Nature | News

Steppe migration rekindles debate on language origin

Eurasian region gains ground as birthplace of Indo-European tongues.


Article tools

W. Haak et al. (2015)

From Icelanders to Sri Lankans, some 3 billion people speak the more than 400 languages and dialects that belong to the Indo-European family. Two fresh studies — one of ancient human DNA, the other a newly constructed genealogical ‘tree’ of languages — point to the steppes of Ukraine and Russia as the origin of this major language family, rekindling a long-standing debate.

Scholars have long recognized an Indo-European language group that includes Germanic, Slavic and Romance languages as well as classical Sanskrit and other languages of the south Asian subcontinent. Yet the origins of this family of tongues are mired in controversy.

Some researchers hold that an early Indo-European language was spread by Middle Eastern farmers around 8,000–9,500 years ago (see ‘Steppe in time’). This ‘Anatolian hypothesis’ is supported by well-documented migrations into Europe, where agriculturalists replaced or interbred with the existing hunter-gatherers. In 2012, a team led by evolutionary biologist Quentin Atkinson of the University of Auckland in New Zealand produced a family tree of Indo-European tongues that also pointed to an Anatolian origin more than 8,000 years ago.

A competing theory posits that the languages emerged on the Eurasian steppe some 5,000–6,000 years ago, when the domestication of horses and invention of wheeled transport would have allowed herders there to rapidly expand their range. Proponents of the ‘steppe hypothesis’ note that linguistic reconstructions of a proto-Indo-European tongue include words associated with wheeled vehicles, which were not invented until long after Middle Eastern farmers had reached Europe. “Most linguists have signed up to the steppe hypothesis,” says Paul Heggarty, a linguist at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany.

One knock against the theory was a lack of compelling evidence for a large-scale migration from the Eurasian steppe at this time.

A study of ancient human DNA posted to the preprint server on 10 February now plugs that gap (W. Haak et al.; 2015). A team led by David Reich, an evolutionary and population geneticist at Harvard Medical School in Boston, Massachusetts, analysed DNA from the bodies of 94 individuals who lived across Europe between 8,000 and 3,000 years ago. The data confirmed the arrival of Middle Eastern farmers in Europe between 8,000 and 7,000 years ago. But they also revealed evidence for a second migration that began several thousand years later. DNA recovered from steppe herders called the Yamnaya, who lived in what are now Russia and Ukraine around 5,000 years ago, closely matched that of 4,500-year-old individuals from present-day Germany, who were part of a group known as the Corded Ware culture that encompassed most of northern Europe. The similarities suggest “a massive migration into the heartland of Europe from its eastern periphery”, the team writes.

Yamnaya ancestry survives in the genomes of modern Europeans, with northerners such as Norwegians, Scots and Lithuanians maintaining the strongest link. The geographical extent of the Yamnaya migration is not clear, but the researchers note that the eastern migrants could have completely replaced existing populations, at least in what is now Germany. It is impossible to know the language these migrants spoke, but it is likely to have originated in the steppe homelands of the Yamnaya.

“This seems like very striking support for at least part of the traditional steppe model of Indo-European diversification,” says Andrew Garrett, a historical linguist at the University of California, Berkeley, whose own work adds further support. When he and his team re-analysed the data from Atkinson’s 2012 family tree, this time taking into account the approximate ages of ancient Indo-European languages, they dated the origin to around 6,000 years ago, in line with the steppe hypothesis (W. Chang, et al. Language; in the press).

Atkinson says, however, that the analysis assumes that ancient languages such as Latin and Old Irish are direct ancestors of modern languages, instead of side-branches of a common ancestor. This makes it appear that these languages evolved faster than they did, he says, and would argue incorrectly for a more-recent common tongue.

Heggarty points out that Reich’s ancient DNA study is not the final word on the steppe hypothesis either. He suspects that the Yamnaya spoke a language that later developed into Slavic, Germanic and other northern European tongues, but he doubts that the group imported the predecessor of all Indo-European languages: “For me, these data look like the steppe population was speaking a branch of Indo-European.”

Reich and his team acknowledge that attributing the origin of all Indo-European languages to the Yamnaya migration would require the discovery of their genetic signatures in samples from further east, such as from India and Iran. But Carles Lalueza-Fox, a palaeogeneticist at the Institute of Evolutionary Biology in Barcelona, Spain, notes that the climates of the Middle East and southern Asia do not augur well for preservation of ancient DNA: “It could be difficult to find good samples from the right time frame.”

Journal name:
Date published:



An earlier version of this story misspelt the name of Paul Heggarty.

For the best commenting experience, please login or register as a user and agree to our Community Guidelines. You will be re-directed back to this page where you will see comments updating in real-time and have the ability to recommend comments to other users.


Commenting is currently unavailable.

sign up to Nature briefing

What matters in science — and why — free in your inbox every weekday.

Sign up



Nature Podcast

Our award-winning show features highlights from the week's edition of Nature, interviews with the people behind the science, and in-depth commentary and analysis from journalists around the world.