Human Biodiversity

How race was rebranded for the twenty-first century

It was 1998 or thereabouts that an invitation arrived in Jonathan Marks’s email inbox. He can’t recall the precise date because until I asked him, he hadn’t thought about it for years. What he does remember is that the sender was a little-known science journalist and former writer for the conservative National Review, Steve Sailer. The invitation was for Marks to join a mailing list of people interested in the subject of human variation.

“I knew absolutely nothing about him,” Marks recalls. “He just seemed to be someone who was organizing something.”

At the time, Marks was teaching at the University of California, Berkeley, a few years on from having written a popular textbook on race, genes, and culture, titled Human Biodiversity. Wedging these two words together, he had neatly coined a phrase to describe biological and social variation across the human race. Part of the reason he chose it, he says, is that “biodiversity” had become something of a buzzword. He never guessed it would cause any problems. And why would it? Diversity was being celebrated; both the glorious biodiversity of the natural world and the rainbow of cultural and physical diversity in human societies. It was the proud label of liberal antiracists, of the good guys like himself.

“America’s answer to the intolerant man is diversity,” Robert F. Kennedy affirmed in 1964 at the dedication of an interfaith chapel in Georgia. “United in diversity” was the motto of the European Union. It goes without saying that although different cultures have different qualities to offer, the breadth of human variation does nothing to undermine the general consensus that we’re biologically pretty much all the same beneath the skin. This is a truth that’s been universally acknowledged since the end of World War II.

At least that’s what Marks thought.

The invitation itself appeared perfectly innocent. The 1990s marked the early years of electronic mailing lists, and Steve Sailer apparently wanted to use one as a way of pulling together scientists, intellectuals, and fellow journalists to start a private conversation about human difference. “He said, ‘Hey, I’m interested in human variation, and I like your work. Let’s get an email list together,’” says Marks. “It seemed pretty straightforward and harmless.” So he signed up. What intrigued him especially was that Sailer happened to be brandishing Marks’s own neologism, calling his list the “Human Biodiversity Discussion Group.”

Others joined in their dozens. By the summer of 1999, Sailer’s roster of members was astounding. Along with prominent anthropologists such as Marks, there was psychologist Steven Pinker, political scientist Francis Fukuyama, and economist Paul Krugman. In hindsight, the large number of economists in the group might have been a warning. There in the mix, too, was the controversial author of The Bell Curve, political scientist Charles Murray. That should have been another red flag.

When Marks had talked about human biodiversity, he meant the superficial variations we see among individuals, right across the species—not variation among human groups. “There isn’t really room for three, or four, or five biologically distinct kinds of people,” he tells me. He certainly didn’t expect to see people on the list reinforcing old-fashioned racial stereotypes of the kind that had long been debunked. That school of racism was long dead, he assumed. Yet here on this email list, something strange was happening. Observing the conversations that Sailer steered through the group, Marks noticed the term “human biodiversity” being used differently from the way he had originally intended. Members were using it to refer to deep differences between human population groups.

Among the people added to the group that summer was Ron Unz, a Harvard graduate and founder of a financial services software company who had recently run as the Republican candidate for the governor of California. His introduction to other members was pasted alongside a 1994 article Unz had written in the Wall Street Journal, “How To Grab the Immigration Issue,” on the changing demographics in California’s politics. “Conservatives and Anglos have become enormously angry and frustrated over the growth of crime, welfare, affirmative action and the general decay of their society,” he stated. Another addition to the group was Deepak Lal, a professor of international development at the University of California, Los Angeles, whose work explored the reasons the West was economically more successful than the rest of the world.

It dawned on Marks that Sailer’s seemingly innocent email list was not so much a way to discuss science in an objective way but more about tying together fresh science and economics with existing racial stereotypes. One debate that sticks in Marks’s mind today, for instance, took place between him and a journalist who claimed that black people were genetically endowed to be better at sports. Marks insisted that this was a scientifically shaky argument, not to mention one with dangerous political implications. The two experts clearly disagreed. But rather than help reach a consensus, “Steve Sailer clearly took his side,” he tells me.

“At which point I realized, ‘Ah! This isn’t an impartial scholarly discussion.’”

Another time, Sailer defended a writer on the list who suggested that different ethnic groups had their own particular strengths—for example, saying that Turks were born physically more powerful than other groups. This was, Sailer suggested, one reason why affirmative action policies to hasten racial equality weren’t a good idea. Different ethnicities should instead be encouraged to do what they do best. When Sailer talked about human biodiversity, he didn’t appear to be using the phrase in a politically neutral way, but as a euphemism. He had spun the language used by liberal antiracists to celebrate human cultural diversity to build a new and ostensibly more acceptable language around racism.

In email correspondence with me, Sailer denies duping anyone, although he does admit that the group leaned toward the heretical.

Whatever his intentions, this wouldn’t have been the first time someone had tried to subvert the idea of diversity. In their conclusion to The Bell Curve—a book which just a few years earlier had notoriously claimed that black Americans were innately less intelligent than whites—Charles Murray and Richard Herrnstein similarly undermined the political push towards racial equality by arguing that biological differences between groups made racial equality practically impossible. “We are enthusiastic about diversity—the rich, unending diversity that free human beings generate as a matter of course, not the imposed diversity of group quotas,” they wrote. Every person in a diverse society had a valued place, they implied, just not the same place.

Realizing his error, Marks left the group. “If somebody said the same to me today I would probably be a little bit more suspicious,” Marks admits. “I would look at who else is on the invite list. But even that can be misleading because he may be inviting people who are just as confused about this as I am.” At the time it seemed harmless. Just a bunch of people with some marginal political ideas trying to convince others of the same.

Instead, Marks was about to find out just how prophetic the existence of the list was. “That was my introduction to what became the alt-right.”

For those sucked into Sailer’s electronic arena for the intellectual discussion of race, his email list was just a taste of the virulent racism that would later be seen far more often in the shadowy areas of the internet, then more openly on social media and right-wing websites, and finally in mainstream political discourse. Many more soon took hold of the phrase “human biodiversity,” giving it a life of its own online. Today it’s nothing short of a mantra among self-styled race realists.

“Blogs have made the dissemination of wacko ideas much more efficient,” says Marks. “Actually, in academia, we don’t really know what to do with blogs because they have a really short shelf life. You forget about them a day after you read them; it’s hard to cite them. But today, they are out there. They’re these markers of people with wacko ideas.”

To be fair, few could have guessed that the email list was a precursor to something bigger. But as the group slowly went defunct, Steve Sailer’s political convictions became increasingly obvious. He and other members of the list went on to become prominent conservative bloggers, writing frequently on race, genetics, and intelligence. As a columnist for American website, which describes itself as a news outlet for patriotic immigration reform (and, like the Mankind Quarterly, is automatically blocked by my internet service provider), Sailer once argued passionately for the biological reality of race, stating that it was a fundamental aspect of the human condition. In 2009 the same right-wing website published Sailer’s first book, about Barack Obama, titled America’s Half-Blood Prince. In 2013 the software mogul Ron Unz founded his own blogging platform, the Unz Review, as an alternative to the mainstream media; he recruited Sailer as one of his most prolific columnists.

Sailer rose to true prominence, though, in the US presidential election of 2016. Six years earlier he had proposed focusing heavily on immigration to draw in white working-class voters as a single bloc. Having subverted the language of diversity through his email list, he did the same with identity politics. If ethnic minorities, such as black and Hispanic Americans, could assert their rights and defend their interests, he reasoned, then why not white voters who felt they were losing out to cheap immigrant labor and globalization? It turned out to be an unexpectedly successful campaign strategy for Donald Trump. Dog-whistle politics was reframed as a pushback to liberal elitism.

“I think the big problem this country has is being politically correct,” Trump said in the first Republican Party presidential debate, in Ohio in 2015. A similar approach was adopted by some campaigners in the UK in favor of leaving Europe during the Brexit debate around the same time.

An article in New York magazine in 2017 described Sailer’s string of prophetic political insights as a new wave of populist thinking on the right, dubbing it “Sailerism.” In the United States at least, he was credited with inventing a form of identity politics for disgruntled poorer whites, a group that had been neglected by politicians. But at the same time, political observers couldn’t help but notice that his ideology often looked suspiciously like white nationalism.

For Keith Hurt, in Washington, DC, who investigated right-wing intellectual racism in the 1980s and has kept a close eye on the immigration debate ever since, none of what happened during that time should have come as a surprise. “The election of Trump made it impossible for many people to any longer overlook this stuff,” he tells me.

Hurt explains that the racist ideologies that existed at the start of the twentieth century, which manifested themselves in the eugenics movement and then German nationalism, survived to the end of the century. The only difference was that those who held these views were later forced to keep them private. “The post–World War II ideological consensus sort of rested on an implicit social contract that said, on the one hand, overt racist language will not be deployed in the public square in the way it was before. On the other hand, if this kind of talk is pushed out of the public square, society will refrain from making accusations of racism against all but the most extreme fringe,” he says. And so the intellectual racists, the ideologues, communicated with each other through their own tight networks and disseminated their ideas through their own publications, some of which were so marginal and private that they were almost invisible to the outside world. It was easy for the public to assume that they didn’t exist, that the only genuine racists left were the ones they could see and hear, the skinheads and thugs.

When the time was right, however, political and intellectual racism slowly resurfaced in the mainstream. Sensing the rising tide, people like Steve Sailer waded in. “They couldn’t exist as they do now if there hadn’t been the intellectual ideological continuity,” adds Hurt. They weren’t starting from scratch; they were just repackaging the old ideas, the existing racial assumptions that had been around for many decades.

But it all came as more of a surprise to academics like Jonathan Marks. “I was working on the assumption that these guys were a lunatic fringe. If you had told me twenty years later that they would be part of a political mainstream wave, I would have said you are absolutely crazy. These guys are antiscience. These guys are positioning themselves against the empirical study of human variation and they are clearly ideologues for whom empirical evidence isn’t important,” he says with a laugh. “But I think they were a lot cleverer than us professors.”

Offline, in mainstream academia and among respected scientists, a different debate was taking place—one that would, similarly, come to reshape the way people thought about race.

It was 1991, shortly after the launch of the multi-billion-dollar Human Genome Project, which was steadily working to build a map of the genetic data shared by our species for the first time by sequencing the entire length of human DNA. At the Stanford University School of Medicine, an influential geneticist by the name of Luigi Luca Cavalli-Sforza, then around seventy years old, spotted an opportunity. His own long and illustrious career had been dedicated to studying human variation across the world, work that had taken him from Italy to the United States and brought him to the forefront of a field known as population genetics. Cavalli-Sforza and some of his colleagues—a small team of anthropologists and geneticists, based mostly in the United States—wondered whether the same kind of data being collected by the Human Genome Project could also be used to pick through the fundamental differences between human groups, the genetic variations that make us who we are. Little did he guess that it would turn out to be one of the most controversial scientific initiatives of its time, leading to decades of debate about the true biological reality of race.

Unlike Steve Sailer, Cavalli-Sforza and his team didn’t come to their new project with race in mind. Indeed, quite the opposite. They were avowed antiracists, fully signed up to the UNESCO statements on race in the 1950s, and firm believers that science would do nothing but prove racial stereotypes incorrect. They thought that their work might even help free the world from the scourge of racism. “This was, at least on the American side, in general political terms, a quite left-wing group of people,” recalls Henry Greely, a professor at Stanford Law School who became involved in the initiative later on.

Cavalli-Sforza himself was used to receiving hate mail from people who didn’t agree with his outspoken belief that genetics didn’t support old-fashioned notions of race. In 1973 he had publicly debated William Shockley, a Stanford University physicist and joint Nobel Prize winner, who in later life became a notorious racist. Shockley believed that black Americans had intellectual shortcomings that were hereditary, and that black women should therefore be voluntarily sterilized. He was among the most prominent race theorists to receive support from the Pioneer Fund. When they met for a debate at Stanford, Cavalli-Sforza coolly demolished Shockley’s claims, fact by fact.

The field of population genetics itself was also born of post–World War II efforts to move away from traditional race science and eugenics. In the 1950s and 60s, when geneticists stopped talking about race, they turned their attention instead to “populations,” the “human variation” that existed between these populations, and the “frequencies” of alleles—different versions of the same gene—that existed within these groups. It was a more rigorous, mathematical, molecular approach to studying human difference. In the course of this study, population geneticists such as Cavalli-Sforza quickly noticed that there were no hard genetic boundaries around human groups, but rather continuous statistical variation, with a good deal of overlap. What differences there are exist along gradients, not borders.

That said, variation isn’t random either. Depending on how you look at it, it can fall into clusters in which certain genes are statistically more common in some groups than in others. Population geneticists became fascinated by these clusters. They assumed that they would be more obvious in places where people had been geographically isolated for hundreds of years, living on islands or at the tops of mountains, mating almost exclusively within their own group through many generations. “Primitive groups” were believed to be particularly distinct genetically because of the length of time they had spent away from others, shielding their genomes from the effects of intermixing. Being so remote, their lineage made conspicuous thanks to their long isolation, maybe their genes could offer special insights into how humans evolved and adapted.

Cavalli-Sforza was among those who believed that by studying the genomes of these primitive groups it might be possible to track historic patterns of migration. If they could see how gene frequencies varied in them and their closest neighbors, they could perhaps track where their ancestors once lived in the distant past. In 1961 he became one of the first scientists to apply modern statistical methods to see how frequencies of major human blood types varied among large human groups; he created a “family tree” of blood types to show how these groups were related.

The search for what they thought would be genuine difference at the far reaches of the planet took researchers to the most remote places. In 1964 the World Health Organization picked out Eskimos in the Arctic, Guayaki hunter-gatherers in the forests of eastern Paraguay, pygmies in the Central African Republic, Aboriginal Australians, and the tribes of the Andaman Islands in India as potentially interesting “primitive groups.” Scientists who couldn’t make the trips to see them would settle on studying immigrant groups in their own backyards, particularly Jewish and Roma communities, also known to be ancient and tight-knit. There was no point in studying very large human populations—say, just Africans or Europeans—because there was too much variation within them as a result of migration and intermixing. Small, old indigenous communities were thought to be special, and more distinct.

On the back of all this research, Cavalli-Sforza formed a new plan. He mooted the idea of using the same revolutionary gene sequencing technology as the Human Genome Project to map not just one genome, but to travel the world and draw genetic data from lots of individuals of different ethnicities. This Human Genome Diversity Project, he and his collaborators announced in the journal Genomics in 1991, would “supplement and strengthen findings from archaeology, linguistics, and history.” Through their work, genetics would take its place alongside the study of culture, language, and history to help paint the whole human story.

The Human Genome Diversity Project wanted initially to zoom in on four hundred to five hundred different small populations, especially geographically remote and apparently dwindling indigenous ones—including the Basques in Europe, the Kurds of eastern Turkey, and Native Americans—which they referred to as “isolates.” They expected the groups’ genetic data to provide clear patterns that might shed light on prehistoric migration and social structure. Calling for urgent funding from the world’s governments, their main argument was that this was a race against time to document the breadth of human variety before we all entered the great big melting pot, when migration and assimilation would leave us all so genetically similar that such an effort might not be quite so worthwhile.

But what they were proposing didn’t sit easily with everyone. Some outside observers couldn’t help but be a little uncomfortable. After all, it was hard not to wonder whether in the nineteenth century, this might have just been called race science. The Human Genome Diversity Project was technically more precise, of course, more scientific. It wasn’t sampling skin and hair color, or slotting people into racial hierarchies. It was using genetics. But in some ways it was hardly distinguishable from the study of human difference a hundred years earlier. The word “race” had been prudently replaced by “population,” and “racial difference” by “human variation,” but didn’t it look suspiciously like the same old creature?

Then again, could it be called race science if the scientists involved were obviously antiracists? As Henry Greely told me, they were left-wing liberals committed to stopping racism, whose public lives had been dedicated to fighting scientific racists and eugenicists. How could there be anything to fear?

“They won’t use the term ‘race,’” says the Yale University historian Joanna Radin, who has studied the progress of the Human Genome Diversity Project from 1991 to the present day. She notes that there was always a deliberate effort to avoid the word “race,” mostly because of its obvious political baggage but also because the scientists didn’t see genetic “isolates” as being “races” in the traditional sense. They were different enough to merit study, but not in the way racial difference had been described in the past. They were small groups, not large continental-scale ones. “They don’t necessarily map onto existing racial hierarchies.”

Those behind the project insisted that their research was countering racial myths. Their stated intention was to replace ignorance and prejudice with hard scientific facts, and make it clear that we are one single human species, united in our common origins. Their plan was not to look for difference as racists had in the nineteenth century, to prove inferiority or superiority, but to use the tiny difference there deep in our genomes to help build a picture of human migration. They would be like archaeologists, digging through our genes, looking for clues about our history. Their aim was simply to understand our past.

Given the unimpeachable political credentials of scientists such as Luigi Luca Cavalli-Sforza, the scientists behind the Human Genome Diversity Project might have expected it to go forward without a hitch. All they needed was funding, and permission from their “isolates,” the indigenous communities whose blood and DNA they wanted to sample.

But things had become more complicated. This wasn’t the sixties anymore, when foreign researchers could pick the communities they wanted to study and just expect them to comply. They had become more wary. “Luca Cavalli-Sforza was an old school anthropologist,” geneticist Mark Jobling at the University of Leicester explains. “I mean, I went to talks by him in the nineties where he would show old slides of him collecting DNA, blood samples in Africa from Pygmy groups, and offering glass beads and cigarettes in return, things like that.” This wasn’t the way things were done now. This was the 1990s, the dawn of the internet age, identity politics, and the fight for indigenous rights.

Scientists who took an interest in indigenous communities found that now, the same communities were taking an interest right back. They weren’t as trusting as they had been in the past. And they were organized. And they had good reason to be organized. Remote tribes and ethnic groups had been exploited throughout history, their land and cultural artifacts pillaged by Western colonizers, their bodies targets for unethical experimentation. Between 1946 and 1948, for example, the United States government ran secret experiments on thousands of people in Guatemala in which they deliberately exposed them to sexually transmitted diseases. Before and during World War II, British scientists deliberately sent Indian soldiers fighting for Britain into gas chambers to study the soldiers’ response to exposure to mustard gas. There was a long and bloody tradition of scientists abusing other populations for their own ends, particularly those populations deemed at the time to be racially inferior. But people were now prepared. Rights activists, alert to the risk of exploitation, were ready to defend the communities targeted by the Human Genome Diversity Project.

These activists warned of the possibility that DNA analysis might damage how these communities chose to understand their past, might reveal something that could be used commercially to extract profit, or even be used as a weapon by racists. They weren’t prepared to hand over their biological data, their blood and tissue, knowing that there was a possibility that it could end up misused.

Even so, their resistance to join in the project baffled some of the scientists involved. As Yale’s Joanna Radin explains, “What had changed—and this is what caught the scientists by surprise—was that the indigenous groups they had imagined disappearing, that they didn’t have to reckon with once they left, these purported isolates, were organizing in indigenous movements. They had access to the web; they were in touch with activists.”

As the future of the project came increasingly under threat, Cavalli-Sforza and his colleagues called in Henry Greely from Stanford Law School to help navigate the ethical dilemmas and deal with critics. Greely, fascinated by the project and taking a personal liking to the participants, agreed. At the beginning, he knew very little about the science itself. “I knew how to spell DNA but that was about it.” But from the outset, he could see that there were likely to be problems. The scientists were of course aware that science did not have a great historical record when it came to race. “They knew this was an issue,” he says. “They knew that it had a bad past.” But they didn’t see themselves as part of that past. As far as they saw it, “They were the break from the past. They were the good guys. They were the ones who understood the rights-enhancing, equality-enhancing potential of genomics, and they were going to bring it to the world.” Greely adds that this was the first place he had heard “that all humanity is more similar to each other than a band of chimpanzees that lives in a particular region of Africa. They expected to see that people were quite similar.”

But at the same time, they were oddly naïve about how the project might be seen from the outside. One scientist “talked about the need to sample ‘isolates of historical interest,’ a term that indigenous populations did not care for,” Greely admits. “It struck me that that was not likely to be well received because it’s a very clinical, bloodless way of referring to people who are alive, and cultures that are living now. Historical interest is sort of something you find in a museum. It was tone-deaf. Naïveté is always easily diagnosed through the retrospectoscope.”

Greely’s job, to navigate the ethical quandaries posed by the Human Genome Diversity Project, turned out to be a poisoned chalice. Partly because of the political controversy surrounding the project, it didn’t attract the funding they wanted. Yet, throughout, the scientists struggled to understand why. “Some were so comfortable with their own knowledge of their own moral bona fides that it was hard for them to imagine being attacked from the left,” says Greely. “They would have imagined that any remaining racists would be attacking them, not that they would be attacked as being racists.” If anything, the scientists were on the same side as those attacking them. “I was the most conservative person on the North American committee, and I’m a Carter-Clinton-Obama Democrat!”

The activists representing indigenous groups turned up the dial on their protests. At a meeting of the World Council of Indigenous Peoples in Guatemala in 1993, Greely even found himself facing down the untrue accusation that he was a CIA agent, intent on committing genocide. “We took some lumps. Some of them were deserved and some of them weren’t.”

In an address at a special meeting of UNESCO in 1994, Cavalli-Sforza turned his focus to the charge of racism, insisting that his project would help combat prejudice, not perpetuate it. But in 1995 another political storm was created when scientists funded by the United States National Institutes of Health tried to patent a virus-infected cell line from people belonging to the Hagahai tribe in the highlands of Papua New Guinea for the purpose of developing a new treatment for leukemia. Activists accused them of stealing people’s biological samples and attempting to profit from them.

It was in this charged political environment that Greely wisely drew up a model ethical protocol, setting in stone that samples wouldn’t be taken unless entire groups, not just individuals, had given their consent. For scientists, it seemed obvious that they were not ill intentioned.

But if people were slow to see their good intentions, there was another, deeper reason. At the same time that the project claimed to be antiracist, it was hard to escape the paradox that this was also all about finding out how people differed. If the genetic variation between us was already known to be trivial, then why embark on a multi-million-dollar international project to study it at all? In what way did this reinforce that we were all the same underneath?

For the geneticist and critic Mark Jobling, the way the project was structured, deliberately going after the DNA of isolated populations rather than scanning people all over the world wherever they happened to be, was the thing that ultimately undermined it. “How you define the population in the first place, these are culturally loaded things in themselves. So there was a lot of cultural discrimination in the original aims.” The isolated communities that scientists such as Cavalli-Sforza believed were unique were actually never all that isolated nor unique, but they were treated as though they were.

When I raise this issue with him, Greely admits that there was “this sort of uneasy recognition that if your project is about looking for differences then it’s sort of counterintuitive to say it’s showing similarities.” As antiracist as the scientists behind the project were, they had somehow fallen into the trap of treating groups of people as special and distinct, in the same way that racists do. They were still forcing humans into groups, even if they weren’t calling these groups races. They were using similar intellectual frameworks to pre-war race scientists, but with fresh terminology.

That’s not to suggest that Cavalli-Sforza or his colleagues were being duplicitous. “I don’t think he believed himself to be engaged in a racist enterprise,” says Joanna Radin. “But I also don’t think he really had any more sophisticated a sense of how this was going to fight racism than just being able to show we’re all connected, we’re all cousins or something.” In truth, they saw human differences as meaningful, but they didn’t want to focus on that uncomfortable fact in a world in which saying so could have political consequences. Antiracism seemed to be more of a political ideal tagged on as an afterthought. Mark Jobling adds, “They did have a slightly happy-clappy narrative to it, you know, ‘joining up the whole human family’ kind of thing.”

It would have been perfectly possible to study human variation without grouping people. As Jobling explains, the divisions between us are so blurry that humans can theoretically be grouped any way you like. “You could do a thought experiment where you just said we will take Kenyans, Swedes, and Japanese, and will just proportion everybody into those three things.” If this were done, because we are all genetically connected to the average Kenyan, Swede, or Japanese person, either directly or by historic migration, then everyone on earth could theoretically be fully assigned to a group based on just these three nationalities. You could say that you were so many percent Kenyan, so many percent Swedish, and so many percent Japanese.” This may seem meaningless, but actually it is no more meaningless than dividing the world into black, brown, yellow, red, and white. “The definitions of those populations are cultural, and the choice of population is driven by expediency.”

Other geneticists have also warned against dividing up the world. It imposes a certain order onto our species and ignores the actual fuzziness. If the Human Genome Diversity Project had proposed sampling people more systematically, in a grid pattern across the globe perhaps, the true overlapping nature of human variation would be easier to see. Scientists would have been able to map gradual, continuous variation across regions, rather than tight knots centered on very small communities. It’s hard not to imagine that this approach—which was mooted at the time but then discarded—might also have been a more effective way of fighting racism. But it wasn’t the one the researchers chose.

In the end, most governments, including in the United States, were unwilling to invest in the Human Genome Diversity Project. It never quite got off the ground in the way it was envisioned. And to this day, it remains something of a cautionary tale. In hindsight, part of the problem was that scientists, however well intentioned they were, failed to connect what they were doing with people’s real-life experience of race, with the history and politics of this deadly idea. They thought they were above it all, when in fact they were always central to it.

Luigi Luca Cavalli-Sforza died in 2018 at the age of ninety-six. When I emailed him shortly before his death, he was retired and living in Milan. I found someone whose commitment to his science hadn’t waned, and neither had his personal politics. “There are simply no races in humankind,” he wrote back. He was still a hero to biologists in his field, an inspiration, someone who had helped build his scientific discipline into one that today has enormous importance. It was impossible not to admire him.

And, yet, it was also difficult to read his work and come away convinced that his generation of scientists had fully abandoned race science after World War II. Although they had ditched race in name, it wasn’t clear that they had necessarily shed it in practice.

In 2000, after the controversy around the Human Genome Diversity Project had been playing out for almost a decade, the project no closer to being realized, Cavalli-Sforza published a book titled Genes, Peoples. and Languages. In it, he deftly sketched his grand plan for how genetics could be used to reconstruct human history. It’s also a story, the book’s back-cover blurb added, that claimed to reveal “the sheer unscientific absurdity of racism.”

In a section titled “Why Classify Things?” Cavalli-Sforza wrote eloquently about taxonomy, and how humans have always felt compelled to categorize objects. Yet somehow, he managed to do this without any reference to politics or social history. He never mentioned that humans were classified in large part because it was politically and economically useful to those who did it. He completely glossed over colonialism and slavery, and the ways in which they fundamentally shaped how European scientists thought about race in the nineteenth and early twentieth centuries (indeed, within his own lifetime). Instead, as he saw it, racism was just a scientific idea that turned out to be incorrect. “It seems wise to me,” his chapter concluded, “to abandon any attempt at racial classification along the traditional lines.”

It’s easy to miss the catch. And it lay in the final four words of his statement: “along the traditional lines.

“A race is a group of individuals that we can recognize as biologically different from others,” he said later. Clearly, then, he hadn’t abandoned the use of the word “race” at all. Going by this statement, there may be no room for three, four, or five old-fashioned racial types, with hard divisions between them—the way we usually think about race—but there could certainly be thousands of “social groups” all over the globe, each characterized by their gene frequencies, and therefore having some biological distinctiveness to them. This is the obvious result of relatedness. People who are related are of course closer to each other genetically, and historically we have tended to live near our relatives, which is how clustering of genetic similarity happens. This means that even neighboring towns may be genetically different from one another in some statistically significant way. The small clusters produced by the fact that we don’t mate completely randomly could, by Cavalli-Sfroza’s definition, be considered “races.” So according to his definition, there are races, except the number of them is practically endless.

This is an old idea, which owes itself to early geneticists who wanted to move the study of race away from vague generalizations to something more precise. As early as the 1930s, when the field of population genetics was just emerging, Theodosius Dobzhansky—an evolutionary biologist who was later an inspiration to Cavalli-Sforza—was the one to substitute the old-fashioned idea that races were fixed types with the more modern idea that they were populations sharing certain gene frequencies. Like Cavalli-Sforza, Dobzhansky was an outspoken antiracist. But while being actively involved in antiracist efforts within the scientific community, Dobzhansky retained the concept of race. He just redefined it. The way it was redefined squared the circle of how it was possible for all humans to be practically the same while also being different. Under this definition, there’s no contradiction in my having possibly more in common genetically with my white neighbor than with my Indian one. My population group as a whole (say, North Indians) will share genes in frequencies that her population group (say, white Britons) doesn’t. In other words, if you want it to, race can exist, but you must remember that it’s a statistical phenomenon. Not every individual will fit.

“They basically redistributed race,” argues Joanna Radin. Race wasn’t ditched at all, just the parameters were changed. According to Radin, the problem with the new statistical “population” approach to studying human difference is that even though it may look different in some ways, it hasn’t fully shed the baggage of the past. “An interesting analogy would be colonialism,” she tells me. “A colonial nation declares independence and they have to forge a new nation with the structures of the old colonial regime, and it’s very, very hard to transcend that.” Even if the word “race” isn’t being used, the idea of race is still there, buried in the bedrock.

The Canadian philosopher Lisa Gannett has similarly warned about the ethical limits of thinking about race in this new way. To some it may not seem racist to think about statistically average “populations” rather than distinct “types” of people. Certainly early population geneticists such as Dobzhansky believed that racism was rooted in the assumption that within one ethnic group people are all the same, whereas those like him believed that, within these groups, people are actually very different. But in the racist mind, as Gannett explains, it doesn’t necessarily matter how differences are distributed, so long as they are there in some form or another. This conceptual loophole in population genetics—the fact that we’re all different as individuals but that there is also some apparent order to this diversity—is what has since been seized upon by people with racist agendas. Gannett calls it “statistical racism.”

The question all this raises is a slightly odd one: Is race still a problem if we redefine race? And even odder: Can science be racist if the people doing it are antiracists? For Radin, intention does matter, but it doesn’t fix the underlying problem. “If you look at the UNESCO statements on race, people often think that they declared that race is a social construct and that race doesn’t exist. But really what they did is try to constrict use of the term “race” to biologists who could be seen to use it responsibly, and not equate it with inequality.” The concept of race was thought to be safe in the hands of liberal, left-wing, antiracist population geneticists because their politics were beyond reproach. “I do think that their sense of virtue really emboldened them to feel like they were creating a transcendent mode of science, that might be able to leave race behind,” she explains.

In Genes, Peoples, and Languages, Cavalli-Sforza added a humorous aside when talking about the fact that researchers looking hard enough could spot average genetic distinctions between neighboring populations, even at the village level: “People in Pisa and Florence might be pleased that science had validated their ancient mutual distrust by demonstrating their genetic differences,” he wrote jokingly. But then, isn’t this exactly what racism is? A dislike of other groups in the belief that they are biologically different? In the mind of the racist, it probably doesn’t matter how big the groups happen to be, or if the differences are gradual or sharp. It presumably means equally little if it’s all about gene frequencies or population averages, so long as the differences are real. If the people of Pisa and Florence could have their mutual distrust validated by population genetics, then why not the people of any other two places?

For Radin, the problem is obvious. It lies in the need to group in the first place, to separate even when that separation means having to zoom in on the very tiniest bits of the genome that might differ, and even then only on average. This need to separate, to treat people as different, is how race was invented. “What happens is that you’ve got a large community of very well meaning, self-described antiracist scientists seeking to find a way to move beyond race into population genetics, which seems to be incredibly neutral. It’s numbers, it’s statistical, it’s objective,” she says. “What they have a more difficult time reckoning with is that even something like population genetics is a science done by people, working with the assumptions and the ideas that are available at the time.” They may believe themselves to be free of racism, but they can’t help thinking about humans in racial terms.

In his correspondence with me, too, Cavalli-Sforza made one comment that surprised me. He observed that interracial relationships—relationships that, in the early twentieth century, eugenicists feared might lead to offspring with strange physical and mental deformities—has turned out to be no bad thing. Miscegenation, as it was called, is now obviously recognized to be no threat to human health. You only have to look at “the beauty and vitality of hybrids, children of partners coming from genetically distant groups,” he wrote in his message. To use the phrases “hybrid” and “genetically distant” might have seemed at one time scientifically acceptable, but is it anymore? To make this observation and use this language in the twenty-first century just feels plain odd.

There is a gray area in which well-meaning people make what in other contexts might be considered incendiary statements, and we overlook them because we know they are well-meaning. In reference to reproduction rates, Cavalli-Sforza’s book stated, “Europeans are largely at a standstill while populations in many developing countries are exploding; thus blonds and light-skinned people will decline in relative frequency.” If the superficial differences between us don’t matter, then why should this? To the population geneticist, that people with blond hair are disappearing may be as much of a concern as the possibility that Native Americans might dwindle, or that indigenous Andaman Islanders living in an archipelago in the Bay of Bengal might be subsumed into the wider Indian population. To the antiracist, objective scientist, there is no value judgement in this. It’s a problem to science only because researchers are losing some interesting subjects of study, some statistical corners of human diversity, perhaps a few rare blue-eyed-blond gene combinations. But to someone with alternative politics, this might be seen instead as an argument in favor of racial purity, of preserving distinct population groups against the threat of miscegenation.

It’s easy for academics to imagine that the language they use, and the frameworks they operate in, don’t really matter. They are just words, not data. “I think that in the real world what the scientists say has about as much influence as turning on a fan has on El Niño. It’s what throwing a cobble into the English Channel has on Atlantic weather,” Henry Greely tells me near the end of our interview. “We’re just not that important.” But it does matter, because their frameworks and language contribute to our understanding of ourselves. If scientists call people of mixed ancestry “hybrids,” this implies that race is real because we are different enough to warrant using that word. If they talk about “isolates,” this sounds like there are groups who are more “racially pure” than others. Dismantling the edifice of race is about more than just tweaking language; it is about fundamentally rewriting the way we think about human difference, to resist the urge to group people at all.

It takes some mental acrobatics to be an intellectual racist in light of the scientific information we have today, but those who want to do it, will. Racists will find validation wherever they can, even if it means working a little harder than usual. And this is the reason that good scientists who do reliable research, ones who are also well-intentioned and antiracist, like Cavalli-Sforza, can’t afford to be cavalier or leave too much roomfor misinterpretation. There’s an uncertain space between recognizing that there is a gap of knowledge and actually filling that gap. It’s a place where speculation thrives, where the racists reside. Racists adopted the same concepts as good scientists—and the same language as antiracists—to assert that, if certain groups can show some average differences to other groups then, by that logic, certain groups might be better on average than others at certain things. When Steve Sailer and his followers talk about “human biodiversity,” this is what they mean. This wolf in sheep’s clothing is twenty-first-century scientific racism.

And we were told this might happen. A caustic report on the Human Genome Diversity Project released in 1995 by UNESCO’s International Bioethics Committee sounded precisely this warning. It argued that the project, whether it meant to or not, could give racists some basis to believe that certain groups were inferior or superior to others. In particular, the committee was concerned that by bringing genetics to the fore in telling the human story, people would ignore culture and history, and return to the kind of simplistic biological thinking that propelled the eugenics movement in the early twentieth century. It advised scientists to resist the temptation to use their work to shore up any kind of political ideology, whether racist or antiracist. “Racism,” it reminded them in case they had forgotten, is “socially and politically constructed.”

Science is just a pawn in the bloody game.

Although it never got off the ground, the concept behind the doomed Human Genome Diversity Project did survive. In the years that followed, other teams stepped in to achieve essentially the same outcome in other ways. The Center for the Study of Human Polymorphisms in Paris today keeps a bank of DNA samples from populations all over the world, ready for researchers who want to tap it. In 2002 the United States National Human Genome Research Institute introduced a $100 million initiative to study human variation. And in 2015 the United Kingdom launched its own project to make a genetic map of the people within its own borders, named People of the British Isles.

The project had one more unintended consequence. In 2005 the National Geographic Society in Washington, DC, the one behind the famous magazine and satellite TV channel, decided to dip a toe into the world of population genetics. Naming its effort the Genographic Project, it chose Spencer Wells, an anthropologist, geneticist, and television presenter, to lead the project. Wells had spent a portion of the 1990s studying under Luigi Luca Cavalli-Sforza, seeing the difficulties around the Human Genome Diversity Project at close quarters. His solution to the controversy was simple. National Geographic would sell easy-to-use kits that members of the public could use to help them understand the history of migration that might be hidden in their DNA, and in the process the organization would build a data bank from their genetic information.

“We put together this consortium of scientists with the goal of sampling the world’s DNA, and at the same time wanted to enable anybody, any member of the general public who was curious about their own genetic ancestry, to get themselves tested,” Wells tells me. The idea that people might want to spit into a cup and have their ancestry tested didn’t seem at the time like a highly profitable venture. Even the CEO of National Geographic was doubtful, warning Wells before the launch that nobody was going to spend a hundred bucks to test their DNA.

The project turned out to be a money maker. “The day we announced, we sold ten thousand. It had hit a hundred thousand by the end of the year,” says Wells. “It launched the consumer genomics industry”—which got a noticeable boost in 2006 when the media legend Oprah Winfrey had her DNA tested for a television show, revealing ancestral links to people now living in Liberia, Cameroon, and Zambia. It also turned out that Winfrey, unlike many black Americans, shared no recent ancestry at all with Europeans. “I feel more connected to where I’ve come from,” she told the show’s host, Henry Louis Gates Jr., a Harvard professor of African American studies, who wrote a book about her experience.

Before long, companies such as 23andMe and AncestryDNA were selling their own kits, turning over billions of dollars. In 2018 it was announced that AncestryDNA alone had sold a total of around ten million kits around the world.

Spencer Wells has since left the Genographic Project, become the owner of a nightclub in Texas, and moved on to new genetic testing ventures. He tells me that the personal ancestry testing industry flourishes in the United States because of the rootlessness of so many of its citizens. “In societies like the present-day US, where we have a lot of hyphenated Americans—Irish Americans, Italian Americans, Hispanic Americans—people feel somewhat disconnected from the entity that comes before the hyphen. So they want to figure out who those people were. Who were the people who migrated to the US?” For black Americans in particular, most of whose ancestors were transported as slaves and ripped away from any connection to their families or homelands, the kits have offered the only means they may have of tracing their genealogy. But the psychological effect on the public of sequencing the human genome, of convincing people that our differences are identifiable in our DNA, is that this now appears to be a reliable way to define who we are.

In reality, genetic testing is only an educated guess about where your relatives may have lived, based on the data fed into the models in the first place. As geneticist Mark Jobling explained, it’s possible to group people any way we like. Ancestry tests scan portions of people’s genomes to find those who have genetically a little more in common, then pool them together. Theoretically, they could pool them using any measure. But, of course, companies most often use old-fashioned racial categories or nationality.

This also means that if there are no DNA samples from the modern-day people related to you, you’re stuck. For example, one of the reasons Winfrey is linked to Liberia may be that this is where former slaves were long ago sent by white American leaders who couldn’t bear the thought of slaves living freely among them. Ancestry testing doesn’t show you your past as much as it reveals the people you are distantly related to in the present who have had similar tests done. Oprah has some connection to people who now live in Liberia, but this is not necessarily her ancestral homeland, from where her relatives originally came.

Mark Thomas, a leading geneticist at University College London who has seen his own research recruited into these models, tells me he has always been skeptical of ancestry testing firms. “It’s not that they’re cynical, it’s not that they’re nasty, it’s not that they’ve got particularly racist agendas. They want to make money, and you make money by servicing peoples prejudices.” Not just prejudices, but also people’s desire to know who they are. Thomas and his colleagues were threatened with legal action by BritainsDNA, one such company based in Scotland, after they challenged the firm’s wide-ranging claims in the press, for instance, that the actor Tom Conti is “Saracen” in origin, and that Prince William has Indian ancestry. Neither claim could be established by genetic data.

Ancestry testing has taken the work of well-meaning scientists who only tried to do good in the world and inadvertently has helped reinforce the idea that race is real. Using the methods and data of scientists such as Cavalli-Sforza, an entire industry has achieved exactly the opposite of what these scientists once set out to do. The true way that variation works, the nuances, are rarely explained by those selling ancestry-testing kits. Having seen how “black” Oprah really was, white supremacists in the United States began using the very same tests to prove how “white” they were, sometimes sending off vials of spit to various companies until one came back with the desired result and established beyond doubt that they were of purely European ancestry. By tracking history through our genes, by dividing up individual bodies into proportions of nationalities—so much European, so much African, so much Asian—the tests fortify the assumption that race is biologically meaningful. If it’s possible to categorize, we assume, there must be something to the categories.

The irony is that as more research has been done into our origins since the launch of the Human Genome Project and consumer ancestry testing, it has only undermined these measures of identity. Within the last decade, as scientists have uncovered exponentially more genetic evidence about us and our ancestors, even they have been surprised by the results. Nothing has matched expectation. Our roots, it turns out, are very rarely where we think they are.