The Invisible History of the Human Race

by Christine Kenneally

  • • •

  Despite its potential for great good, many critics have spoken forcefully about the negative impacts of reading history in DNA. The concerns are real, and much of the commentary reflects a sense of responsibility in the scientific community. Still, the collective response can give the impression of a great, looming threat.

  In 2007 a group of scientists published a policy discussion about ancestry testing in Science, in which they stated, “Genetic ancestry testing . . . has serious consequences. Test-takers may reshape their personal identities and they may suffer emotional distress if the results are unexpected or undesired.” The American Society of Human Genetics examined the topic with some optimism but much hand-wringing in 2010, noting that “the very concept of ancestry is subject to misunderstanding in both the general and the scientific communities.” An earlier paper in the British Medical Journal stated, “Tracing genetic identity can lead to the resolution of uncertainty but can cause more problems than it solves.” Some critics view the motives of anyone who wishes to look at his own genome as potentially suspect.

  Writing about early-stage misinterpretations of genetic data, the biological anthropologist Jonathan Marks worried, “With genetic data, it seems, one could find entities that did not really exist, or impose cultural assumptions on the data and mistake them for patterns inherent in the data, yet still cloak oneself unimpeachably in the mantle of modern science.” In a caption to a racist depiction from 1842 of the different skull shapes of a Caucasian, an African, and a chimpanzee, he wrote, “As the authoritative voice on identity and descent, science’s track record is hardly blemish free”—a sentiment with which few scientists would disagree.

  Kim TallBear, an assistant professor of American Indian studies at Arizona State University, wrote that genomic scientific techniques, specifically those of the Genographic Project, have emerged out of the racial science dating back to the seventeenth century. The hope expressed by Genographic and genetic genealogy companies that this science was different—that it sought not to affirm racist categories but to demonstrate that they were cultural and unsupported by nature—was, to her mind, “naive at best.” TallBear also wrote that the claim that genetic proof of humanity’s African origins was “anti-racist” was complicated by the fact that it involved “portraying Africa and Africans as primordial.”

  The intent, of course, is to ward off the most venal use of biology, the specter hanging over all genetic research and all genealogy as well. Men like Francis Galton, Madison Grant, Heinrich Himmler, and Adolf Hitler have left in their wake a deep fear about the abuse of science, including the notion that merely considering the human genetic network will either make us all racist or justify our latent racism. I once spoke about the genetics of ancestry with a Holocaust historian who had hunted some of the last surviving Nazis in the 1990s. When I told him that little letters in our genetic code might testify to the ethnicity of our parents and grandparents, he said, “The Nazis would have loved this.” They would certainly have seized upon the idea, but in the end the full picture would have let them down just as badly as all the other dubious measures of race they tried to develop. Comparing the volumes of people’s skulls proved pointless, as did trying to formulate an objective measure of beauty. Likewise, the bits of DNA that genetic historians study do not indicate what a person will look like, think like, or live like. They are records only of ancestry; they tell us that groups once existed that, for whatever reason, lived together long enough that they ended up with genetic commonalities.

  • • •

  Around the same time that the Jefferson research was taking place, another hugely ambitious genetic project was stalling spectacularly. In the early 1990s, Luigi Luca Cavalli-Sforza founded the Human Genome Diversity Project (HGDP), the goal of which was to rebuild the biological and language trees of humanity by sampling the DNA of tens of thousands of people from all over the globe. It would be hard to overstate the contribution that Cavalli-Sforza has made, not just to the science of population genetics but to the imaginations of generations of population geneticists and historians. His early projects in the 1960s were the first attempts to rebuild the history of the world through the distribution of traits found in blood. His popular book, Genes, Peoples, and Languages, first published in 2001, outlined his ambitious work, bringing together evidence from languages and genes to unearth the history of humans. The Human Genome Diversity Project would not only enrich history, argued Cavalli-Sforza, but would also have medical applications and, by demonstrating that there was no such thing as biological race, would be a tool against racism.

  For all its apparent idealism, the project’s organizers did not pay enough attention to addressing the contexts in which the project’s requests for blood, and ultimately for knowledge, were made. It was one thing for middle-class westerners who have access to the educational and medical outcomes of the project to share their blood, and the information in it. But many indigenous groups of interest to the project were still fighting for basic human rights and health care. They struggled with extreme poverty, terrible health, and overrepresentation in the criminal justice system. Many had experienced exploitation and even medical experimentation at the hands of scientists.

  In response, small, well-organized activist groups created many impediments to the progress of the project. They pointed out that it was unclear whether pharmaceutical companies would be able to access the DNA and not only use the information it contained to make enormous profits but even to patent people’s genes. They claimed that some indigenous groups were afraid that the scientific story of the world would be used to rewrite their own cosmologies. Their objections were connected to issues of genetic ownership too, such as the question of whether an individual had a right to share his own DNA with the project if other people in his community did not want to. Ultimately, the contradiction that millions of dollars were being invested in the HGDP but not in the people whose blood it would use became overwhelming. Among some indigenous groups it became known as the Vampire Project. (At the same time, many of the same political battles were being fought over genetic and other research conducted on ancient remains.)

  A small group of cultural anthropologists accused Cavalli-Sforza and his colleagues of racism and hubris. Some even questioned the scientific merit of the project, suggesting that its questions about human history were obscure and asking whether the genetics of a single group would ever shed light on any history but its own. Years later in a United Nations address, Cavalli-Sforza said, “Ignorance can breed fear and hate, but I have discovered that it is most dangerous when mixed with the personal political agenda of science haters.”

  For obvious reasons, genetic research with subjects whose history is less troubled provokes less overt concern. Recall that the genomes of the people in the British genetics project were so overwhelmingly similar that as far as current medical genomics is concerned, they were effectively the same. Yet there were still identifiable differences that told a historical tale: The subjects’ ancestors had lived in different regions and they had left a mark in the genetics of their modern-day descendants. As far as we can currently tell, these differences were shaped by the same neutral evolutionary mechanisms that result in some groups having differently shaped skulls.

  What is perhaps most confusing about the criticism of this kind of genetic research is that its detractors often cite one of the most popular ideas of the human genome era: namely, that DNA reveals that race is a myth and that beneath the skin we are all fundamentally the same. But how can this be true when another consequence of the human genome era is that we can now have our genome analyzed and our racial history quantified? Does race exist in our genes or just in our heads?

  • • •

  The modern notion that there is no such thing as biological race and that we are more alike across populations than within them can be traced to 1972, when Richard Lewontin, an evolutionary biologist and geneticis
t at Harvard, conducted a landmark experiment that continues to shape how people think about the subject. Lewontin looked at seventeen places in the genome where a single letter might be different between people, and he showed that for each of these spots there were more differences within populations—or what we often think of as racial groups—than between them. The implication was that the largest differences that exist between people occur along all sorts of spectra, none of which is racial. As Lewontin wrote:

  Human races and populations are remarkably similar to each other, with the largest part of human variation being accounted for by differences between individuals. Human racial classification is of no social value and is positively destructive of social and human relations. Since such racial classification is now seen to be of virtually no genetic or taxonomic significance either, no justification can be offered for its continuance.

  These findings were generally taken to mean that the ways we differ from one another across racial or ethnic divides are far, far smaller than everything we have in common; that even though individuals from different parts of the world may look different from us, they are generally going to be more like us than not; and that there is no biology of race. The findings were also taken to mean that you cannot identify a person’s background from his DNA. The Human Genome Project announced that “two random individuals from any one group are almost as different as any two random individuals from the entire world.”

  However, if you look at even more bits of the genome, the picture changes. For example, in 2007 a team led by D. J. Witherspoon, a researcher at the University of Utah, examined the same question. They confirmed that if you were comparing a few hundred bits between people across traditional racial divides, you would still find that many individuals may have more in common across the divide than within their own group. However, if you took people from different populations in different parts of the globe and you compared thousands of bits of DNA, the picture changed: Increasing the resolution by looking at more DNA shows that people will tend to have more in common with an individual from their own population than from a distant one.

  Witherspoon and his colleagues refined the approach even further and demonstrated that actually, when comparing populations that have been separated geographically for a long period of time, you need only a hundred bits of DNA to tell which person came from which population, which may make it seem as if race could be detected in DNA. Actually, the researchers did find something in this data, but it was not race.

  There are many reasons for this, some of which have nothing to do with genes. The first reason they didn’t find race in the genome is because it was never there to be found; race is an imprecise and ultimately unhelpful notion in biology. Part of its power comes from the implication that the divisions it refers to are absolute and eternal, yet “race” is one of the shiftiest words in language. Sally Hemings’s mother had a white father and a black mother, and Hemings’s father was white (Hemings shared a father with Martha Wayles Jefferson, the president’s wife). Yet had Hemings been included in consecutive censuses since 1790, when then–Secretary of State Thomas Jefferson instituted the first American census, at first she would have been noted without name or race, and she later may have been classified at different times as “mulatto,” or “black,” or “white.” The collection of traits that are supposed to distinguish different races changes in different eras, depending on who has power and who doesn’t. Not only is race defined with a good deal of arbitrariness, but who gets to define it changes too. Sometimes a racial category is imposed on people, and sometimes it’s one that people choose for themselves. Race fuses cultural traits with physical traits, and it presumes either that what is cultural is determined by what is physical or that they always go hand in hand.

  Again and again science has shown through both its failures and its successes—from the grotesqueries of eugenic “science” and failed Nazi attempts to quantify race to the positive revelations of the Y chromosome—that the categorical boundaries we draw between people when we talk about race are always in part culturally determined; they never exactly fit onto real populations. There is simply no predetermined set of genetic or other physical divisions into which different human groups throughout space and time can be discretely assigned. Modern-day racists may wish to believe that some DNA is more privileged than others, but nothing in the human genome can be explained by the age-old foils of racism, such as platonic intelligence or beauty or purity.

  Still, as unhelpful as the idea of race is, it is a hard one to shake off. Simply asserting that race does not exist doesn’t appear to be changing people’s minds or lives, perhaps because it seems to be so flatly contradicted by their more vivid daily encounters with different groups of people that look quite dissimilar. Insisting that race is a cultural construct doesn’t help people understand the common experience of meeting a person who appears to be, say, Chinese or northern European and finding out that she was indeed born in China or northern Europe. If race is not the thing we see in other people, then what is it?

  The confusion arises because when we use the term “race,” we often include the idea of ancestry. This becomes a problem when people who want to reject race, or at least prove that it has no biological underpinning, effectively reject the idea of ancestry too. Early on critics attributed interest in the genetics of ethnicity to an unmoored “faith” in genetics as a solution to disease. In response to a 2005 New York Times op-ed about the medical utility of “race,” one political scientist argued that staff and grantees of the Department of Health and Human Services and National Institutes of Health should not publish or cite anything suggesting that genetics is associated with any population category, including nationality or ethnicity, unless the finding was statistically significant and the description would “yield clear benefits for public health.”

  While these proposed guidelines are extreme, they exemplify a widespread anxiety that is not often so boldly articulated. Declaring statistically significant information off-limits by fiat is unscientific, and it has worrying implications for free speech. Far more important, the measure would endanger public health more than protect it. Studies of genetic correlates of disease are easily confounded by markers of ancestry. Medical research that hopes to identify genetic causes will risk being misled by false positives if it ignores the ideologically neutral markers of ancestry.

  Ancestry is real and it can’t just be defined away. You can see it on people’s faces, and you can definitely identify it in their DNA. The it that makes letters of the genome fall into different patterns in different groups is in fact the ancestry of the people carrying them.

  • • •

  In 2011 Eran Elhaik was hired to solve one of the biggest jigsaw puzzles in the history of the human race; in fact, the puzzle was the history of the human race itself. Following the failed Human Genome Diversity Project, National Geographic launched the Genographic Project in 2005 to develop a way of reading people’s Y chromosome and mtDNA, and half a million people eagerly participated by contributing samples of their DNA. From the beginning, indigenous communities were approached in a different way. It was made clear that they retained ownership of their own DNA. Whereas the HGDP had proposed to keep the cell lines from samples alive in perpetuity, the new project committed to not do that, as the idea that cells would live on after their owners’ deaths was disturbing for many groups. A mouthwash was developed for sampling so that people who objected to the idea of giving blood did not have to do that. While it is hardly the case that the problems of indigenous groups have been solved, there is generally a more mutually respectful and appreciative relationship between this project and the groups it engages with.

  In 2012 Genographic decided to include all the chromosomes and analyze autosomal DNA as well. Elhaik was asked to design a method that would extract the most information from a sample but at the same time extract only historical information and not anything to do with
an individual’s health or features. (See chapter 14 for more about health and the genome.) Before he could do that, Elhaik had to collect a big enough set of data to survey as many different populations as possible, as he needed to see the whole in order to understand its parts. The trickiest part of the challenge was that whatever series of letters made any two populations distinct from each other was probably going to be a different set from the series that made any other two populations different.

  Elhaik spent years obsessively collecting data. He collected as much as he could from publicly available data sets, and he was given data as well. “There were a lot of data-rich scientists who were kind enough to share their data with me,” he recalled. Ultimately he was able to gather the genetic data of tens of thousands of people from almost five hundred different populations, amassing the largest set of its kind in the world.

  Elhaik worked out how all the populations differed from one another by comparing each group to another one by one and working out the minimum number of letters he needed to be able to distinguish the pair. “If you have a Lebanese and a Syrian, you ask, Do I need a hundred, two hundred, a thousand, two thousand genetic markers so I can correctly classify a Lebanese as a Lebanese and a Syrian as a Syrian?”

  “You cannot do it for every population,” Elhaik clarified, “because some of them are genetically indistinguishable. I had a lot of Indian groups, including different linguistic groups and castes, but no matter how many genetic markers you are going to use, they were not separable.” Elhaik found that he needed between five hundred and two thousand letters of DNA to tell most of the subject groups apart.

  Elhaik’s subject populations roughly corresponded to what we think of as different ethnic or racial groups, but it was actually ancestry that he measured. This is not a semantic trick, an attempt to replace an incendiary word (“race”) with a more neutral one (“ancestry”). Elhaik’s analysis was based on the knowledge that individuals in each group uniquely carried a particular pattern of DNA because they descended from a particular population of people. The concept of biological race is of no help here, not only because it is imprecise but also because it carries a fatally incorrect implication—the idea that people can be sorted into completely distinct genetic buckets. Ancestry does not work that way. Elhaik now works at the University of Sheffield, and the company Prosapia Genetics has been created based on his analyses.