Human Diversity

by Charles Murray

  Allele frequency. The evolution of population differences in traits that already exist is driven by changes in allele frequency. Suppose that some trait is determined by 100 SNPs and that the average allele frequency for the alleles that promote that trait is .20 among Bretons and .60 among the Hakka. The trait in question will probably be more pronounced or more common among the Hakka than among Bretons, but the trait is not exclusively Hakka. On the contrary, the luck of the draw means that some Bretons will have a more pronounced expression of that trait than some Hakka.

  CALCULATING AND EXPRESSING AN ALLELE FREQUENCY

  Allele frequency is a proportion ranging from 0 to 1. It can also be expressed as a percentage—an allele frequency of .15 is equivalent to 15 percent of genomes—but I will express it as a proportion instead because most people instinctively interpret a percentage as referring to the percentage of the population. That’s incorrect. An allele frequency is the proportion of gene copies within a population, which means the denominator is based on chromosomes, not people.

  Since each individual carries two copies of each gene, the total number of gene copies in a population of 100 people is 200. Suppose that the two alleles in a particular site are A and a and we want to calculate the allele frequency of the A-allele. An individual could carry the genotypes AA, Aa, or aa. Suppose that 20 people have the AA combination. They contribute 40 copies of the A-allele. Another 20 people have the Aa combination, contributing 20 copies of the A-allele. The remaining 60 people have the aa combination. So while 40 percent of the population carries at least one A-allele, the allele frequency is (40 + 20)/200, or 30 percent, which I will express as .30.

  The five main candidates driving evolution through either mutations or changes in allele frequencies are natural selection, sexual selection, migration, introgression, and genetic drift. You have already encountered genetic drift in chapter 6. Here are quick summaries of the other four:

  Natural selection. The most famous of the mechanisms for translating infusions of new genetic variations into effects on traits is the principle of natural selection, the momentous insight achieved independently by Charles Darwin and Alfred Russel Wallace. Here is the way Darwin put it in On the Origin of Species: “If variations useful to any organic being do occur, assuredly individuals thus characterized will have the best chance of being preserved in the struggle for life; and from the strong principle of inheritance they will tend to produce offspring similarly characterized. This principle of preservation, I have called, for the sake of brevity, Natural Selection.”6

  Both Darwin and Wallace were inspired by the same empirical observation: All species are so fertile that their populations should increase exponentially, and yet the sizes of populations are relatively stable. The necessary implication is that significant numbers of any generation fail to reproduce.

  Another empirical observation was that no two members of a species are exactly alike. Furthermore, it was obvious to Darwin and Wallace that many of the variations within a species are heritable. Hence Darwin’s and Wallace’s key inference: Some variations facilitate the transmission of traits to the next generation; some variations impede that transmission. Variations increase or diminish reproductive fitness. Evolution selects for reproductive fitness.

  Sexual selection. Sexual selection is a form of natural selection, but it doesn’t necessarily have anything to do with an organism’s abilities to survive threats to life. On the contrary, some signals that attract the opposite sex can reduce reproductive fitness. For example, bright coloring in male birds makes them more attractive to females but also makes them more visible to predators.

  Sexual selection is all about finding someone to mate with and having offspring that survive to the age of reproduction. Reproductive fitness is irrelevant. A frivolous example, but one that makes the point: Whether someone is left-handed or right-handed is genetically determined, with left-handers being in a small minority—about 10 percent of the population.7 There is no reason to think that left-handedness in and of itself augments reproductive fitness. But if for some reason women were to develop a strong sexual preference for left-handed men, it would not take long on an evolutionary time scale for the proportion of left-handed men to increase drastically.

  What’s your best chance of passing on your genes? It depends on your role in mating versus parenting, which in turn depends on your sex, which in turn depends (you may be surprised to learn) on the size of your gametes. Gamete refers to a germ cell that is able to unite with another to sexually reproduce. By biological definition, the male sex is the one that produces the smaller gametes (in human males, spermatozoa), and the female sex is the one that produces the larger gametes (in human females, eggs).8

  Why is it that so many living things are characterized by two sexes that produce gametes of very different size? Because as an empirical matter, that’s the arrangement that gives the best odds that progeny will survive. The process can be modeled mathematically. Assuming a positive correlation between the size of the fertilized cell and its survival, a stable equilibrium is often a bifurcation of the population into two types characterized by extremely different gamete sizes. The intuitive explanation of this result is that any fertilized cell resulting from a large gamete begins the struggle for life with the advantage of a stockpile of nourishment; the production of many small gametes is advantageous because the number of fertilizations is maximized. Two intermediate-sized gametes don’t work as well because neither is fully committed either to nourishing the fertilized cell or maximizing the number of fertilizations, and this mediocrity tends not to prosper.

  Now comes the evolutionary kicker that is at the source of so much contemporary argument about sex roles and child-rearing: Among almost all living things that reproduce sexually, the sex with the smaller gametes (males) provides less care after fertilization than the sex with the larger gametes (females).

  It makes evolutionary sense. In most mammalian species, the most efficient way for a male to ensure that his genes survive is to impregnate as many females as possible.9 The most efficient way for a female to ensure that her genes survive is to make the most of her limited opportunities to produce offspring, which means being choosy about the quality of the male’s fitness and devoting a lot of effort to keeping the offspring alive. In his book Male, Female, evolutionary psychologist David Geary makes the point with an extreme case: A male elephant can impregnate several females in a single day. A female elephant must carry a single offspring for 22 months before it is even born, let alone weaned. Female elephants don’t get many chances to pass on their genes. The female elephants who succeed tend to be the ones who are most adept at mating with healthy males and who are devoted to the welfare of their babies.10 Homo sapiens is an outlier among mammals, with the male providing more paternal care than is customarily observed—but it’s still not a lot compared to the burden carried from fertilization onward by the female.

  Migration. Different human populations have lived in contact with each other as long as there have been humans, and occasionally their members have had sex with each other and produced children. Over a long period of time, these offspring alter the gene pool.

  Such intermingling through migration has been going on for tens of thousands of years. And yet contemporary human groups, including different ethnicities within the same continental ancestral population, still have distinctive genetic signatures in cluster analyses and are still visually distinguishable to a greater or lesser degree. Why haven’t humans long since become a uniform shade of beige with a common set of physiological features?

  It’s not mysterious. Even though the interbreeding of populations has gone on for so long, usually it has not happened on a large scale over many generations, which means it is easily possible for populations to mingle some aspects of their gene pools and yet remain visually distinct. Suppose that back in the thirteenth century a Chinese woman had borne a child with a visitor from Italy, Marco Polo. That child was half Italian and ha
lf Chinese and probably looked like a mix. But the second generation of that initial union was only a quarter Italian, the third generation one-eighth, the fourth generation one-sixteenth—and by that time none of Marco Polo’s descendants had the slightest visible trace of their Italian ancestry.

  This has been a common experience throughout history: Interbreeding produces a visible blend in the first generation of progeny, but the heritage of one of the parents wins out over the long run. It remains surprisingly true even today: America has one of the most ethnically diverse populations in the world, with the most opportunities for children of mixed parentage to mate with other children of mixed parentage, and yet, for example, among American women who have a European American and a Chinese American parent, 82 percent marry a European American husband, putting their children (now only a quarter ethnically Chinese) on the road toward eventual indistinguishability from fully European Americans.11

  Even though encounters among human populations have not led to the degree of visible blending one might intuitively expect, those encounters can nonetheless introduce changes that persist in the gene pool forever after. Suppose, for example, that Marco Polo had passed on an allele that conferred protection against a deadly pathogen afflicting the Chinese. In this hypothetical case, it would still be true that several generations later his descendants in China would look exactly like any other Chinese and would carry few distinctively Italian alleles—but one of them could be that highly valuable antipathogen allele, already in the process of spreading among the Chinese population without leaving a visible trace.12

  Introgression. The fastest way to introduce completely new genes into a species, orders of magnitude faster than mutation, is interbreeding with another species—introgression.13 It’s not common. A breeding wall usually prevents different species from producing offspring. If they succeed, those offspring are often infertile (e.g., mules, produced by the interbreeding of horses and donkeys). But some species are interfertile, meaning that introgression produces fertile offspring. Introgression is likely to produce significant evolutionary results because of the sheer quantity of new genetic variants introduced into both genomes.

  The 1000 Genomes Project. In addition to the foregoing evolutionary terms, you need to know something about the 1000 Genomes Project that will be referenced in the rest of this chapter and will figure prominently in chapter 9.

  The 1000 Genomes Project had its inception in September 2007, when an international group of geneticists convened at Cambridge University to plan a collection of sequenced genomes from individuals around the world. The goal was to find genetic variants with allele frequencies of at least .01 in a broad range of populations. The Phase 1 database for the 1000 Genomes Project assembled information on more than 39 million SNPs and other genetic variants for 1,094 persons grouped into 14 ancestral subpopulations. Five of them came from Western and Northern Europe, three from East Asia, three from sub-Saharan Africa, and three from the Americas. In the subsequent discussion and the next chapter, I limit my use of the Phase 1 data to the subpopulations representing the Big Three continental ancestral populations that are at the center of discussions about race differences: sub-Saharan Africa (hereafter “Africans”), Western and Northern Europe (“Europeans”), and East Asia (“Asians”). The subpopulations for Africa are the Luhya in Western Kenya, Yoruba in Nigeria, and African Americans in the American Southwest. The subpopulations for Europe are British samples drawn from England and Scotland, Finns in Finland, Tuscans from Italy, and Americans of Northern and Western European ancestry from Utah. The subpopulations for East Asia are Han in South China, Han students in Beijing, and Japanese in Tokyo. Details on the populations are given in the note.[14]

  Rethinking the Nature of Recent Human Evolution

  How much evolution took place after humans left Africa? Before the genome was sequenced, geneticists had few tools for exploring that question. Some things looked as if they must have evolved after the dispersal—most obviously light skin among Europeans and Chinese and lactase persistence in Europeans. But otherwise the story of recent evolution was inaccessible.

  Then came the sequencing of the genome, “a turning point in the study of positive selection in humans,” as evolutionary geneticist Pardis Sabeti put it.15 It changed just about everything about the study of human evolution since humans left Africa: the type of natural selection at the center of geneticists’ attention, the strategies they used to identify the specific genes and SNPs involved, and the methods for implementing those strategies.

  Early on, a team of anthropologists and geneticists (first author was John Hawks) took advantage of the newly sequenced genome to test the provocative hypothesis that adaptive human evolution has recently accelerated. The transition to agriculture in some populations about 10,000 years ago led to drastic changes in diet, population density, technology, economics, and culture in general. Evolutionary biologists are still trying to understand the many intense evolutionary pressures that were generated. Among the most urgent was the need to adapt to the lethal epidemic diseases—smallpox, malaria, yellow fever, typhus, and cholera—that followed the introduction of agriculture. The Hawks study concluded that “the rapid cultural evolution during the Late Pleistocene created vastly more opportunities for further genetic change, not fewer, as new avenues emerged for communication, social interactions, and creativity.”16 It was a radical departure from the conventional wisdom, but just the beginning of a wholesale rethinking of recent evolution.

  From Darwin’s Insights to the Modern Synthesis

  The word genetics wasn’t even coined until 1905, soon after Gregor Mendel’s pioneering work was rediscovered after having been ignored for almost half a century. The first half of the twentieth century saw a series of landmark discoveries about the biology of genetic transmission, led by Thomas Hunt Morgan in the early decades and culminating in the discovery of the double-helix structure of DNA by James Watson and Francis Crick in 1953.17

  From the beginning, scientists were aware of the potential importance of genetics for explaining how evolution worked at the molecular level. One of the earliest theoretical findings was genetics’ equivalent to Newton’s first law of motion: identification of the circumstances under which the frequencies of alleles at a given site would remain stable.18 Discovered independently by Godfrey Hardy and Wilhelm Weinberg, who both published in 1908, it became known as the Hardy-Weinberg equilibrium. It provided the baseline against which to identify and measure disturbances to the equilibrium—the process of evolution.

  But there was a problem in reconciling Darwinian evolution and Mendelian genetics. The Darwinian model posited excruciatingly slow and imperceptibly small modifications in continuous phenotypic traits. In Mendelian genetics, the phenotype changed abruptly depending on a few discrete alleles—for example, the alleles determining the color or shape of the seeds among Mendel’s pea plants.

  The tension was resolved by three giants born within three years of each other: Ronald Aylmer Fisher (1890–1962), J. B. S. Haldane (1892–1964), and Sewall Wright (1889–1988). From 1915 through the mid-1930s, they laid the mathematical foundation for what has become known as the modern evolutionary synthesis, or, within the field, simply “the modern synthesis.”

  The first among equals was probably Fisher, an authentic genius (in the process of making his seminal contributions to genetics, he also made seminal contributions to modern statistics). His 1918 article “The Correlation Between Relatives on the Supposition of Mendelian Inheritance” is foundational, demonstrating that gradual variation in a trait could be the result of Mendelian inheritance.19 In so doing, he incorporated the polygenic nature of traits and the importance of allele frequency (though he did not use those terms). I have already mentioned J. B. S. Haldane’s coefficient s for expressing the probability that a mutation will go to fixation and Sewall Wright’s work in the theory of genetic differentiation among populations. But these are just examples of a cascade of contributions that F
isher, Haldane, and Wright had made by the mid-1930s. They built the theories of population and quantitative genetics that still guide these disciplines today.

  A Shift in Focus from Hypothesis-Driven Candidate Gene Studies to Hypothesis-Generating Genome-Wide Studies

  Candidate gene studies. From the 1930s through the rest of the century, the number of things that geneticists could do with this magnificent body of theory was limited. Perforce, geneticists had to adopt strategies that worked despite only fragmentary knowledge of what was going on within the genome. One was the “candidate gene” approach.20

  It began with Anthony Allison’s observation back in the early 1950s that sickle cell anemia was geographically limited within Africa to areas where malaria was endemic. Why should people who survived in an environment afflicted by malaria be peculiarly vulnerable to a blood disease unknown elsewhere? Allison hypothesized that a mutation with a huge s value because it protected against malaria had swept through the population even though it had the lesser deleterious effect of susceptibility to sickle cell anemia. A mutation that did both of those things was probably in that part of the genome involving blood. Allison picked the Hemoglobin-B gene as his candidate gene and subsequently identified a specific mutation that had been the target of natural selection.21