Ancient genomes from Caucasus inc. Maykop

[IronSide]

I see. Do you know if there is an (apparent) common pre-IE substrate in all European branches of IE? I think there were some hypothesis, including the "Vasconic" one, but most of them were discredited by most linguists, and AFAIK the mainstream position has been that there is no demonstrated common thread between the non-IE substrates in all of Europe... I remember reading that there were some very similar non-IE borrowings in Germanic and Greek, but they were also not particularly similar to Basque. These results are the main reason why I think it is perfectly plausible that the Anatolian Neolithic spread to Europe could have involved two or more linguistically differentiated populations (we already know, at least, that they took two totally different routes of expansion, maybe that indicated also a different tribal/cultural identity?), or maybe the linguistic evolution was very rapid and intense, creating highly diverged subfamilies, in the 3,000-3,500 years before the IEs absorbed the EEF.

Well, I'm no linguist, but I noticed that "gods" in Etruscan means "aisar", reminded me of the æsir of Norse mythology.

The other theory on Etruscan origins is a migration from Anatolia in the Iron Age, but Etruscan territory has a striking one to one correspondence to the previous Villanovan culture, which theoritically came from the north, no major disruption is found that would support the IA Anatolian model, so it does seem Etruscan descends from a farmers language, since its not IE.

 

[CrazyDonkey]

Razib Khan's inconclusive conclusions (my emphases):

It is hard to deny that the steppe migrations between 4 and 5 thousand years ago had something to do with the distribution of modern Indo-European languages. But, it is harder to falsify the model that there were earlier Indo-European migrations, perhaps out of the Near East, that preceded these.

So, we are left with the Steppe Hypothesis being still probable ("hard to deny") and what might be called the Pre-Steppe Hypothesis being not impossible (harder to falsify). The only thing that is quite certain is that languages, cultures, and genes are muddy affairs.

To what degree are they pussyfooting around what can only be called "nationalistic" sensitivities?

 

[ToBeOrNotToBe]

The ancient presence of CHG north of the Caucasus does indeed complicate the pattern. I wish that team of geneticists would conduct a similar scrutiny of ancient genomes from Northern Mesopotamia and the Lake Van area.
Their kind of rather impermeable genetic frontier between Maikop and Steppe runs counter to an R1b link between NW Iran and the Steppe. What of that R1b guy in Hajji Firuz ? A stray lone wanderer ?

I’m not commenting specifically on this(!), but don’t discount the idea of social hierarchies in cultures actually involving genetic divisions between the classes. People always seem to assume that these cultures were uniform within their hierarchies - that DEFINITELY will not be the case considering the nature of many of these societies (e.g. invaders as a ruling elite, and perhaps in "the early days" where societies relied much more on intensive subsistence farming, assimilation by e.g. metallurgists, who will naturally be of a high social standing), and in many examples I expect these differences to be pronounced. Another thing to consider is that if burials are involved, you’ll be sampling more from the higher standing individuals of that culture.

We all accept that there would have been an extremely strong caste system in the midst of the Indo-Aryan invasion of South Asia, so I think the community as a whole should be more open to these kind of ideas.

 

[A. Papadimitriou]

I see. Do you know if there is an (apparent) common pre-IE substrate in all European branches of IE? I think there were some hypothesis, including the "Vasconic" one, but most of them were discredited by most linguists, and AFAIK the mainstream position has been that there is no demonstrated common thread between the non-IE substrates in all of Europe... I remember reading that there were some very similar non-IE borrowings in Germanic and Greek, but they were also not particularly similar to Basque. These results are the main reason why I think it is perfectly plausible that the Anatolian Neolithic spread to Europe could have involved two or more linguistically differentiated populations (we already know, at least, that they took two totally different routes of expansion, maybe that indicated also a different tribal/cultural identity?), or maybe the linguistic evolution was very rapid and intense, creating highly diverged subfamilies, in the 3,000-3,500 years before the IEs absorbed the EEF.

I don't want to state what I believe exactly without more data but if we assume that the following map is correct (I don't know if it is), maybe related languages expanded from Anatolia to Thessaly to Danube to Northern Europe (incl. Danemark, S. Sweden), maybe also to S/SW Spain and distinct languages expanded to Italy, S. France, E Spain, Slovenia (?)

 

[Saetrus]

The haplogroups in this set of Chalcolithic/Bronze Age South Caucasus correspond to the haplogroups that I have associated with the Kura-Araxes expansion (J2a, J1-Z1828, L1b, T1a-P77 and G2a-L293) except that they didn't find any T1a among those 12 samples (but it's not surprisingly considering the low frequency of this haplogroup in any region today).

What time frame and cultures would you associate with the arrival of south Caucasus people in Italy?

 

[CrazyDonkey]

I’m not commenting specifically on this(!), but don’t discount the idea of social hierarchies in cultures actually involving genetic divisions between the classes. People always seem to assume that these cultures were uniform within their hierarchies - that DEFINITELY will not be the case considering the nature of many of these societies (e.g. invaders as a ruling elite, and perhaps in "the early days" where societies relied much more on intensive subsistence farming, assimilation by e.g. metallurgists, who will naturally be of a high social standing), and in many examples I expect these differences to be pronounced. Another thing to consider is that if burials are involved, you’ll be sampling more from the higher standing individuals of that culture.

We all accept that there would have been an extremely strong caste system in the midst of the Indo-Aryan invasion of South Asia, so I think the community as a whole should be more open to these kind of ideas.

The traditional IE tri-partate division: Jarls, Karls, and Thralls. The Scandinavians attributed phenotypic, not just status, differences to them.

 

[holderlin]

if there was still bridal exchange after Anatolian farmers had arrived south of Caucasus, then there also should have been autosomal Anatolian farmer in the steppe, quod non

Yeah I don't know how people are missing this.

Bronze Age steppe is Eneolithic steppe + GAC. If any "pure CHG" was flowing into the steppe to form the steppe eneolithic populations it was long before the Caucasian samples shown in this paper, and actually it looks like its been there since MA-1. So there may be nothing "Caucasian" about it.

We should call it "Siberian Teal"

 

[holderlin]

Their Eneolithic steppe sample in Admixture is far more than 50% CHG. In Samara Eneolithic, the CHG gets cut down a bit by WHG and, what, East Asian? Still, more than 50%.

CHG was on the steppe very early indeed. Perhaps that's why they put in all that language about pre-existing clines.

Look at AG3 and MA-1. It's not that CHG came into the steppe early, it's that its been there since MA-1.

 

[holderlin]

This is one thing that is bothering me in this whole "Transcaucasia vs. Steppes" discussion. I mean, if the bulk of the CHG in the Pontic-Caspian steppe populations is very ancient, having being absorbed in high proportions before 5000 BC (the lack of ANF in the steppes also indicate that early introgression), then it is, from a linguistic point of view, very hard to accept the possibility that Early PIE was a "CHG language" from the southern slopes of the Caucasus, because that would imply that Anatolian IE (or also possibly, according to this study, even much less diverged languages like Armenian and especially Greek) and the Pontic-Caspian steppe IE would've split in the 6th millennium BC. Either scientists assumed a much faster pace of linguistic evolution, or it is just impossible that the initial split of PIE happened so early.

Now, if we could demonstrate that there was a significant amount of extra CHG in the Chalcolithic Pontic-Caspian region, that could suggest a new demographic and presumably linguistic layer onto the older ethnic makeup.

As for the lack of EHG in those few "Hittite" (or at least "near to the Hittite") samples, I was thinking (okay, speculating) a bit about it from the assumption (pretty much mainstream among linguists) that Anatolian IE split much earlier than the others and probably in a very different historic context (certainly not the mobile horse-driven pastoralism of Yamnaya and descendants). It also seems from this paper that the North Caucasus, right next to the steppes, had a very different genetic structure with a much higher CHG and ANF, so I'd think it is plausible that the southernmost portions of Sredny-Stog and/or Khvalynsk in direct contact with the North Caucasus could have some substructure in a transition zone to the steppes. We know now that Maykop had EHG, which is not found further to the south, so there was some cline. If the ancestors Hittites came from this region and expanded more or less in the fashion of later IE branches into regions that were already very populated (like Greece and South Asia), then they could've migrated south becoming a relevant and dominant minority with an increasingly diluted DNA makeup, and given their very early separation from the steppe or North Caucasus populations it's probable that by the time they established in former Hattic-speaking lands to form their kingdom and empire their EHG portion was just too small to make a significant presence in the genomic makeup of the region's average inhabitant. Mere genetic drift and regional substructure could make EHG virtually invisible after a few centuries unless we had many more samples. Doesn't this type of thing happen when migrations were not that powerful to trigger an appreciable population replacement? Just playing a bit with this speculation, let's imagine this totally hypothetical (and admittedly baseless for now) scenario:

From CHG-enriched southern steppes just north of the North Caucasus (South_Steppe): 40% EHG, 55% CHG, 5% ANF
Admixture with (north or south?) Caucasians during the Proto-Anatolian phase: 25% South_Steppe + 75% Local Caucasians (5% EHG, 70% CHG, 25% ANF) >>> 13% EHG, 66% CHG, 8% ANF.
Admixture with North-Central Anatolians during the Hittite phase: 20% Proto-Anatolian + 80% Hatti and other natives (0% EHG, 30% CHG, 70% ANF) >>> 2.5% EHG, 37% CHG, 60.5% (2.5% EHG - and that's assuming that the Caucasians already had some EHG even before Maykop and that the Proto-Anatolians still made a reasonable demographic impact of ~25% and the Hittites one of ~20%, not too shabby)

I'm just entertaining all the possibilities, especially considering that apparently the CHG component in the steppes is MUCH older than even the 1st Indo-European split around ~4000 BC, and not just some Late Neolithic/Chalcolithic influx bringing not just a new (presumably more advanced) people but potentially a new language family.

Yeah I don't know what all these people are celebrating. The CHG in steppe eneolithic could not have come from the groups that everyone is trying to say spoke PIE. wtf

*sigh*

 

[ToBeOrNotToBe]

I don't want to state what I believe exactly without more data but if we assume that the following map is correct (I don't know if it is), maybe related languages expanded from Anatolia to Thessaly to Danube to Northern Europe (incl. Danemark, S. Sweden), maybe also to S/SW Spain and distinct languages expanded to Italy, S. France, E Spain, Slovenia (?)

But isn’t the Neolithic expansion in Europe mediated by a cline of G2a farmers with I2 hunter-gatherers, originally keeping to themselves but inevitably mixing over time?

The keys to this, in my opinion, are to try and track L51 and Z2103, but also to note the clear relationship between the Balkan and Mesopotamian Chalcolithic cultures (The oldest examples of Swastikas and metallurgy are found here) - based on modern distributions, you’d expect this to be mediated by Z2103.

What I don’t understand is if Z2103 went on such a rampage from the Steppe, why was the post-CW IE expansion mostly mediated by L51? If Z2103 and L51 split from L23 on the Steppe (unlikely to me), what explains that Copper Age Z2103 from NW Iran?

 

[Ownstyler]

I'm not sure how the expansion of the Greek language is impacted by these findings. Greeks have both Neolithic farmer (south of Caucasus) and Bronze Age Steppe ancestry at substantial levels, so in my opinion genetic data does not change much for linguistics.

 

[Ygorcs]

I’m not commenting specifically on this(!), but don’t discount the idea of social hierarchies in cultures actually involving genetic divisions between the classes. People always seem to assume that these cultures were uniform within their hierarchies - that DEFINITELY will not be the case considering the nature of many of these societies (e.g. invaders as a ruling elite, and perhaps in "the early days" where societies relied much more on intensive subsistence farming, assimilation by e.g. metallurgists, who will naturally be of a high social standing), and in many examples I expect these differences to be pronounced. Another thing to consider is that if burials are involved, you’ll be sampling more from the higher standing individuals of that culture.

We all accept that there would have been an extremely strong caste system in the midst of the Indo-Aryan invasion of South Asia, so I think the community as a whole should be more open to these kind of ideas.

Well, the effects of the strong and longlasting South Asian caste system can actually be very easily seen in the samples from within the very same population, because there is a substantial genetic structure even in the same place and cultural milieu. As Razib Khan recently pointed out, some of the samples from the very same region look like they came from totally different and distant regions. On the contrary, these steppe samples, as early as the Eneolithic, look heavily EHG-CHG mixed (proportions vary, but the trend is the same) and without a very starkly different autosomal makeup denoting a strongly enforced endogamy and significant internal social hierarchy based on distinct genetic/ethnic origins. At least I can see nothing suggesting that in these data.

 

[Ygorcs]

I don't want to state what I believe exactly without more data but if we assume that the following map is correct (I don't know if it is), maybe related languages expanded from Anatolia to Thessaly to Danube to Northern Europe (incl. Danemark, S. Sweden), maybe also to S/SW Spain and distinct languages expanded to Italy, S. France, E Spain, Slovenia (?)

I also lean towards that position, though of course it probably can't be demonstrated. But it's plausible that such massive migrations didn't came exactly from just one place and the different patterns of expansion may have been correlated with different cultures/languages (and linguistic diversity in early Neolithic societies must've been really big if the example of Pre-Columbian Americas- essentially Neolithic or Chalcolithic-like - is any indication).

 

[Ygorcs]

About the presence of ANF-related ancestry in Yamnaya samples, it's important to notice that it probably did not come from the Caucasus populations bordering on the steppes, but from WHG-admixed ANF i.e. EEF in and near the westernmost steppe (I add: possibily via earlier Sredny-Stog influence, since there was one, some time ago, one published Dereivka sample with some 30% of EEF???).

The study says: "Importantly, our results show a subtle contribution of both Anatolian farmer-related ancestry and WHG-related ancestry (Fig.4; Supplementary Tables 13 and 14), which was likely contributed through Middle and Late Neolithic farming groups from adjacent regions in the West. A direct source of Anatolian farmer-related ancestry can be ruled out (Supplementary Table 15). At present, due to the limits of our resolution, we cannot identify a single best source population. However, geographically proximal and contemporaneous groups such as Globular Amphora and Eneolithic groups from the Black Sea area (Ukraine and Bulgaria), which represent all four distal sources (CHG, EHG, WHG, and Anatolian_Neolithic) are among the best supported candidates (Fig. 4; Supplementary Tables 13,14 and 15)."

To me the results of this study seem to point out that by the time of Maykop and even before Yamnaya (considering their Eneolithic steppe samples) the overall genetic makeup of the region, as far as Caucasian influence/influx is concerned, was already consolidated. So, if a southern source of CHG is also the source of PIE spoken in the steppes, that was an earlier Neolithic phenomenon - and then I have strong doubts that any split of a Common PIE (incuding pre-Proto-Anatolian) could've happened even before that immigration into the steppe, pushing the existence of PIE even further back.

 

[halfalp]

This is one thing that is bothering me in this whole "Transcaucasia vs. Steppes" discussion. I mean, if the bulk of the CHG in the Pontic-Caspian steppe populations is very ancient, having being absorbed in high proportions before 5000 BC (the lack of ANF in the steppes also indicate that early introgression), then it is, from a linguistic point of view, very hard to accept the possibility that Early PIE was a "CHG language" from the southern slopes of the Caucasus, because that would imply that Anatolian IE (or also possibly, according to this study, even much less diverged languages like Armenian and especially Greek) and the Pontic-Caspian steppe IE would've split in the 6th millennium BC. Either scientists assumed a much faster pace of linguistic evolution, or it is just impossible that the initial split of PIE happened so early.

Now, if we could demonstrate that there was a significant amount of extra CHG in the Chalcolithic Pontic-Caspian region, that could suggest a new demographic and presumably linguistic layer onto the older ethnic makeup.

As for the lack of EHG in those few "Hittite" (or at least "near to the Hittite") samples, I was thinking (okay, speculating) a bit about it from the assumption (pretty much mainstream among linguists) that Anatolian IE split much earlier than the others and probably in a very different historic context (certainly not the mobile horse-driven pastoralism of Yamnaya and descendants). It also seems from this paper that the North Caucasus, right next to the steppes, had a very different genetic structure with a much higher CHG and ANF, so I'd think it is plausible that the southernmost portions of Sredny-Stog and/or Khvalynsk in direct contact with the North Caucasus could have some substructure in a transition zone to the steppes. We know now that Maykop had EHG, which is not found further to the south, so there was some cline. If the ancestors Hittites came from this region and expanded more or less in the fashion of later IE branches into regions that were already very populated (like Greece and South Asia), then they could've migrated south becoming a relevant and dominant minority with an increasingly diluted DNA makeup, and given their very early separation from the steppe or North Caucasus populations it's probable that by the time they established in former Hattic-speaking lands to form their kingdom and empire their EHG portion was just too small to make a significant presence in the genomic makeup of the region's average inhabitant. Mere genetic drift and regional substructure could make EHG virtually invisible after a few centuries unless we had many more samples. Doesn't this type of thing happen when migrations were not that powerful to trigger an appreciable population replacement? Just playing a bit with this speculation, let's imagine this totally hypothetical (and admittedly baseless for now) scenario:

From CHG-enriched southern steppes just north of the North Caucasus (South_Steppe): 40% EHG, 55% CHG, 5% ANF
Admixture with (north or south?) Caucasians during the Proto-Anatolian phase: 25% South_Steppe + 75% Local Caucasians (5% EHG, 70% CHG, 25% ANF) >>> 13% EHG, 66% CHG, 8% ANF.
Admixture with North-Central Anatolians during the Hittite phase: 20% Proto-Anatolian + 80% Hatti and other natives (0% EHG, 30% CHG, 70% ANF) >>> 2.5% EHG, 37% CHG, 60.5% (2.5% EHG - and that's assuming that the Caucasians already had some EHG even before Maykop and that the Proto-Anatolians still made a reasonable demographic impact of ~25% and the Hittites one of ~20%, not too shabby)

I'm just entertaining all the possibilities, especially considering that apparently the CHG component in the steppes is MUCH older than even the 1st Indo-European split around ~4000 BC, and not just some Late Neolithic/Chalcolithic influx bringing not just a new (presumably more advanced) people but potentially a new language family.

Well they actually have done it... look at their graph, CWC have more CHG than EHG, Yamnaya have more CHG than EHG. Genetically, everything is going in they way they have meant to, meaning they have the result of those samples between a long time now and that they already have create a story around them. Thats a huge meli-melo, because EHG is ANE + WHG but CHG is Iran_Neolithic + something WHG / EHG. Meaning there is no way to say what is what, Satsurblia is older than Iran_Neolithic, so what's the ancestor of Iran_Neolithic ( that have ANE ) that contribute to the early CHG ? This is really a mind breaker at this point, CHG needs to be clearly defined without any Iran_Neolithic because it is technically older than Iran_Neolithic. The response of CHG can only come from paleolithic samples of eastern europe and iran / armenian plateau. And they have create that new ANF for refute the possibility than anatolian neolithic ancestry in steppe came from EEF, i mean i'm not a complotist really, but in this study everything mingle a little too much good, more than a hundreded yamnaya samples didn't have any anatolian_neolithic at all and now in that study it pops from everywhere, but it didn't came from europe, so frome where, the caucasus ? that study is not clear at all, where in all previous studies it let place to controversy, this one tries to underground the steppe one for good, with newly created genetic notions.

 

[halfalp]

Well, I'm no linguist, but I noticed that "gods" in Etruscan means "aisar", reminded me of the æsir of Norse mythology.

The other theory on Etruscan origins is a migration from Anatolia in the Iron Age, but Etruscan territory has a striking one to one correspondence to the previous Villanovan culture, which theoritically came from the north, no major disruption is found that would support the IA Anatolian model, so it does seem Etruscan descends from a farmers language, since its not IE.

Etruscan have a lot of obscure words that is related with obscure IE languages words. For exemple The Son is Clan like in Gaelic ( Irish ) the word Clann means The Childs. I'm pretty sur there is way more relative words that exists. There is a theory of the Nordwesternblock being related to a Rheatic language, itself related to Etruscan. And if i'm not wrong, etruscans were mostly y-dna J2a. If thats the case, we have two possibilities, an autochtonous origin going way back to neolithic europe and the LBK ( where substantial y-dna lineage were J2a ) wich would have borrowing with IE languages at the time of the Bell Beaker expansion from central europe to north-western europe. Or a way more younger origin in anatolia like the ancient texts says, somewhere around the bronze age collapse ( 1400BC ), maybe related to the antic Gasgas of northern anatolia ? But the fact that they borrow words related to obscure words in IE only found in continental proper europe goes more for a neolithic origin for the core of the etruscan language.

 

[halfalp]

Razib Khan is also grappling with the timing questions:

"The close relationship of Indo-Aryan and Iranian languages is obvious to any speaker of either of these languages (I can speak some Bengali). A divergence in the range of 4 to 5 thousand years before the present seems most likely to me. But the relationship of the other Indo-European languages is much less clear.
One of the arguments in Peter Bellwood’s First Farmers is that the Indo-European languages exhibit a “rake-like” topology with the exception of Indo-Iranian, which forms a clear clade. To him and others in his camp, this argues for deep divergences very early in time.
It is hard to deny that the steppe migrations between 4 and 5 thousand years ago had something to do with the distribution of modern Indo-European languages. But, it is harder to falsify the model that there were earlier Indo-European migrations, perhaps out of the Near East, that preceded these. Only a deeper understanding of linguistic evolution, and multidisciplinary analysis of regional substrates will generate the clarity we need."


On the relevance of steppe in Hittites:

"More interesting are the results in West Asia, and the linguistic supplement. In the authors note that tablets now indicate an Indo-Aryan presence in Syria ~1750 BC. Second, Assyrian merchants record Indo-European Hittite, or Nesili(the people of Nesa), as early as ~2500 BC.

"The main aspect I’d bring up with this is that in other areas steppe ancestry has spread deeply and widely into the population, including non-Indo-European ones. It is certainly possible that the sample is not needed enough to pick up the genuinely Hittite elite, but I probably lean to the likelihood that the steppe signal won’t be found. It seems that the Anatolian languages were already diversified by ~2000 BC, and perhaps earlier. Linguists have long suggested that they are the outgroup to other Indo-European languages, though this could just be a function of their isolation among highly settled and socially complex populations."

What would happenned autosomaly if a 50/50 CHG / EHG people would go in a 50/50 Anatolian_Neolithic / CHG population for the EHG signal ?

 

[halfalp]

Ok so i might have been completely wrong with the graph, but what is the green component because it pop as a majority in Iran_Neolithic ? And what is the red " Caucasus " component, because CHG is mainly green ? How i understand the graph is, in southern steppe, the CHG element become dominant over the EHG one and also a new component wich is the orange " Anatolian_Neolithic " pop up. But for what i see the orange component is only relevent for Maikop wich is not steppe but from south caucasus ?

 

[ToBeOrNotToBe]

Do these results support the hypothesis of R1b moving from Anatolia through the Caucasus and taking CHG women, and then further moving up and taking some EHG women, with the Kura-Araxes expansion resulting in mainly J2 taking the R1b women (yeah, I know, Y DNA is only found in men) from the stage of R1b taking women in the Caucasus?

Because that's what I think happened. It could be the other way round, with K-A forcing the R1b guys upwards instead of just expanding from the power vacuum left by R1b moving upwards.

I could also be completely wrong, but I'm just trying to model in two things I see as very likely: R1b coming from (Eastern) Anatolia, and the Kura-Araxes picking up R1b-like ancestry.

 

[halfalp]

Do these results support the hypothesis of R1b moving from Anatolia through the Caucasus and taking CHG women, and then further moving up and taking some EHG women, with the Kura-Araxes expansion resulting in mainly J2 taking the R1b women (yeah, I know, Y DNA is only found in men) from the stage of R1b taking women in the Caucasus?

Because that's what I think happened. It could be the other way round, with K-A forcing the R1b guys upwards instead of just expanding from the power vacuum left by R1b moving upwards.

I could also be completely wrong, but I'm just trying to model in two things I see as very likely: R1b coming from (Eastern) Anatolia, and the Kura-Araxes picking up R1b-like ancestry.

Probably not, that's a very complicate hypothesis you present here. I mean people have already difficult to assume that R1b have conquered most of western europe from eastern europe in millenia, so your hypothesis looks like a Master Race Rampage over the world. It also sound a little bit, correct me if im wrong, that you would like middle-eastern to have culturally and genetically conquer europe isn it ?