The genomic history of southeastern Europe-Mathiesen et al

[bicicleur 2]

In all models it is highly related to the Vestonice cluster. A sister clade. It is entirely possible that is spawns from it. As I said, it can be modeled as a mixture.



That is a matter terminology. Fu et al lobbed Loschbour in their Villabruna cluster



True. However, in formal stats WHG picks Anatolian over any other old Middle-Eastern.

I agree, Villabruna is a sister clade of El Miron, which got extinct when Villabruna expanded.
Paleolithic Europe was in part WHG.
In other words, WHG is a combination of paleo Europe with admixture Villabrunans picked up in their southeastern refuge.

 

[MarkoZ]

You are talking about <a href="https://www.nature.com/articles/ncomms9912/figures/2" target="_blank" rel="nofollow">this figure (fug. 2)</a> I presume? Cause that doesn't have an "<em>influence that unambiguously appears in Villabrunna</em>", that shows both split off a long time ago from each other. It does OTOH show exactly what I proposed, WHG (Bichon in this case) influx in EEF. However, the EEF they used is Stuttgart and that had extra WHG over Anatolians so it could be that. After the Anatolian paper came out people played with those and D-stats showed WHG affinity to those.

No, I mean the Fu paper. The admixture they detect is similar to Kotias-Satsurbalia.

Anatolians are chosen over Natufians as well.

That's not too surprising given that Natufians are at the extreme end of the WHG <-> Meso-Neolithic Levantine continuum in Lazaridis 2016 paper. The Natufians are estimated to have ~45% Basal Eurasian, whereas the Anatolians have roughly 25%, only marginally more than the EN_European cluster (~22%-23%). These seem to be the incipient stages of West Asian diversity, i. e. what differentiates present day Anatolians from either Iranians or Arabs.

>

 

[bicicleur 2]

It makes one wonder how far ANE stretched before LGM.

EDIT: It makes one realise how very few true unadmixted ANE samples we have. One 1 Y-DNA (R*) and 2 mtDNA's (U and R3)

I don't think it was in Europe before LGM.
IMO it was brought by Q from Siberia.
Khvalynsk had Q1a, and in this paper there was a Q1a2.

And I don't know whether this is correct :

Körös

Hungary

Vörs 52

5500 BC

C5

Guba 2011

MtDNA C5 in Hungary?

 

[A. Papadimitriou]

Notice that one of the two samples from the Balkans that dates to the time period for Mycenaean expansions is R1a. And remember that known historical Greek speakers extended North. I would like to see Shaft Grave samples that have no steppe, unfortunately I think we only have mtDNA.

What do you mean exactly?

 

[MOESAN]

the Cardial foreign type which took foot in Southern France (among HGs) and climbed up towards North along the Garonne and the Rhône was a bit different from the 'danubian' type come from/across Southern Balkans until Alsace, Parisian bassin and Normandy; not fully foreign, but different as an average : so even damned physical anthropology did not tell us too much bullshit, finally, helped by archeology -

 

[berun]

Don't forget the expansion of the Megalithic cultures in a good chunk of Western Europe from Portugal (linked to I2a). Reich's lab is not capable to find a trace of it, or as with the Catalan BB, have not so much interest to know.

 

[holderlin]

The steppe 'works' because there's ample opportunity for Kartvelian & Semitic (or Afrasian?) contacts - i. e. the only languages that had definite influence on PIE per historical linguistics. These would be hard to explain in the case of a Baltic homeland.

Anthony believes that these influences arrived north of the Caucasus with Maikop.

Yes I'm aware of these, but the only "definite" we have is loan words or theorized loanwords, which only prove contact. I notice you conveniently leave out Uralic as well, which of course is also disputed. Please don't try to Tr0ll me with "Afrasian". My views really have nothing to do with white supremacy, and the Semitic connection is the weakest. I don't know why you would include that in your point, much less among the "definite".

I understand the problems with the Baltic though. This is for the most part based on an archaeological argument, and fitting that to the larger picture. The Y-HG lines and the WHG-EHG cline certainly suggest long standing contacts from the Vistula to the Urals, then an interaction zone as the farmer expanded into the North Balkans, which also fits right into the archaeology.

 

[holderlin]

What do you mean exactly?

Macedonians.

 

[holderlin]

afaik the early Hittites didn't have charriots, the first ones in the area were the Mittani
the Hittites and the Egyptians learned about the charriots from the Mittani
they started a cold war and a war industry to produce charriots and swords, culminating in the Kadesh battle

IMO the Anatolians, when they arrived in Anatolia were simple herders, but with some social structure which finaly made them rule over or make alliances with other local farmer tribes. The Hittites were a federation with more or less independant allies. Some of them spoke even other non-IE (Hurrian) languages.

Yes, yes this is highly likely as well, but there is an early record of a siege (1700BC or around there) where Hittites speak of "teams of horses". So who knows.

 

[Angela]

Don't forget the expansion of the Megalithic cultures in a good chunk of Western Europe from Portugal (linked to I2a). Reich's lab is not capable to find a trace of it, or as with the Catalan BB, have not so much interest to know.

Could we cut out the conspiracy theory references, Berun? Nobody is out to get Iberians. I see a lot of Iberians on both the Olalde and Rui Martiniano teams. The data is what it is.

To the Board as a whole, could we have a little less heat and no more insults? We can respectfully agree to disagree.

 

[LeBrok]

On the contrary. Did you notice they model KO1 (Hungarian HG) as ANE admixted? That is either mistake or there is going to be an update. KO1 has mtDNA R3 it shares with AG3

It was a big surprise to me too. Till now all WHG (together with SHG) were presented as variations of same European source, without "foreign" admixtures like, SW Asian, Caucasian, Baloch or gedrosia. EHGs were the ones, WHGs who ventured East and mixed with ANE Baloch/Gedrosia and American.
Soon, population geneticists need to get together and define admixtures! I don't want every time figure out "their interpretations" of ancient genomes.

 

[LeBrok]

I've been telling many times that CHG admixture on the steppe had already started with Khvalynsk.

Can you tell how old is this S2 sample (samara)? The rest guys with green are from Yamnaya. S2 must be from the time the Neolithisation process started in steppe, together with mixing with Transcaucasian people.

 

[epoch]

No, I mean the Fu paper. The admixture they detect is similar to Kotias-Satsurbalia.



That's not too surprising given that Natufians are at the extreme end of the WHG <-> Meso-Neolithic Levantine continuum in Lazaridis 2016 paper. The Natufians are estimated to have ~45% Basal Eurasian, whereas the Anatolians have roughly 25%, only marginally more than the EN_European cluster (~22%-23%). These seem to be the incipient stages of West Asian diversity, i. e. what differentiates present day Anatolians from either Iranians or Arabs.

>

That is a Basal versus Neanderthal graph. I fail to see how that represents a "WHG <-> Meso-Neolithic Levantine continuum". I also fail to detect incipient grades of differentiation in it. Even worse, the oldest Iranian sample from Hotu has the largest distance to WHG, the neolithic Iranians are a tad closer and chalcolithic Iranians even more closer.

Listen, Marko, you're not defending your case very well. Shall I help a bit. I assume that what you want to defend is something like this: At the onset of the Gravettian, or even earlier, the whole area from Iran to Europe had a similar Kostinki-14 like population. Drift caused the Iranians to differentiate from Europeans. Later, but LGM and arrival of Basal Eurasians, part of these Iranians moved to, say, the Balkans, weathered the LGM there and thus became isolated only to remerge to become WHG. So they share a part drift with the descendant of the remaining Iranians.

That about it?

 

[Angela]

The hypothesis that the Anatolian languages (Hittite, Luwian, Lycian, Lydian etc.) came into Anatolia from the steppe via the Balkans is looking increasingly dubious imo.

I don't know of a single linguist, or even non-linguist, like Anthony, who doesn't date Anatolian to around 4500-4000 BC. based on many factors, including relative chronology.

There was never any archaeological evidence for any movement from the Balkans to Anatolia around that time; indeed, the movement went in the opposite direction.

Now we discover that there was very little steppe ancestry in the Balkans around that time. (You have to look at the dates and Admixture among other things.)

The argument has been made that perhaps these languages didn't enter Anatolia until long after 4000 BC., when there was some more steppe. Ok, let's look at that scenario.

If the Anatolian speakers weren't in Anatolia, where were they? We have to have a rational alternative. They can't have been on the steppe, because then these languages wouldn't be so archaic; they would exhibit the changes present in late PIE. They couldn't have been in the Balkans either for the same reason. Plus the diversity of the Anatolian languages is in Anatolia, and it took place by 1800 BC. It takes at least 1-2000 years for that kind of diversity to develop. I don't see anything in what we know about how languages develop, or what we know about the Anatolian languages that would lend support to the idea that they weren't in Anatolia by 4000 BC.

Now, one could say that maybe it was a very small group which moved through the Balkans into Anatolia. However, could a very small group of simple herders, in a migration too small to leave any archaeological trace, have spawned so many Indo-European speakers? It's starting to seem like special pleading.

The other part of the genetics piece of the puzzle is that steppe has to show up in Anatolia at an appropriate time.

As Hittite is attested in actual documents in the 19th century BC, any steppe found in the Balkans or Anatolia after that time isn't probative of anything. The authors of the paper state they find neither WHG nor EHG in their Bronze Age Anatolian samples. Now that's a problem because you can't have steppe without them. It's true that Hittites burned their dead. However, they weren't the only Anatolian language speakers. There are the Luwians, for example. Does anyone know if any of the samples come from any of the Indo-European speaking areas?

I find it hard to believe that the Reich Lab, given the thousands of samples they have, and given the Lazaridis West Asian paper that is soon to come out, is deliberately steering us in the wrong direction.

I used to think that the Anatolian languages could have come south from the steppe through the Caucasus, but if they find no steppe in any Indo-European speaking areas at the proper time, what then?


https://en.wikipedia.org/wiki/Anatolian_languages

Here's a direct link to the map below so it's larger.

 

[Fire Haired14]

With the data in this paper there's no reason to support or not support a IE Steppe migration from to Anatolia through the Balkans. There simply aren't enough genomes. Basically all of the "Balkan" genomes in this paper come from Bulgaria. Furthermore they pretty much only come from three time periods; 6000-5500 BC, 4500 BC, and 3400-2900 BC. They sampled nothing from Bulgaria dating between 2900 and 1700 BC and nothing between 1700BC and 500 BC. That's a huge gap in time.

One of the Bulgarian genomes dating 3400-2900 BC did have Steppe ancestry! And coincidentally he was buried under a tumulus, indicating he was an early Indo European. He dates to 2900 BC and is therefore older than any Corded Ware genomes.

And the single Bulgarian genome from 1700 BC is basically a Srubnaya man. Also, all three of the Croatian genomes dating to about 1500 BC had Steppe ancestry.

So yes there was a Steppe presence in the Balkans. It may not have been as strong as in Corded Ware or Bell Beaker but it was definitely there. Until we get DNA from ancient Hittites we have no reason to doubt IE speaking Steppe groups went through the Balkans to Anatolia.

 

[MarkoZ]

That is a Basal versus Neanderthal graph.

I know, that's why I posted it :bigsmile:

I fail to see how that represents a "WHG <-> Meso-Neolithic Levantine continuum". I also fail to detect incipient grades of differentiation in it.

It can be seen in this plot from Lazaridis (2016):

WHG and Natufian form an almost perfect cline despite the extremely divergent Basal Eurasian component in the latter. Most European farmers plot neatly on this cline in accordance with their varying degrees of additional WHG admixture. Some have slight pull towards highland West Asians due to their minor Iran-Caucasus admixture.

Even worse, the oldest Iranian sample from Hotu has the largest distance to WHG, the neolithic Iranians are a tad closer and chalcolithic Iranians even more closer.

The Iranian samples are hardly relevant to the West Eurasian <-> Levantine continuum since they derive large chunks of their ancestry from another source (i. e. not Basal Eurasian or WHG). Can we agree on this point? I don't see what's controversial about this - recent publications don't leave room for doubt in this regard.

Listen, Marko, you're not defending your case very well. Shall I help a bit.

The only case I have is my agreement with Fu et al. with regards to additional West Asian admixture starting with Villabruna or Miron as you correctly pointed out to me. But please do try to help me a bit.

I assume that what you want to defend is something like this: At the onset of the Gravettian, or even earlier, the whole area from Iran to Europe had a similar Kostinki-14 like population. Drift caused the Iranians to differentiate from Europeans.

No, I don't think any of this works. The component that makes Kotias-Satsurbalia distinctive must have arisen further to the east. Since it seems to be associated with P lineages which weren't present in West Eurasians before its arrival I assume the distant origins should be sought in South-East Asia or Central Asia. South-East Asia would be the place where the ancient paternal ancestor of most West Eurasian must have lived at one point. Iran is merely the crossroads where such migrations naturally end up, as seems to have happened early with the West Eurasians. Without DNA from these pivotal regions the details are murky.

Later, but LGM and arrival of Basal Eurasians, part of these Iranians moved to, say, the Balkans, weathered the LGM there and thus became isolated only to remerge to become WHG. So they share a part drift with the descendant of the remaining Iranians.

I don't think it's possible to narrow it down in any meaningful way without ancient samples. All we can say is that this type of ancestry appears in Villabrunna when before it wasn't there.

 

[MarkoZ]

With the data in this paper there's no reason to support or not support a IE Steppe migration from to Anatolia through the Balkans. There simply aren't enough genomes. Basically all of the "Balkan" genomes in this paper come from Bulgaria. Fu

I think what Gimbutas & Mallory meant by an Anatolian migration from the Balkans is actually an Anatolian migration from Bulgaria specifically. So there's some method to their sampling with regards to the Indo-European question. I wish they'd focus on other things for once, but they seem to know what they're doing.

 

[Angela]

With the data in this paper there's no reason to support or not support a IE Steppe migration from to Anatolia through the Balkans. There simply aren't enough genomes. Basically all of the "Balkan" genomes in this paper come from Bulgaria. Furthermore they pretty much only come from three time periods; 6000-5500 BC, 4500 BC, and 3400-2900 BC. They sampled nothing from Bulgaria dating between 2900 and 1700 BC and nothing between 1700BC and 500 BC. That's a huge gap in time.

One of the Bulgarian genomes dating 3400-2900 BC did have Steppe ancestry! And coincidentally he was buried under a tumulus, indicating he was an early Indo European. He dates to 2900 BC and is therefore older than any Corded Ware genomes.

And the single Bulgarian genome from 1700 BC is basically a Srubnaya man. Also, all three of the Croatian genomes dating to about 1500 BC had Steppe ancestry.

So yes there was a Steppe presence in the Balkans. It may not have been as strong as in Corded Ware or Bell Beaker but it was definitely there. Until we get DNA from ancient Hittites we have no reason to doubt IE speaking Steppe groups went through the Balkans to Anatolia.

Most of what you wrote is irrelevant. We have written Hittite from 1800 BC. The spoken language has to be older. So, any steppe ancestry after 2000 BC doesn't prove anything. Neither is steppe ancestry in the Balkans 3400-2900 BC probative of anything in terms of this hypothesis. Anatolian formed around 4000 BC. So, the only data about steppe in the Balkans that's pertinent would be from before that time.

Nobody said there was no steppe in the Balkans, certainly not me. That's a straw man argument.

Please read my last post.

@Marko,
Indeed.

 

[Fire Haired14]

Most of what you wrote is irrelevant. We have written Hittite from 1800 BC. The spoken language has to be older. So, any steppe ancestry after 2000 BC doesn't prove anything. Neither is steppe ancestry in the Balkans 3400-2900 BC probative of anything in terms of this hypothesis. Anatolian formed around 4000 BC. So, the only data about steppe in the Balkans that's pertinent would be from before that time.

All I'm saying is there aren't enough Balkan genomes across space and time to say there's evidence or a lack of evidence of a Steppe movement to Anatolia through the Balkans.

There definitely could have been Steppe groups moving through the Balkans that lived side by side native groups without outbreeding. Hungarian genomes dating to the 3rd millenium BC demonstrate this is possible. One is 73% Yamnaya(and carries R1b Z2103), one is something like 90% Anatolia Neolithic, another looks like a UkrainNeolithic and AnatoliaNeolithic mixture.

Nobody said there was no steppe in the Balkans, certainly not me. That's a straw man argument.

Reich's team seems think if there wasn't a Corded Ware or Neolithic Farmer-like mass migration then there was no migration. That's where I have a problem.

 

[epoch]

The hypothesis that the Anatolian languages (Hittite, Luwian, Lycian, Lydian etc.) came into Anatolia from the steppe via the Balkans is looking increasingly dubious imo.

I don't know of a single linguist, or even non-linguist, like Anthony, who doesn't date Anatolian to around 4500-4000 BC. based on many factors, including relative chronology.

There was never any archaeological evidence for any movement from the Balkans to Anatolia around that time; indeed, the movement went in the opposite direction.

Now we discover that there was very little steppe ancestry in the Balkans around that time. (You have to look at the dates and Admixture among other things.)

The argument has been made that perhaps these languages didn't enter Anatolia until long after 4000 BC., when there was some more steppe. Ok, let's look at that scenario.

If the Anatolian speakers weren't in Anatolia, where were they? We have to have a rational alternative. They can't have been on the steppe, because then these languages wouldn't be so archaic; they would exhibit the changes present in late PIE. They couldn't have been in the Balkans either for the same reason. Plus the diversity of the Anatolian languages is in Anatolia, and it took place by 1800 BC. It takes at least 1-2000 years for that kind of diversity to develop. I don't see anything in what we know about how languages develop, or what we know about the Anatolian languages that would lend support to the idea that they weren't in Anatolia by 4000 BC.

Now, one could say that maybe it was a very small group which moved through the Balkans into Anatolia. However, could a very small group of simple herders, in a migration too small to leave any archaeological trace, have spawned so many Indo-European speakers? It's starting to seem like special pleading.

The other part of the genetics piece of the puzzle is that steppe has to show up in Anatolia at an appropriate time.

As Hittite is attested in actual documents in the 19th century BC, any steppe found in the Balkans or Anatolia after that time isn't probative of anything. The authors of the paper state they find neither WHG nor EHG in their Bronze Age Anatolian samples. Now that's a problem because you can't have steppe without them. It's true that Hittites burned their dead. However, they weren't the only Anatolian language speakers. There are the Luwians, for example. Does anyone know if any of the samples come from any of the Indo-European speaking areas?

I find it hard to believe that the Reich Lab, given the thousands of samples they have, and given the Lazaridis West Asian paper that is soon to come out, is deliberately steering us in the wrong direction.

I used to think that the Anatolian languages could have come south from the steppe through the Caucasus, but if they find no steppe in any Indo-European speaking areas at the proper time, what then?


https://en.wikipedia.org/wiki/Anatolian_languages

Here's a direct link to the map below so it's larger.

One - I admit rather farfetched - option could be that Anatolian was spoken by a group in the Balkans who were themselves only slightly Steppe admitted. More or less like the Turks today. We *do* have slight Steppe admixture in a Varna sample. After 10 generations 1 persons genetic legacy is about 0.1%. If a slight tiny ruling class managed to impose their language on a large neolithic populace, and that populace would enter Anatolia after a few 100 years we may explain the lack of Steppe admixture.