the location of what is effectively a 'regional Adam', in much the same way that we identified our African Adam as being an ancestor of the San Bushmen. The deepest branches in the M45 genealogy are found today only in central Asia - not India, or Europe, or east Asia. Thus, M45 arose in central Asia.

The limited distribution of the oldest descendants of the central Asian clan suggests that the population where it arose was isolated from people living in the surrounding parts of the continent. While the Hindu Kush provides a ready explanation for why there was no easy migratory path to India, it is not clear why this population had no contact with groups living in the Middle East. After all, our Eurasian clan had migrated into central Asia along this route - why couldn't the central Asian clan make the return trip? The inference is that another bollard had entered the story, and given that it hadn't been an insuperable barrier several thousand years before when the central Asia clan's ancestors first migrated to the heart of the continent, it was likely to have appeared after that first migration.

Today, the Dasht-e Kavir and Dasht-e Lut deserts of central Iran are scorched, parched wastelands. The tiny population living there ekes out a meagre living using a highly developed system of agriculture, complete with miles of underground irrigation channels known as ghanats that have been in use for thousands of years. During the heat of the day the residents of cities such as Yazd retire to subterranean chambers cooled by wind channelled down long pipes, creating a haunting wail that can be heard from miles away. It is inconceivable

{p. 115} that anyone could survive for long in this harsh climate without such a well-adapted lifestyle. Hunting and gathering would be impossible - at least today. Similarly the Karakum and Kyzylkum deserts of central Asia are harsh, desolate places with very few inhabitants apart fro few nomadic shepherds.

There are, however, two belts of continuous steppe across the deserts of central Iran, one to the north of the deserts, near the Caspian, one to the south, near the Arabian Gulf. When the world was in midst of its climatic schizophrenia around 40,000 years ago, it is likely that the steppelands and deserts of Iran and central Asia went through periods when the amount of moisture in the atmosphere would have been similar to, or perhaps greater than, today. This could have been aided by changes in the prevailing winds, bringing moisture in off the Arabian Sea. During these relatively wet periods, which may have been brief, humans would have been able to migrate fairly easily across the Iranian plateau and into central Asia - again, the prey and hunting methods would be virtually identical throughout the entire journey. We know that they did so because of the genetic trail they left in their descendants, which traces a direct path from the Levant to central Asia.

Once the ice age reached a threshold temperature, though, there was a significant decrease in precipitation and humidity as evaporation stalled and water became frozen into the expanding ice sheets of the far north. This seems to have happened between 40,000 and 20,000 years ago, and it resulted in the creation of a new desert bollard on our route. The continent was now split into northern, southern and western populations, all headed into the coldest part of the ice age. The people living in India and the Levant had the benefit of the sea which served to mitigate the effects of the increasingly cold and arid conditions. Those trapped north of the Hindu Kush, however, had to adapt to the increasingly harsh lifestyle of the Eurasian steppes - or die.

It is likely that these early central Asians would have stayed in the relatively warm environs of the southern steppes had encroaching desertification not forced them on. Some stayed behind, retreating into the foothills of the Hindu Kush where the water supply from glacial melting, and the number of animals, were sufficient for survival. Most

{p. 116} though, appear to have followed the migrating herds of game to the north - into the face of the storm, as it were. It is likely that they first reached Siberia during the early part of this period, around 40,000 years ago, when Upper Palaeolithic tools make their appearance in the Altai Mountains. The conditions would have been unimaginably different from those their ancestors had left behind in Africa 10,000 years before. Winter temperatures dropped to -40°C or lower, and much of their time would have been spent hunting for food and keeping warm. But the animals they hunted would have made the difficulties worthwhile.

We saw earlier that one of the defining features of species living at high latitudes is their great size - Bergmann's rule. The reason is that large animals have less surface area relative to their volume than small ones, and heat is lost through the surface. Shrews must eat constantly to maintain their hyperactive metabolism, in part because their tiny size makes it extremely difficult for them to retain heat. In cold environments, then, there is selection for large animals with slower metabolisms (since the food resources are not as plentiful as they are in warmer regions) - big, lumbering beasts that aren't particularly clever. This is how natural selection created animals such as the woolly mammoth.

{p. 117} The Eurasian interior was a fairly brutal school for our ancestors. Advanced problem-solving skills would have been critical to their survival, which helps us to understand why it was only after the Great Leap Forward in intellectual capacity that humans were ready to colonize most of the world. During their sojourn on the steppes, modern humans developed highly specialized toolkits, including bone needles that allowed them to sew together animal skins into clothing that provided warmth at temperatures not unlike those on the moon, but still allowed the mobility necessary to hunt game such as reindeer and mammoth successfully. They had to venture beyond sheltering hills and caves, out on to the icy open steppe and tundra, necessitating the development of portable shelters. Their migrations would have taken them far beyond ready sources of the fine-grained stone they used to make weapons, so they had to become more economical in their tool-making. This led them to develop microliths, small stone points (such as arrowheads) that were hafted on to wooden shafts and used as weapons.

The problem-solving intelligence that would have allowed Upper Palaeolithic people to live in the harsh northern Eurasian steppes and hunt enormous game illustrates something that could be called the 'will to kill'. Survival depended on finding sufficient food resources, whatever the obstacles - and the steppes were a veritable meat locker. It was the necessity of obtaining food that led them into the freezer, but it would take them well beyond central Asia. The Steppe Highway gave them a straight shot to the extreme ends of the continent, and once they had adapted to the harsh conditions a new world lay open to them.

{p. 118} The genetic composition of these first Siberians was a mixture of both central Asian and ancestral Eurasian clan lineages. While M45 is the marker that we use to infer the migrations of the early central Asian steppe hunters, there were still many men alive who did not have Y-chromosomes marked with M45 - they would have had unmarked Eurasian M9 Y-chromosomes. This is because new markers do not immediately increase in frequency to the point where all other markers - such as the ancestral M9 lineage - are lost. All of the Y-chromosome markers we study originated in a single man at some point in the past, so their original frequency was one (that individual) divided by the total number of men in the population - a very low frequency in all but the smallest groups. Over time, they become more common primarily due to the effect of genetic drift - the random changes in frequency that characterize all human populations. Thus the earliest people to colonize southern Siberia would have had members of both the central Asian M45 and the older Eurasian M9 clans, although drift appears to have caused them to lose most of their ancestral Middle Eastern chromosomes by this point.

As with the Eurasians who entered India on the other side of the Hindu Kush, some of these Eurasian clan members would have migrated to the north and east, guided in their journey by the Tien Shan mountains. Some of them, perhaps taking advantage of the so-called 'Dzhungarian Gap' used thousands of years later by Genghis Khan to invade central Asia, made it into present-day China. It is likely that the majority were migrants along the Steppe Highway further to the north, avoiding the harsh deserts of western China by detouring through southern Siberia. However, make it they did. We know this because they left descendants from another Y-chromosome marker that is almost completely limited to east Asia, and is entirely absent from western Asia and Europe - M175.

Today, M175, which arose on a Eurasian M9 chromosome, is found at highest frequency, around 30 per cent, in Korean populations. Based on absolute dating methods, it appears to be roughly 35,000 years old, coinciding very closely with the appearance of the Upper Palaeolithic

{p. 119} in Korea and Japan. There are several more recently derived markers that have M175 as an ancestor (particularly M122, which will play a significant role in Chapter 8), and together these related lineages account for 60-90 per cent of the Y-chromosomes in east Asia today. Like a collection of soup recipes that all have a common ingredient, M175 unites most Asian men living east of the Hindu Kush and Himalayas, defining an east Asian clan.

When these modern humans reached east Asia, they found themselves in an area that had been inhabited by their distant relatives Homo erectus for nearly a million years. Dubois' missing link had relatives in China, called (before being united with their Javanese cousins to the south) Peking Man. But mysteriously, no erectus remains from Chinese sites are found after 100,000 years ago - there is a gap in the record until fully modern Homo sapiens make their appearance around 40,000 years ago. What caused this hominid gap is unclear, although the likely culprit is - once again - the steadily deteriorating climate. For example, the cave at Zhoukoudian, where many erectus remains have been found, is located in north-eastern China, near Beijing - a region that experiences extremely cold winters even today. ...

We know that erectus didn't change substantially for I million years in east Asia, perhaps the result of stable selection pressures. Isolation from other hominids and a penchant for relatively uniform climatic conditions would have favoured continuity rather than change, and there is no evidence for an erectus Great Leap Forward. While some Chinese scientists argue for an evolutionary model known as 'regional continuity', in which east Asian erectus evolved into a local variant of Homo sapiens independently of what was happening in Africa, there is absolutely no genetic evidence for this. Moreover, the genetic results show that there was not even any interbreeding between modern human immigrants to east Asia and erectus - if in fact any populations still existed 40,000 years ago that are invisible to today's archaeologists. In a recent analysis of over 1,000 men from throughout east

{p. 120} Asia, geneticist Li Jin and his colleagues found that every single one traces his ancestry to Africa within the past 50,000 years - because every man has our old friend M168 on his Y-chromosome. Everyone....

If the story ended there, it would be very tidy and self-contained. But unfortunately, life is never that simple. In this case, the spanner in the works comes in the form of the presence of our Coastal lineage at high frequency in some east Asian populations. The Coastal lineage is found at a frequency of 50 per cent in Mongolia, and it is common throughout north-east Asia. How it reached this location remains a mystery, but it is likely that the early coastal migrants to south-east Asia gradually moved inland, migrating northward over thousands of years. The M130 chromosomes in the south are older than those in the north, consistent with such a migration. At some point, perhaps 35,000 years ago, they would have met the descendants of the other, main line of migrants - our incoming Eurasians. The presence of both Eurasian and Coastal lineages in east Asian populations attests to the extensive admixture that occurred between them.

The picture that emerges is that east Asia was settled by modern humans from both north and south, like migrational pincers or 'chop- sticks'. The northern route, which was characterized by Eurasian clan members, probably entered around 35,000 years ago from the steppes of southern Siberia. The southern route, which was composed primarily of members of the Coastal clan, was probably in place before this - perhaps as early as 50,000 years ago. The present composition of east Asia still shows evidence of this ancient north-south divide. Luca Cavalli-Sforza, working with Chinese colleagues, examined several dozen non-Y-chromosome polymorphisms in east Asian populations. In their analysis, they saw a clear distinction between the northern and

{p. 121} southern Chinese. Even members of the same ethnic group, such northern and southern Han, are most closely related to their geographic rather than their ethnic neighbours; northern Han group with other, non-Han northern populations, and the southerners form separate group. It seems that the ancient evidence of a two-pronged settlement is still visible in the blood of today's Chinese.

So, our Middle Eastern clan had made it to the eastern extreme of the continent. Along the way it had acquired additional marker producing the widespread Eurasian clan, the Indian clan, and the central Asian clan. The mountain ranges of central Asia served as effective barriers to migration 40,000 years ago, as they continue to do today. The effect of this was to produce an isolated east Asia Y-chromosome clan that only occasionally pops up in the west. But while the route to eastern Asia was clear, that to Europe required more circuitous tour. As we saw, modern Europeans contain rather too many ingredients in their soup to have been the direct descendants of the Middle Eastern clan. The search for the ancestors of the first Europeans is where we are headed next.

{p. 122} One evening in the autumn of 1997 as I was driving up 25th Avenue, I heard a news announcement that almost caused me to swerve into an oncoming bus. I pulled over and listened, hanging on every word.

The announcer was reporting that a team of scientists led by Professor Svante Paabo of the University of Munich had just published the first DNA sequence from a Neanderthal.

{p. 123} The field of ancient DNA research was pioneered in the 1980s by Svante Paabo and his colleagues (including Allan Wilson, of mitochondrial Eve fame) in Berkeley and Munich. The impetus behind this work was to do the impossible - to go back in time by examining the DNA that existed in a long-dead individual. It was, in effect, an attempt to develop a genetic time machine that would allow us to answer questions about our ancestors directly. One of the first applications was in the analysis of DNA from Egyptian mummies, but soon people were trying it on fossils that were millions of years old. Michael Crichton's novel Jurassic Park was based on the heady early days of the field, when it seemed that anything would be possible - even getting intact dinosaur DNA from bloodsucking insects embedded in amber!

While the claims for successful retrieval of DNA from sources that were tens of millions of years old eventually proved unfounded, usually resulting from minute amounts of contamination by modern DNA, it was sometimes possible to retrieve DNA from more recent samples, or those that had been preserved in ideal conditions for tens of thousands of years. The frozen bodies of mammoths and ancient alpine

{p. 124} travellers yielded analysable DNA, as did the dried remains from mummies and other desert-dwellers. Even then, the analysis was almost always limited to mitochondrial DNA, present in huge numbers of copies in every cell - making it more likely that one copy would have survived the Russian roulette of molecular degradation over the centuries. It was still extremely difficult to do this sort of analysis, though, because in most cases the molecules had completely disintegrated after death. This meant that negative results were far more common than positives - but the stories revealed by the tiny fraction of cases where DNA could be successfully extracted made the effort worthwhile. It was with this in mind that Paabo's group had developed reliable ways of evaluating and extracting DNA from ancient samples, and his laboratory represented the state of the art in the early 199OS - they were the undisputed experts in the field.

The scientific coup that led to my near-death experience in San Francisco actually began with the very first Neanderthal bones to be unearthed. Comprising the so-called type specimen - the one against which all the others were judged by palaeoanthropologists - these bones had sat in a museum in Bonn for nearly 140 years when the Munich group was approached to do the analysis on them. Paabo jumped at the chance, and his graduate student Matthias Krings performed the DNA analysis as part of his PhD thesis work. In over a year of tedious trial-and-error work, Krings gradually managed to extract enough intact mitochondrial DNA to create a 105-base-pair sequence. What he saw when he pieced it together was extraordinary. ...

After painstakingly reproducing the result from a separate bone fragment, and duplicating the experiment in a laboratory on a different continent (to be certain that a contaminant in the Munich laboratory was not producing an experimental artefact), he accepted the validity

{p. 125} of the sequence. By repeating the procedure several times, he eventually managed to obtain 37 base pairs of mitochondrial DNA sequence from the remains - enough to generate a statistically significant estimate of its evolutionary divergence. The sequence was clearly not from modern human mtDNA, but it didn't belong to an ape either. Rather, it came from a hominid that last shared a common ancestor with modern humans around 500,000 years ago. This date was consistent with what was predicted by palaeoanthropologists who had studied the dispersal of so-called 'archaic humans' from Africa into Europe, and it proved that Neanderthal was not the direct ancestor of modern humans. Rather, Neanderthals represented a local population of archaic hominids who were later replaced by modern Homo sapiens - with no detectable admixture. Of the thousands of human mitochondrial sequences that have been obtained from people all over the world, not one is anywhere near as divergent as Krings' Neanderthal sequence. Neanderthals fall well outside the range of genetic variation found in the human species - and therefore they represent a separate species. This early result has been confirmed by two additional genetic studies of Neanderthal remains from different parts of Europe, showing that the Neanderthals were closely related to each other, but very distantly related to us. The genetic data is incontrovertible - modern Europeans trace their recent ancestry to Africa, in common with everyone else in the world.

Along with the study of 1,000 Asian Y-chromosomes discussed in the last chapter, the Neanderthal results placed the final nail in the coffin of multiregionalism. Our hominid relatives were clearly replaced by modern humans who spread out of Africa within the past 50,000 years.

{p. 126} We saw earlier that the most obvious location from which to enter

{p. 127} Europe, the Middle East, appears to have contributed little to the gene pool of modern Europeans. The Y-chromosome lineage defined solely by M89, which would have characterized the earliest Middle Eastern populations around 45,000 years ago, is simply not very common in western Europe. It is such a tiny hop across the Bosporus from the Middle East to Europe that we might ask why it took so long - perhaps 10,000 years - for modern humans to make a significant foray into western Europe. To solve this riddle of where the majority of Upper Palaeolithic Europeans came from - we need to examine the genetic markers in western Europe and ask which Eurasian lineage they could have come from, and when.

I said at the beginning of Chapter 5 that my Y-chromosome is defined by a marker known as M173. It turns out that this marker is not unique to me - in fact, it is found at high frequency throughout western Europe. Intriguingly, the highest frequencies are found in the far west, in Spain and Ireland, where M173 is present in over 90 per cent of men. It is, then, the dominant marker in western Europe, since most men belong to the lineage that it defines. The high frequency tells us two things. First, that the vast majority of western Europeans share a single male ancestor at some point in the past. And second, that something happened to cause the other lineages to be lost.

Desperate for a date

The first thing most of us want to know when we hear that almost all western Europeans trace their family line back to one man is 'when did he live?' This is where our absolute dating methods come in. If we examine genetic variation - polymorphisms - on the M173 chromosomes, we can estimate how long it would have taken for our mutational clock to create it. But if all of the chromosomes are M173, how can we study variation? Surely they are all identical?

Fortunately for us, they are not. While all of them are very closely related, and thus share the M173 marker, there are other markers that help to distinguish them. Unlike the stable markers we have studied to define the order - or relative dates - of the Y-chromosome lineages, these other markers do not involve simple one-letter changes in the

{p. 128} genetic code. Rather, they exist because of a biochemical speech impediment. When we replicate our DNA, the double strands of the molecule open up and tiny machines known as polymerases actually do the hard work of assembling the complementary copy. Remember that if we know the sequence of one strand of the double-stranded DNA molecule, then we also know the other, because of the inviolable rules of molecular biology. A always pairs with T, and C always pairs with G. This works very well for more than 99 per cent of the genome, where the letters occur in a unique order and it is easy to tell how the pairing should work. Unfortunately, a small fraction of our genome is not so simple. It consists of what are known as tandem repeats - short sections of the same sequence, repeated several times in a row in the DNA strand. These often take the form of a couple of letters, such as CACACACACA ..., but there can be three, four or more letters in the motif that is repeated. As you might expect, the polymerase can become confused when it encounters these parts of the genome. After all, if there are a dozen or more repeats, how can you tell where you are in the sequence - is it repeat number ten or eleven? So, in a reasonable number of cases (about one in every thousand), the polymerase makes a mistake when it is assembling the complementary strand, and adds or subtracts a repeat. If the original strand had twelve repeats, the copy may have eleven or thirteen - simply by chance, due to an error at the molecular level. It is a process that Luca Cavalli-Sforza has called genetic 'stuttering'.

One in a thousand may not seem like a very common event, but it is when we are talking about the work of DNA copying. If that was the rate at which polymerases made single-letter copying mistakes, then we would introduce an average of over a million mistakes, or mutations, into our DNA every time it was copied. Since genetic copying takes place when we are having offspring, this means that each child would be born with over a million new mutations. Biology takes a dim view of this level of mutation, and it is likely that the child would die of a horrible inherited disease - if it were born at all. Thus, the usual rate at which new mutations appear is much more sedate, perhaps twenty or thirty per generation. This is around 100,000 times lower than the mutation rate we see for repeats, which means that new mutations in 'regular' sequences are much less common than those in

{p. 129} repeats. The repeats are on an evolutionary speedway, accumulating diversity at an extraordinary pace.

While this has very little effect on the health of the child, since repeats are usually found in regions of the genome that do not affect well-being, it does give us a tool for studying diversity. These repeats, known as microsatellites, are particularly valuable when we want to ask questions about variation on lineages where we do not have much single-letter variation - such as the M173 chromosomes. They give us a way to determine absolute dates that we can use to test our hypotheses about the timing of human migrations. The rate at which mutations occur is roughly constant, so the level of variation tells us how long they have been mutating. This tells us how old the chromosome is, because all of the chromosomes descend from a single chromosome at some point in the past. By definition, the level of variation at this point was zero, since there was only one copy.

When several microsatellites from M173 chromosomes are examined, the level of variation is consistent with an age of around 30,000 years. Of course, as with any estimate of time, this has a substantial range of error, but the most likely date for the origin of M173 is around 30,000 years ago. This date means that the man who gave rise to the vast majority of western Europeans lived around 30,000 years ago - consistent with a recent African diaspora, and again showing that Neanderthals could not have been direct ancestors of modern Europeans.

Significantly, it is around this time that the Upper Palaeolithic becomes firmly established in Europe - and the Neanderthals disappear. ... By 30,000 years ago the Neanderthals had been nearly eradicated, or perhaps 25,000 years ago they had disappeared entirely. The coincidence of the genetic and archaeological dates, as well as the increase in population size implied by the large number of Upper Palaeolithic sites from around 30,000 years ago, suggests that the invading moderns actually

{p. 130} displaced the Neanderthals.

{p. 132} As we saw, lineages belonging to the Middle Eastern clan - which we would expect to find if there had been a straight shot out of Africa to Europe, via the Middle East - are hardly found at all in Europe. M173, our 30,000-year-old marker, has the advantage of being present at very high frequency in the most isolated European populations (including the Celts and the Basques), and its age corresponds roughly to the inferred date of modern human settlement based on archaeology. Other major Y lineages present in Europe are younger than M173, and thus arrived later, or descended from M173 itself. Thus M173 is the likely marker of the first modern Europeans, defining the European clan. Of course, it is simply the terminal marker in a long line of genealogical descent that traces back to M168 and our African Adam. The penultimate marker, though, actually solves the mystery of where the earliest Europeans came from. This marker, a stepping-stone on the way to M173, is M45 - making Europeans a subset of the central Asian clan.

As we discussed earlier, the steppelands of 30-40,000 years ago stretched across a vast swathe of the Eurasian landmass. To Upper Palaeolithic hunters, this ecosystem would have been a land of plenty and migration along it would have allowed modern humans to disperse well to the west, into Europe proper, as well as to the east into Korea and China. During this period, the steppe zone extended well into present-day Germany, and may have reached France. We know from bones that have been found in French caves of 30,000 years ago that reindeer - a species adapted to the cold steppe and tundra of norther Eurasia - were common in France around this time. The climate had opened a window into Europe that allowed these central Asian steppe hunters to enter. As we have seen, they soon took over, dominating the region within a few thousand years. ...

{p. 149} Middle Eastern archaeologists have found that the end of the last ice age was a period of intense climatic variation in the eastern Mediterranean, with a general pattern of change from a continental to a Mediterranean climate. As archaeologist Brian Fagan summarizes it, this had the effect of producing an ecological zone with long, dry summers and short, wet winters. The effect of this climatic change was to favour grasses, which produce seeds in the spring and then lie dormant over the summer. Early humans would have exploited the relative plenty of food during the spring by harvesting large quantities of seed, then storing it for the rest of the year. This concentrated gathering behaviour would have favoured a settled lifestyle, which set the stage for the revolution that was to follow.

After 9000 BC the eastern Mediterranean summer was becoming increasingly drier as the full effects of rising global temperatures kicked in. This reduced the yield of cereals, and (as with arid periods in the distant past) would have favoured mobility. However, the necessity of storing their gathered grain would have tied the Natufians to one

{p. 150} location. The pull of these two forces - reduced yields and a relatively immobile lifestyle - would, within a few hundred years, lead some Natufian settlements such as Jericho to try a new innovation: planting some of the gathered cereals (which are really seeds) in order to simplify the gathering process. Kenyon's work at Jericho traced the development of the Neolithic, or New Stone Age, following this early innovation. Archaeologists and anthropologists continue to debate what happened after the first crops were cultivated - whether the need for a reliable water supply in order to grow crops led to the development of irrigation, which may have fostered water-rights issues and social hierarchies, and so on. What is clear is that the end of the ice age appears to have set in motion a series of events that were to culminate with the development of settled, agrarian communities within a thousand years. What archaeologist V. Gordon Childe called the 'Neolithic Revolution' had arrived.

The second Big Bang

The Neolithic marked a turning point for the human species. It was at this point that we stopped being entirely controlled by climate - as we were during our Palaeolithic wanderings - and began to assume control of our own destinies. ...

The second development was that of greatly increased population density. One of the consequences of cultivating food and settling in

{p. 151} one place is that the necessity of not over-exploiting limited resources is relaxed - after all, if you want to have more children, you can simply plant and harvest more crops. While this does oversimplify the situation, it is true that settled, agrarian societies are more densely populated than those of hunter-gatherers. Coupled with the freedom to choose where to live, this can lead to very rapid population expansions, with agriculturalists spreading throughout a region. It is estimated by palaeodemographers, who study past population sizes using archaeological and anthropological methods, that the entire population of the globe was around 10 million at the time agriculture originated; by the dawn of the Industrial Age, around 1750, world population had risen to over 500 million. If Palaeolithic hunter-gatherer populations had taken over 50,000 years to increase from a few thousand individuals living in sub-Saharan Africa to a few million scattered around the globe, clearly the agriculturalists of the last ten millennia were making up for lost time.

The final new feature of the Neolithic revolution is that it demonstrates the importance of new technology to human migration. ...

{p. 152} While agriculture played a critical role in the development of modern society, the genetic effects of agriculture were equally pronounced. While Upper Palaeolithic hunter-gatherers tended to maintain a relatively stable population size, except via the settlement of new territory, agrarian societies were able to expand massively without leaving home. As the first farming communities increased in size, their inhabitants gradually moved further afield in search of cultivable land. When they did so, they carried with them their genetic markers. One of the consequences of this is that we see the expansion of certain genetic lineages, giving us a glimpse of the origin and spread of agriculture. In the case of the Middle East, the genomes of today's western Eurasians still retain a signal of those events at Jericho 1,000O years ago.

Archaeologists had long known that agriculture spread from its origin in the Middle Eastern 'Fertile Crescent' to Europe over the course of several thousand years. The earliest evidence is found in the Balkans, and it appears later and later as you move to the north-west. It is only relatively recently that ancient Britons left behind their hunter-gatherer lifestyle, several thousand years after their cousins in Jericho had done the same. Crucially, it is exactly those plant species initially cultivated in the Fertile Crescent that make their appearance in the advancing wave of agriculture as it moved into Europe. It seems that the European hunter-gatherer lifestyle was replaced by the new Middle Eastern invention.

In the 1970s Luca Cavalli-Sforza, along with fellow geneticists Alberto Piazza and Paolo Menozzi, initiated a study of the genetic effects of agriculture. The question that they asked was about the way in which agriculture had spread. In particular, they wanted to know if the migration of agriculture into Europe marked the migration of people, or simply the spread of a sexy new cultural development - the MTV of its era. In effect, they were asking a question about the genetic composition of modern Europeans. Was there evidence for an expansion of certain genetic markers out of the Middle East, or did modern Europeans appear to be relatively free of Neolithic markers?

At the time the study was done, the only data available was that on

{p. 153} the 'classical' markers we learned about in Chapter - blood groups and other cell-surface protein markers that served as convenient polymorphisms but gave little information about their underlying DNA sequence changes. The analysis of these markers led Cavalli-Sforza and his colleagues to conclude that there had been a mass migration of genes out of the Middle East, and the genetic pattern was very similar to that observed for the first appearance of agriculture: the genetic signal decreased regularly as you moved from south-east to north-west across Europe. The methods of analysis used in this study limited what the researchers were able to infer, since it wasn't possible to obtain an accurate date for this migration, but their findings did corroborate the theory that agriculture had spread with farmers as their population had expanded, rather than as a purely cultural phenomenon that 'migrated' as Palaeolithic Europeans learned farming skills.

Cavalli-Sforza's results became accepted wisdom, leading to what they called the 'Wave of Advance' model for the diffusion of agriculture. The assumption made by many (although not Cavalli-Sforza and his colleagues) was that the majority of the European gene pool was Neolithic in origin, since it was the most pronounced genetic pattern in Europe (although Cavalli-Sforza's later work showed that it still accounted for less than a third of the genetic variation). Many anthropologists remained sceptical, but it was to be over twenty years before the model received a serious re-evaluation. This came in the late I99Os with the detailed analysis of mtDNA lineages in Europe and south-west Asia by Martin Richards and his colleagues at Oxford University. In a series of scientific papers they analysed mtDNA lineages from a selection of populations across Europe and the Middle East, carefully dating the lineages using the absolute methods similar to those we learned about earlier. This allowed them to evaluate the relative contributions of various migrations to the European gene pool. Their results suggested that, rather than having a significant genetic effect on the population of Europe, the expansion of agriculture involved very few Middle Eastern migrants. Most of the lineages in Europeans seem to have been present since the time of the Upper Palaeolithic, between 20,000 and 40,000 years ago.

One of the objections to the Richards study, raised by Cavalli-Sforza and others, was that mtDNA actually provided very little resolution

{p. 154} between European populations. It was difficult, for instance, to distinguish between eastern and western Europeans with mtDNA data alone - they have very similar patterns of mtDNA markers. Nonetheless, the mtDNA result was suggestive. What was needed was to look at the male lineage, with its greater inherent resolution, in order to see if it showed the same pattern.

This was finally done in 2,000, when Ornella Semino and her colleagues (among them Cavalli-Sforza) analysed the Y-chromosomes of over 1,000 men from Europe and the Middle East, looking specifically for evidence of the agricultural expansion. What they found was that lineages defined by Neolithic Middle Eastern markers are found in a minority of modern Europeans. In fact, the results from the Y agree almost perfectly with the mtDNA data, suggesting that 80 per cent of the European gene pool traces back to other waves of migration, primarily during the Palaeolithic. In western Europeans, this Palaeolithic component is largely defined by our friend from the last chapter M173, which links Europe back to central Asia. Only 20 per cent of European Y-chromosomes - defined by more recent markers, particularly one known as M17 - derive from Neolithic Middle Eastern immigrants. In effect, modern Europeans are largely genetically Cro-Magnon on both their maternal and paternal sides.

This is not to say that the advent of agriculture had no effect on Europe - far from it. There is clear genetic evidence for a significant European population expansion after the end of the last ice age, almost certainly aided by the onset of food production. Evidence for this comes in the form of a recent analysis by David Reich and his colleagues at Massachusetts Institute of Technology. They studied markers from many independent regions of the genome and found a pattern of variation suggesting that the European population underwent a substantial decrease in population size between 30,000 and 15,000 years ago, as Europe was moving into the depths of the last ice age. This was then followed by a population expansion from the few survivors after the end of the last ice, producing the relative dearth of variation seen in Europe today. In other words, the human population had been through what is known as a bottleneck - a reduction in size followed by a period of growth. Patterns of mtDNA variation also support this model of postglacial population growth. Archaeological evidence

{p. 155} suggests that the Palaeolithic population of Europe was confined to Iberia, southern Italy and the Balkans during the period of most extensive glaciation around I6,000 years ago, and that human populations then expanded northward during the postglacial period. Agriculture almost certainly contributed to the end of this population expansion, because it allowed much higher population densities.

How do we reconcile the pattern seen for the Y-chromosome and mtDNA, of a Palaeolithic European population relatively unaffected by Neolithic immigration, with the Wave of Advance? The pattern seen by Cavalli-Sforza and his colleagues clearly exists, but they were studying large-scale patterns across the entirety of Europe and the Middle East. The agricultural expansion was simply one population movement into Europe - there is clear archaeological evidence for several others. As their later analysis showed, it still accounted for a minority of the genetic variation in Europe. Furthermore, because the Wave of Advance had no estimate of age, the Neolithic component could have been confounded with Palaeolithic immigration from the Middle East. Finally, since central Asian populations were not included in their analysis (there was no data available at the time their study was conducted), it is possible that the pattern reflects a general trend of migration from Asia to Europe during the Upper Palaeolithic. After all, if we simply had Y-chromosome data from the Middle East and Europe, we would infer that M89-bearing populations had migrated into Europe via the Balkans, giving rise to M173 in Europe. It is only because we know that M173 arose on an M45-containing lineage that we trace the Upper Palaeolithic settlement of Europe back to central Asia.

The Y data actually provides a partial solution to the apparent conundrum. It seems that southern European populations experienced a greater influx of Neolithic farmers from the Middle East, carrying lineages such as M17, than did northern Europeans. One possible scenario is that farming spread first around the Mediterranean, with Neolithic Middle Eastern immigrants favouring its climate, similar to that of the Levant. Only later did indigenous Palaeolithic Europeans take up agriculture in the interior, gradually spreading the culture - but only a small percentage of the genes - of the Neolithic throughout. The Cro-Magnons of northern Europe appear to have made a

{p. 156} conscious decision to leave behind the Palaeolithic for a new Middle Eastern lifestyle with a small minority of Middle Eastern immigrants.

... The pattern of settlement and intense exploitation of a few plant species that characterized the Middle East was seen at around the same time in China. Northern Chinese sites such as Banpo and Zhangzhai in Shaanxi province show early evidence of millet agriculture, around 7,000 BC. Millet, a cereal crop like wheat, seems to have been domesticated around the Yellow River, spreading from there to the rest of northern China. The remains at Pengtoushan, on the Yangtze River in central China, indicate that rice was being cultivated there independently around the same time. At both sites pottery was used for storing grain, and the people lived in carefully constructed clay houses, suggesting that the Neolithic lifestyle was well developed, even at this early date. Agriculture soon spread throughout China, with rice dominating in the south, where the wet, humid conditions favoured this grain. Rice agriculture spread down the Yangtze, and was widespread in southern China by 5000 BC, perhaps helped by a second, independent domestication along the south coast. By 3500 BC it was being cultivated in Taiwan, and by 2000 BC in Borneo and Sumatra. It reached the rest of the Indonesian archipelago by 1500 BC. Overall, the archaeological evidence suggests that rice agriculture spread from an origin in central-southern China to the islands of south-east Asia in the space of roughly 3,000 years - timing similar to that of the agricultural expansion into Europe. However, unlike in Europe, there is a very strong genetic signal of this expansion, suggesting that it was people, and not merely the culture, that moved.

In Chapter 6 we learned that one descendant lineage of M9, defined by a marker known as M175, is widespread in east Asia. Based on its present distribution, this marker probably arose initially in northern China or Korea. Looking at the pattern of Y variation in modern

{p. 157} Chinese populations, it is now clear that the first agriculturalists in China were descendants of M175. In fact over half of the entire male population of China have Y-chromosomes defined by a marker that shows evidence of a massive expansion in the past 1O,000 years. M122, which first appeared on an M175 chromosome, is now widespread throughout east Asia. It is hardly found west of the great central Asian mountain ranges, and does not occur at all in the Middle East or Europe. This is the pattern we expect to see with a recent expansion, rather than an ancient event that typically leaves a more widespread trail.

The genetic data shows that the development of rice agriculture in east Asia created a Wave of Advance. However, while the wave leaving the Fertile Crescent for Europe seems largely to have dissipated after inundating the Mediterrapean, the one leaving China was to saturate the entirety of east Asia. Today, M122 - marking the descendants of the first Chinese rice agriculturalists - is found from Japan to Tahiti. A recent study by David Goldstein and his colleagues at University College London shows that microsatellite diversity on M122 chromosomes is very high in China and Taiwan, but drops significantly moving southward into peninsular Malaysia and Indonesia. This is precisely what would be expected from a population expansion originating in China within the past 10,000 years - and exactly parallels the archaeological evidence for the spread of rice agriculture. Together, MI122 and a related Chinese haplotype (also a descendant of M175) defined by marker M119, account for nearly half of the Y-chromosomes in south-east Asia. In Europe, on the other hand, Neolithic immigrants account for only o per cent of the present Y diversity. In comparison to Europe, the Wave of Advance in east Asia appears to have been more of a tsunami.

{p. 158} Early agriculturalists were taking on a new set of risks when they committed themselves to a settled existence. The most important was a decrease in the breadth of their resource base. By focusing cultivation on a few species, they were reducing their choices in the event of a climatic shift. Droughts, intense periods of cooing (such as the Dryas periods at the end of the last ice age) and shifts in watercourses were all very easy to deal with for Palaeolithic hunter-gatherers. Their response to any of these changes was to move into another area with better resources. ...

The second main worry for our Neolithic agriculturalists was the increase in disease. While hunter-gatherers may appear to have had a difficult life, relying as they did on apparently 'primitive' technology and the necessity of killing or gathering enough food to survive, in fact they were surprisingly healthy. ...

Infectious diseases do not arise spontaneously as a by-product of a settled lifestyle, but rather from exposure to disease-causing organisms in such a way that transmission occurs from one infected individual to another. Most diseases can exist only in large populations, where a threshold number of people remain infected, allowing the disease to remain in the population. These are so-called endemic diseases, such as smallpox or typhoid. A population of several hundred thousand is

{p. 159} necessary to maintain the disease - otherwise it is lost because not enough people remain susceptible to infection. Populations of this size only arose after the development of agriculture. Other diseases can be introduced from an outside source, such as an animal. While human had contact with animals as hunter-gatherers, the sort of prolonged. close contact that encourages the spread of disease occurred only aftel the domestication of animals in the Neolithic. ...

The final negative aspect of a sedentary lifestyle was the growing stratification of society. In general, hunter-gatherers are remarkably egalitarian, having few social divisions.

{p. 160} The onset of the Neolithic established many of the regional patterns of cultural diversity we see in the modern world. Expanding waves of agricultural migrants in east Asia spread rice cultivation to Indonesia and beyond, and today their descendants still carry the genetic traces of this event. As we saw earlier, the first inhabitants of south-east Asia may have been more similar to today's Andamanese or Semang Negritos. It is likely that most of these groups were engulfed by the wave of expanding rice-growers, their culture subsumed into the agricultural mainstream. Similarly, hunter-gatherer groups in Europe, the Americas and Africa all gave up their Palaeolithic lifestyle in favour of the new way of feeding themselves. But culture is defined by far more than eating - it encompasses social traditions, clothing and tool-making styles, means of transport and thousands of other things. And one the most important aspects of culture is language.

{p. 161} Language similarities had been recognized since Classical times, particularly among such well-studied European examples as Latin, French, Spanish and Greek. By the eighteenth century scholars had begun to take a broader view, focusing on the languages of Asia, Africa and the Americas. For instance Janos Sajnovics, in his obscure I770 treatise 'Demonstration that the language of the Hungarians and Lapps is the same', arrived at the conclusion given in the title. We now know that both Hungarian and Lapp belong to something known as the Uralic language family, uniting them with more obscure languages such as Khanty, Nenets and Nganasan. Sajnovics, though, was unaware of these more distant relationships. And while he, like many other scholars, recognized the similarities uniting different languages, he crucially failed to explain how they had arisen.

An explanation for the similarities among members of a language family arrived a few years after the Sajnovics study. In a 1786 address to the Royal Asiatic Society, Sir William Jones - then a judge in India - noted that Sanskrit (the religious language of Hinduism) bore a closer resemblance to Greek and Latin 'both in the roots of verbs and in the forms of grammar, than could possibly have been produced by accident'. So much so, he concluded, that they must 'have sprung from a common source'. It was this last statement that was to be his most lasting contribution, since it implied a mechanism for the generation of linguistic diversity. Languages change over time, he was saying, and if there are enough deep similarities among a group of languages, then they must have had a common ancestor in the past and subsequently diverged from each other. It was an evolutionary explanation for linguistic diversity, presaging Darwin by over sixty years.

The languages that Jones described all belong to what became known as the Indo-European language family, after the geographic

{p. 162} locations of the languages. There are 140 separate languages in the family, ranging from those belonging to the Celtic branch, spoken in the extreme north-western parts of Europe (Gaelic and Breton are two examples), to Sinhalese, spoken in Sri Lanka. English is a member of the Germanic branch of Indo-European, although its complicated history has left it with many words borrowed from French. Clearly, this is a widespread and diverse collection of languages.

Today the hypothesis that Jones advanced - that all of the Indo- European languages trace their descent from a common ancestor - is widely accepted by linguists. In fact it is one of the few language families to have received universal acceptance. The implication of his model, known as the genetic model of language classification, is that at some point in the past there was a group of people who spoke an ancestral form of Indo-European, which later evolved into the languages we see today. Like our soup recipes, additions and modifications of ingredients have produced local linguistic varieties, which eventually became distinct languages. The parallels with DNA evolution seem obvious. But is it possible to learn anything about language diversity - and to understand the present distribution of the world's languages - from the study of genetics?

The subject of language change has always been a key interest of Luca Cavalli-Sforza, particularly its overlap with genetic patterns. Instead of drawing vague comparisons between genetic and linguistic diversity, in 1988 he decided to test the hypothesis directly - much like Dick Lewontin had done with the genetic data from different races. He and his colleagues examined genetic data from forty-two worldwide populations and drew a tree of their relationships based on minimizing the differences in marker frequencies between them. The tree that resulted - in effect, a genealogical tree of the populations - corresponded very well with known linguistic relationships. So, for instance, speakers of Indo-European languages tended to group together in the genetic tree, as did speakers of Bantu languages in Africa. There were obvious inconsistencies, such as the deep split between northern and southern Chinese (almost certainly resulting from the pattern of early migrations discussed in Chapter 6), but overall the genetic and linguistic groups seemed to be very similar to each other. This suggested

{p. 163} that genetic data could be used to study the origin and dispersal of languages.

There were two caveats made by Cavalli-Sforza and his colleagues in their study. The first is that the genetic markers they were studying did not cause the pattern of linguistic diversity - there was no Bantu gene that forced its hapless carriers to speak those languages. Rather, similar genetic markers reflected the common history of the speakers of that language, as markers of descent. The second caveat is that in many cases relationships suggested by genes and languages disagreed, showing that the correspondence wasn't absolute. The reasons for this might be language replacement, in which people learn to speak a new language without a corresponding influx of outside genes, or gene replacement, in which there was a significant influx of genes but the language stayed the same. The first explains the difference between northern and southern Han Chinese, while the second may explain the close genetic similarity between linguistically unrelated groups, such as Na-Dene-speaking Native Americans and neighbouring Amerind speakers. Thus, genes were often markers of linguistic relationships, except when they weren't. Either way, the genetic data should help to shed light on language relationships, by illuminating the way in which languages have spread.

In search of a homeland

If we accept that William Jones was right, and that all Indo-European languages descend from a common source, then the implication is that there must have been, at some point in the past, a single group ol people who spoke this ancestral form of Indo-European. The search for the identity of the first Indo-Europeans, and their geographic location, has been one of the main areas of archaeological and linguistic investigation over the past 200 years. It has become a sort of quest - although like all good quests, it is in some ways rather quixotic. The attempt to disentangle the web of conflicting evidence surrounding the location of the Indo-European 'homeland' illustrates a particularly exciting new anrlication of genetics to our understanding of human history.

{p. 164} Gordon Childe, who coined the term 'Neolithic revolution', proposed in the 1920s that the Indo-European homeland should be identified with a culture originating north of the Black Sea that had distinctive 'corded' pottery - marks that resembled impressions left by cord or twine. The theory was revived by archaeologist Marija Gimbutas in a series of articles published in the 1970S. Gimbutas argued that the remains left by nomadic horsemen of the southern Russian steppes, dating from around 6,000 years ago, mark the earliest signs of a culture that can be identified as proto-Indo-European (PIE), which included Childe's Corded Ware people. The Kurgan culture, as she called it, left enormous burial mounds (known as kurgans) that are still dotted across the entirety of the Eurasian steppe, from Ukraine to Mongolia and south to Afghanistan. The golden treasure hoards recovered from kurgan excavations in the twentieth century confirmed the existence of a people who were known to Herodotus as the Scythians - fearsome horsemen of the Asian grasslands, and previously thought by many scholars to be mythical.

The evidence that the Kurgan people spoke PIE is based on an analysis of words common throughout Indo-European languages. If a word can be shown to derive from the same root, then it is likely (although not certain) that it was inherited from the common ancestor. For instance, the English word ox has cognates in the Sanskrit uksan and the Tocharian (an early Indo-European language spoken in western China) okso. Similarly, many words for animals and plants are common throughout the Indo-European languages, as are those for tools and weapons. Perhaps most interestingly, there is a rich vocabulary for horses and wheeled vehicles in common among all the languages, suggesting that the PIE speakers had domesticated the horse as a draft animal. Coupled with the archaeological remains showing that the horse was domesticated in the southern Russian steppes, this pointed toward the kurgan-builders being the PIE people.

But while the evidence in favour of the Kurgan people being early Indo-Europeans was compelling, there was no archaeological evidence for the spread of their culture into western Europe. Their culture, dominated by horses, was ideal for the steppes, but it was not well suited to European forests and mountains. It was difficult to see why the steppe horsemen would have been able to conquer Europe and

{p. 165} impose their language upon its inhabitants. For this reason, Colin Renfrew proposed in his 1987 book Archaeology and Language that the Kurgan culture did not mark the origins of Indo-European, but rather a later, eastern extension of it. Renfrew suggested that PIE had been a Middle Eastern language, originally spoken 9,000 years ago, which spread with the agricultural Wave of Advance into Europe. He identified Anatolia as the Indo-European homeland, on the basis of it being roughly central in the modern distribution of Indo-European languages, and also the home of several extinct examples. The hypothesis he advanced is that the early farmers carried their language - PIE - with them as they expanded their population, and thus the linguistic inundation of Europe should have involved a genetic wave as well. It was a bold suggestion, which had little initial support from the linguistics community. As we have seen, the Wave of Advance actually contributed little to the gene pool of modern Europeans, and its influence seems to have been largely limited to the Mediterranean region. The Indo-European speakers living in Ireland, for instance, have virtually no Neolithic Y-chromosome markers, while Greeks have a substantial Neolithic component. What this suggests is that, it farming spread Indo-European languages throughout Europe, it must have done so largely without the actual spread of farmers - thereby reducing the strength of Renfrew's argument.

Of course, as the name suggests, Indo-European languages are spoken not only in Europe. Modern Iran, Afghanistan and the Indian subcontinent all have a majority of Indo-European speakers. How did they come to speak languages related to Irish Gaelic, thousands of miles away? Again, there are competing hypotheses. The first, advanced by Childe, Gimbutas and others, is that the early steppe horsemen carried their language from central Asia into India when they invaded around I 500 BC. The Rig Veda, an early Indian religious text, records the conquest of India by mounted warriors from the north. This received corroboration in the 1920s when Sir John Marshall and his colleague excavated Mohenjo Daro and Harappa in the Indus Valley. These great cities date from around 3500 BC and by the second millennium BC they were massive settlements with thousands of houses, extensive agriculture and enormous populations. Then, around 1500 BC they entered a period of decline, and by AD 1,000 the Harappan culture had

{p. 166} disbanded, its cities abandoned. What caused this sudden cultural collapse? To the archaeologists, it seemed to correlate perfectly with an invading force of Aryans from the Steppes. Archaeology seemed to be reinforcing Childe's argument, and corroborating the Rig Veda.

More recent research has suggested that there were probably indigenous causes for the collapse of the Harappan civilization. Perhaps a river changed course, or social decay had set in (think of the Romans, 2,000 years later). Whatever the cause(s), the invading Aryans were not necessarily the all-powerful conquerors that early archaeologists thought they were. In the wake of this reinterpretation, Renfrew suggested two models for how the Indo-European languages could have come to India.

Renfrew's first model is that of an early Neolithic migration from the Middle East, with the settlers carrying their PIE language with them. In this model, the Harappans would already have been Indo-European, and thus there is no reason to infer an Aryan invasion in order to account for the languages of India. The second model, giving more credence to the Rig Veda, is that there was an invasion of the Indus region by Indo-European speaking nomads from central Asia, but it was carried out by relatively few individuals. Thus it had little impact on the population of the subcontinent, aside from the imposition of a language and culture. In both cases, the Indian genetic data shows a minor contribution from the northern steppes.

The test of the Childe-Gimbutas and Renfrew hypotheses awaited the development of markers that were capable of distinguishing between populations from the steppe and the indigenous Indian gene pool. As we saw in Chapter 6, M20 defines the first major wave of migration into India from the Middle East, around 30,000 years ago. It is found at highest frequency in the populations of the south, who speak Dravidian languages - a language family completely unrelated to Indo-European. In some southern populations, M20 reaches a frequency of over 50 per cent, while it is found only sporadically outside India. Thus, for our purposes, it is an indigenous Indian marker. What was needed to complete the analysis was a steppe marker, in order to see what contribution it may have made to the genetic diversity present in India.

This came with the discovery of a marker known as M17, which is

{p. 167} present at high frequency (40 per cent plus) from the Czech Republic across to the Altai Mountains in Siberia and south throughout central Asia. Absolute dating methods suggest that this marker is 10-15,000 years old, and the microsatellite diversity is greatest in southern Russia and Ukraine, suggesting that it arose there. M17 is a descendant of M173, which is consistent with a European origin. The origin, distribution and age of M17 strongly suggest that it was spread by the Kurgan people in their expansion across the Eurasian steppe. The key to solving our language puzzle is to see what it looks like in India and the Middle East.

The answer is that M17 in India is found at high frequency in those groups speaking Indo-European languages. In the Hindi-speaking population of Delhi, for example, around 35 per cent of men have this marker. Indo-European-speaking groups from the south also show similarly high frequencies, while the neighbouring Dravidian speakers show much lower frequencies - 10 per cent or less. This strongly suggests that M17 is an Indo-European marker, and shows that there was a massive genetic influx into India from the steppes within the past 10,000 years. Taken with the archaeological data, we can say that the old hypothesis of an invasion of people - not merely their language - from the steppe appears to be true.

And what of the Middle East? Interestingly, M17 is not found at high frequency there - it is present in only 5-10 per cent of Middle Eastern men. This is true even for the population of Iran, speaking Farsi, a major Indo-European language. Those living in the western part of the country have low frequencies of M17, while those living further east have frequencies more like those seen in India. What lies between the two regions is, as we learned in Chapter 6, an inhospitable tract of desert. The results suggest that the great Iranian deserts were barriers to the movement of Indo-Europeans in much the same way that they had been to late Upper Palaeolithic migration.

The Y-chromosome results from Iran and the Middle East also suggest that early Middle Eastern agriculturalists did not spread Indo-European languages eastward as they moved into the Indus Valley. The marker M172, associated with the spread of agriculture, is found throughout India - consistent with an early introduction from the Middle East, most likely during the Neolithic. But the frequency is

{p. 168} comparable in Indo-European and Dravidian speakers, suggesting that the introduction of agriculture pre-dated that of the Indo-European languages. Thinking in terms of actual behaviour, many Indian descendants of Neolithic farmers have learned to speak Indo-European languages, while fewer M17-carrying Indo-European speakers - up to this point- have given up their language in favour of Dravidian.

The low frequency of M17 in western Iran suggests that, in this case, exactly the sort of scenario envisaged by Renfrew in his second model has occurred. It is likely that a few invading Indo-European speakers were able to impose their language on an indigenous Iranian population by a process Renfrew calls elite dominance. In this model, something - be it military power, economic might, or perhaps organizational ability - allowed the Indo-Europeans of the steppes to achieve cultural hegemony over the ancient, settled civilizations of western Iran. One candidate for this 'something' was their use of horses in warfare, either to pull chariots or as mounts. Cavalry and chariots, both steppe inventions, would have given the early nomadic Indo- Europeans a distinct advantage over their adversaries' infantry. The use of horses would provide a major technological advantage to armies over the next three millennia. It is not difficult to imagine that it gave an early advantage to the people of the Eurasian steppe.

Thus, while we see substantial genetic and archaeological evidence for an Indo-European migration originating in the southern Russian steppes, there is little evidence for a similarly massive Indo-European migration from the Middle East to Europe. One possibility is that, as a much earlier migration (8,000 years old, as opposed to 4,000), the genetic signals carried by Indo-European-speaking farmers may simply have dispersed over the years. There is clearly some genetic evidence for migration from the Middle East, as Cavalli-Sforza and his colleagues showed, but the signal is not strong enough for us to trace the distribution of Neolithic lineages throughout the entirety of Indo-European- speaking Europe. Cavalli-Sforza has suggested that an initial migration of Neolithic pre-PIE speakers from the Middle East could have introduced a language to Europe, including our Kurgan people, which later became PIE. There is nothing to contradict this model, although the genetic patterns do not provide clear support either.

There is another possibility, which comes from the distribution and

{p. 169} relationships among extinct languages in the Middle East and Europe. What if the language of the first farmers was not Indo-European, but another language entirely? The Basques, who live in north-eastern Spain, speak a language unrelated to any other in the world. Jared Diamond, in his book The Rise and Fall of the Third Chimpanzee. suggested that it might be a remnant of the agricultural Wave of Advance from the Middle East. Interestingly, some linguists have suggested that Basque is related to languages spoken in the Caucasus. while others find similarities to Burushaski, a language isolate spoken in a remote part of Pakistan. Similarly, there were other now-extinct languages spoken throughout the Mediterranean world, in south-eastern Spain (Tartessian and Iberian), Italy (Etruscan and Lemnian and Sardinia (there is a non-Indo-European source for many place names). Place names in southern France similarly suggest that Basque was much more widely spoken in the past than it is today, and Greek place names indicate the presence of a pre-Indo-European element there as well. Overall, there is reasonable evidence for a 'Mediterranean' collection of pre-Indo-European languages that were later replaced by the expansion of Greek and Latin.

Taken at face value, then, we have a set of languages that were once widespread around the Mediterranean and Middle East, extending eastward into Pakistan. This is precisely the territory colonized by early Neolithic farmers during the period between 10,000 and 7,000 years ago. One possibility is that these early farmers spread 'Mediterranean' languages as they expanded their populations. The Palaeolithic populations of Europe took on the language of farming, and its culture, even if (as in the case of the Basques) there was hardly any genetic influx. These languages also spread to the east, introducing farming throughout the river valleys of central Asia and Pakistan. Later migrations, of Dravidian and Indo-European speakers in the case of Pakistan, and Indo-Europeans in the case of Europe, would have reduced the current speakers of the Mediterranean languages to the isolated pockets we see today.

Of course, this scenario is purely speculative, but it may be a plausible alternative to Renfrew's Indo-European farmers and Cavalli-Sforza's pre-PIE farmers. Furthermore, the genetic data shows some correlations: most of the regions mentioned, from the Mediterranean

{p. 170} to the Caucasus to Pakistan, have substantial frequencies of M172, our canonical Neolithic marker. This is particularly true for populations from the Caucasus, some of which have frequencies of M172 in excess of 90 per cent. The generally close genetic similarly between Caucasian populations and those from the Middle East suggests that there was a substantial influx of people during the Neolithic, who may have introduced languages related to Sumerian to the region.

{p. 173} The spread of languages is a special case of cultural diffusion, or change. Unfortunately, the attempt to identify cultural change with the migration of people is now seen as old-fashioned in many archaeological circles. Instead, modern archaeologists stress indigenous reasons for the development of cultural attributes, or their borrowing from other cultures. The old school of diffusionism, which attempted to trace the expansion of particular cultures from a single place of origin, has fallen out of favour. However, the genetic results show that in some cases, this has clearly occurred. If genetic and cultural patterns overlap, as in the case of the eastern Dene-Caucasian languages, it is likely that there has been an ancient expansion of people carrying their culture with them. However, it is quite possible to expand a culture

{p. 174} without a concomitanmt movement of people. this may have happened with the expansion of farming into north-western Europe.

As geneticists, we are limited by what we study. While we take history, archaeology and languages into account in our interpretations; our unique contribution is our ability to trace genealogy - actual biological relationships. Thus, we can find evidence to support human migration, as in the case of M17 and the steppe culture, as well as to refute it. Language is a good cultural attribute to study, since there are often written records. Even when there aren't, the relationships among languages can be examined systematically. Most cultural processes are not like this, making their interpretation more problematic. ...

Sexual politics

The Karen people of northern Thailand and Burma are perhaps not as well known as their neighbours the Padaung, with their neck-extending rings of brass, but they are fascinating to ethnographers. This is because their social system runs counter to the pattern common in the vast

[p. 175} majority of the world. Over 70 per cent of the world's societies practise something known as patrilocality. In this type of society, men control the wealth. Inheritance - and group membership - is passed through the male line. When two people marry, the wife goes to live with her husband and assumes a new identity in her husband's clan. The European custom of a wife changing her name to that of her husband traces its origin to this type of patrilocal behaviour.

One of the effects of patrilocal behaviour is that men tend to stay in one place while women are constantly immigrating into the family or clan. This may seem counter-intuitive - after all, don't men sow their wild oats more than women? - but it is the rule in most societies. The Karen, in contrast, do things differently. In Karen society, everything is turned upside-down. Women control the wealth, and group identity is passed through them to their daughters. In Karen marriages, men immigrate to the woman's village, taking over the care of her fields. Their society is what anthropologists term matrilocal, in that the women stay put and the men move. While the Karen may seem like ethnographic curiosities, in fact they have been instrumental in revealing the effect of culture on human genetic diversity. Like a bespoke experiment, they provide a social contrast to the prevailing pattern in human populations around the world.

We have used the Y-chromosome for most of our studies of human migration. This is because the Y shows greater differences in frequency between populations than most other genetic markers. As Dick Lewontin's analysis showed, most of the genetic diversity in the human species is found within populations, with a tiny fraction - 10 to 15 per cent - distinguishing between. For the Y, 30 to 40 per cent of the diversity is found between populations. Greater genetic contrast provides better resolution, which is why the Y is so good at tracing migrations.

When the Y was first studied as a marker of population affinity, one of the results that kept popping up again and again is that it connected people to a particular location. With a few DNA polymorphisms, it was possible to achieve incredible geographic resolution - there were even Y-chromosome polymorphisms that were limited to particular villages. If you imagine population genetics as a game of twenty questions, most genetic systems, including blood groups and mitochondrial DNA, needed all twenty to identify even the coarsest pattern,

{p. 176} such as which continent the individual came from. In contrast, the Y could typically identify subcontinental regions with a few questions. The observation, then, was that Y-chromosome lineages were geographically localized - they tended to define people as coming from a particular place. A fantastic tool for studying population movements, but the explanation for the pattern remained elusive.

In 1998 Mark Seielstad, then a graduate student working with Luca Cavalli-Sforza and Dick Lewontin, published a paper that proposed a solution to the Y mystery. Seielstad studied Y-chromosome markers in fourteen African populations, finding that the fraction of variation distinguishing between populations was much greater than that seen for other genetic markers. In a sample of European populations, the divergence between populations as a function of geographic distance increased at a much higher rate for the Y than for other genetic systems, such as mtDNA. Seielstad's interpretation of these two patterns was that women moved more than men, dispersing their mitochondrial lineages among neighbouring populations, producing a relatively homogeneous mtDNA distribution. The men, meanwhile, stayed at home - and their Y-chromosomes diverged independently in the different populations. The finding led Cavalli-Sforza to quip that Verdi was right when he wrote 'la donna e mobile' (the woman moves).

Seielstad's publication created quite a stir, even attracting the attention of activists such as Gloria Steinem, who requested a copy. It seemed to undermine the ancient notion of peripatetic Lotharios wandering the globe, sowing their wild oats and dispersing their Y-chromosome lineages. What the activists failed to take into account, though, was that it actually reinforced the notion that women make a minor contribution to group identity. In a patrilocal society, it makes little difference who your mother was - it is your father who gives you your family or clan affiliation, and your inheritance. What Seielstad had found was that human culture has had a significant effect on the pattern of genetic variation in our species. Simple, local decisions about marriage and property, summed over hundreds of generations, had produced profound differences in the pattern of genetic variation on the male and female sides. Hindu castes show clear evidence of this pattern, with much greater Y-chromosome than mtDNA divergence

{p. 177} between the castes, suggesting that women could move between castes while men were locked into theirs.

The real test of this theory, as Seielstad pointed out, was to examine the pattern of variation in matrilocal societies. The prediction is that these would show greater divergence for mtDNA, with the Y lineages tending to be homogenized among them. This was finally done in 2001, when Mark Stoneking and his colleagues published a study on the Karen, as well as a sample of patrilocal Thai tribes from the same area. They found Seielstad's predicted pattern of greater Y diversity in the Karen, providing strong evidence that patrilocality had produced the geographic clustering in Y-chromosome variation seen in most human societies.

While this helped to explain the localization of Y-chromosome lineages, it skirted round another odd observation. As we saw in Chapter 3, the coalescence time - the time elapsed since our common ancestors, Adam and Eve - is much more recent for the Y-chromosome than it is for mtDNA. Patrilocality can explain the high degree of Y divergence between populations, but the overall coalescence time should still be the same for Y and mtDNA. In effect, the Y pattern should be fragmented into many deeply divergent populations, all of which trace their ancestry to a single African man who lived around 150,000 years ago. Instead, we see many fairly divergent populations, all of which seem to coalesce to a common ancestor as soon as they are traced back to Africa: the data points to an African Adam who lived only a few thousand years before humans started to leave the continent. This result suggested another factor at work.

The rate of genetic drift - random changes in marker frequency due to small population size - depends on the actual size of the population, as we saw earlier. In large populations drift is negligible, while in small populations the effects of drift are significant. In the smallest populations, such as those first Beringeans who colonized the Americas, tiny-population sizes can lead to a few lineages reaching frequencies of IOO per cent in a very short period of time. This is the explanation for why Native Americans are almost uniformly blood group 0 - types A and B were lost during their journey through the Siberian ice age.

This same pattern can be used to explain the recent dates for our

{p. 178} Y-chromosome ancestor. If there are fewer men than women in a population, then the rate at which Y-chromosome lineages are lost will be greater. But this can't be true, you might be saying - the birth ratio is 50: 50. Surely there are the same number of men and women in every population? Surprisingly, while this is true in terms of numbers, it is not true for the number that pass on their genes by leaving offspring. In the genetic sense, those who don't reproduce don't count, and should be excluded from the equation. What we are interested in, then, is what is known as the effective population size - the number of breeding men and women. This is where we see the difference.

The likely explanation for why there is a greater rate of lineage loss for the Y-chromosome is that a few men tend to do most of the mating. Furthermore, their sons - who inherit their wealth and social standing - also tend to do most of the mating in the next generation. Carried through a few generations, this social quirk will produce exactly the sort of pattern we see for the Y-chromosome: a few lineages within populations, and different lineages in neighbouring populations. It will also produce a very recent coalescence time for the Y, since the lineages that would have allowed us to trace back to an Adam living 150,000 years ago were lost while our ancestors were still living in Africa. The definitive proof of this hypothesis will come only from careful studies of traditional societies, where the same social patterns have been practised for hundreds or thousands of years, but my prediction is that it will be confirmed by the data. As with the search for the language of Adam and Eve, the study of the effects of culture on human genetic variation promises to be one of the most exciting areas of enquiry in anthropology over the next few decades. Unfortunately, we may be racing against the clock, as we'll see in the next chapter.

Back to the sea

We've been through a tour of how culture, from the development of agriculture to local marriage patterns, has had an effect on human genetic diversity. We are now ready to re-evaluate the Hawaiians who were 'discovered' by Captain Cook in the late eighteenth century.

{p. 179} Where did they come from, and why had they conquered the Pacific in the last few thousand years?

The first question we can ask is whether there is a linguistic relationship among the Polynesian languages that suggests a source population. The answer is that there is. While Thor Heyerdahl favoured a South American origin for the Polynesians, their languages are more closely related to those spoken in south-east Asia. As early as the nineteenth century, scholars had linked the languages of Polynesia to those spoken in Taiwan (then Formosa) and Malaysia. Today, Taiwan is inhabited by Han-speaking Chinese, but prior to the seventeenth century it was home to aboriginal groups speaking completely different languages. All of these languages were united into one family, Malayo-Polynesian, which became known as Austronesian in the early twentieth century. So, there is clear linguistic data tracing from Hawaii back to Asia, rather than the Americas.

The overlap between the Austronesian languages and the spread of agriculture in east Asia is striking, and the theory which emerged for the peopling of Polynesia is that agriculturalists who had perfected the art of sailing simply hopped from island to island through south-east Asia, eventually heading into the open ocean. The 'Express Train' model, as it became known, predicted a close genetic link between aboriginal Taiwanese and the Polynesians. MtDNA seemed to support this model, although its resolution - as we have seen elsewhere - is often limited. Recent results from the Y-chromosome, though, have suggested that the theory needs to be modified.

The pattern seen for the island south-east Asians is that, while agriculturalists of (ultimately) Chinese origin did have a significant impact on the gene pool, there are a substantial number of indigenous lineages (particularly M130) found throughout Indonesia and Melanesia. These are also present at high frequency in the Polynesians What this suggests is that after agriculture was introduced to island south-east Asia, it went through a maturation phase as it was adapted to local crops that were better suited to the environment there. Instead of flying past on their express train, the agriculturalists dawdled and dabbled, gradually adapting their culture to its new home. Archaeologist Peter Bellwood has pointed out that the crop yield of Chinese rice strains drops significantly if they are planted near the equator, since

{p. 180} they need the variation in day length found only outside the tropics in order to mature. These sorts of pressures would have encouraged agriculture to change as it passed through south-east Asia, in some cases replacing millet and rice with other crops. The Polynesian taro root, ubiquitous throughout the Pacific and used to make Hawaiian poi, reflects this change. The genes also show evidence of a sojourn in south-east Asia before heading out to sea.

The answer to our question of timing, then, can be found in the maturation phase of agriculture. It was only after a fully mature tropical variant of agriculture had taken root that the proto-Polynesians were able to set sail for undiscovered lands. They took with them their crops, confident in their ability to survive wherever they came ashore. Hunter-gatherers would never have been able to make this leap into the unknown ocean - repeatedly - because they had no idea what lay beyond the horizon. The Polynesians, though, as inheritors of a well-adapted agricultural tradition, were in control of their own destinies. They may have been encouraged to set sail by an expanding population at home (another consequence of agriculture), but their unique solution was only possible because they had the choice of sailing into the unknown. ...

{end of quotes}

See the associated map at wells-Y-marker-map.jpg.

I have omitted Wells' material on the Americas, East Asia and the Pacific.

Material such as the above is so relevant to human history that, in a sense, it should belong in a global Commons.

It is never my purpose to deprive an author or publisher of income. Rather, I hope to entice the reader to buy the book.

Apart from Wells' book The Journey of Man , there is a DVD of the same name. When buying, check which one you are obtaining.

Buy The Journey of Man (book) (new or 2nd hand) at Amazon Books: http://www.amazon.com/Journey-Man-Genetic-Odyssey/dp/0812971469/ref=sr_1_1?ie=UTF8&s=books&qid=1212712417&sr=1-1.

Buy a 2nd hand copy through Abebooks: http://www.abebooks.com/servlet/SearchResults?an=spencer+wells&bi=0&bx=off&ds=30&sortby=2&sts=t&tn=journey+of+man&x=66&y=13.

(2) Deep Ancestry: Inside the Genographic Project, by Spencer Wells

Spencer Wells' latest book is Deep Ancestry (2006).

It goes into greater detail about haplogroups (of both mtDNA and Y Chromosome). Persons sharing a genetic marker eg M9 or M52 are assigned to haplogroups.

The major Y chromosome haplogroups for Europe are R1a1, R1b, I1a, I1b, J2, N and E36. Major mitochondrial haplogroups for Europe are H, K, T, U, V and J.

Haplogroup J is a genetic signature from the first Neoloithic agriculturalists in the Middle East about 10,000 years ago; they expanded outwards from there, east as far as the Indus Valley, and also into east and central Europe, but the lineages carried by these Neolithic expansions are found today at low frequencies.

For Haplogroup J, the ancestral line is "Eve" -> L1/L0 -> L2 -> L3 -> N -> R -> J.

For R1b, the ancestral line is "Adam" -> M168 -> M89 -> M9 -> M207 -> M173 -> M343. Persons in R1b are descendants of the Cro-Magnon people, some of whom created the cave paintings in southern France. M343 is the defining marker of haplogroup R1b.

For R1a1, the ancestral line is "Adam" -> M168 -> M89 -> M9 -> M45 -> M207 -> M173 -> M17. M17 is the defining marker of haplogroup R1a1; its bearers were the Aryan invaders from the steppes.

Deep Ancestry: Inside the Genographic Project

Spencer Wells

National Geographic, Washington DC 2006

{the Appendix on the haplogroups is presented first here}

{p. 175} Appendix

MITOCHONDRIAL DNA AND Y-CHROMOSOME HAPLOGROUP DESCRIPTIONS

{p. 176} mtDNA HAPLOGROUPS

{p. 180} HAPLOGROUP M

Ancestral line: "Eve"" -> Ll/LO" -> L2" -> L3" -> M

{p. 181} ... This haplogroup is prevalent among populations living in the southern parts of Pakistan and northwest India, where it constitutes around 30 to 50 percent of the mitochondrial gene pool. The wide distribution and greater genetic diversity east of the Indus Valley indicates that these haplogroup M-bearing individuals are descended from the first inhabitants of southwestern Asia. These people underwent important expansions during the Paleolithic, and the fact that some East Asian haplogroup M lineages match those found in Central Asia indicates much more recent migration into Central Asia from the east.

{p. 187} HAPLOGROUP N1

Ancestral line: "Eve"" -> Ll/LO" -> L2" -> L3" -> N" -> N1

In addition to a wide geographic distribution similar to N, this haplogroup is significant because its members constitute one of the four major Ashkenazi Jewish founder lineages. "Ashkenazi" refers to Jews of mainly central and eastern European ancestry. Most historical records indicate that the founding of this population took place in the Rhine Basin and subsequently underwent vast population expansions. The Ashkenazi population was estimated at approximately 25,000 individuals around A.D. 1300,

{p. 188} whereas that number had increased to about 8,500,000 individuals by the turn of the 20th century.

Around half of all Ashkenazi Jews trace their mitochondrial lineage back to one of four women, and haplogroup N1 represents one. It is seldom found in non-Ashkenazi populations, although it does appear at frequencies of roughly 3 percent or higher in those from the Levant, Arabia, and Egypt, which indicates a strong genetic role in the Ashkenazi founder event. Today, haplogroup N1 is the second most common in Ashkenazi Jews and is shared by around 800,000 people.

{p. 197} HAPLOGROUP H

Ancestral line: "Eve" -> L1/L0 -> L2 -> L3 -> N -> R -> pre-HV -> HV-> H

As humans began to repopulate western Europe after the ice age, by far the most frequent mitochondrial lineage carried by these

{p. 198} expanding groups was haplogroup H, which came to dominate the European female landscape.

Today haplogroup H comprises 40 to 60 percent of the gene pool of most European populations. In Rome and Athens, for example, H is found in about 40 percent of the entire population, and it exhibits similar frequencies throughout western Europe. Moving eastward the frequencies of H gradually decrease, illustrating the migratory path these settlers followed as they left the Iberian Peninsula after the ice sheets had receded. Haplogroup H is found at around 25 percent in Turkey and around 20 percent in the Caucasus Mountains.

While haplogroup H is considered fhe western European lineage due to its high frequency there, it is also found much farther east. Today it comprises around 20 percent of southwest Asian lineages, about 15 percent of people living in central Asia, and around 5 percent in northern Asia.

Importantly, the age of haplogroup H lineages differs quite substantially between those seen in the West compared with those found in the East. In Europe its age is estimated at 10,000 to 15,000 years old, and while H made it into Europe substantially earlier (30,000 years ago), reduced population sizes resulting from the ice age significantly reduced its diversity there, and thus its estimated age. In Central and East Asia, however, its age is estimated at around 30,000 years old, meaning the lineage made it into those areas during some of the earlier migrations out of the Near East.

HAPLOGROUP V

Ancestral line: "Eve" -> L1/L0 -> L2 -> L3 -> N -> R -> pre-HV -> HV-> V

Today, haplogroup V tends to be restricted to western, central, and northern Europe. Its age is estimated at around 15,000 years old, indicating that it likely arose during the 5,000 years or so

{p. 199} that humans were confined to the European reftigia during the last ice age. It is found in around 12 percent of Basques, an isolated population in northern Spain, and around 5 percent in many other western European populations. It is also found in Algeria and Morocco, indicating that these humans migrating out of the Iberian Peninsula also headed south across the Strait of Gibraltar and into North Africa. Its genetic diversity reduces gradually moving west to east, indicating the migratory direction these groups followed during the recolonization.

Interestingly, haplogroup V attains its highest frequency in the Skolt Saarm of northern Scandinavia, a group of hunter-gatherers who follow the reindeer herds seasonally from Siberia to Scandinavia and back. While V makes up about half the mitochondrial lineages in the Saami, its genetic diversity is considerably reduced compared to that observed in western Europe, and was likely introduced into the Saami within the past several thousand years.

{p. 199} HAPLOGROUP J

Ancestral line: "Eve" -> L1/L0 -> L2 -> L3 -> N -> R -> J

This group of individuals also descended from a woman in the R branch of the tree. The divergent geenetic lineage that constitute haplogroup Jindicates that she lived sometime around 40,000 years ago. Haplogroup J has a very wide distribution, and is present as far east as the Indus Valley bordering India and Pakistan, and as far south as the Arabian Peninsula. It is also common in eastern and northern Europe. Although this haplogroup was present during the early and middle Upper Paleolithic, J is largely considered one of the main genetic signatures of the Neolithic expansions.

While groups of hunter-gatherers and subsistence fishermen had been occupying much of Eurasia for tens of thousands of years, around 10,000 years ago a group of modern humans living in the

{p. 200} Fertile Crescent (present-day eastern Turkey and northern Syria) begin domesticating the plants, nuts and seeds they had been collecting. What resulted were the world's first agriculturalists, and this new cultural era is typically referred to as the Neolithic.

Groups of individuals able to support larger populations with this reliable food source began migrating out of the Middle East, bringing their new technology with them. By then, humans had already settled much of the surrounding areas, but this new agricultural technology proved too successful to ignore, and the surrounding groups quickly copied these new immigrants. Agriculture was quickly and widely adopted, but the lineages carried by these Neolithic expansions are found today at low frequencies.

Haplogroup J has greater diversity in the Near East than in Europe, indicating a homeland for J's most recent common ancestor around the Levant, a coastal region in Lebanon., It reaches its highest frequency in Arabia, comprising around 25 percent of the Bedouin and Yemeni. But genetic evidence indicates that the higher incidence is more reflective of low population sizes or the occurrence of a founder event than this region actually being the geographic origin of haplogroup J.

HAPLOGROUP K

Ancestral line: "Eve" -> Ll/LO -> L2 -> L3 -> N -> R -> K

K individuals also descend from a woman in the R branch of the tree. Because of the great genetic diversity found in haplogroup it is likely that she lived around 50,000 years ago. ...

{p. 201} While some members of this haplogroup headed north into Scandinavia, or south into North Africa, most members of haplogroup K stem from a group of individuals who moved northward out of the Near East. These women crossed the rugged Caucasus Mountains in southern Russia, and moved on to the steppes of the Black Sea.

Like the N1 lineage, haplogroup K is very significant because it and its subgroups also constitute three of the four major Ashkenazi Jewish founder lineages. Around half of all Ashkenazi Jews trace their mitochondrial lineage back to one of four women, and haplogroup K represents a lineage that gave rise to three of them. While this lineage is found at a smaller frequency in non-Ashkenazi Jews, the three K lineages that helped found the Ashkenazi population are seldom found in other populations. While virtually absent in Europeans, they appear at frequencies of 3 percent or higher in groups from the Levant, Arabia, and Egypt. This indicates a strong genetic role in the Ashkenazi founder event, which likely occurred in the Near East. Today, K has given rise to three of the four most common haplogroups in Ashkenazi Jews and is currently shared by more than three million people.

Haplogroup T

Ancestral line: "Eve" -> L1/L0 -> L2 -> L3 -> N -> R -> T

The divergent genetic lineage that constitutes haplogroup T indicates that the woman they descended from lived sometime around 40,000 years ago. Haplogroup T has a very wide distribution, and is present as far east as the Indus Valley.and as far south as the Arabian Peninsula. It is also common in eastern and northern Europe. Although this haplogroup was present during the early and middle Upper Paleolithic, T is largely considered one of the main genetic signatures of the Neolithic expansions.

{p. 204} Y-CHROMOSOME HAPLOGROUPS

{p. 211} HAPLOGROUP H

Ancestral line: "Adam" -> M168 -> M89 -> M69

Ancestors of haplogroup H migrated along the steppe super-highway from the Middle East around 45,000 years ago, continuing toward India. On this journey, a boy was born approximately 30,000 years ago with the M69 marker, which came to define this new lineage. Although M69 is known as an "Indian Marker," this male ancestor may have been born in southern central Asia. His descendants were part of the first major inland settlement of India.

Geneticists believe that haplogroup H might have originated somewhere along the migration route of peoples carrying the

{p. 212} M20 Y-chromosome marker (see "Haplogroup L"). The peoples of these clans migrated along the steppe highway from the Middle East, and then moved south into India.

These were not the first humans to arrive in India, but they likely undertook the first major settlement of the region some 30,000 years ago. The earliest waves of African migrants had traveled along the Indian coastline 50,000 to 60,000 years before; some settled along the coastal route, but inland areas were largely populated by H members coming down from the north.

HAPLOGROUP H1 Ancestral line: "Adam" -> M168 -> M89 -> M69 -> M52

The specific genetic marker that defines haplogroup H1, M52, is part of a largely Indian lineage. This marker made its first known appearance some 25,000 years ago in India. M52 was part of the second major wave of human migration into India, long after a large wave of African migrants traveled along the Indian coastline 50,000 to 60,000 years ago.

Haplogroup H1 ancestors appear to have arrived from the north on a journey from the Middle East and were likely the first to establish significant settlement in India. It became quickly established throughout the area and was successfully passed on to subsequent generations.

Today H1 is found at frequencies as high as 25 percent in some Indian populations. Members of haplogroup H1 are also found in lower frequencies in Iran and throughout much of southern central Asia. ...

{p. 214} HAPLOGROUP J

Ancestral line: "Adam" -> M168 -> M89 -> M304

{p. 215} The patriarch of haplogroup J was born around 15,000 years ago in the Fertile Crescent, a region today that includes Israel, the West Bank, Jordan, Lebanon, Syria, and Iraq. Today the M304 marker appears at its highest frequencies in the Middle East, North Africa, and Ethiopia. In Europe, it is seen only in the Mediterranean region. The early farming successes of the J lineages spawned population booms and encouraged migration throughout much of the Mediterranean world. In fact, both haplogroup J and its subgroup J2 are found at a combined frequency of around 30 percent among Jewish individuals.

HAPLOGROUP J1

Ancestral line: "Adam" -> M168 -> M89 -> M304 -> M267

Haplogroup J1 emerged during the Neolithic Revolution in the Middle East. Members of the J1 clans shared the farming successes of the other J haplogroups. In particular some J1 individuals moved back into North Africa and were quite successful, as evidenced by the fact that this haplogroup is currently seen there at its highest frequency. While other members of haplogroup J1, bearing the characteristic genetic marker M267 remained in the Middle East, some moved northward into western Europe, where today J1 is found at low frequencies.

HAPLOGROUP J2

Ancestral line: "Adam" -> M168 -> M89 -> M304 -> M172

The M172 marker defines a major subset of haplogroup J, which arose from the M89 lineage. Haplogroup J2 is found today in North Africa, the Middle East, and southern Europe. In southern Italy it occurs at frequencies of 20 percent, and in southern Spain, 10 percent of the population carries this marker.

{p. 216} HAPLOGROUP K

Ancestral line: "Adam" -> M168 -> M89 -> M9

The marker M9 first appeared in a man born around 40,000 years ago in Iran or south-central Asia, which marked a new lineage diverging from the M89 Middle Eastern clan. His descendants spent the next 30,000 years populating much of the planet.

This large lineage, called the Eurasian Clan, dispersed gradually over thousands of years. Seasoned hunters followed the herds ever eastward, along the vast superhighway of the Eurasian steppe. Eventually their path was blocked by the massive mountain ranges of south-central Asia: the Hindu Kush, the Tian Shan, and the Himalayas.

The three mountain ranges meet in the center of a region known as the Pamir Knot, located in present-day Tajikistan. Here the tribes of hunters split into two groups. Some moved north into central Asia, others moved south into what is now Pakistan and the Indian subcontinent. These different migration routes through the Pamir Knot region gave rise to more separate lineages. Most people native to the Northern Hemisphere trace their roots to the Eurasian Clan. Nearly all North Americans and East Asians are descended from this man, as are most Europeans and many Indians.

HAPLOGROUP K2

Ancestral line: "Adam" -> M168 -> M89 -> M9 -> M170

Not all M9 descendants challenged the problem of the Pamir Knot. Others stayed in the relatively fertile environment of the Near East. There, some 30,000 years ago the marker M70 appeared and today defines this haplogroup, K2. Ancient members of haplogroup K2 dispersed across the Mediterranean world. They traveled west along the coast of North Africa and

{p. 217} also along the Mediterranean coastline of southern Europe. These movements suggest an intriguing possibility that the M70 marker may have been carried by Mediterranean traders such as the Phoenicians. These seafariing people established a formidable, first-millennium B.C. trading empire that spread westward across the Mediterranean from its origins on the coast of modern Lebanon. M70 is found today throughout the Mediterranean, but it shows its highest frequency (about 15 percent) in the Middle East and in northeast Africa. Members of this haplogroup are also found in southern Spain and France.

HAPLOGROUP L

Ancestral line: "Adam" -> M168 -> M89 -> M9 -> M20

This part of the M9 Eurasian clan migrated south once they reached the rugged and mountainous Pamir Knot region. The man who gave rise to marker M20 was possibly born in India or the Middle East. His ancestors arrived in India around 30,000 years ago and represent the earliest significant settlement of India. For this reason haplogroup L is known as the Indian Clan. More than 50 percent of southern Indians carry marker M20 and are members of haplogroup L, even though they were not the flrst people to reach India.

{p. 224} HAPLOGROUP R1A1

Ancestral line: "Adam" -> M168 -> M89 -> M9 -> M45 -> M207 -> M173 -> M17

Sometime between 10,000 to 15,000 years ago, a man of European

{p. 225} origin was born in present-day Ukraine or southern Russia. His nomadic descendants would eventually carry his genetic marker, M17, from the steppes to places as far away as India and Iceland. Archaeologists speculate that these people were the first to domesticate the horse, which would have eased their distant migrations.

In addition to genetic and archaeological evidence, the spread of languages can trace prehistoric migration patterns of the R1A1 clan, whose descendants may be have spread the Indo-European languages. The world's most widely spoken language family, Indo-European tongues include English, French, German, Russian, Spanish, several Indian languages such as Bengali and Hindi, and numerous others. Many of the Indo-European languages share similar words for animals, plants, tools, and weapons.

Some linguists believe that the Kurgan people, nomadic horsemen roaming the steppes of southern Russia and the Ukraine, were the first to speak and spread a proto-Indo-European language, some 5,000 to 10,000 years ago. Genetic data and the distribution of Indo-European speakers suggest the Kurgan, named after their distinctive burial mounds, may have been descendants of M17.

Today a large concentration - around 40 percent - of the men living from the Czech Republic across the steppes to Siberia, and south throughout central Asia are descendants of this clan. In India, around 35 percent of the men in Hindi-speaking populations carry the M17 marker, whereas the frequency in neighboring communities of Dravidian speakers is only about 10 percent. This distribution adds weight to linguistic and archaeological evidence suggesting that a large migration from the Asian steppes into India occurred within the last 10,000 years.

The M17 marker is found in only 5 to 10 percent of Middle Eastern men. This is true even in Iranian populations where Farsi, a major Indo-European language, is spoken. Despite the low frequency, the distribution of men carrying the M17 marker in Iran provides a striking example of how climate conditions, the spread

{p. 226} of language, and the ability to identify specific markers can combine to reveal migration patterns of individual genetic lineages. In the western part of the country, descendants of the Indo-European Clan are few, encompassing perhaps 5 to 10 percent of the men. However, on the eastern side, around 35 percent of the men carry the M17 marker. This distribution suggests that the great Iranian deserts presented a formidable barrier and prevented much interaction between the two groups.

HAPLOGROUP R1B

Ancestral line: "Adam" -> M 68 -> M89 -> M9 -> M207 -> M173 -> M343

Around 30,000 years ago, a descendant of the clan making its way into Europe gave rise to marker M343, the defining marker of haplogroup R1b. These travelers are direct descendants of the people who dominated the human expansion into Europe, the Cro-Magnon. The Cro-Magnon created the famous cave paintings found in southern France, providing archaeological evidence of a blossoming of artistic skills as people moved into Europe. Prior to this, artistic endeavors were mostly comprised of jewelry made of shell, bone, and ivory; primitive musical instruments; and stone carvings. The cave paintings depict animals and natural events important to Paleolithic life such as spring molting, hunting, and pregnancy. The paintings are far more intricate, detailed, and colorful than anything seen prior to this period.

HAPLOGROUP R2

Ancestral line: "Adam" -> M168 -> M89 -> M9 -> M207 -> M124

About 25,000 years ago, a man from southern central Asia first displayed the genetic marker M124. His descendants migrated

{p. 227} to inhabit what is now Pakistan and also faAer east in modern India..Today they are found in Northern India, Pakistan, and southern ~central Asia at frequencies of 5 to 10 percent. The k2 lineage also belonged to the second major human migration into India, long after the first large wave of African migrants 50,000 to 60,000 years ago.

Members of the haplogroup R2 are also found in Eastern Europe among the Gypsy populations. Their genes tie these wandering peoples back to their origins on the Indian subcontinent. These ancient migrations and the distribution of their genetic lineages still remain mysterious as scientists search for more data with which to uncover the history of this haplogroup.

{end Appendix: Haplogroup Descriptions}

{p. 76} Occasionally, when the DNA is being copied to pass on to the next generation the cellular "typist" will add or subtract a repeat. ...

These short, repetitive parts of the genome are known as microsatellites. They typically have around a dozen copies of the repeated nucleotides (the ISS in MISSISSIPPI). When the DNA is copied in each generation to pass on to the next, there is a slight chance - around one in a thousand - that a repeat unit will be added or subtracted. By measuring the length of the DNA fragment using molecular sizing techniques, it is possible to infer how many repeat units are present. And because this one in a thousand mutation rate is much greater than the one in a hundred million rate for normal DNA sequences, it means that we will rapidly generate diversity in a genetic lineage. Since we know that the clan-defining change only occurred once, given the level of variation we see today we can infer how long these mutation changes have been taking place - in other words, the age of the lineage - just like we do for the mtDNA. The older a lineage, the more variation it has accumulated through this stuttering process. Lineages that are one thousand generations old will have more accumulated stuttering variation than those that have only been around for ten or one hundred generations.

When applying this stuttering analysis to the Y clans, a pattern similar to the one for the mtDNA clans emerges. Two Y haplogroups (J2 and E3b) seem to have

{p. 77} expanded in Europe in the past 10,000 years from an origin in the Middle East. These men likely accompanied the J clan women with those first farming communities. Like the mtDNA patterns, they account for only a small fraction of the genetic lineages in Europe - also around 20 percent; 80 percent stem from other migrations. However, the Y changes reveal details that the mtDNA mutations do not.

The Y-chromosome R1A1 clan also dates from this time, but its unusual distribution suggests it spread across the steppes of southern Russia into central Europe, not from the Middle East. We don't see a similar pattern for the mtDNA, but the high frequency of R1A1 in central Europe (where as many as 40 percent of men are members of this clan) begs an explanation. The most likely answer is that the populations living on the steppes of eastern Europe were the first to domesticate the horse, and this advance may have allowed them to spread across the steppe region. It's interesting that R1A1 has such an abrupt drop in frequency as we move west into the forested regions of Europe - precisely what we would predict if the advantages conferred by horses were what allowed the clan to become so successful on the open grasslands. Consistent with this interpretation, R1A1 is also found throughout central Asia and down into northern India. It's possible that the early speakers of Indo-European languages (which include English, French, and the other languages of Europe, as well as those spoken in Iran and much of India) could have been R1A1 clan members, and their nomadic lifestyle could account for the spread of this lineage. The lack of a similar pattern on the female side suggests that conquest - a largely male-driven process since soldiers are typically men - could have contributed to the spread.

{p. 69} The second factor to consider about the distribution of haplogroup J is that, while it is found in many European populations and sometimes reaches fairly high frequencies, it is still not the dominant haplogroup in any population. This suggests that the members of the J clan did not simply sweep other people aside as they migrated into Europe, but rather mixed with other clans. We need to study these other

{p. 70} lineages to answer the question of how Europe became agricultural - the J story isn't enough.

{p. 71} Rather than dating to the last 10,000 years, most of the major mtDNA lineages are much older (Figure 6). Instead of the relatively low level of diversity we see in haplogroup J, we see far more. This is particularly true for haplogroup U, which seems to have been accumulating diversity for at least 50,000 years. Haplogroups U, H, T, and V seem to predate the arrival of the first farmers. In fact, the only lineage that appears to have arrived in Europe after 8,000 years ago is haplogroup J, as well as some less common lineages (mostly subgroups of H and T). The majority seem to date back to the time before agriculture arrived, a period known as the Paleolithic. Most European women - around 80 percent - have genetic lineages that did not spring from the Middle East with the expansion of farming. Rather, their ancestors have been living in roughly the same locale for tens of thousands of years, tracing their ancestry back to the pre-Neolithic hunting and gathering populations of early Europe. The theory of Ammerman and Cavalli-Sforza, of farming being carried into Europe on a wave of humanity, seems to have been proven wrong. Rather, it looks like most European hunters voluntarily chose to give up their existing lifestyle for a sheaf of wheat - or at least the women did. But what about the male version of the story? Can we conduct a similar analysis for the Y chromosome?

{p. 159} In a paper published in 2000, more than 13 years after the discovery of Eve, Underhill, Oefner, and 21 others (myself included) described the clearest view yet obtained of our common male ancestor. The study presented a genealogical tree - the basis for the Y-chromosome trees throughout this book - based on the new collection of Y-chromosome polymorphisms. It gave us a first glimpse at the fascinating geographic patterns we have explored in this book. It also showed how recently our common Y-chromosome ancestor lived - only 60,000 yars ago.

The date was stunningly recent, since it revealed that all of the Y-chromosome diversity in the world had been generated in that comparatively short amount of time. We were surprised because it differed so much from the date for Eve. If she lived 170,000 years ago, and Adam lived 60,000 years ago, that's a long time to wait around for your mate to show up. Where were the men 170,000 years ago?

The reason we don't find male lineages coalescing at the same time as Eve is because of early human sexual behavior. In most traditional societies, a few men do most of the mating - think of chieftains and warlords, for instance. Some men never get to have children, while others have more than their fair share. This is known as the variance in reproductive success, and it is higher for men than women, which means that women have more equal opportunities to have children. Since women are passing on their mtDNA, the result is that - in general - mtDNA

{p. 160} lineages have a more equal chance of being passed on to the next generation than Y-chromosome lineages. This behavioral quirk tends to reduce the elective population size of the Y chromosome, since not every man will pass on his Y, while most women will pass on their mtDNA. Since genetic drift acts more quickly in small populations, changes in the lineage composition of the Y-chromosome pool occurred more rapidly. Over time, this meant that Y lineages were more likely to be lost than mtDNA lineages. The result is that the deeper Y lineages were lost-over the past 170,000 years, and the only ones left date to around 60,000 years ago.

While the difference between the Y and mtDNA dates are important technical issues, their only real significance for our story lies in placing the origin of modern humans in Africa. The more recent date - that of the Y-chromosome coalescence - tells us that modern humans were still living in Africa until 60,000 years ago, and only after this time did they leave the continent to populate the rest of the world. We made our journey around the world and generated the dizzying amount of diversity that defines us today - colors, sizes, languages, and cultures. This adds up to only 2,000 generations, which is the blink of an eye on an evolutionary timescale. Such a date also helps to make sense of the earlier results, such as Lewontin's finding that we all appear to be part of an extended family at the genetic level. If we only started diverging so recently, it's no wonder.

The complex tapestry of human diversity is woven together by genetic threads, connecting us through the migrations of our ancestors. Deciphering the complex patterns in this tapestry is the goal of the Genographic Project. The end result will, we hope, serve to unite the world's people while respecting the incredible diversity that defines us as a species. If we can achieve this, the Project will have succeeded.

{end of quotes}

If a few men did most of the mating, it means that Karl Marx's idea of Primitive Communism is wrong. In hunter-gatherer society, men killed other men and acquired their women.

Such behaviour is described from first-hand observation in a sympathetic book about the Kreen-Akrore, one of the last wild peoples in the Amazon: The Tribe that Hides from Man, by Adrian Cowell (one of the best books I have ever read; it cost me 20c at Bundaberg rubbush tip).

The Kreen-Akrore have been saved from extinction, have been brought into the modern world, and are now asserting indigenous rights. They are known today as the Panara: http://en.wikipedia.org/wiki/Panar%C3%A1.

Buy Deep Ancestry from Amazon (new or 2nd hand): http://www.amazon.com/Deep-Ancestry-Inside-Genographic-Project/dp/1426201184/ref=sr_1_2?ie=UTF8&s=books&qid=1212712417&sr=1-2.

Buy Deep Ancestry from abebooks (2nd hand): http://www.abebooks.com/servlet/SearchResults?an=spencer+wells&bi=0&bx=off&ds=30&sortby=2&sts=t&tn=deep+ancestry&x=0&y=0.

Marija Gimbutas on the Aryan invasions: gimbutas.html.

Write to me at contact.html.

HOME