Monday, July 20, 2015

Progress In Understanding Kalash Genetics And Its Linguistic Implications

Eurogenes reports progress in understanding Kalash genetics (from his own analysis of public datasets including newly available ancient genomes).
A couple of years ago Moorjani et al. concluded that present-day Georgians of the Transcaucasus were the best available proxy for the ancient West Eurasian population that mixed into the South Asian gene pool. This was a solid statistical fit. . . . But it was also a big fat coincidence . . . . 
Thus, the Indo-Iranian and hence Indo-European speaking Kalash no longer looks very similar to the Kartvelian speaking Georgian. In fact, [the Kartvelian speaking Georgian] appears to be most closely related to the supposedly Indo-European speaking Afanasievo and Yamnaya nomads of the Early Bronze Age Eurasian steppe. The rest of his ancestry is probably best described as South Central Asian, which is an unknown quantity to me at this stage, but probably in large part of indigenous South Asian origin (see here).

I'm only able to show this thanks to the ancient samples that are on the tree, for which, as far as I know, there aren't any useful substitutes among present-day populations. Obviously, Moorjani et al. didn't have this luxury, so they ended up with a model that was statistically sound, but didn't make much sense otherwise, especially in terms of linguistics.
My . . . model is easily reproducible with most of the other South Asian samples from the Human Origins, and it gels nicely with uniparental marker data too. For instance . . . not only do Pathans cluster among these ancients from the Eurasian steppe, but most of them also carry the same Y-chromosome haplogroup: R1a-Z93, which is derived from R1a-M417, and in all likelihood first expanded in a big way with the Proto-Indo-Iranians of the Trans-Ural steppe.
In another his own posts, linked above in the block quote, the key conclusion is that:
One of the toughest nuts to crack in population genetics has proved to be the story of the people of the Hindu Kush. However, using Treemix and ancient genomes from the recent Allentoft et al. and Haak et al. papers, I'm seeing most of the Kalash and Pathan individuals from the HGDP modeled as ~65% Late Neolithic/Early Bronze Age (LN/EBA) European and ~35% Central Asian. . . . [T]he Kalash and Pathans come out ~65% LNE/EBA European (which includes substantial Caucasus or Caucasus-related ancestry), ~12% ASI, and ~23% something as yet undefined. If I had to guess, I'd say the mystery ~23% was Neolithic admixture from what is now Iran. But ancient DNA has thrown plenty of curve balls at us already, so that's a low confidence prediction, even though it does make good sense.
Kalash Y-DNA is about 45% West Eurasian, and the percentage of Kalash mtDNA that is potentially West Eurasian in character (a somewhat less definitive geographic attribution than for the available Y-DNA data) is about 43%.  The absence of a strong gender imbalance in uniparental markers is notable.  Also, LN/EBA Europeans, themselves, aren't necessarily purely European and have a significant indigenous steppe component.  So those percentages aren't necessarily inconsistent and given the small effective size of the Kalash population, genetic drift and founder effects are also to be expected.

It has long been recognized that the Kalash may look like a high level branch of the population genetic history of non-African modern humans, when in fact, they are merely an admixed population that has been isolated and inbred for a sufficiently long time to look like something unique. (The alternative view that the Kalash were isolate for 11,800 years that was expressed in Qasim Ayub, et al.  (2015) is a completely implausible interpretation of the genetic data that they examined in what was otherwise a useful paper.)  But, this analysis is starting to finally establish precisely what is happened to form this genetically distinctive people of the Hindu Kush with more than guess work.

This puts the oldest possible point of Kalash ethnogenesis at about 4500 years ago, a few centuries before the earliest archaeological evidence (Cemetery H), of Indo-European appearance in South Asia, but after the replacement of the Afanasievo culture with a genetically distinct successor culture in the Central Asian steppe around 2500 BCE to 2000 BCE.  The Yamna culture is contemporaneous with it and more or less contiguous to the west of the Afanasievo culture.

The collapse of these cultures in favor of Y-DNA genetically distinct cultures that are otherwise quite similar to their north (who are the proto-Indo-Iranians) as the northern cultures penetrate South Asia appears to be another remarkable untold story of prehistory.  The timing, however, strongly suggests that the 4.2 kiloyear climate event was almost surely an important cause of this sudden upset.

The boundaries on the possible youth of Kalash ethnogenesis aren't quite as specific, but the fact that their Dardic language is very basal within the Indo-Aryan languages (or alternately, its status as a fourth basal branch of Indo-Iranian) suggests that the earliest possible date is an appropriate place to expect to find Kalash ethnogenesis.  Asko Parpola suggests in a 1999 scholarly anthology that the Dardic languages broke off from proto-Rig Vedic Sanskrit around 1700 BCE based upon Rig Vedic linguistic features found in Dardic languages and absent in other Indo-Aryan languages.

While Eurogenes understates the point a bit, I will underline it:

The long standing hypothesis that the Y-DNA R1b dominated Afanasievo and Yamnaya peoples were linguistically Indo-European is increasingly ill supported.  The Afanasievo and Yamnaya peoples had closer ties to their Caucasian neighbors than their definitely linguistically Indo-European neighbors to the North of them.

This also tends to support my hypothesis that heavily Y-DNA R1b people of Western Europe were probably part of the same Vasconic language family as the modern Basque until around the time of Bronze Age collapse, when there was a mass language shift to Germanic, Celtic and Italic language, respectively, in Western Europe, with only a fairly modest population genetic impact.

And, it also supports the argument that the Vasconic languages are distant relatives (at a time depth of about 4000-5000 years ago) of languages spoken in the highlands of the Caucasus Mountains, Iran and/or Anatolia, perhaps with a strong Atlantic Megalithic linguistic substrate, whose closest surviving relatives are one or more of the modern languages of the Caucasus mountains.

I remain agnostic regarding which of those languages are the closest relative and it could be that the proto-Vasconic languages were a sister language family to all of them.


Nirjhar007 said...

// a few centuries before the earliest archaeological evidence (Cemetery H), of Indo-European appearance in South Asia//
Cemetery H is a local derivative nothing else with some new innovations, please read things scientifically before concluding.
//The long standing hypothesis that the Y-DNA R1b dominated Afanasievo and Yamnaya peoples were linguistically Indo-European is increasingly ill supported//
Afanasevo is R1b dominated? do we yet have any Y-DNA from there yet? the more likely Scenario is that Afanasevo will be R1a Dominated.
BTW where can i find tour new hypothesis on PIE?.

Nirjhar007 said...


andrew said...

Re Cemetery H, one of the real distinguishing factors here is the litmus test of cremation which appears almost everywhere at just around the moment that Indo-Europeans start to arrive.

See, e.g., R1b and also the identity of the two contemporaneous cultures genetically using autosomal DNA. I believe that there were was direct typing in one of the two big ancient DNA studies from Nature.

I've written a lot about PIE and my hypothesis isn't really new. One of the better ways to examine it is to click on the "language" tag on this post and to skim through the posts with that tag that are relevant.

Nirjhar007 said...

''Re Cemetery H, one of the real distinguishing factors here is the litmus test of cremation which appears almost everywhere at just around the moment that Indo-Europeans start to arrive.''
It is a local development with no trail from outside its people also show similar biological affinity and ''About cremation, in Harappan sites there are often pots with ashes and bones, but rarely human. A work on Kalibangan's cemetery observes that the tombs were too few for all the population, so cremation could be a way of managing the dead, but there were also fractional burials after exposure. Clear cremation is present in Cemetery H at Harappa, and it can be connected with a different ideology, as in Europe and the Near East. It is thought that cremation is connected with a more spiritual view, where the body has no more importance and the soul has to leave it. It is possible that it is connected with the belief in reincarnation, but not only, in Homeric Greece it was believed that cremation made possible that the soul enters the Hades. But in various cultures inhumation and cremation are found together, for instance the Etruscans introduced again the inhumation burials only for the elite, while the common people continued cremation as in the Villanovan period. Also in Hittite Anatolia inhumation and cremation are found sometimes in the same cemeteries, and also Greeks and Romans practiced both. See here:
In Rigveda it is also mentioned as you will see in that Wiki that Both Cremation and Burial were likely practiced by the Aryans so its most likely a cultural thing nothing much for any invisible migration.
David recently Posted on Afanasevo and CWC with R1a and in most probable scenario Aafanasevo will be R1a Dominant.
Okay but i will need more time...

andrew said...

Cremation appears in Greece with the Mycenians, in the Balkans with the first Indo-Europeans there, in Anatolia with the Hittites, in France with the Indo-European Urnfielders, in South Asia with the Aryans, and in Iran with the Indo-Iranians. Yes, cremation was a cultural thing, but it was in particular an Indo-European specific cultural thing. This is a practice that the Indo-Europeans brought to the party that was not practiced by the Harappans. The Rig Veda and the Cemetery H mix indicate that it was present at the moment of transition. The Etruscans were a pilot wave, who were one of the last non-Indo-European cultures to survive in Europe because they emulated key Indo-European practices.

Cemetery H is also coincident with the arrival of iron goods, just as in the case of the Hittites.

Nirjhar007 said...

Yes they appear but for example not in India from outside as i said its a local development with a change in the philosophy although traces of Cremation in Harappa is there as i mentioned, Avestans practiced open burials or Sky Burials if i'm not wrong and even the Sintashta had Burial customs, yes Rigveda is mostly of 2000 bc-1500 bc period and the Practice of Cremating is not of PIE origin and it developed later roughly in similar time frame and it is related mostly to Philosophy not migration.
I don't know but Indian Iron age likely started in South India around 2200 bc as per latest discoveries where we also find most of the Iron Deposits and another thing is that we don't find Iron Mentioned in Rigveda but only in later texts, the term Ayas means simply metal or something strong.

andrew said...

Avestans were Indo-Iranian and it is fair to say that Sky Burial is a piece with cremation philosophically.

How can you know that it is a local development with a local change in philosophy when it happens throughout the IE world, a roughly similar time frames, and nowhere the IE has not yet arrived (for example, Western Europe) until IE arrives. You need evidence to support that position.

Nirjhar007 said...

Inhumation Practices were common earlier but later Cremation became popular and dominant like around 1500 BC Greeks preferred inhumation but some centuries after it changed,Phoenicians practiced both cremation and burial. Roman philosopher Cicero considered Inhumation to be an archaic and probably original though in his time Cremation was very dominant. Yes It happened roughly in similar time frame and in some manner it was when the IE spread happened in various areas and it was already the dominant practice by that time.

Nirjhar007 said...

BTW Andrew I have something to ask you, Which is the oldest example of Cremation in Europe is it in the Carpathian Basin around 2000 BC?.

andrew said...

I would have said the Balkans around 2000 BCE, but that geographic description is not very specific. I would have took look to references to determine precisely where in SE Europe the oldest cremations were located.

Nirjhar007 said...

So we have Cremation Appearing Both In Europe and India around 2000 BC as earliest examples.
BTW Regarding which seems now Confirmed On Yamnaya R1b-
Those Yamnaya Nomads who came from Somewhere around Caucasus, is it logical to propose that Yamnaya Spoke a Caucasus related language rather than Indo-European?.

andrew said...

Yes, it is. Indeed, I currently believe (although I can't prove definitively) that the Yamnaya spoke a non-Indo-European language, probably an ergative one.

I also believe, but can't prove, that all of the ergative non-Indo-European languages of the Caucasus, Anatolia, and the highlands of Iran, as well as Sumerian, Minoan and Basque, are probably part of the same macro-language family at a time depth not older than the late Mesolithic era.

I won't venture to say precisely which Caucasian language family would be most similar, or that Yamnaya was definitively the linguistic source for the Vasconic languages, although I don't rule out that possibility.

I'm inclined to think that Caucasus to steppe language transmission is more likely, but can't rule out the reverse.

I am thoroughly agnostic on the nature of the Harappan language, and on the nature of the first farmer LBK/Cardial Pottery languages and their relationships to each other.

I am inclined to think the Etruscan language (and a few other non-Indo-European Aegean languages in the same family) are not related to Yamnaya and Vasconic and Caucasian languages, but not every confident of that conclusion.