Over a hundred years ago, a skeleton was found in Gough’s Cave in Somerset, England, and dubbed "Cheddar Man." He lived around 10,000 years ago, and is thought to be one of the first people to come from continental Europe to Britain as the last ice age came to an end.
Britons have assumed that if they are the descendants of Cheddar Man, then he would have looked like them, with pale skin and eyes. Until very recently, we wouldn't have had any way of verifying this. But as the techniques for isolating ancient DNA from skeletal remains gets better and better, we can examine these remnants of our past like never before.
The Cheddar Man, new DNA analysis showed last month, probably did have blue eyes, but also a very dark brown or black complexion and dark curly hair. This means that lighter skin may not have been present in Europe until much later than originally thought.
We now have genome-wide information on more than a thousand ancient people, David Reich, a geneticist at Harvard Medical School who was not involved in the Cheddar Man research tells me. In just the past three years, Reich's lab has contributed greatly to that: He's helped publish the DNA from 938 ancient humans whose lives, migrations, and breeding fundamentally influenced who we are today.
Past writers have been critiqued when discussing how our genetic pasts can be used to understand human differences. In Nicholas Wade’s 2014 book, A Troublesome Inheritance, he explained racial stereotypes through what he claimed were innate genetic differences, and received backlash from the scientists’ whose work he referenced.
But Reich says, in his new book, Who We Are and How We Got Here: Ancient DNA and the New Science of the Human Past, that geneticists can talk about genetic differences without implicating racial stereotypes. In fact, like the Cheddar Man findings, we can use ancient DNA to disrupt our ingrained assumptions about our current populations, how long they've been around, and what they look like. I spoke with Reich about what we can truly learn from this growing trove of genetic information.
What information does ancient DNA provide us, and how's that different from what we could tell about our ancestors using only present-day DNA?
Present-day people are descended from ancient people, and our genes are affected by that history in profound ways. But we're so mixed up that we have a lot of difficulty reconstructing how we're related to each other, deeply. With ancient DNA you can get high-quality DNA from people who lived long ago, in association with known archeological cultures, and see how they're related to other ancient samples, as well as people today. It's like having a time machine that brings you right to the place, and it tells you what was actually happening. It's totally transformative. It's much more powerful than just being able to look at things from the perspective of the present, which has driven all DNA research of history until now.
What did people first hope ancient DNA could tell us, and has it lived up to those expectations?
When people first started to look at ancient DNA, I think the hope was that ancient DNA would teach us what makes humans distinct as organisms from apes, Neanderthals, and other archaic humans. By comparing the DNA from us to those, we might see the changes that make us biologically different or special, and confirm our belief that our behaviors are special.
DNA results, so far, have not actually shed much clear light on that question, or if we are even special at all. But it has been a runaway success with respect to the question of how humans got to where we are today from migration mixtures. That is how DNA is telling us who we are and how we got here in rich technicolor detail. It’s shown us how our ancestors migrated, and how they mixed to form who we are today. It's been very, very surprising and not what any of us expected.
One surprising finding that ancient DNA gave us was about our past interaction with Neanderthals. What happened there?
This is work that came out of Svante Pääbo's lab. He was really motivated to try and sequence the DNA from Neanderthals, our archaic human cousins who lived in Europe as recently as about 40,000 years ago, had big brains like us, but were very different in other ways. There was a big question about whether they interbred with modern humans and the DNA work, in which I was involved, showed definitively that they did interbreed with the ancestors of non-Africans (non-Africans contributed about 2% of the DNA over Africans living today).
His question was: Was there interbreeding or was there not, and if there was interbreeding, how much interbreeding was there? What wasn't anticipated in his findings was a whole Pandora's box of additional archaic humans and mixtures amongst those archaic humans. In addition to the Neanderthal genome, there was a finger bone from a cave in the Altai mountains of southern Siberia, called the Denisova cave, that was thought by the excavators to be that of a modern human, or maybe a Neanderthal. It actually yielded a genome from a third group, that was distinct by half a million years to either of them.
For Neanderthals, we knew they were interesting and we went out as a community to go find the genome. In this case, we had no idea who these people were. All of these complex additional relationships between different archaic humans have really changed our picture of deep relationships amongst humans. It shows that interbreeding amongst these groups was common in history, and it's shown that Eurasia was full of groups that lasted for a long time, and interacted in profound ways.
There’s also a group that you call "humanity's ghosts." Who were they?
One of the chapters of the book is called "humanity's ghosts" because it talks about the idea of ghost populations: when we are analyzing ancient DNA data and we find evidence of mixture from a population that existed in the past but don’t have any descendants or evidence of them today. That's very exciting when we find it.
The experience again and again from genetic data has been that once ghosts get predicted, they get found through ancient DNA. A really amazing example of this is the discovery of the ancient north Eurasian, which is a group that my laboratory predicted in 2012, when we found that northern Europeans are a mixture of two groups–one related to ancient near easterners and southern Europeans, and the other related to , of all people, Native Americans. To explain this, we suggested that northern Europeans today are a mixture of southern Europeans–rural European farmers basically–and an ancient north Eurasian group that was once distributed across northern Eurasia, and whose ancestors also contributed more than 15,000 years ago to the ancestors of Native Americans. That prediction was borne out two years later by a group in Denmark, which was sequencing DNA from a 24,000-year-old human in Siberia, and found a perfect match to this prediction.
That population was no minor population. It was a population that's contributed hundreds of millions of full genomes worth of people today. About a third of the ancestry of Native Americans, about a sixth of the ancestry of Europeans, about a sixth of the ancestry of south Asians is from this group.
What other things can we uncover that are not visible to us now, or were erased intentionally?
One of the other powers of DNA is to recover stories that are gone, because there's no writing, or because due to power differentials, those histories have been forcibly taken away or lost somehow.
There’s two really striking examples of this. One is that of African-Americans who are using DNA to reclaim the past that's been stolen from them from the trans-Atlantic slave voyage and systematic attempts to take away their people's cultural roots by those who enslaved them.
Another example is Native Americans, who, when Europeans came, were relegated to second or third-class positions in society. A lot of the rich native histories and cultures and traditions, they're not completely lost, but they were not recorded as clearly. There’s a big open question about how large American populations were before Europeans and Africans got here and brought diseases and took away a lot of people's land.
Genetics can, in principle, reconstruct those sizes and reconstruct the degree of density and sophistication that some of these groups must have had. The work that we're doing right now for Peru that shows that over four or five thousand years, we can see that the populations were extremely large, with almost no evidence of small populations in this period, at this time. So this is telling you that what you're seeing is really a grand, high-density population.
There is some nervousness existing about studying or pointing out biological differences of different populations and different races. Nicholas Wade’s book provoked a lot of criticism for doing so when it came out. What’s the general fear that people have in this field in determining biological or genetic differences between races, and how do you think about overcoming that when you're doing this work day to day?
In the early 1970s, for the first time, genetic measurements became good enough to measure how much variation there was across populations. A geneticist named Richard Lewontin showed that the average difference between human populations was a sixth of that between two individuals. So, if you take two individuals that you see around the street today, they look quite different. They behave quite different. Some of that is genetic, and what he showed is that the average differences between a pair of populations are a sixth of that size. So, most of the differences you see are amongst individuals, not amongst populations.
That was a profound insight because it said that on the scale of the differences amongst humans perceived around us, the differences between populations, on average, are modest. That contributed to a kind of orthodoxy that developed over the subsequent decades through the work of geneticists and anthropologists, that the degree of biological differences amongst human populations was so small that it could not really predict any substantial differences in traits we might care about across populations, and that differing traits like skin color were very unusual.
But now we can measure those smaller differences, and I think we need to contend with the findings that it's bringing. We need to prepare ourselves for having to talk about those.
Why is it problematic at all to say that there are differences between populations, even at this minute level that we're now able to measure?
It shouldn't be problematic, because there are differences on average across different populations, as we've found for many, many decades. Just like there are differences among individuals, although they're much smaller. Just like there are differences between men and women genetically, although those are much, much larger than it is between populations. It's not problematic. It's the truth, so we need to figure out a way to discuss them.
I think that we don't really address the issues head-on because some anthropologists become uncomfortable when we do. In the 2000s, I was focused on trying to find risk factors for disease in African-Americans, who have higher rates of certain diseases and lower rates in other diseases than, for example, European Americans. I focused on trying to find risk factors for prostate cancer, which is a disease that occurs 1.7 times more often in African-Americans than in Europeans. The methods I was using involved going through the genome, and identifying in an African-American person–who on average is about 20 percent European in ancestry–what sections of their genomes derived from African ancestors 500 years ago, versus European ancestors.
That was very successful. We found seven genetic risk factors that entirely explained the difference in risk between the two populations, but when I talked about this to anthropologist colleagues, there was a huge amount of anxiety and discomfort with the idea of even talking about these issues. The reason was that there was a concern that by talking about differences amongst US populations, we would somehow be giving biological reality to the idea of race. When in fact, as everybody knows, and as anthropologists have shown, race is not defined biologically. It's defined based on social definitions and changes over time and place. But somehow there was a concern that by doing this, we would make race more “real” and we were going to contribute to discrimination or racism based on an impression of biological differences across groups.
That’s what the criticism of Nicholas Wade was. In his book, he talked about genetic differences between races to explain things like why Chinese people are better workers, or why black people have lower IQs. How did he come to those conclusions?
Nicholas Wade's work was really intellectually problematic. The first part of the book is totally fine, and it makes the correct point that academics, scholars in genetics and anthropology have adhered to an orthodoxy, and have perpetuated an orthodoxy, that there are no meaningful differences between human populations worth talking about, and that there can be none. Which is incorrect. The average differences amongst human populations are small, but they are real, and there's been enough time separation from pairs of different human populations, including ones that correspond to self-identified racial groups, for substantial differences to arrive.
So that was right, and it's not something that geneticists have been really comfortable talking about, because there's so much flack that comes from talking about it. But what was very dishonest and problematic in Wade's work was that he jumped from that to something which is completely unsupported. He was saying, okay, there's a space for differences amongst groups, and I know what those differences are: They correspond to traditional stereotypes of differences amongst human populations.
There is absolutely no reason to think that traditional stereotypes are based in genetics at all. In fact, what the genome revolution has told us again and again is that whenever we look at real data when we measure things, our stereotypes and our expectations are blown apart. What he did was like a bait-and-switch. He started with a correct statement about something that wasn't being said, highlighted a group of academics who were perpetuating an orthodoxy and sort of confusing the public, and then he put up a "truth" that has no support or validity at all. That's the problem with the book in my view.
Back to your work on ancient DNA learning about migration patterns and population divisions, how does that free us from the way that we currently think of how we're divided, rather than enforce all of the current population divisions that we have?
I think it frees us in two ways. One is it explodes the intuitive understanding many people have of the current racial structure of the world, or the current population structure of the world. For example, if you like to think of "European" as an age-old grouping, that's deeply wrong. We now know from genetic data that the genetic group that exists today that we think of as northern Europeans or Europeans, really was only formed 5000 years ago.
The great majority of people of European ancestry today didn't exist more than 5000 years ago, and are the result of a mixture of four extremely different populations, each of them as different from each other as Europeans and Chinese, that existed 10,000 years ago or more. If you think that the present population structure of the world is somehow an immutable, unchangeable thing, you're deeply wrong and ancient DNA proves you wrong in a way that nobody expected prior to this ancient DNA revolution.
It de-centers us and destabilizes us, and makes us realize that actually, we should be humble, and not think that our gut impressions of the world around us are reliable. I think that anybody who listens to the gushing fire hose of information that's coming out of the ancient DNA revolution and the genomic studies of the past has to realize that we need to have some humility and that we can't pre-judge what we're going to find.