The Genetic Archaeology of Race

DNA analysis is explaining where "racial difference" comes from—and what it does and doesn't mean. The study of human genetic variation has become the most contentious area in modern science

Over the past decade or so genetics researchers have been undermining the widespread belief that groups of people differ genetically in character, temperament, or intelligence. They have shown that all human beings are incredibly similar genetically—much more so than other species of large mammals. They have revealed the folly of attributing group behavioral differences to biology rather than culture.

But that's not how many of the news stories have read. On the contrary, here are the kinds of headlines you might have seen: "RESEARCHERS FIND GENETIC MARKER UNIQUE TO AFRICANS." "ASIANS BIOLOGICALLY LESS SUSCEPTIBLE TO ALCOHOLISM." "ALL NATIVE AMERICANS DESCENDED FROM A SMALL NUMBER OF FOUNDERS." In other words, given how journalists, pundits, and bigots have interpreted genetics research, people are probably more convinced than ever that group differences are significant.

We will continue to be inundated with DNA-sequence information—and with interpretations of that information—for many years to come. These genetic data have immense medical potential (though that potential will probably take much longer to realize than most people suspect). By studying the genetic differences among individuals, researchers will eventually find many DNA variants that contribute to health or disease.

But genetics research is also producing results of an entirely different kind. Differences in DNA sequences from person to person reflect the cumulative effects of human history. The patterns of genetic variation in the world today therefore carry a record of that history. They document the evolution of an African ape that began walking on two legs about four million years ago. They record the existence, sometime between 100,000 and 200,000 years ago, of a small group of people who are the ancestors of every person alive today. They chronicle the origins of "races" and "ethnic groups" and describe how those groups have both blended and separated over time.

Most genetics researchers are well aware of the historical dimensions of their work. But because these considerations raise uncomfortable issues, particularly issues of race, they tend to be downplayed. The White House news conference held on June 26 of last year to celebrate the sequencing of the human genome was filled with stirring but vague homages to human unity. "The human genome is our shared inheritance," said Francis Collins, the head of the publicly funded Human Genome Project, at the National Institutes of Health. "Race has no genetic or scientific basis," said Craig Venter, whose company, Celera Genomics, has been sequencing and analyzing human DNA.

The reality is considerably more complex. Genetics research is demonstrating that the differences in appearance among groups are profoundly incidental, but these differences do have a genetic basis. And although it's true that all people have inherited the same genetic legacy, the genetic differences among groups have important implications for our understanding of history and for biomedical research. These complications in an otherwise reassuring story have thoroughly spooked the leaders of the public and private genome efforts. The NIH has been collecting information about genetic variants from different ethnic groups in the United States, but it has refused to link specific variants with ethnicity. Celera has been sequencing DNA from an Asian, a Hispanic, a Caucasian, and an African-American, but it, too, declines to say which DNA is which.

This strategy of avoiding the issue is almost sure to backfire. It seems to imply that geneticists have something to hide. But the message emerging from laboratories around the world should be hailed, not muzzled. It is one of great hope and promise for our species.

One reason for the caution displayed by the genome sequencers is a largely overlooked drama that has been playing out on the fringes of the Human Genome Project for the past ten years. Unlike most scientific dramas, this one has a clear protagonist: Luigi Luca Cavalli-Sforza, a genial but sharp-tongued professor of genetics at Stanford University. Now seventy-nine, with thick silver hair and a perpetually inquisitive expression, Cavalli-Sforza combines the demeanor of a man accustomed to respect with a natural openness that has won him many friends. Born and educated in Italy, he went to Stanford University in 1971. There he quickly became a leader in the fledgling field of anthropological genetics, which draws inferences about the past based on the patterns observed in human DNA. His former students are now scattered around the world, carrying on the work he began. Yet he still rides his bike to campus each morning to write, read, and analyze the latest results from the lab. "He'll work until he drops," an associate says. "He's curious about the science."

In 1991 Cavalli-Sforza and a group of colleagues proposed a comprehensive study of human genetic differences, which they called the Human Genome Diversity Project. The study would involve gathering cells from several thousand people around the world, "immortalizing" the cells by converting them into laboratory cell lines, and using the cells' DNA to reconstruct human evolution and history. For Cavalli-Sforza, the Human Genome Diversity Project was to be the culmination of a lifetime of work.

The proposal loosed a flood of controversy. Aboriginal groups in the United States, New Guinea, and other countries accused the HGDP of stealing their genes, destroying their culture, and even contributing to genocide. Academic critics claimed that the project could encourage racist thinking, by oversimplifying issues of great complexity. "The idea of studying human genetic diversity is a good one," says one outspoken critic, Jonathan Marks, an anthropologist at the University of North Carolina at Charlotte. "But the way that Cavalli-Sforza has conceptualized it has problems at all levels."

Cavalli-Sforza has been baffled by the reaction to his proposal. He has always believed that the HGDP will help to end racism, not inflame it. In the 1970s he participated in public debates with the Stanford physicist William Shockley to dispute Shockley's racist ideas. He has worked closely with various African groups and cares deeply about their well-being. He protests that his intentions have always been purely scientific.

For almost a decade Cavalli-Sforza has been trapped in the paradox at the heart of human genetics: The only way to understand how similar we are is to learn how we differ. Yet any study of human differences seems to play into the hands of those who would accentuate those differences. Researchers might claim that the genetic differences they identify among groups have no biological significance. Yet simply by dividing human beings into categories—whether sub-Saharan Africans, Jews, Germans, or Australian aborigines—they reinforce the distinctions they would seek to minimize. How to resolve this dilemma is quickly becoming one of the most difficult problems facing the study of human genetics.

A Theory of Mutation

More than 10,000 years ago, on the frigid, wind-swept plains of northeastern Siberia, a genetic accident occurred in a testicle of a particular man. As one of the man's sperm cells divided, the Y chromosome in the cell underwent a copying error. One of the chemical units making up his DNA changed from a molecule called cytosine to one called thymine. An elaborate biochemical proofreading apparatus is supposed to correct such copying errors, which geneticists call mutations. But there are so many individual chemical units, or nucleotides, in human DNA—about 60 million in the Y chromosome, and about three billion in the other chromosomes in a human sperm or egg cell—that a few mutations inevitably creep in every time a cell divides.

Within the next couple of months the man impregnated a woman. The sperm cell that combined with her egg was the one with the mutated Y. The woman gave birth to a son, each of whose cells had the mutated Y he got from his father. The son was no different from the other men in his tribe (the mutation in his Y had no effect on his body), yet he was a pivotal figure in human genetic history.

At some point, according to one interpretation of events, the son of the man whose Y had mutated crossed what was then a broad plain leading from Asia to North America—presumably with a small band of others. Before him stretched a continent that was largely, or perhaps completely, devoid of human beings. This man had sons himself, and his sons had sons. Over subsequent centuries his descendants spread down the length of North America, across the Isthmus of Panama, and into South America. All of them carried their forebear's distinctive Y chromosome, to which they added their own mutations. Today more than half of Native American males have this mutated Y chromosome.

Genetic reconstructions of historical events can always be interpreted in somewhat different ways, observes Peter Underhill, the geneticist in Cavalli-Sforza's lab who first detected this and many other variations in Y chromosomes. The mutation could have occurred in Siberia some generations before the migration along the Bering land bridge, or it could have occurred in North America. Nevertheless, we know that this man existed and that his Y chromosome differed from any previous Y chromosome in this way. His particular mutation could not have originated in more than one of the ancestors of today's Native Americans, and the mutation occurred in no other group in the world. Yet DNA is such a long and complex molecule that every act of human procreation produces at least some unique mutations. These mutations spill across the generations like an unusually shaped jaw or distinctively colored eyes. The result is an elaborate human genealogy, an intricately branching tree of genetic alterations.

When Cavalli-Sforza first became interested in genetics, the detailed mechanisms of human heredity were completely unknown. In 1938, at the age of sixteen, he had enrolled in medical school at the University of Pavia, in Italy, largely because of a fascination with microscopes. "It turned out to be a very lucky choice," he says. "Had I not gone to medical school, I would have been conscripted at the beginning of the war."

But when he graduated, in 1944, he found that he disliked the practice of medicine. So he took a part-time faculty position at the University of Parma and did research on bacterial genetics. He is still well known among geneticists for his contributions to the discovery that microbes have sex—or at least can engage in the kind of genetic exchanges involved in human couplings.

In 1951 a chance remark by one of his students redirected his research toward human beings. "Throughout my life I've been very lucky in finding good students," he recalled recently, when I met with him in his office at the Stanford medical school. "This particular student was also a priest. He mentioned to me that there were some data he thought would be of interest for human genetics." For more than three centuries the Catholic Church had collected information on births, marriages, and deaths in many Italian parishes. It was the ideal data set, Cavalli-Sforza realized, for the study of a particularly contentious issue in twentieth-century genetics—the role of genetic drift in evolution.

Most descriptions of evolution emphasize natural selection, in which a beneficial mutation becomes more common over time because bearers of this mutation are more likely to survive and procreate. But if an organism just happens to have lots of descendants, its genetic variants will become more common whether they are selected for or not. That's what happened with the Siberian forefather of many Native Americans. His Y chromosome did not have an advantage over any others—it simply prospered through the vagaries of genetic chance.

In the 1950s many geneticists believed that natural selection would almost always squelch genetic drift. The parish records gave Cavalli-Sforza a way to test the idea. They showed that most people in the mountain valleys high above Parma married within their own small villages. Genetic drift is more obvious in small, relatively insular populations, because an individual who happens to have lots of children can flood a population with his or her distinctive genetic variants. In contrast, on the plains around the university, for example, the genetic effects of any one person are reduced, because the population is larger and more mixed. There genetic drift should be less obvious.

One way to measure genetic drift is to look at the relative proportions of blood types within certain communities. So Cavalli-Sforza and a few assistants took needles and test tubes and fanned out over the countryside. With the help of parish priests they gathered blood samples, often in sacristies after Sunday mass. They found that the distribution of blood types varied much more from village to village in the mountains than in the valley, just as predicted by the theory of genetic drift.

His success led Cavalli-Sforza to consider the matter more broadly. If he could link genetics with mating and migrations among the people around Parma, why couldn't he do the same thing on a larger scale? In fact, he ought to be able to determine the genetic relationship between any two groups of people. A group should carry many of its predecessors' variants, just as children bear the genetic legacies of their parents. By detailing the genetic similarities and differences among groups, Cavalli-Sforza could trace humanity's spread across the planet.

More and more kinds of blood were being discovered in the 1950s. In 1961 Cavalli-Sforza decided that he had enough data to try his idea. He and a colleague analyzed published data on blood types in fifteen populations—three each from Europe, Africa, Asia, and the Americas, and one each from Australia, New Guinea, and New Zealand—and produced a tree showing how the various groups were related.

The results looked reasonable. The Native Americans from Venezuela and Arizona were related to Eskimos and Koreans in the sample, squaring with a migration across what is now the Bering Strait. The Africans and Europeans were genetically close, reflecting their continents' relative proximity. But to find out more, Cavalli-Sforza needed new kinds of data.

Decoding Human History

The best way to determine the genetic relationships among people is to compare the sequences of the nucleotides in their DNA. But in the early 1960s those sequences were inaccessible. Manipulating DNA in the laboratory at that time was like playing the piano with a baseball bat—existing tools were far too awkward to examine individual nucleotides. Cavalli-Sforza therefore turned to the next best thing: the many thousands of proteins in the human body. The sequence of nucleotides in DNA dictates the sequence of the amino acids that constitute proteins, though the translation between the two is a convoluted process that partially obscures the underlying DNA sequence. Still, by studying proteins Cavalli-Sforza could learn at least a little about the DNA differences among people.

The result was a decades-long string of remarkable, though in some cases still hotly contested, discoveries. In the early 1970s, for example, Cavalli-Sforza and the archaeologist Albert Ammerman proposed a radical new hypothesis for the peopling of Europe. At that time most anthropologists believed that modern Europeans were descended largely from the continent's Stone Age inhabitants, who replaced the Neanderthal people starting about 40,000 years ago. By analyzing the genetic variation of modern Europeans, Cavalli-Sforza and Ammerman came to a different conclusion. They decided that Europeans are descended largely from populations of farmers who started migrating out of the Middle East 9,000 years ago. As the sons and daughters of farming families left their parents' farms and moved into new territory, they interbred with the existing hunter-gatherer populations, which produced gradients of genetic change radiating from the Middle East. Only in mountainous areas unattractive to farmers—the Pyrenees homelands of the Basques, for example—were the genes of the indigenous peoples comparatively intact.

Other historical events, too, appeared to have influenced the European gene pool. For example, a genetic trail leads from the area north of the Black and Caspian Seas into the rest of Europe. Cavalli-Sforza linked this trail to the spread of the descendants of nomadic warriors and herders who first domesticated the horse, about 4,000 B.C.

Similar traces of historical events showed up on every other continent. The genetic history of China is dominated by a split between northern and southern people, despite the official position that all ethnic Chinese are descended from common ancestors. Native American proteins point to three major waves of migration, suggesting connections among groups that had never considered themselves related. And wherever Cavalli-Sforza looked, the fragmentary protein evidence hinted at much-greater detail to come.

Cavalli-Sforza didn't know it when he moved to Stanford in 1971, but a series of experiments then going on at the university and elsewhere were about to transform anthropological genetics—and much of the rest of biology. The Stanford biochemist Stanley Cohen and the University of California at San Francisco biochemist Herbert Boyer were figuring out how to cut DNA in precise locations, combine DNA from different organisms, and grow the resulting hybrid DNA in bacteria. For the first time, human beings could control DNA nucleotide by nucleotide. The age of genetic engineering had begun.

New tools followed rapidly, with names like fluorescent DNA sequencing and polymerase chain reaction. By the mid-1980s biologists had realized that they had the means to pursue a goal that would have been unthinkable just a decade earlier: with sufficient effort they could read the entire multibillion-nucleotide sequence of human DNA. In 1990 an international consortium of governments launched the Human Genome Project to determine the sequence. Spurred by competition from the private sector, the project is today more or less finished (some cleanup work is still needed to complete the sequence).

Elsewhere on the Web
Links to related material on other Web sites.

Automated DNA Sequencing
A description of automated DNA sequencing methods—including fluorescent sequencing—in use at the Joslin Diabetes Center facility in Boston.

"Polymerase Chain Reaction-Xeroxing DNA" (1999)
An explanation for laypeople of how polymerase chain reaction works. Posted by the National Health Museum, "the site for health and bioscience teachers and learners."

As far as Cavalli-Sforza was concerned, the Human Genome Project was a good idea but didn't go far enough. The DNA that was sequenced is a pastiche of chromosomal fragments from various unidentified donors, and the project has produced just a single generic sequence to represent all of humanity. To trace human history Cavalli-Sforza needed to know how DNA sequences vary among people from different parts of the world. He began to think about the best way to gather such data, and he traded ideas with friends and colleagues. The result was the Human Genome Diversity Project, which was proposed in 1991 and fleshed out over the next two years. According to the planning document for the project, the goal would be to collect DNA from several hundred distinct groups, including many indigenous groups. Immortalized in cell lines, this DNA would be used primarily to study human history, but it would also have more-practical applications. Medical researchers could use it to investigate the connections between genetics and disease. And according to the planning document, by demonstrating the nature of the genetic differences among people, research on genetic variation would "help to combat the widespread popular fear and ignorance of human genetics and will make a significant contribution to the elimination of racism." Who could be opposed to that?

Bioethics and Biopiracy

In 1993 an odd-looking document appeared on the desks of the Human Genome Diversity Project's organizers. Under the heading "RAFI communiqué" it was titled "Patents, Indigenous Peoples, and Human Genetic Diversity." An artful combination of analysis and innuendo, the document was unambiguous in its conclusion: "The Human Genome Diversity Project should immediately halt any collection efforts."

The campaign against the HGDP marked the first foray into human genetics for RAFI—the Rural Advancement Foundation International. A small organization based in Canada, RAFI had previously targeted corporations that removed indigenous plants from developing countries, repackaged their genetic material in hybrid seeds, and then offered the seeds to Third World farmers for exorbitant prices. Now it accused the HGDP of a similar form of "biopiracy." The DNA of indigenous people would be mined for valuable information. Pharmaceutical companies would then use this information to make drugs far too expensive for Third World people to buy.

The actions of the U.S. government seemed to bear out RAFI's claims. Federal agencies had applied for patents on cell lines containing DNAfrom several Third World groups. The patent applications, which were all aimed at medical uses, had nothing to do with the HGDP, and the project repeatedly dissociated itself from them. But opponents eagerly conflated the uproar over the patents with what they called the "vampire project."

Many Native Americans were especially critical of the HGDP. "The benefits for Native people are elusive or nonexistent, and the risks are tremendous," says Debra Harry, a member of the Northern Paiute Nation and the director of an organization called the Indigenous Peoples Council on Biocolonialism. Native American creation stories say that Native Americans have always lived in the Americas, Harry observes, not that their ancestors migrated across the Bering land bridge sometime before the arrival of Europeans. Furthermore, many Native American cultures hold that biological materials are sacred. "DNA is not ours to manipulate, alter, own, or sell," Harry says. "It was passed on from our ancestors and should be passed on to our children and future generations with its full integrity."

Native Americans and many other groups were also quick to raise the specter of biological warfare. What would stop a government from developing lethal microbes tailored to particular groups? This criticism especially galls geneticists, who say that it is too far-fetched to take seriously. Even if it were possible someday to devise a genetically based weapon, it would simply kill varying percentages of the people in different groups.

The tendency of critics to mix fact and fancy led to dark mutterings among the project's organizers. "These groups have to support themselves somehow," Cavalli-Sforza says, "and they've decided to use us." But the critics were much more skilled than the geneticists at ad hominem attacks. "Nice, smart, liberal—but naive and a bit arrogant—white guys firmly ensconced at reputable universities" is how a RAFI organizer once described Cavalli-Sforza and his colleagues to me.

Cavalli-Sforza quickly proved to be a liability as spokesman for the project. He is not one to weigh the effect of his words before speaking. With European frankness he simply says what he thinks. "He's never been very politically astute," one colleague says. "Sometimes we can't help cringing, because we know [what he says is] going to come out condescending or arrogant."

Realizing that his lifelong ambition was slipping from his grasp, in the fall of 1993 Cavalli-Sforza turned to a colleague to help represent the project: a Stanford law professor named Hank Greely. With a background in bioethics, Greely spearheaded the creation of a "Model Ethical Protocol" to govern the collection and use of DNA samples. He also reminded geneticists of the sensitivity of their work. But the controversy was spiraling out of control. Dozens of organizations representing indigenous peoples were protesting the project. They decried spending millions of dollars to collect the blood of groups that were disappearing because of poverty, disease, and official neglect. Project organizers responded by suggesting that they could provide medical assistance to participating groups. Then a new objection arose: indigenous peoples were being bribed for their DNA.

For Greely, the low point came in December of that year, when he traveled to Quetzaltenango, Guatemala, to represent the project at a meeting of the World Council of Indigenous Peoples. Scheduled to speak twice at the meeting, Greely spent most of his second session listening to angry speeches from the audience. By the end of the conference he was standing at the podium being accused of being a CIA agent. "I had the opportunity to respond after each comment and almost always did so," he later recounted in a report about the meeting.

I noted that this was not a project about indigenous peoples, but about all the world's peoples ... I stressed that populations would not be involved in the project unless they wanted to be. I agreed that the West and western science had done terrible things to indigenous peoples, but said that science could also do some good things. I urged that our project was different—that we were trying to do things right.

His protestations were to no avail. "My statements were uniformly either ignored or dismissed as lies," he recalls. "The only faintly positive feedback from the crowd was that two speakers complimented me for my courage."

Opponents of the HGDP have succeeded in linking the project with existing qualms about genetic testing. Say that a group has an unusually high frequency of a genetic variant associated with a disease. Would every member of that group be stigmatized, whether or not he or she had the variant? Or what if genetic studies conflict with a group's belief about its origins?

The traditional way of dealing with such qualms has been through informed consent. People must be fully informed about what a test can reveal and must be given the choice of whether or not to be tested. But the HGDP adds a twist to the idea of consent. Studies of genetic variation usually focus on groups—Han Chinese, New Guinea highlanders, Hopis—rather than on individuals (in fact, individual DNA donors are usually anonymous). By the logic of informed consent, groups ought to have some say over whether their members participate in the research. Some bioethicists argue strongly against this position, saying that it denies individual autonomy, but the committee set up to oversee the HGDP in North America agrees with it. Before the project collects any DNA in North America, the groups being studied must understand and okay their involvement. It's not always obvious who speaks for a group, Greely admits, but when a representative body can be identified, it should be consulted.

Of course, giving groups the right to consent means that some will decline. "That's fine," Greely says. "We're under no illusion that we're going to get samples of every human population. Our goal is to get just five percent—so nineteen of twenty could not participate."

How Different Are We?

RAFI's efforts to derail the Human Genome Diversity Project can be seen as a noble attempt to protect the powerless. They can also be seen as tragically misguided. If activists succeed in fomenting widespread distrust of genetics research, they could ruin an opportunity to discredit the essential notion underlying racism—that human groups have innate and fundamental biological differences.

This idea has deep roots. In 1758 the Swedish botanist Carolus Linnaeus gave the human species its formal name, Homo sapiens. He also divided the species into subcategories: red Americans, yellow Asians, black Africans, and white Europeans. He described Homo sapiens americanus as "ill-tempered, ... obstinate, contented, free." Homo sapiens asiaticus was "severe, haughty, desirous." Homo sapiens afer was "crafty, slow, foolish." And Homo sapiens europaeus was—of course—"active, very smart, inventive."

Similar prejudices characterized biological thinking well into the twentieth century. From 1910 to 1930 most of the leading human geneticists in the United States actively supported the campaigns of eugenicists to limit reproduction by those deemed biologically inferior, including Eastern Europeans, Jews, and people with mental disabilities. This line of thinking led directly to the gas chambers of Nazi Germany, which effectively ended the eugenics movement. Yet the basic tenet of eugenic thinking—that the mental attributes of human groups differ for genetic reasons—remains firmly embedded in the popular imagination.

Until very recently most geneticists professed agnosticism on this issue. They said that not enough was known to assess the contributions of genetics to behavioral differences among groups. The data collected by Cavalli-Sforza and other population geneticists have been making that position less and less tenable. People are too closely related—and have mixed too much throughout history—to differ in fundamental ways.

Interpretations of this research have been controversial in the past, but the genetic evidence is now overwhelming. It clearly indicates that sometime in the period 100,000 to 200,000 years ago our ancestors went through a severe genetic bottleneck. Perhaps an environmental change drove ancient people to the brink of extinction. A more likely scenario, however, is that a relatively small group, numbering fewer than 20,000 at times and probably living in eastern Africa, was isolated for many thousands of years from the many groups of archaic human beings scattered throughout Africa, Europe, and Asia. The people who emerged from this genetic bottleneck had traits never before seen in human beings. They had lighter builds, new ways of interacting among themselves, and perhaps a greater facility with language. Eventually the descendants of these people spread throughout Africa and beyond. They reached Australia at least 60,000 years ago, probably traveling from the Horn of Africa and then along the South Asia shoreline. They arrived in the Middle East a bit more than 40,000 years ago. By 35,000 years ago anatomically modern people had spread into Europe from the Middle East and into East Asia from Southeast Asia. Sometime more than 12,000 years ago they entered the Americas.

Elsewhere on the Web
Links to related material on other Web sites.

Evolution of Modern Humans: The Biological and Cultural Evolution of Archaic and Modern Homo sapiens
A site offering descriptions, charts, and timetables documenting the evolution of modern humans. Created by a professor at Palomar College, San Marcos, California. Whenever modern human beings came in contact with their archaic counterparts, the latter eventually disappeared. Many questions surround this process. The genetic evidence indicates that modern people bred very little with the archaic people living throughout the Old World. According to Joanna Mountain, a Stanford anthropologist who has been studying the genetics of African groups since the 1980s, when the Peace Corps took her to Kenya, "If the ancestral modern human population fifty thousand years ago had been highly diverse genetically, as it would have been if there had been a lot of mating with archaic peoples, we would still see evidence of that diversity today, but we don't." One possibility is that modern and archaic human beings did mate, but were so different genetically that such matings were infertile. Or maybe some matings between archaic and modern human beings did produce offspring, but at such a low rate that the archaic genes have been diluted out of existence.

How modern human beings replaced their predecessors also remains a mystery. Archaeological evidence indicates that bands of modern and archaic people sometimes lived near each other for thousands of years. Yet no remnants of warfare have been found. The cave paintings of Europe, some of which date from the period when modern people were replacing Neanderthals, evince plenty of violence against animals but not against other people.

Fewer than 10,000 generations separate everyone alive today from the small group of Africans who are our common ancestors. That's much more than the twenty or so generations mentioned in Genesis, but it's the blink of an eye in evolutionary terms. Even over thousands of generations human groups have not differentiated in any substantial way. Rather, the genetic evidence indicates that modern human beings have expanded as a single, relatively well mixed population without subsequent genetic bottlenecks (bottlenecks tend to erase the evidence of previous bottlenecks, which is how geneticists know that the bottleneck in Africa was the most recent one). Our comparative youth as a species accounts for our extreme genetic homogeneity. The chimpanzees living on a single hillside in Africa have twice as much variety in their DNA as do the six billion people scattered across the globe.

There's another reason for our biological homogeneity. Modern human beings have never been able to resist for long what Noël Coward called "the urge to merge." A person traveling due east from Madrid to Beijing (both at about 40°N latitude) would pass Italians, Greeks, Turks, Armenians, Uzbeks, Tajiks, Kyrgyz, Uighurs, Mongolians, and Han Chinese, among others. All these groups resemble their immediate neighbors more than they do groups farther away because of the continual exchange of mates across group boundaries.

There's a simple way of describing our genetic relatedness. Not only do all people have the same set of genes, but all groups of people also share the major variants of those genes. Geneticists have never found a genetic marker that is of one type in all the members of one large group and of a different type in all the members of another large group. That's why ethnically targeted biological weapons would never work. Every group overlaps genetically with every other.

The extreme interpretation of this observation, now popular in academia, is that biological groups do not exist. That's obviously absurd. The ways in which typical Nigerians, Koreans, and Norwegians differ physically belie any claim that all human groups are somehow "socially constructed." But the development of morphological differences in a widely distributed species is a biological commonplace. Whenever the members of a group are more likely to mate inside the group than outside, the frequency of particular genetic markers within that group can become higher or lower. In most cases these changes are entirely random, as with the blood-type distributions that Cavalli-Sforza studied in Italian villages. But natural selection can also be a factor. To take the classic example, as modern human beings moved from equatorial regions into more-northern latitudes, dark skin was no longer needed to protect the body from the sun's ultraviolet rays, and light skin made it possible for the body to produce more vitamin D. The resultant lightening of skin color seems to have occurred at least three times during human history: when Africans moved north into the Middle East and then into Europe; when dark-skinned people living on the islands and mainland of Southeast Asia migrated into what is today China; and when people from southern India moved north into the Punjab (genetics research is demonstrating that migrations of European people into the Subcontinent have had much less biological significance than is commonly assumed).

"What we see is the surface of the body, but the surface of the body is determined by climate," Cavalli-Sforza says. "Adaptations to climate have to be superficial, because those are the parts of the body that are exposed to the outside world." New Guinea highlanders and sub-Saharan Africans are about as different from each other genetically as any human beings on earth. Yet they have physical similarities because of where they live, including dark skin to protect against the rays of the sun.

Biologists have several options in describing the variation they observe in people. They can label an individual purely in geographic terms, based, for instance, on the distance he or she lives from the Equator. Or they can sort people entirely in terms of the genetic differences they find. Instead they tend to use the labels generated by our highly fractious culture, even though many of these labels have little to do with the underlying genetics. In this way biologists endow those differences with much more weight than they deserve.

From Skin Color to Intelligence

People convinced that human groups differ for genetic reasons in intelligence, aggressiveness, or other complex behaviors have one last recourse. They can assume that the same forces leading to differences in appearance could somehow have influenced mental attributes. Maybe, for example, cold climates exerted some sort of selective pressure on people moving north from the tropics, favoring individuals with greater initiative or intelligence. Or maybe some other genetic process divided cognitive traits unevenly among groups.

The argument doesn't work, for two reasons. First, no mechanism has been identified that could sort complex attributes within such a genetically homogenous species, causing the behavior of one group to differ from that of another. The idea that natural selection favored different cognitive traits on different continents—that selective forces on colder continents led to greater intelligence, for example—seems designed more to justify social prejudice than to establish testable hypotheses. After all, Neanderthals were much better adapted to the cold than modern human beings, yet they weren't able to compete with the newcomers from Africa and eventually became extinct.

Even if a potential differentiating mechanism could be identified, the case for group differences fails for a second reason. There is a fundamental difference between a simple trait such as skin color and a complex attribute such as intelligence. Skin color is determined by a handful of genes and does not depend on the experiences a child has in the womb and while being raised. The development of the brain involves thousands of genes and is indissolubly linked to experience. For this reason it is impossible to parse a particular complex trait into purely genetic and purely environmental components.

Of course, at some point genetic differences must override the effects of experience. Human beings and chimpanzees differ in intelligence. If Neanderthals had survived, their behavior would probably be genetically different from ours. But the genetic differences among modern human beings are so small that group differences in behavior fall entirely within a range attributable to culture.

From the archives:

"Who Owns Intelligence?" (February 1999)
Three unresolved issues will dominate the discussion of intelligence: whether intelligence is one thing or many things; whether intelligence is inherited; and whether any of its elements can accurately be measured. By Howard Gardner

"Academic Ignorance and Black Intelligence," (June 1972)
"The myth of verbal deprivation is particularly dangerous because ... it leads its sponsors inevitably to the hypothesis of the genetic inferiority of black children, which the verbal-deprivation theory was designed to avoid." By William Labov

Take IQ tests as an example. In Japan the Buraku are a caste of people discriminated against in education, housing, and employment. Their children typically score ten to fifteen points below other Japanese children on IQ tests—about the average black-white difference in the United States. Yet when the Buraku emigrate to the United States, the IQ gap between them and other Japanese vanishes.

What if geneticists were to find a particular genetic variant that seems to be associated with a given behavior? Wouldn't that provide evidence for those who believe there are significant genetic differences among groups? In fact, just such an association is about to be made. Within the next few years geneticists will find the specific genetic variants that determine skin color. In the United States, because of the black-white gap on average IQ scores, these variants will be statistically correlated with our most commonly used measure of intelligence. But the genetic variants affecting skin color have nothing to do with the functioning of the brain. They exert their effects entirely through the influence of culture—by sorting people according to the color of their skin.

The complexity of the relationship between genes and behaviors will always confound simple-minded efforts to link the two. Even if a genetic variant seems to cause a particular behavior—such as extroversion or verbal fluency—in one environment, it may have no effect, or the opposite effect, in a different environment. The importance of a person's experiences makes it a fallacy to cite the frequency of certain genetic variants as the cause of group behaviors.

Our knowledge of history argues against any such association. Did the Vikings somehow lose a marauding gene in becoming the laid-back Scandinavians of today? Did the Arabs gain a religiosity gene with the coming of Mohammed? How could Hispanics have any kind of innate characteristics when the group consists of people with varying percentages of European, African, and Native American ancestry? Group attributes change with the alacrity of culture, not with the languor of genes.

There is no need to posit unknown genetic forces to explain the differences among groups. Individuals have unique experiences from the moment of conception. They receive varying levels of nutrition and medical care. They are treated differently every single day by adults, teachers, and peers. They are born into cultures with particular histories and beliefs.

Of course, human history could have worked out differently. Suppose that an archaic species of human beings such as Homo erectus had become firmly established in the Americas during a time of lower sea levels. As sea levels rose and the Bering land bridge became submerged, modern human beings might have ceded the hemisphere to their archaic predecessors and never left the Old World. In that case, when Columbus came ashore in the West Indies, he would not have encountered modern human beings separated from his own ancestors by just a few thousand generations. He would have met slope-browed, linguistically primitive people with a cranial capacity about two thirds of our own. Then we would have a race problem. Instead we have cultural differences masquerading as race problems.

We are highly visual creatures, which makes us very good at spotting differences among people. But researchers who study human fossils and our nonhuman ancestors have a different perspective. They point to the attributes that modern human beings share: our high foreheads, our nuanced languages, the light of intelligence in our eyes.

The Genetic Information Age

Because of the controversy over the Human Genome Diversity Project, the federal government has so far granted the project relatively little funding. The large-scale effort envisioned by Cavalli-Sforza has not occurred. Paradoxically, the prospects for achieving the project's goals have never been better. Small-scale investigations of genetic variation are thriving. Researchers in other countries have been much more successful than have U.S. investigators at skirting issues of race and genetics. The Center for the Study of Human Polymorphism, based in Paris, now offers DNA samples from populations around the world for research. And the growth of the Internet is changing how genetics research is done. According to Mark Weiss, who oversees a gradually increasing budget for anthropological genetics at the National Science Foundation, the HGDP could go forward as a collection of networked efforts rather than as a single grand undertaking.

But the biggest boost to studies of genetic variation is coming not from anthropology but from medicine. When the project to sequence the human genome began, its organizers anticipated that having a single generic DNA sequence would satisfy most researchers. They have since recognized their short-sightedness. Much of the medical interest in the human genome lies not in the similarities among people but in the differences. If an individual is genetically susceptible to heart disease, cancer, or some other ailment, it is because of particular variants in that person's genes. To understand that susceptibility, researchers need to know about those variants.

Public and private organizations are now feverishly gathering data about human genetic variation. Essentially, they are looking for the same kind of variants involved in skin color and other simple genetic traits: discrete mutations in particular genes that have spread more or less widely in different groups. "Researchers will look at, say, heart patients and controls, or African-Americans with high blood pressure and controls, to find the genetic variants related to those diseases and to find effective therapies," says Lisa Brooks, the director of the Genetic Variation Program at the National Institutes of Health.

Not surprisingly, biomedical researchers are coming up against the same problem that has plagued the Human Genome Diversity Project. How can researchers describe genetic variances among groups without implying that groups are fundamentally different? So far the biomedical world has dealt with the problem largely by wishing it would go away. Investigators using data from the primary database on human genetic variation at the National Institutes of Health, for example, must sign a form saying that they will not try to determine the ethnicity of the people who contributed the DNA (though population geneticists could easily derive the ethnicity by comparing the samples with known sequences).

Maintaining a firewall between anthropology and biomedicine may be politically expedient, but it is intellectually dishonest. To understand the patterns of genetic variation in groups, researchers have to study the history of those groups. "The same evolutionary and historical factors that underlie genetic variation in general underlie the variations responsible for genetic diseases," says Lynn Jorde, a human geneticist at the University of Utah. "The only question is whether we want to be efficient in studying those factors, and to me the answer is sure, why not?"

The privacy and rights of groups need to be protected in these studies, just as do the privacy and rights of individuals. Groups should be involved in the planning and conduct of research, and they need a negotiated measure of control over the uses made of genetic data. Geneticists and anthropologists are in a position to do great harm to groups through irresponsible and unethical actions (though many groups are threatened by societal forces much greater than those any scientists can unleash).

At the same time, groups run a risk by refusing to become involved in genetics research. Such research can have direct medical benefits if it involves diseases affecting the group, or indirect benefits if researchers provide participants with medical care. There can also be commercial dividends if a group retains control over its genetic materials. Besides, there's more than one way to study the genetics of a population. If a group declines to be studied, geneticists can always find more-cooperative members who are outside the group's control.

The rocky history of the Human Genome Diversity Project has demonstrated many of the pitfalls that need to be avoided. First, we have to keep in mind the extreme fluidity of human groups. The word "race," for example, can't begin to capture the commonalities and differences of our shared history. Most African-Americans have European ancestors. All European-Americans have African ancestors. It makes no sense to talk about "races" when we are all complex mixtures of many different peoples.

We also have to remember just how small the genetic differences among groups are. The genetic variants affecting skin color and facial features are essentially meaningless—they probably involve a few hundred of the billions of nucleotides in a person's DNA. Yet societies have built elaborate systems of privilege and control on these insignificant genetic differences.

Finally, as we learn more about our genetic susceptibilities to disease and our relationship to the past, we need to find better ways of putting genetics in context. People tend to attribute great importance to the findings of geneticists. But the striking homogeneity of our DNA actually emphasizes the centrality of the environment and our experiences in determining who we are. Because culture exerts such a profound influence on complex traits, our genetic heritage has little importance in considerations of ethics or public policy.

Cavalli-Sforza has always believed that if people understood genetics, they couldn't possibly be racists. For that he must certainly be judged naive. Then again, geneticists think in terms of generations, and over that time scale even the staunchest opinions can change. In the 1950s only four percent of white Americans approved of marriages between members of different races. Today the number is close to 50 percent, and it is undoubtedly much higher among young people.

Changing attitudes and new social forces are already having an influence on the collective genome of our species. Barriers that in the past have limited intermarriage among groups are breaking down. Cavalli-Sforza believes that many societies are moving toward what he calls the American model. "Two hundred years from now," he says, "people all over the world will be mixing in the same way that people in the United States are today."

A greater rate of intermarriage will generate great cultural upheavals. Genetically it will matter not a whit. Human beings are so similar that it makes no difference biologically for a white to marry a black, or for an Asian to marry an Australian. More intermarriage will make it harder to figure out an individual's ancestry. But it can only hasten the approach of a color-blind society.

Today most people still bear some traces of group biological history in their faces and skin. Perhaps someday our species will lose those distinguishing characteristics, through either intermarriage or genetic engineering. Until then the study of our genetic differences, if interpreted with care and understanding, could be one of the best ways to appreciate our biological unity.