Weird phenomena in Southeast Europe during EBA and implications for IE spread

[Ryukendo]

Just something I noticed.

From previous papers, we already know that while the introgression of Steppe ancestry in Corded Ware (>70%) and Bell Beaker (~50% in most cases) is very great, the introgression into Southeastern Europe almost seems deccelerated, plateauing at around 30% in the middle to late Bronze Age overall. This makes the BA Hungarians look almost like French and Spanish autosomally, which is visible from David's PCA as the olive squares.

Image

This is corroborated by further data from Mathieson et al, who give us the following:

Other Copper Age (~5000-4000 BCE) individuals from
the Balkans have little evidence of steppe ancestry, but Bronze Age (~3400-1100 BCE)
individuals do (we estimate 30%; CI: 26-35%).

Their PCA is projected, which makes it difficult to predict the final clustering of the Balkan samples, but you can already see that the Balkan Bronze Age samples are between the Mako Hungary and Hungary LBA, and cluster exactly with Vatya, which in David's plot will make them cluster between Basque, Spanish and French, which is a very low level of Steppe ancestry considering how close they are to the steppe.

More information about Vatya, including their low level of Steppe ancestry, can be found in the Allentoft et al Supplementary materials.

This quite surprising, given that the early Yamnaya migration into the Carpathian basin was given great importance in many theories. It also means that the migration through the balkans probably was not the source of Steppic ancestry into BB and CW, since they were less Steppic than the BB and CW themselves.

The situation is made even murkier if we look at the paternal haplogroups of the EBA Balkans. Here are their haplogroups from Supplementary Table 3 of Mathieson et al, along with their level of Steppe ancestry:

I2520: H2-L279. ~7% Steppe ancestry, 3336-3028 calBCE.
Grave goods, including crimson spondylus shell necklace, silver objects, and ceramic vessels, present. Head facing East.

I2176: I2a2a1b. ~10% Steppe ancestry, 3338-3025 calBCE.
I2175: I2a2a1b. ~6% Steppe ancestry, 3338-3015 calBCE.
Relatively tall pair, silver objects and ceramic vessels present, red ochre and sandstone present, head facing east.

Bul10: female; ~12% Steppe ancestry, 3090-2924 calBCE
Infant in an EBA ditch with vessels.

I2165: I2a2a1b. ~22% Steppe ancestry, 3020-2895 calBCE
Head to the east, colored with red ochre.

I2510: G2a2a1a2. 0% Steppe ancestry. 2906-2710 calBCE
Grave disrupted by another bronze age grave.

Bul6: I2a3. ~5% Steppe ancestry. 3400-1600 BCE
Bul8: I. ~5% Steppe ancestry. 3400-1600 BCE
Unearthed from beneath an actual round barrow.

These samples stand in great contrast to the Bulgarian Yamnaya sample, who is ~85% Yamnaya or so, and was geographically proximal to the Bulgarian samples. Not a single one of them are R1b or R1a.
The rest of the Balkan Bronze Age samples are >1 millenium later, and are comfortably Yamnaya-rich (>40%), with an R1a 80% Yamnaya individual, but in these earlier burials not even those from beneath the barrows are rich in Steppic ancestry, and the rest of them are not low social status individuals either.

This looks very weird and uncomfortable to me, in the context of the Kurgan theory as propounded by Mallory, Heyd and co--or at least their interpretation, with an early (and massive) migration into the Balkans playing a pivotal role.

What is going on? Anyone have any ideas? Perhaps Southeast Europe was late to be Kurganized, or kurganisation was fleeting, or there was a lot of social differentiation during the Kurganisation process, because the density of farmers was greater than in Northern and Western Europe? Or maybe the Kurganisation of SE Europe was unique in being less mediated demographically and more culturally, maybe due to a long history of contact?

P.S.
On the other hand, in the Vucedol, we have:
I4175: Female with U4a. ~50% Steppe ancestry. 3000-2700 BCE.
I3499: R1b1a1a2a2. ~25% Steppe ancestry. 2884-2666 calBCE.
I2792: G2a2a1a2a. 0 steppe ancestry. 2872-2617 calBCE

It seems like Vucedol has more steppic connections than the intervening cultures to the east of them.

[Agamemnon]

Bul10 is from Ezero.

[Ryukendo]

Bul10 is from Ezero.

A note adressing this topic from the point of view of Heyd et alia:

Text.jpg

From Prehistoric Mobility and Diet in the West Eurasian Steppes 3500 to 300 BC, By Claudia Gerling, Walter de Gruyter GmbH & Co KG, Jul 1, 2015

Now we have a very interesting phenomenon where a small amount of Yamnaya autosomal introgression has taken place, and barrows at least have come in, in terms of ritual and ideology, but the major Y haplogroup is not seen (a minor component of the Steppic pool is seen); the autosomal introgression is extensive across the balkans, unlike before where it is sporadic, but it is still not massive like in BB and CW; and other cultural and ideological developments cannot be securely attributed to Yamnaya influence. Perhaps the red ochre and etc will join the package of traits thought to reflect Kurganisation but actually do not, e.g. the changes in GAC. Or perhaps we will converge with Heyd's observations in Antiquity magazine where ideological Kurganisation can happen with the bits of the ideological package disseminating in a disaggregated manner without implicating wholesale genetic change.

Fascinating. I wonder what Heyd would think of it.

Is there any way to salvage the Western Anatolian route from this patchwork? I know linguistics is its own field, but one should at least be able to use it with archaeology and genetics to arrive at a synoptic view of past social processes. What do you think?

[Agamemnon]

A note adressing this topic from the point of view of Heyd et alia:
Text.jpg
From Prehistoric Mobility and Diet in the West Eurasian Steppes 3500 to 300 BC, By Claudia Gerling, Walter de Gruyter GmbH & Co KG, Jul 1, 2015

Now we have a very interesting phenomenon where a small amount of Yamnaya autosomal introgression has taken place, and barrows at least have come in, in terms of ritual and ideology, but the major Y haplogroup is not seen (a minor component of the Steppic pool is seen), the autosomal introgression is extensive across the balkans, unlike before where it is sporadic, but it is still not massive like in BB and CW, and other cultural and ideological developments cannot be securely attributed to Yamnaya influence. Perhaps the red ochre and etc will join the package of traits thought to reflect Kurganisation but actually do not, e.g. the changes in GAC. Or perhaps we will converge with Heyd's observations in Antiquity magazine where ideological Kurganisation can happen with the bits of the ideological package disseminating in a disaggregated manner without implicating wholesale genetic change.

Fascinating. I wonder what Heyd would think of it.

Is there any way to salvage the Western Anatolian route from this patchwork? I know linguistics is its own field, but one should at least be able to use it with archaeology and genetics to arrive at a synoptic view of past social processes. What do you think?

From a linguistic standpoint, a migration from the steppe to Anatolia via the Balkans makes much more sense than a migration through the Caucasus. I think the genetic evidence so far does not necessarily discredit the Western Anatolian pathway, the steppe component does make its appearance after all. In fact, I think it would be naive to expect some sort of sudden blitzkrieg in Anatolia with the kind of major steppe contribution we see in northern Europe. Anatolia and the Balkans were diverse and heavily populated starting from the Bronze Age (even before that in fact). The linguistic data suggests that the Proto-Anatolians were nothing more than an intrusive minority, by the time Hittite made its appearance in the epigraphic record (in the form of personal names) it already was mutually unintelligible with other Anatolian languages, suggesting that rapid contact-induced change took place. It's possible that the steppe migrants simply did not have the kind of demographic impact their cousins had in Northern Europe.

That being said, I'm not totally opposed to a migration through the Caucasus, although there are a lot of limitations associated with that route.

[Bas]

Could it be that it's the geography of the Balkans which made it harder to settle down for these new Yamnaya incomers?

It could be a mixture of a couple of reasons to that region to explain less steppe genetic influence-for Hungary and the Pannonian plain, the geography making it harder for immigrants to claim land and a foothold in the region doesn't seem to be a factor as it could have been in the Balkans,rather it could be related to the Pannonian plain being prime farming country, and somewhere capable of supporting a large pop density for the Neolithic farmers.

[Gravetto-Danubian]

Nice summary

But first a correction – Heyd never claims a massive migration to the Balkans. In fact, very few people do this side of 1965. Rather, he outlines a ‘come and go’ scenario: the steppe migrants came, occupied steppe like niches, then “disappeared” c. 2400 BC, with the aridization of the steppe. Thus he is in contrast to Mallory & Anthony, who aren;t specialists, but offer a, quite frankly, a dumbed down version of prehistory for the general public. Indeed, the entire scenario outlined by Anthony borders on creative writing, and he resorts to distortion of facts (and this takes nothing away from his work on the Samara region, for which he appears to have a better grasp of).

Contrary to Agamemnon, there is no linguistic evidence that requires Anatolian to come via Balkans; the reality is, the IE phylogenetic tree only requires Anatolian to branch off first, it doesn't mandate specific locations.

I won;t go into the full details as to your request for further ideas because that'll require an entire manuscript. Suffice to say, the Bronze Age Anatolians alluded to in this paper come from southwest Anatolia, and show CHG/ Iran shift, as predicted by archaeology, and mandated by prehistoric reality

Your comments about Y haplogroups is notable. It erodes the simplistic, north-Euro centric idea propagated by genealogists that R1a and R1b are THE indo-European markers.

[Ryukendo]

Nice summary

But first a correction – Heyd never claims a massive migration to the Balkans. IN fact, very few people do. Rather, he outlines a ‘come and go’ scenario: the steppe migrants came, occupied steppe like niches, then “disappeared” c. 2400 BC, with the aridization of the steppe. Thus he is in contrast to Mallory & Anthony, who in blunt honestly cannot hold a torch to Heyd. Their work is like pop reading for the general laity infatuated with a dumbed down version of history. Indeed, the entire scenario outlined by Anthony borders on creative writing, and he resorts to distortion of facts (and this takes nothing away from his work on the Samara region, for which he appears to have a better grasp of).

Contrary to Agamemnon, there is no linguistic evidence that requires Anatolian to come via Balkans. Supporters of the kurgan scenario will continue to plead this version of history, because it is the platform they cannot afford to be eroded. But the reality is, the IE phylogenetic tree only requires Anatolian to branch off first, it doesn;t mandate specific locations.

I won;t go into the full details as to your request for further ideas because that'll require an entire manuscript. Suffice to say, the Bronze Age Anatolians alluded to in this paper come from southwest Anatolia, and show CHG/ Iran shift, as predicted by archaeology, and mandated by prehistoric reality

Your comments about Y haplogroups is notable. It erodes the simplistic, north-Euro centric idea propagated by genealogists that R1a and R1b are THE indo-European markers.

Thanks for looking this over. Its time to ask some other archaeologists what the interpretation will be. Let's see what heyd and kristiansen say themselves.

Once the Anatolian Bronze Age genomes are out in Lazaridi's revised manuscript, the IE homeland debate will perhaps heat up again. I've always been more agnostic about the ultimate origins of Indo-hittite.

Then again Krause's theory may still be incorrect, because the genomes from India will probably clearly demonstrate that IIr derives from the steppe and not directly from West Asia, with anatolian the only exception to steppic origin.

[Gravetto-Danubian]

Thanks for looking this over. Its time to ask some other archaeologists what the interpretation will be. Let's see what heyd and kristiansen say themselves.

Once the Anatolian Bronze Age genomes are out in Lazaridi's revised manuscript, the IE debate will perhaps heat up again. I've always been more agnostic about the ultimate origins of Indo-hittite.

Then again Krause's theory may still be incorrect, because the genomes from India will probably clearly demonstrate that IIr derives from the steppe and not directly from West Asia, with anatolian the only exception to steppic origin.

Why would Kristiansen know anything about the Balkans ? He couldnlt even manage to get proto-Germanic correct.
Perhaps the bes tpeople to ask are people who are most familiar with the material - people from SEE- which is the avenue I took.

[Ryukendo]

Why would Kristiansen know anything about the Balkans ? He couldnlt even manage to get proto-Germanic correct.
Perhaps the bes tpeople to ask are people who are most familiar with the material - people from SEE- which is the avenue I took.

Hmm agree. It seems like close knowledge about archaeology is one of those fields where expertise is still very regional.

What do you make of Heyd's idea of 'Yamnaya ideological-cultural package' in the Balkans?

[parasar]

Nice summary

But first a correction – Heyd never claims a massive migration to the Balkans. In fact, very few people do this side of 1965. Rather, he outlines a ‘come and go’ scenario: the steppe migrants came, occupied steppe like niches, then “disappeared” c. 2400 BC, with the aridization of the steppe. Thus he is in contrast to Mallory & Anthony, who aren;t specialists, but offer a , quite frankly, a dumbed down version of history for the general public. Indeed, the entire scenario outlined by Anthony borders on creative writing, and he resorts to distortion of facts (and this takes nothing away from his work on the Samara region, for which he appears to have a better grasp of).

Contrary to Agamemnon, there is no linguistic evidence that requires Anatolian to come via Balkans. Supporters of the kurgan scenario will continue to plead this version of history, because it is the platform they cannot afford to be eroded. But the reality is, the IE phylogenetic tree only requires Anatolian to branch off first, it doesn;t mandate specific locations.

I won;t go into the full details as to your request for further ideas because that'll require an entire manuscript. Suffice to say, the Bronze Age Anatolians alluded to in this paper come from southwest Anatolia, and show CHG/ Iran shift, as predicted by archaeology, and mandated by prehistoric reality

Your comments about Y haplogroups is notable. It erodes the simplistic, north-Euro centric idea propagated by genealogists that R1a and R1b are THE indo-European markers.

They are definitely THE IE markers as far as spread of IE is concerned. Nothing north-Euro centric about that part at least. Otherwise the sharp Y boundaries we see in eastern India where IE, Tibeto-Burman, Dravidian, and A. Austric meet would not make any sense. Y-R1a1 (and to a very diffuse extent R2) marks an abrupt change by just vanishing to ~0 in the TB, Dravidian, and AA speakers in that region. With early PIE (as opposed to Patterson's LPIE) I would agree with you.