Monday, May 7, 2018

Did Hittites and Other PIE Anatolian Language Speakers Lack Steppe Ancestry?

Iosif Lazaridis, in the pre-print "The evolutionary history of human populations in Europe" (May 4, 2018) casts doubt on the Steppe hypothesis as applied to the Anatolian language family within the Indo-European languages. While I understand the basis for his speculation, I think that this possibility is improbable and that late Bronze Age samples from Hittite individuals will show steppe ancestry.

The most prominent of the languages in this family was the Hittite language of the Bronze Age Hittite people who conquered essentially all of Anatolia and some of the Northern Levant and Northern Mesopotamia in the Bronze Age before collapsing in the Bronze Age collapse.

Contemporaneous historical accounts from Akkadian traders suggest that this conquest started shortly before 1740 BCE. Advocates for an early arrival of Anatolian language speakers (on linguistic grounds) argue that these languages should have been pervasive in Anatolia for more than a thousand years before this expansion, despite evidence that the territory conquered by the Hittites mostly spoke a non-Indo-European Hattic language.

But, the archaeology does not support this conclusion and there are not early historical accounts to support this claim either. Physical anthropology of remains from the region also point to a late arrival of physically distinct people in Anatolian linguistic communities.

The only other pre-Hittite empire Anatolian language which is attested is Luwian, a sister language to Hittite possibly spoken by the Trojans. The other Anatolian languages are only attested after the collapse of the Hittite empire and could have a relationship to it similar to that of Latin to the Romance languages.

They key passage states:
Steppe migrants Steppe populations during the Eneolithic to Bronze Age were a mix of at least two elements[28], the EHG who lived in eastern Europe ~8kya and a southern population element related to present-day Armenians[28], and ancient Caucasus hunter-gatherers[22], and farmers from Iran[24]. Steppe migrants made a massive impact in Central and Northern Europe post5kya[28,43]. Some of them expanded eastward, founding the Afanasievo culture[43] and also eventually reached India[24]. 
These expansions are probable vectors for the spread of Late Proto-Indo-European[44] languages from eastern Europe into both mainland Europe and parts of Asia, but the lack of steppe ancestry in the few known samples from Bronze Age Anatolia[45] raises the possibility that the steppe was not the ultimate origin of Proto-Indo-European (PIE), the common ancestral language of Anatolian speakers, Tocharians, and Late Proto-Indo Europeans. In the next few years this lingering mystery will be solved: either Anatolian speakers will be shown to possess steppe-related ancestry absent in earlier Anatolians (largely proving the steppe PIE hypothesis), or they will not (largely falsifying it, and pointing to a Near Eastern PIE homeland).
Note 45 is a reference to Lazaridis I, et al.: "Genetic origins of the Minoans and Mycenaeans." 584 Nature 214 (2017), which contains three Bronze Age Anatolian samples from Southwest Turkey. The one male and two female Bronze Age Anatolian samples relied upon are as follows (from the Supplemental Materials):

(3925±35 BP,

(4040±35 BP,

The man has Y-DNA J1a, mtDNA H and no steppe ancestry and is from 2558-2295 BCE. One woman is from 2836-2472 BCE, has mtDNA K1a2 and has no steppe ancestry. The other woman is from 2500-1800 BCE, has mtDNA T2b and has no steppe ancestry.

We don't have heaps of information about the archaeological context, but two of the three samples clearly pre-date the Hittite expansion that is attested historically and archaeologically in Anatolia and the third, a woman, could pre-date or post-date that expansion which in every other case was male dominated.

This data is a blow to the theory that the Anatolian languages broke away especially early form the Indo-European languages and were present in Anatolia long before 2000 BCE. 

But, it is not at all in conflict with my longstanding hypothesis that the Anatolian languages arrived in Anatolia with steppe people at a time similar to their arrival in Europe, the Aegean, South Asia and West Asia (i.e. Iran) and that the divergence of the Anatolian languages from other PIE language families is due to stronger and more divergent substrate influences rather than from great antiquity.

If my hypothesis is right, these data are a perfect fit and steppe ancestry should start to become widespread especially in specifically Anatolian language speaking men, and more generally, in late Bronze Age Anatolia. It will take future ancient DNA finds from those periods to determine if this is correct.

I also disagree with his conclusion that an absence of steppe ancestry would imply a Near Eastern PIE homeland. Instead, it would imply that the Anatolian languages arrived in Anatolia largely via elite transmission which could also help explain why it is more divergent than Indo-European languages that arrived with large communities of native speakers via mass migration.

UPDATE May 14, 2018: Davidski does some analysis of the Bronze Age Anatolian genomes discussed above, and agrees with me that these are likely Hattian and not Hittite individuals, albeit for slightly different reasons. He also notes, in data completely consistent with my late Anatolian expansion hypothesis that: "I am actually seeing a minor, but persistent, signal of steppe ancestry in one of the two Old-Hittite Period (~1750–1500 BCE) samples: Anatolia_MLBA MA2203." Also, not inconsistent with the historical record or a late Anatolian expansion hypothesis is his observation that: "None of the other Anatolia_MLBA individuals, three of whom are from the Assyrian Colony Period (~2000–1750 BCE), show such obvious steppe ancestry."


bellbeakerblogger said...

I'm a bit surprised at the arguments to relocate the PIE urheimat based on a few random souls lacking EHG. The thing that sticks out most is the fact that PIE essentially requires two components that make intense contact with each over a reasonably large time and space.
The genetic formulation of steppe people gives a very satisfying explanation for the identity and formation their language.

bellbeakerblogger said...

Agree with you totally.

I wanted to add also that the Hattians are definitely not IE. The Hittites may have genetically no more IE than the average Peruvian.

Jaap said...

'The most prominent of the languages in this family was the Hittite language of the Bronze Age Hittite people who conquered essentially all of Anatolia and some of the Northern Levant and Northern Mesopotamia in the Bronze Age before collapsing in the Bronze Age collapse.'
Aren't you confusing language and political control here? The Hittites may well have come as elite guests (coveted brides), to take over within a few generations with the support of a Hattian clacque. They could be from Iran or Caucasus (Majkop), and may have been EHG-less altogether. Hittite-speakers were never very numerous, as later Luwian-speakers obviously were.
Couldn't Hittite be an uncle to PIE rather than off-spring? I don't know, it's too big for me. But Majkop for me isn't out as IE-homeland. Spoken language tends to be female-mediated (as opposed to liturgic language, and/or story-telling language).
I totally agree to your last point.

andrew said...


Language and political control aren't the same thing, but about 500 years passed from Hittite expansion to the collapse of the Hittite empire, and somewhere along the line Hattic had become a purely liturgical language.

The timing for Hittite to be a PIE uncle isn't good and it is awfully divergent to hold that spot.

I'm agnostic about gender and language transmission.

Maijkop is a likely source of at least some of the cultural and technological package of the PIE people, and there does seem to be female mediated introgression into steppe cultures which were EHG first. So, it isn't a stretch to think that it could have had a linguistic influence as well. But, there isn't much solid to support that hypothesis either.

Palacista said...

It seems strange to say that a pre IE population in Anatolia having no EHG component
has anything to do with IE origins.

andrew said...

Lazaridis is operating on the assumption, popular in linguistic circles, that Anatolian languages were widespread in Anatolia from something like 4000-3000 BCE, which is reached based upon the great divergence linguistically of the Anatolian IE languages from the other IE languages (except possibly Tocharian).

A lack of an EHG component in Anatolia is inconsistent with that hypothesis.

I think that hypothesis is wrong and that almost everyone in Anatolia prior to ca. 2000 BCE was pre-IE and not IE, in which case it isn't surprising to see a lack of EHG in Anatolia.

If Anatolian IE languages were widely spoken in Anatolia ca. 4000-3000 BCE and these individuals were IE individuals, however, it would have potential implications for IE origins, although an elite dominance language shift to Anatolian IE languages is a much easier explanation than a Near East IE origin theory, IMHO.