Hittite from the north-east

Hittite from the north-east

This is a response to a recent high-powered paper based on genetic data which offers a new solution to questions about Indo-European origins and specifically the division between the Anatolian languages (such as Hittite) and the rest. The paper is “The genetic history of the Southern Arc: A bridge between West Asia and Europe” and it has 205 authors. It was published in Science in August 2022.

A few years ago David Reich’s groundbreaking book on ancient DNA observed that the DNA of early hunter-gatherers on the steppes included an influence from the other side of the Caucasus Mountains, from a genetic group which lived in the South Caucasus and also further south. This had the implication that the horse hunters who evolved into the early Indo-European (the 'Yamnaya' culture, referring to their pit-grave interments) were partly the product of an earlier migration from the South. This migration preceded the adoption of farming. We are talking about perhaps 5000 BC. "Anatolia is remarkable for its lack of steppe ancestry down to the Bronze Age. The ancestry of the Yamnaya was, by contrast, only partly local; half of it was West Asian, from both the Caucasus and the more southern Anatolian-Levantine continuum. Migration into the steppe started by about 7000 years ago, making the later expansion of the Yamnaya into the Caucasus a return to the homeland of about half their ancestors."

The new work follows up Reich’s work, or rather the work of the world-leading DNA lab which he heads, to say that the DNA of remains located in the area where we know the Hittites lived is different from the DNA of steppe Indo-European areas and so also of the DNA of parts of Bronze Age Europe which were invaded (! or at least settled) by people from the steppes who putatively spoke Indo-European. It follows that the Hittites, Luvians, and related groups came into Anatolia from the north-east, and not via a tortuous migration along regions to the north of the Black Sea and through the Balkans. Their history is separate from that of all other Indo-European groups. Anatolian entered Anatolia from the north-east (or conceivably had been spoken south of the Caucasus since very ancient times).

There is a very interesting paper by Craig Melchert (“western affinities of Anatolian”), following up a 1994 paper by Jaan Puhvel, which traces matches between Hittite and specific other languages which are not matches with the reconstructed Indo-European lexicon. That suggested a shared (and late) geographical history which the new work puts seriously in question. He was thinking of convergence in a shared contact zone after the migrations. This data and the pattern which it supports are now of great interest, but Melchert only saw a tentative pattern in it. “Puhvel (1994) argued for Anatolian as a western dialect sharing features with Italic, Celtic, and Germanic (plus or minus Greek and Baltic). However, his paper was both initially and subsequently universally (but wrongly) ignored.”

Since several other IE languages are known from the area of Anatolia, and points east, one has to ask if it is the Anatolian group only which missed out on a long trip around the Black Sea to end up in that region. Armenian is certainly a candidate, perhaps also Phrygian (a “rubble language”). The Science paper describes the Armenians as the product of a migration from the steppes into Anatolia. I don’t think anyone is going to propose Iranian as such a candidate, which only geographical logic (not linguistic) would suggest.

The earliest written records of Hittite are quite far south, in Kanesh, but this is an artefact of the way in which writing reached Anatolia, evidently from the south-east and originally in the Akkadian language and script, having nothing to say about where the predecessor forms of the Hittite language had been spoken or what migration routes their ancestors followed. We now have the possibility of dating the split between ancestral Anatolian and the other Indo-European stem, from archaeological data. So this may be as early as 5000 BC. The whole history of Indo-European studies has assumed that there was a nuclear area from which Indo-European spread into territories speaking (fundamentally) different languages. But what we now know about the Anatolian branch makes it possible that the area around the Black Sea was populated at least in part by peoples speaking languages related to Indo-European, as distantly as Hittite and Luvian, and even that this facilitated the rapid spread of Indo-European. It is a puzzle that all IE languages lost the laryngeals and yet they still existed just before the break-up. Substrate influences may explain this, at least speculatively.

It remains possible that Greek came to Greece from the east, along the southern shores of the Black Sea (and initially through the Caucasus?), but this has always been a minority view and the genetic data now make it unlikely. It was not the view of the ancient Greeks and they had tales of the settlements in Ionia being founded from what we now think of as Greece.