The 10,000 Year Explosion

by Gregory Cochran

  So it may be that the Sarmatians introduced key elements of the Arthur mythos, the Matter of Britain, the subject of books and poems and movies for hundreds of years. A good story can go a long way, as can a good allele. There is a real similarity. A slight contact can transmit an idea, if it falls on fertile ground— if people like the idea and repeat it. In the same way, a long- forgotten Roman transfer of troops may have played a key role in the genetic history of Britain. A few copies of a favorable allele can increase tremendously, given time: The average Englishman has only a dab of Sarmatian ancestry, but might be mostly Sarmatian in a key gene or two.

  BLUE EYES

  A strong state fosters gene flow within itself through trade, free movement, and sometimes forced movement. At the same time, it tends to limit gene flow from outside, particularly when it upholds its borders with military force. When such a state disintegrates, there can be a massive movement of peoples—in part because the borders are no longer defended, but even more in those cases in which the inhabitants have lost their military habits over the course of a long imperial peace. In classical times, one obvious sign of that kind of relative domestic tran- quility was unwalled cities.

  The fall of the Roman Empire followed this pattern. As the state weakened, many ethnic groups entered. In the early days they were often mercenary soldiers, but eventually some came as plundering bands. Those invaders also carried new al- leles. We will discuss the Vandals, one of the most spectacular of these groups. We think they may have played a part in spreading a particular far-flung new allele, the one responsible for blue and green eyes.

  Blue eyes are common in Europeans and their descendants and are found to some extent in adjacent populations, but they are essentially nonexistent in most of the world. Some 10,000 years ago there seems to have been no such thing. There are other shades of eye color, and other genes have some influence, but most of the story boils down to a single new allele of the gene named OCA2 (for oculocutaneous albinism II, also discussed in Chapter 4). To be exact, blue eyes are caused by a change in a DNA sequence that regulates the expression of OCA2, a sequence that is embedded in HERC2, the gene next to OCA2.16That allele accounts for 75 percent of the variation in eye color in Europe. It's the third longest haplotype in Europeans and therefore can't be very old: Analysis of the unshuffled region associated with OCA2 suggests that it originated about 6,000 to 10,000 years ago. Blue eyes are most common in northern Europe, centered around the Baltic. The simplest assumption

  A member of the Tuaregs, nomadic Berbers of the Sahara

  is that the allele originated in the center of the region, where its frequency is very high today, so our best guess is that it first occurred in a Lithuanian village about 6,000 years ago.

  Clearly the spread of this allele involved more than just men marrying the girl next door. You can see light eyes in Berbers from the Atlas Mountains of Morocco, and even in Tuaregs living in the Sahara. Light eyes are fairly common in Kurds living in the Zagros Mountains along the border between Iraq and Iran, and we can find people with light eyes as far away as Afghanistan, over 3,000 miles from Vilnius. In order to explain these patterns, we're going to have to make a little excursion through history. Again, a gene is the center of attention rather than battles and kings.

  Light-eyed Afghan girl

  Start with the Berbers. Clearly the OCA2 allele didn't get there solely by peasants marrying their neighbors, since the Mediterranean Sea blocks that kind of connection. It could be that blue eyes in Morocco are caused by some local mutation, not the same one that has spread over Europe. But if it is the same mutation, as seems likely, since we have found the same OCA2 haplotype around the Mediterranean, the prime candidates for bringing that gene to North Africa are barbarians and pirates.

  Barbarians came first. During the twilight of the Roman Empire, whole tribes, mostly some flavor of German, began wandering across the borders. The Vandals were among the most troublesome. They are thought to have originated insouthern Sweden and had moved to Silesia (now eastern Poland) by 120 BC. By the third century, they had moved on to western Romania and Hungary, then into Roman territory. Around AD 400 they began moving westward (along with their allies, the Alans and Suebians), crossing the frozen Rhine on the last day of 406. They went on to devastate France, plundering their way through Aquitaine and crossing the Pyrenees in late AD 409. The Alans (an Iranian-speaking people off the steppe) set up a kingdom in Portugal, while the Suebians settled down in Galicia.

  Facing pressure from the Visigoths (yet another medieval gang, which had already crushed the Alan kingdom), the Vandals and the surviving Alans, some 80,000 souls altogether, crossed from Spain into Africa in 429. This Vandal kingdom dominated the western Mediterranean with its stolen fleet, extorting tribute from seaports and sending out raiding parties every year. The Vandals sacked Rome in 435, ensuring that their name would live in infamy. Finally, in 533, Justinian, the emperor of the remaining eastern half of the Roman Empire, sent a force to attack the Vandals. Led by Belisarius, the best general of the age and one of the best of all time, the Imperials landed in Carthage and made short work of the Vandals—some of whom, when finally defeated, may have blended into the countryside. As Edward Gibbon wrote, "When every resource, either of force or perfidy, was exhausted, Stoza, with some desperate Vandals, retired to the wilds of Mauritania, obtained the daughter of a Barbarian prince, and eluded the pursuit of his enemies, by the report of his death."17

  Seems like a lot of trouble just to inject a few thousand copies of the new OCA2 allele into the Rif.

  Later came pirates. From 1500 to 1800, Muslim corsairs captured and enslaved many Europeans: by one estimate, more than a million. Mainly they took captives from the Mediterranean coasts of Italy and Spain, but occasionally they ranged farther afield, raiding Cornwall, Ireland, and even Iceland. Most male slaves were worked to death and made little contribution to the gene pool, but women could and often did end up in harems. You see a similar pattern with the slaves the Arabs took from sub-Saharan Africa: About 5 percent of the maternal ancestry of Arabs in the Middle East is African, judging from mtDNA, but you see very few African Y chromosomes there.

  We're not suggesting that Berbers have a lot of European ancestry: Judging from Y-chromosome and mtDNA data, that does not appear to be the case. The point is that even a moderate degree of admixture can introduce many copies of a beneficial allele, and over time that allele can become common. The earlier the introduction, and the more copies introduced, the more effective this process is. If we had to guess, we'd say that the blue-eyed variant of OCA2 found in Berbers was probably introduced by the Vandals, but it may have happened earlier and involved population movements that we are not aware of.

  This page intentionally left blank

  6

  EXPANSIONS

  History is full of examples of human groups expanding at the expense of their neighbors. Anatomically modern humans expanded and replaced archaic humans, the Bantu expanded at the expense of the Bushmen and other peoples, and Turks and Mongols pushed aside the Iranian-speaking peoples who had previously occupied the steppes of central Asia. In many of those cases, there was some degree of admixture, but replacement dominated. We could cite dozens of other examples. Surely the most obvious question is why those groups expanded.

  In some cases, sheer chance may have played an important role—perhaps some key battle was lost for want of a horseshoe nail. More often, the successful group had some kind ofadvantage that drove their expansion. Anatomically modern humans probably had new and improved technologies like projectile weapons as well as more sophisticated language. The Bantu had iron tools and a set of domesticated plants adapted to Africa, which together gave them a pretty powerful advantage over the hunter-gatherers they encountered. The Turco-Mongol advantage over the steppe Iranians is less obvious, but they may have had stronger political organizations.

  The general assumption is that the winning advantage is cultural—that is to say, lear
ned. Weapons, tactics, political organization, methods of agriculture: all learned. The expansion of modern humans is the exception to the rule—most observers suspect that biological differences were the root cause of their advantage. Biological advantages are particularly potent because they last: Archaic humans such as Neanderthals may have been able to copy some of the cultural attributes of modern humans (exemplified by the Chatelperronian toolkit), but they couldn't become modern humans, couldn't copy or acquire abilities that were consequences of modern human biology. So being an anatomically modern human was an enduring advantage, and thus genetics can explain a replacement process that seems to have taken about 20,000 years (from the original trek out of Africa to the last Neanderthals).

  The assumption that more recent expansions are all driven by cultural factors is based on the notion that modern humans everywhere have essentially the same abilities. That's a logical consequence of human evolutionary stasis: If humans have not undergone a significant amount of biological change since the expansion out of Africa, then people everywhere would have essentially the same potentials, and no group would have a biological advantage over its neighbors. But as we never tire of pointing out, there has been significant biological change during that period—tremendous amounts of change, particularly in those populations that have practiced agriculture for a long time. Therefore, the biological equality of human races and ethnic groups is not inevitable: In fact it's about as likely as a fistful of silver dollars all landing on edge when dropped. There are important, well-understood examples of human biological inequality: Some populations can (on average) deal far more effectively with certain situations than others.

  New alleles that have undergone selective sweeps under agriculture show up randomly—so they might occur more in some groups than in others by sheer chance. We know that they become common first where they're favored—in the new agricultural ecology, which some peoples experienced before others. Thus, the early adopters got a head start on new adaptive genes. And the way these advantageous alleles propagate, people along lines of communication are likely to pick up more of them than people who live off the beaten path.

  Early adopters ought to be better at agriculture than latecomers: They should be better adjusted to the new diet, tougher against the new diseases, and better at tolerating crowding and hierarchy.

  Those new advantages all had to increase individual fitness, but their effects at the level of the tribe or ethnic group varied. Some aided individual survival but didn't have much effect at the group level. For example, a mutation that protected against an infectious disease wouldn't have had much effect on overall population size if food shortages were the major limiting factor. However, a new allele that allowed its bearers to digest a newfood more effectively might well have increased the size of that same group. A mutation that helped its bearers compete with other humans without conferring any other advantage, one that changed the winners without bringing any more money into the game, probably wouldn't have helped the group; it might even have weakened it.

  If a group happened to acquire one (or a few) of those mutations that increase group fitness as well as individual fitness, it would have had a real advantage over its neighbors. Its population would have expanded. Tribes and bands fight, of course— they always have. War goes back well before the birth of civilization.1 But populations with a biological advantage, more often than not, should have won the wars. They would have been able to generate more young warriors than their neighbors. They would have been able to afford to fight more often and recover faster from defeat. If the expanding group's success depended upon some improved tactic or weapon, the defenders could have copied it. But they couldn't copy a gene. It's hard to fight biological superiority, and expansions based on such superiority could have gone on far longer than ones based upon cultural advantages, which are ephemeral.

  THE COLUMBIAN EXPLOSION

  Other writers have discussed the crucial role of epidemic Eurasian and African diseases in the European expansion into the Americas, but most shy away from clearly stating that this was driven by underlying biological differences—differences that conferred a practical advantage, a kind of superiority, in this particular situation. But there is plenty of evidence that these biological differences existed. When Europeans launchedtheir first ships to the "New World," their diseases came along for the ride, and the Amerindians simply lacked the biological defenses they would have needed to withstand that onslaught.

  The Amerindians migrated from Northeast Asia some 15,000 years ago. They did not carry with them crowd diseases that arose after the birth of agriculture, nor did they carry the genetic defenses that later developed against those diseases. Since their path to the New World went through frigid landscapes like Siberia and Alaska, they left behind some of the ancient infectious diseases that were vectorborne or had complex life cycles—malaria and Guinea worm, for example. The world they entered had never before been settled by hominids or great apes, so there were few local pathogens preadapted to humans. Many of the infectious diseases found in the Old World are thought to have originated in domesticated animals, but this does not seem to have been an important factor in the Americas.

  Although Amerindians did develop agriculture indepen- dently—a very effective agriculture that included some of the world's most important crops, such as maize and potatoes— they domesticated few animals, mostly because they had already wiped out most of the species suited to domestication. That happened whenever modern humans, who were competent hunters, entered a land that had never known any kind of humans before, one in which none of the large animals had had a chance to adapt to humans. It happened in Australia, New Zealand, and Madagascar as well as the New World.

  So the selective pressures favoring disease resistance were weaker among the Amerindians than among the inhabitants of the Old World—possibly weaker than that experienced by any of our ancestors for millions of years.

  One sign of this reduced disease pressure is the unusual distribution of HLA alleles among Amerindians. The HLA system (for human leukocyte antigen) is a group of genes that encode proteins expressed on the outer surfaces of cells. The immune system uses them to distinguish self from nonself, so they play an important part in rejection of transplanted organs. But their most important role is in infectious disease. There they present protein fragments from pathogenic organisms such as bacteria to immune system cells that then attack the pathogen. In addition, when a virus infects a cell, HLA molecules display viral proteins on the outside of the cell, so that those infected cells can be destroyed by the immune system.

  HLA genes are among the most variable of all genes. There are ten or more major variants of each HLA gene, and most have more than 100 variants. Because these genes are so variable, any two humans (other than identical twins) are almost certain to have a different set of them. Because the alleles are codominant, having different HLA alleles expands the range of pathogens that our immune systems can deal with. Natural selection therefore favors diversification of the HLA genes, and some alleles, though rare, have been preserved for a long time. In fact, some are 30 million years old, considerably older than Homo sapiens. That is to say, there are HLA alleles in humans that are more similar to an allele in an orangutan than to other human alleles at that locus. Selection favoring HLA diversity— a selective pressure stemming from infectious disease—has existed more or less continuously for tens of millions of years. This is why even small populations in the Old World retain high HLA diversity.

  But Amerindians didn't have that diversity. Many tribes have a single HLA allele with a frequency of over 50 percent.2

  Different tribes have different predominant alleles: It seems as if the frequencies of HLA alleles have drifted randomly in the New World, which hasn't happened since the Miocene in the Old World. A careful analysis of global HLA diversity confirms continuing diversifying selection on HLA in most human populations but finds no evidence of any selection at all favoring diversity in
HLA among Amerindians.3

  And if infectious disease was so unimportant among Amerindians, selection most likely favored weaker immune systems, because people with weaker immune systems would be better able to avoid autoimmune disorders, in which the immune system misfires and attacks some organ or tissue. Type 1 diabetes, in which the immune system attacks the pancreatic cells that make insulin, and multiple sclerosis, where it attacks the myelin sheaths of the central nervous system, are well- known examples—both are rare among Amerindians. A less vigorous immune system would have been an advantage under those conditions.

  So, there is every reason to think that the inhabitants of the Americas were not just behind the immunological times: While the Old Worlders were experiencing intense selection for increased resistance to infectious disease, the Amerindians were actually becoming more vulnerable. They were adapted to the existing circumstances, but not to the coming collision with the Old World.

  These long-term differences in selection pressures had dramatic consequences when Columbus brought the Old and New Worlds into regular contact. Eurasian infectious diseases such as smallpox, measles, diphtheria, whooping cough, leprosy, and bubonic plague were introduced to the Americas in short order. In tropical and subtropical areas, they were eventually joinedby yellow fever, dengue fever, falciparum malaria, lymphatic fi- lariasis, schistosomiasis, and onchocerciasis (river blindness), most of which came from Africa. Relatively few pathogens, however, traveled in the other direction, from the Americas to the Old World. Syphilis4 and tungiasis (a flea that burrows into the skin) are the only human pathogens we are aware of that originated in the New World and have spread to the Old, although there may be others we haven't yet recognized—for example, some epidemiologists suspect that rheumatoid arthritis is caused by a cryptic New World pathogen.