Acclaimed geneticist David Reich specializes in ancient DNA studies, a field that is radically informing, as well as correcting, ideas of our origins, histories of migration, and formation of communities. These findings, outlined in his recent book Who We Are And How We Got Here, have repercussions not only for the identification of disease risks among population groups but also the sociopolitical lives of nations. In an interview with Lounge during his visit to Delhi in January, he spoke about the book and his research. Edited excerpts:
What are the practical applications of ancient DNA studies?
Ancient DNA helps us connect the people of the present to those of the past, whose skeletons we can obtain data from. It also allows us to learn where different people’s ancestors lived and how each group in the world is related to one another. Prior to this technology, it was really impossible to understand these questions. There was a development, for 20-30 years, when people hoped to achieve this goal by studying the ways in which present-day people are related to one another. But it’s quite clear now that this type of work fails to produce an accurate reconstruction of the matter because people move too much.
Much of the research in this field has implications for medical genetic studies, in trying to find genes that cause disease. That’s what I worked on earlier. My specific strength was to understand population structure differences in history and the ways in which these might affect people’s risk to different diseases. One area I’m interested in is the caste structure in India. There are many endogamous groups in South Asia where people don’t marry outside their groups. As a result, the rare mutations (which we all carry) make people here particularly vulnerable to certain disease risks. In that sense, ancient DNA studies can have profound public health interventions.
How does your work intersect with archaeology and anthropology?
Ancient DNA technology has become possible only since 2010, before which there was no genome-wide data. Now it’s possible to generate it, easily and inexpensively, from people who lived tens and thousands of years ago. The raw material we use are ancient human skeletons, which are archaeological resources. The value of our work is only as good as the understanding of the richness of the material we are studying. We are always looking for questions that genetics can help answer. There are long historical debates about who’s related to who, which DNA studies can clarify.
\Can you talk about the population mix of the subcontinent?
We have studied and generated genetic data in the region, in collaboration with our Indian colleagues in Hyderabad, based on more than 2,000 people from over 300 endogamous groups. In the process, we have characterized a lot of the variations among people from different language groups and places and tried to understand them by identifying simple broad patterns.
There are many different language groups here, such as the Tibeto-Burman, which is spoken in the north and eastern fringes of India. Then, there are special groups like the Austroasiatic languages, such as the Mundari languages spoken in the tribal belt of India (by Santhal and Gond people, among others), which are related to some of the languages spoken in the Nicobar Islands, as well as the majority of the languages in Vietnam, parts of Thailand and Cambodia. What we have is an archipelago across South-East Asia extending as far west as South Asia—we have languages that exist in pockets, surrounded by other languages. It almost certainly reflects an earlier spread of people, probably related to the spread of rice agriculture, but that’s hypothetical.
Our focus in India has mostly been on the main pattern variation among people speaking Indo-European and Dravidian languages. The observation we made in 2007-08 was that if you look at genome-wide data in most people in India today, it’s clear that there’s a gradient of people ranging from two extremes—from those related to East Asians and others related to South Asian hunter-gatherers. It’s a striking pattern. We were able to document that this was, to a large extent, the result of the mixture of two very different source populations. More than 4,000 years ago, there were unmixed representations of these groups in India but that no longer is true. It was also clear that what we see of the caste system now goes back much deeper in time, long before the British strengthened these social divisions.
What light do Indo-European languages throw on ancient DNA studies?
William Jones first discovered the Indo-European group of languages in the late 1700s. He argued that the similarities between Sanskrit, Greek and Latin were too high to be accidental. That must mean they all stem from a common source, which we don’t know when it existed. It’s been a mystery how these languages spread from far-flung places, though it’s known that languages, for the most part, spread through movements of people. This suggests that there might be an ancestral population that spoke a shared language, but the question is, where did those people originate? That has been an unresolved open debate for almost 250 years. Now, genetic data offers opportunities to study the similarities between present-day people who speak these languages and also understand the spread of ancestry types that might track the spread of the languages over time.
How can ancient DNA affect the political and social lives of nations?
Ancient DNA has proven to be very disruptive. When you look at data from the skeletons of people who lived in any one place today, it’s extremely rare that those people looked like people who live in those same places now. This seems to be true almost everywhere now where we get ancient DNAs. We can use this microscope to look at the past and apply it to different ancient cultures and ask the question, how are the people living there in ancient times related to the people living there today? The technology we now have is probing and taking measurements about phenomena that wasn’t possible to quantify before.