A breakthrough in North American prehistory: a theory discounted by the majority might be true

A young scientist from the University of Ostrava (Czech Republic) reveals North American prehistory. He will continue his research at Harvard University with the world-famous geneticist David Reich.

Pavel Flegontov,CSc. (University of Ostrava, Czech Republic) and his team have recently published an article in Nature that intends to put an end to a long-standing dispute over North American prehistory combining data and knowledge from paleogenetics and linguistics.

Geneticists, anthropologists and linguists have long suspected that speakers of the Na-Dene language family from North America (Chipewyan and Navajo among them) have admixed with a Siberian group within the last few thousand years, long after the first migration that brought all Native Americans to the continent around 15 000 years ago. In 2012, David Reich’s group revealed that the additional source of ancestry in Na-Dene speakers was related to the so-called Paleo-Eskimos, the first inhabitants of the American Arctic tundra and coasts having arrived there from Chukotka around 5000 years ago. In 2014 a prominent team of archaeogeneticists led by Eske Willerslev sparked controversy by publishing three papers in the journals Science and Nature (2014-2018), in which it was concluded that Paleo-Eskimos and Na-Dene (Athabaskan) peoples are not genetically linked. That team instead argued that the relatives of Koryaks and Chukchi, indigenous peoples of Kamchatka and Chukotka, have admixed with Na-Dene ancestors. The latter hypothesis requires an additional migration across the Bering Strait that is not attested to archaeologically.

“To resolve this debate, we collected a wide arsenal of data analysis methods, both standard and novel, and obtained breakthrough results using two independent graph-based methods,” says P. Flegontov, a 35-year-old geneticist of Russian origin, who has led an international team of scientists to the pages of the world´s leading multidisciplinary scientific journal.

Now, for the first time, scientists have obtained genomic data of the key ancient populations they needed: Aleuts, Athabaskans, and Eskimos. Initially, all possible branching orders were tested for a sub-tree consisting of Chukotko-Kamchatkan speakers, Paleo-Eskimos, ancient Athabaskans, ancient Aleuts, and modern or ancient Eskimos or Inuit.

Then, on a set of 133 thousand population combinations, the two most likely branching orders were tested: “Willerslev’s topology” and the topology obtained from Flegontov’s research. Finally, several of the most likely topologies were tested by an independent method based on rare genetic variants, Rarecoal. Unlike the admixture graph method, Rarecoal allows events to be dated on the graph. Thus, two independent methods based on graphs did not confirm “Willerslev’s topology”, and this result was also supported using other methods.

This new model of genetic history also suggests that a hypotheses discounted by the majority of linguists might actually be true - that the Yeniseian language family from Western Siberia, that is almost extinct, and the Na-Dene languages spoken by Native Americans, might have a common ancestral language. Although genetics cannot prove or disprove linguistic theories, information on population movements can inform us on the relative likelihood of various historical linguistic theories.

The great success of Flegontov has brought him to the top of his field. “I have been invited to continue this and similar projects as a member of David Reich’s team at the Harvard Medical School. I hope to move to Harvard in September and work there as a staff scientist for about 5 years. I plan to keep my research group here at the University of Ostrava and hope to return to Czechia in 5 years,” he said.

Moreover, Flegontov and his team in Ostrava are now working on a large-scale project on the history of Asia roughly covering the last 10 000 years. So far, no detailed graph describing Asian population history has been published.

“For the first time we have developed a robust graph for major Asian lineages and used it for revealing genetic affiliations of hundreds of ancient individuals and present-day populations. So we could know what the genetic composition of Wusun, Kangju, Xiongnu, and dozens of other mysterious peoples mentioned in Chinese chronicles was.”

More information on the historical model proposed by the authors and how it fits with linguistics and archaeology can be found at the official website of the University of Ostrava. Here is also an interview with P. Flegontov on his latest article.


Updated: 07. 06. 2019