search this blog

Monday, September 11, 2017

Comic relief from Russia (Klejn et al. 2017)

I kid you not, the map below was published in a recent discussion paper in the European Journal of Archaeology. It was put together by two highly experienced Russian academics: archaeologist Leo Klejn and geneticist Oleg Balanovsky. Pretty crazy, huh?

It surely must rank as one of the most naive, awkward and inadvertently comical attempts to debunk the Kurgan Proto-Indo-European theory that I have seen anywhere, and I've seen some really dumb sh*t in this context in the comments at this blog.

Klejn and Balanovsky are actually arguing that Yamnaya-related ancestry did not spread from the Pontic-Caspian steppe to most of the rest of Europe, but rather from somewhere around modern-day Finland to most of the rest of Europe, including the Pontic-Caspian steppe.

Obviously, this is impossible, and the reasons for this are explained by Klejn's co-authors and discussion opponents in the paper. Basically, the very specific type of genetic structure fixed in the Yamnaya population of the Early Bronze Age Pontic-Caspian steppe did not exist in Northeastern Europe prior to the arrival of the Corded Ware people in the region, and they, in all likelihood, came from the Pontic-Caspian steppe, because some of the earliest Corded Ware samples are practically identical to those from the Yamnaya horizon.

Here's a figure from the recent Mathieson et al. 2017 preprint that illustrates this very neatly with an ADMIXTURE analysis. Remarkably, the Globular Amphora Culture (GAC) people, who lived on the northwest edge of the Pontic-Caspian steppe just prior to the formation of the Corded Ware Culture across much of Northern Europe, showed, at best, trace amounts of the Yamnaya-specific genetic component. On the other hand, the early Corded Ware individual from what is now Latvia (Latvia_LN) appears almost indistinguishable from the average Yamnaya folks.

Hopefully, Klejn and Balanovsky have now given up on their highly original theory about the expansion of the Yamnaya genetic signal after looking over the data from Mathieson et al. 2017. But from running this blog and having to deal with copious amounts of stupid sh*t in the comments, I know how exceedingly difficult it is for some people to finally bury their pet theories, no matter how at odds with reality they are, so I guess we'll see.


Klejn et al., Discussion: Are the Origins of Indo-European Languages Explained by the Migration of the Yamnaya Culture to the West?, European Journal of Archaeology, Published online: 28 July 2017, doi:10.1017/eaa.2017.35

Mathieson et al., The Genomic History Of Southeastern Europe, bioRxiv, Posted May 9, 2017, doi:

See also...

Late PIE ground zero now obvious; location of PIE homeland still uncertain, but...

A plausible model for the formation of the Yamnaya genotype


Synome said...

What are they trying to use, WHG for Yamnaya?

Davidski said...

They're arguing that the Yamnaya component in Europe is just EHG, so a lot more ancient than the Bronze Age and that it spread from far Northeastern Europe to Central Europe and the steppe.

Obviously, this is insane.

AWood said...

The heat map looks closer to a EHG fit. Nonetheless, those original EHG foragers were unlikely to have been roaming northern Finland. A steppe homeland seems like the best fit. The map looks closer to a Uralic distribution of people which is certainly not Yamnaya, even though they probably overlap.

K33 said...

Is this just an ADMIXTURE heat map? Uralics score high in Yamnaya calculator components because they share deep ancestry (EHG) with Yamnaya-- not because they descend from Yamnaya.

The map's right-hand arrow might be a fairly accurate representation of how EHG entered the Mesolithic steppe (ie, Samara). The Khanty have an ancient myth about how a group of their ancestors migrated to the Black Sea.

IOW, the conflation of para-proto-Uralics with the EHG genotype might be correct. But the Yamnaya genotype clearly involves the CHG component and the fulcrum of its spread was from the steppe.

Shaikorth said...

A hypothesis they bring up in the paper is that folks from Maglemose and post-Maglemose Jutland and neolithic Southern Baltic who supposedly migrated to Dnieper-Donets were pre-IE, and looking at the map, are supposedly responsible for Yamnaya component of Haak et al. It's just that those people in reality were most likely farmers and WHG.

Archaelog said...

Nothing surprising. A lot of archaeologists and historians have been very vocal about their disagreements with the use of genetics in prehistorical studies. I would say these people are feeling really threatened by recent developments in the field of human archaeogenetics. Given that this has the potential to put them out of work and discredit their pet theories, is it really surprising?

Mark B. said...

As someone once said (to paraphrase): Scientific revolutions don't happen when people change their minds - they happen when the opposition dies off.

Ryan said...

I'll buy that a sizable fraction of EHG would have been speaking something related to Proto-Indo-European at least, but not anything like a large scale expansion of pure EHG. If they think IE spread through EHG alone they should point to the groups with EHG but lacking CHG out there - not to mention finding IE languages lacking the Kartvelian borrowings. The only candidate I can think of is the Makó culture and they got absorbed by the Bell Beakers anyways.

I think it would be interesting and worthwhile to try to figure out where IE's older ancestors passed through though. How long were IE's ancestors hanging out in the Ukraine? Since 14,000 kya? Earlier? Later?

Samuel Andrews said...

The most intelligent people can say stupid things about a topic if they are uninformed/educated in it. Even if they are informed they can still come up with stupid theories.

But it does seem archaeologists have begun understanding and accepting discoveries made by ancient DNA. Some more than others obviously.

Anonymous said...

Ok, I'm confused again. I thought Karelia HG was the oldest individual that was EHG?

Arch Hades said...

Why dont these idiots include data from Pushtuns and Kalasha? Then we wont see a gradual cline from Finland.

Matt said...

@ Lenny Dykstra: Is this just an ADMIXTURE heat map?

I gather it's based on these proportions - - with interpolation, and prediction for missing values based on geographic position.

That's wrong in the first place because that method will deform the pattern.

It's wrong in the second because Haak 2015's Yamnaya values are probably not as correct as what we have now. They're based on using a limited selection of outgroups which had relatively weaker power to detect differences between CHG vs EEF and EHG vs WHG and so on.

For instance the proportions of the Balkans and Spain in Haak 2015's method are the wrong way round. From Davidski's experiments and current methodologies, we know the Balkans to be more like a relatively high Yamnaya region with low WHG, while Spain is relatively higher WHG with relatively lower Yamnaya than the Balkans. (Which, by the way, actually makes sense with, y'know, everything we know about what pre-steppe Spain and the Balkans were like, and basic assumptions of isolation by distance). This in itself will deform the patterns.

(Besides pretty much everything Davidski already said first and best!).

Steven said...

Are the proportions in the map correct? I would expect more Yamnaya in the Swiss and southwest Germany.

rozenblatt said...

Regarding Klejn: he does his own ideas about origin of Indoeuropeans. Basically he thinks that Indoeuropeans spread from Central Europe, and sees Funnel Beaker culture as ancestors of IE. So his criticism of Kurgan model stems partially from that.

Regarding Balanovsky: he is mainly specialized in Y-DNA studies, so his main problem with Kurgan model is lack of R1a and western branch of R1b in Yamnaya samples. Moreover, Balanovsky is under heavy influence of Moscow school of linguistics, basically the guys who use glottochronology. AFAIK Moscow linguists are also not very supportive of Kurgan model.

These all leads to weird thing: many Russian scientists don't think that IE homeland was in Russia, while many Western scientists(Anthony, Mallory, Kristiansen, Reich) think that it was in Russia.

Unknown said...

This component we're seeing illustrated is likely some artificial composite of EHG and CHG but in different proportions than the real steppe package.

At any rate, multiple people on Anthrogenica have been pretty vocal from the get-go about Yamnaya ancestry peaking around the Baltic sea in some calculators. It makes no sense historically, archaeologically, or geographically. There is plenty of extra EHG there and, in all likelihood, Swedes and North Germans do outscore them in real steppe ancestry.

Davidski said...

@Anthro Survey

Bronze Age steppe ancestry does peak around the Baltic; south and east Baltic to be precise. It's much higher there than in Germany.

There are good reasons for this, including the migration to the region of Yamnaya clones via the Corded Ware culture, like Latvia_LN on the ADMIXTURE plot above. Yamnaya-derived populations like Corded Ware and Bell Beaker in Central Europe were more admixed with EEF ancestry.

The problem with the spatial map is that it was based on somewhat imprecise figures from Haak 2015, which inflated the Yamnaya-related ancestry in Finns and Saami, probably due to both extra European hunter-gatherer and Siberian admixture.

Romulus said...

I think in this scenario they propose, EHG would have been paternally distributed into the Corded Ware zone from Baltic and other EHG like foragers from what would become the Comb Ware zone. CHG being simultaneously propagated Nothwest around the black sea, primarily maternally?

I find lumping EHG and CHG together as Steppe ancestry just confuses things, when WHG must have some ANE to contain R1b, and EHG is defined as WHG+ANE. I'd like to see models of (WHG+EHG+ANE) / EEF / CHG, until the relationship between WHG/EHG/ANE is properly understood.

Davidski said...


I think in this scenario they propose, EHG would have been paternally distributed into the Corded Ware zone from Baltic and other EHG like foragers from what would become the Comb Ware zone. CHG being simultaneously propagated Nothwest around the black sea, primarily maternally?


Early Corded Ware was identical to Yamnaya and it came directly from the steppe. So the Baltic region was populated by a Yamnaya population from around 2900 BC.

This is laid out in the Mittnik paper in some detail, especially on page 8. No point ignoring this info and debating the same thing over and over when there's nothing to debate anymore.

Shaikorth said...

@Anthro Survey
The frequencies in the map look like they are from Haak et al 2015. supplements, showing Yamnaya frequency when Europeans are modeled as something like Yamnaya+WHG+EEF+Siberian+Near Eastern. Even that model is too simple to work everywhere, and doesn't account for extra hunter-gatherer admixture which would change the numbers. The paper also has fits which replace Yamnaya with EHG, and that doesn't significantly change the quality - though we know CHG (and thus Yamnaya) is all over Europe and that can be verified with haplotype fits like in Broushaki 2016.

Davidski said...

East Asian/Siberian admixture really screws up the Yamnaya component estimates when not accounted for, like in the Saami. But it's easy to account for it now, so it's not a major issue.

Samuel Andrews said...


Thanks for the background info. Makes sense. We shouldn't expect an archaeologist and a Y DNA-expert (who focuses on modern Y DNA) to understand results based on the entire genome. I'm happy Lazardis and others were able to discuss with them.

Something a lot of naysayers seem to not understand is that Neolithic genomes reject the possibility ANE-rich EHG ancestry let alone EHG+CHG types criss crossed across Europe before Corded Ware and Bell Beaker. The authors of this paper obviosuly don't understand this.

Archaelog said...

@Ryan I believe Indo=European proper (including Hittite) came into existence only after the R1a Hunter gatherers interacted with and absorbed a lot of elements from the R1b and G2/J2 populations coming from the south. The Caucasian substrate theory does appear to have some force to it.

Shaikorth said...

Models in Haak 2015 supplements try to account for Siberian and Near Eastern but don't even include Saami, so Balanovsky is operating on guesswork if that map is supposed to represent them.

jv said...

Fascinating! I would think Russian scientists would be proud to have the P.I.E. homeland in their country.

Grey said...

Anthro Survey said...

"At any rate, multiple people on Anthrogenica have been pretty vocal from the get-go about Yamnaya ancestry peaking around the Baltic sea in some calculators. It makes no sense historically, archaeologically, or geographically."

maybe it does *if*
1) there was a phase before yamnaya-ish people had fully developed their pastoral package
2) they had something valuable and heavy to sell

samara to yaroslavl


yaroslavl to baltic

in which case it might be those regions have the largest component because this phase happened early on and those regions were less diluted by later expansions.

any baltic peoples wear kilts?

Blogger said...

Yes. The Neolithic Revolution was as important as the Indo European migrations. I don't know why it doesn't get the same attention.

Blogger said...

Civilization its roots lie in the Neolithic. Indo European's were an addition IMO.

Samuel Andrews said...

"Yes. The Neolithic Revolution was as important as the Indo European migrations. I don't know why it doesn't get the same attention."

It just doesn't get the same attention on this blog and similar places on the internet. In academic circles I'm sure the Neolithic Revolution gets more attention.

"Fascinating! I would think Russian scientists would be proud to have the P.I.E. homeland in their country."

Apparently they aren't ethnocentric at all. Genetically speaking Russians aren't much more Steppe than Irish though. Russians might basically be recent imports from mainland Europe.

Plains Wanderer said...

Many people overlook the Neolithic Revolution because we know so little about the people involved. Most languages and cultures today come from more recent expansions, so they get the most attention (IE, Semitic, etc.). I agree that Indo-Europeans and similar expansions added on to earlier cultures and civilizations and would not have been the same without their developments. You could say they were standing on the shoulders of giants.

Kaltmeister said...

This is very interesting. As Carl-Heinz Böttcher wrote 1999 in: "Der Ursprung Europas", the Viking expansion was just the last chapter of a development, that repeated again and again over the last 6.000 years. There is no good reason to assume that the starting conditions that resulted, for example, in the Viking conquest and foundation of Russia in the middle ages, turned their impact in the opposite direction.
Böttcher developed his model of Maglemose- and Ertebölle-expansion into Europe long before anybody was talking about y-haplogroups. It is fascinating to find out, that he was right.