91av

THE end of the second world war was meant to have spelled the death of race science. Until the 1930s, it had been relatively acceptable for biologists and anthropologists to believe in innate differences between races. Many assumed that certain groups were superior to others. It was only after the war and the Holocaust that the world finally turned its back on this dangerous field of research.

[video_player id=”3oEU1Fnb” access_level=”everyone”]

People thought about race differently following the war. Anthropologists showed that most of what we think of as racial difference is in fact cultural and linguistic difference. Geneticists, starting with Richard Lewontin in 1972, have shown that more than 90 per cent of the genetic variation we see between humans lies within the racial categories we use. Being of the same race doesn’t necessarily make two people more genetically similar to each other than either of them would be to someone of another race. Race is today described as a social construct, its study confined to the social sciences so we can understand the effects of historical and modern-day discrimination. The handful of scientists who have continued to insist publicly on the existence of biological races have often been on the margins of respectability. There was William Shockley, the Nobel prize-winning physicist at Stanford University in California who wanted in the US to be voluntarily sterilised. Then there was Arthur Jensen, a psychologist at the University of California, Berkeley, who claimed that black people had intelligence levels than white people.

We think of Jensen and Shockley as exceptions. We assume that race has been purged from science. But has it?

In truth, it never completely disappeared. There remains a suspicion among some scientists that there could be something tangible to race, that genetics could someday uncover unpalatable truths. Last year, Harvard population geneticist David Reich wrote in The New York Times, “it is simply no longer possible to ignore average genetic differences among ‘races'”. Reich told me in an interview a month later that he doesn’t believe these differences will be large, and genetics will continue to shatter racial myths, but even so, “there are differences amongst people. We don’t know what they are so we have to deal with uncertainty.”

This uncertainty reflects how even well-intentioned, politically neutral researchers such as Reich can’t help but resort to race when thinking about human difference.

“We assume that race has been purged from science. But has it?”

The field of population genetics is a case in point. One of its former leading lights, the late Luigi Luca Cavalli-Sforza was a staunch anti-racist. In 1973, he publicly debated race with Shockley at Stanford, where he was also a professor. Yet even Cavalli-Sforza, whose work focused on understanding how genetic variation is distributed among humans, clung to the belief that race isn’t a completely redundant concept in genetics.

In his book , Peoples and Languages, published in 2000, Cavalli-Sforza wrote: “A race is a group of individuals that we can recognize as biologically different from others.” His was a statistical definition of race, based on the concept that there are clusters of people who share certain gene frequencies. So while our common racial categories may have little meaning, suggested Cavalli-Sforza, certain populations – especially tight-knit ones – could be considered races, even if only statistically. There are no distinct racial “types” with hard boundaries, only statistical similarities between whichever groups we want to identify.

Philosopher , at Saint Mary’s University in Halifax, Canada, has warned of the ethical repercussions of thinking about race in this clustered way. The problem lies in the need to group in the first place, to separate people even when that means having to zoom in on the tiniest bits of the genome that might differ. Even then, those tiny differences are only on average, not applicable to every person in that group. Gannett calls it ““. While scientists may feel they have left race behind, she argues, in reality it is only their language and parameters that have changed. They may not call them races, but researchers today still refer to European and African “populations”.

Population trap

This clustered way of thinking about race culminated in the early 1990s with the , spearheaded by Cavalli-Sforza himself. He and colleagues based mainly in the US wanted to take DNA samples from what they saw as distinct, isolated indigenous groups around the world – including the Basques of Europe, the Kurds of eastern Turkey and Native Americans – and use them to build a picture of the past as well as of human evolutionary change. This new perspective, they announced in the journal Genomics in 1991, would “supplement and strengthen findings from archaeology, linguistics, and history”.

Could this be called race science? Those behind the Human Genome Diversity Project didn’t see what they were doing as race science, but it was hard not to see how the project bore some of its hallmarks.

One scientist, for example, “talked about the need to sample ‘isolates of historical interest’, a term that indigenous populations did not care for”, says Henry Greely at Stanford Law School, who was brought in to navigate the ethical issues around the project. “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 something you find in a museum. It was tone-deaf.”

Politically anti-racist though the scientists behind the project were, they fell into the trap of treating some populations as biologically special and distinct. They forced people into categories that did not necessarily make evolutionary sense, in the same way that race scientists did in the 18th and 19th centuries when coming up with the racial categories we use today. These populations were sometimes defined by little more than cultural or linguistic boundaries.

“The idea that [the people in] these groups are somehow truly genetically similar is a huge assumption,” says sociologist Catherine Bliss at the University of California, San Francisco. Just because communities appear to be tight-knit and ancient doesn’t mean there has never been any mixing between them and others. There has always been mixing between all human populations, which is why we are so similar today.

Even at the outset of the Human Genome Diversity Project, concerns were raised. Before it was launched, an alternative was mooted. Rather than sampling the DNA of population groups thought to be isolated, scientists could instead study people at regularly spaced geographical intervals across the globe.

Culturally loaded

This “grid sampling” approach was championed by Allan Wilson, a biochemist based at the University of California, Berkeley, known for the Mitochondrial Eve hypothesis, placing our most recent common ancestor in Africa around 200,000 years ago. Wilson believed that grid sampling would document human variation as it really was rather than imposing possibly incorrect assumptions on existing groups.

“If you were to grid sample,” says Bliss, “you’re not going to get tidy and neat similarities.” This is because there aren’t any: human variation doesn’t sit along precise boundaries. It is messier than any racial model, with each neighbouring population blending into the next. While some populations may have slightly different gene frequencies to others, there is no gene that appears in all the members of one race and not in another. “Wilson’s approach would’ve gotten us closer to actual similarities,” says Bliss. “It would have been more accurate.”

But Wilson died aged 56, just before the first planning meetings for the Human Genome Diversity Project began. Without him, his grid sampling proposal was shelved. The and protests from indigenous groups. It never secured the funding it needed to get off the ground.

For geneticist Mark Jobling at the University of Leicester, UK, the way the project was structured – deliberately targeting particular populations rather than looking at the DNA of people wherever they happened to be in the world – was what ultimately undermined it. “How you define the population in the first place, these are culturally loaded things in themselves,” he says. “So there was a lot of cultural discrimination in the original aims.”

“A pernicious assumption is that specific races have specific genes”

It is perfectly possible to study global human variation without grouping people, says Jobling. Our genomes have so much in common, the physical variations so subtle, that humans can theoretically be grouped any way we like. The genetic variants associated with light skin, for instance, are common not only in Europe and parts of east Asia, but also in the San hunter-gatherers of southern Africa. You could do a thought experiment where you grouped everyone on Earth based on three nationalities: Kenyan, Swede or Japanese, he says. Theoretically everyone could be placed in one of these groups as we are all genetically connected to the average Kenyan, Swede or Japanese person, either directly or by historical migration. “You could say that you were so many per cent Kenyan, so many per cent Swedish and so many per cent Japanese,” says Jobling.

This may seem bizarre, and of course he’s not suggesting we do it, but it is no more bizarre than dividing the world into black, brown, yellow, red and white – as race scientists once did. And it is only a little less arbitrary than the racial categories we use today. Medical researchers have noticed how poor a proxy for human variation these racial categories can be. Even the gene that causes sickle cell disease, which people often think of as being mainly associated with those of African ancestry, is actually found in all ethnic groups, as the .

This isn’t to say that those behind the project were racist. “Racism is born of ignorance, fear of the stranger and desire for power: it has to be fought, first by showing its nonsense and its real roots,” Cavalli-Sforza wrote in an email interview with me six months before he died in August 2018. He was a force for good, a scientist who fought publicly against racism and fascism in the full understanding of what this meant, having lived in Italy during the second world war.

Yet in the same exchange, he also referred to the “children of partners coming from genetically distant groups” – commonly known as mixed-race or mixed-heritage children – as “hybrids”, a term that many might see as both morally and scientifically inappropriate.

“Cavalli-Sforza was an old school anthropologist,” says Jobling. “He would show old slides of him collecting DNA, blood samples in Africa from pygmy groups and offering glass beads and cigarettes in return.”

Cavalli-Sforza wasn’t a racist. At the same time, it may be impossible to expect anyone, especially a biologist born in 1922, to completely shed the ideas of race prevalent in the early 20th century, particularly when society in general still lives with them today – racial discrimination is still widespread in education and employment, for example.

This is the heart of the problem. It is difficult to be convinced that mainstream scientists fully abandoned race science after the second world war – not because they were racists, but because they were human. They were living in a world that was still ridden with racism, segregation and apartheid. The old ideas about race that had been a firm part of everyday science wouldn’t simply vanish.

“What happens is that you’ve got a large community of very well-meaning, self-described anti-racist 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,” says historian Joanna Radin at Yale University. But “population genetics is a science done by people, working with the assumptions and the ideas that are available at the time”.

Race is useless

According to geneticist Mark Thomas at University College London, who has collaborated with Reich, race is a useless way to think about human variation. “There is no categorical imperative in biology, and no need or value in placing people in biological boxes,” he says. “Unfortunately, that doesn’t stop people ‘racialising’ others, and perhaps that reflects our desire to categorise.”

The recent discovery that other, now-extinct humans such as Neanderthals and Denisovans bred with Homo sapiens has reignited the racial debate, prompting some to wonder if the greater proportion of Neanderthal DNA shared by Europeans, for instance, gives meaning to our racial categories. But as anthropologist John Shea at Stony Brook University in New York notes, “the interbreeding thing is more like a symbolic thing for us than it is of evolutionary consequence”.

Even now, scientists struggle to accept that they may still be working within the frameworks of the past. There remains a tendency to treat population groups as if they are genetically distinct. The in Paris, France, keeps a bank of DNA samples from populations all over the world. And in 2015, the University of Oxford launched a project to make a genetic map of the people within the UK, named . While no members of these projects would see themselves as race scientists, it could be argued that this grouping of people harks back to the same methods.

Population genetics also fuels the trend for consumer ancestry testing. In 2018, AncestryDNA announced that it had sold around 10 million kits. “The Human Genome Diversity Project had these ethnic groupings, and all these ancestry testing firms have their ethnic groupings, which are the same groupings,” says Bliss. “They collect DNA and have reference populations, and they’re interchangeable. The way they communicate those results and map people’s ancestry is in a completely ethno-racialised way.”

Among the public, the tests may reinforce a belief that race has deeper biological meaning. In 2017, it was reported that white supremacists in the US were using them to prove how “white” they were, results as proof of shared European ancestry. These tests are based on the belief that there are distinct ancient populations from which each of us hail, says Bliss.

The answer, perhaps, is for researchers to stop thinking about people in clusters. “One of the most pernicious assumptions that we still hold when it comes to human genetics and genomics is that specific races have specific genes or specific genetic material that’s unique to them,” says Bliss.

This assumption is false. But we will never be free of the fallacy, she says, “if we keep on sampling by racial categories or ethnolinguistic categories, and write the data through the software, and then slap the categories back on afterwards”.

This piece is an edited extract from Angela Saini’s new book Superior: The return of race science (4th Estate), published on 30 May.