search this blog

Saturday, August 12, 2023

Indo-European crackpottery


I've now had the chance to read and digest the following two papers in Science about the origin of Indo-European languages:

Language trees with sampled ancestors support a hybrid model for the origin of Indo-European languages, Heggarty et al.

The genetic history of the Southern Arc: A bridge between West Asia and Europe, Lazaridis, Alpaslan-Roodenberg et al.

The Heggarty et al. paper is pure fluff. It offers nothing useful or even remotely interesting.

For instance, the authors derive some Indo-European languages in Europe from Anatolian farmers and others from Caucasus hunter-gatherers (see here). This is not just exceedingly far fetched, but also obviously forced.

Wolfgang Haak and Johannes Krause, you should be deeply ashamed of yourselves.

I've already commented extensively about the Lazaridis, Alpaslan-Roodenberg et al. paper (for example, see here). But the one thing I need to add is that this paper is what it is due to the inherent bias of some of the lead authors to push the Indo-Anatolian homeland into West Asia. I won't even bother mentioning their names, because we all know who they are.

See also...

Crazy stuff

Dear David, Nick, Iosif...let's set the record straight

The story of R-V1636

409 comments:

1 – 200 of 409   Newer›   Newest»
EastPole said...

I cannot comment on the method. Let’s assume it is OK. Then the problem is with the data. 170 meanings is small. I suspect that with the set of different meanings or with a much larger set of meanings they would get different results.
There are hundreds of very similar words in Slavonic and Indo-Iranian. If they selected these words they would get a different linguistic tree where Balto-Slavic and Indo-Iranian would be closely related.
If the linguistic tree depends on the selection of cognate sets, and is different for different cognate sets, then the question is what is PIE and which tree shows true PIE origin. We cannot have different PIE origins. Contradictory results would imply that PIE never existed.

Assume that their cognate set is perfect and shows true PIE origin. I doubt it, but assume this for the sake of argument. They say:
“For example, Indo-Iranic is an early independent branch in our analyses, with no close relationship to Balto-Slavic (see Box 1 and SM section 7.6.2.1), so that argument in favor of a northern route falls away. Genetically, the ancestry of Indo-Iranic speakers also derives much more heavily from south of the Caucasus and from Neolithic Iran than from the Bronze Age steppe (16) (see Box 2)”.
OK so let’s assume the southern route. How do we solve the problem of the many similarities between Indo-Iranian and Slavonic languages? Genetics suggests the following solution:

https://postimg.cc/FdfLZrrB

vAsiSTha, there is no other way. Either Indo-Iranians came from Eastern Europe in the Bronze Age, or Slavonic tribes went to India and Iran in the Bronze Age.

Olympus Mons said...

Davidski.
I think it’s warranted. - I , Olympus Mons, seem to have been the original [insert preferred insult here], because if you check your logs (which I suppose you have it) I was alone since 2015/2016 stating that The Shulaveri-Shomu was the urheimat of the PIE (obviously not at all so sure these days about R1B-L23). Literally the first thing I ever wrote (!) about anything related to archeo-genetics was to make that statement and affirm the origin of PIE in the Shulaveri heartland.
If you remember I was also stating back then, that the demise of the Shulaveri (4.900BC) spread PIE to northern Caucasus and to Anatolia.

“…suggest an emergence of Indo-European languages around 8000 years before present. This is a deeper root date than previously thought, and it fits with an initial origin south of the Caucasus followed by a branch northward into the Steppe region.”

I just wonder why don’t they just come out and say outright, SHULAVERI-SHOMU!

Virgin_Quilles_Sucks_R1a_Chadvski said...

Always agree with your words, cant understand why some people still supporting those Kind of arguments.That is what happened when scientists use Only statistical Methods and not a single anthropological source. Everybody knows that Iranian Plateau Had a Elamite-IVC cultural continuum(WE could add Kassites Indeed), imagine thinking serious on those those bullshit. Maykop, the Mesopotamian like society? It Is what Happened when people treat science like a Ego's dispute . Since 50's centuries PIE Homeland was Well defined from Archeology, DNA simple confirm what antrhopology Said. Today WE know that Yamnaya wasnt the PIE, obviously Sredny+ Progress groups made It.
Jura araxes Had a few contribution from PIE, buy l'età be honesto, Balkan OIE was the Last LIE , Balkan was the Most populares área on Europe Düring neolithic. .
Even If It wasnt a Lot, probably with WGS WE should catch It on other explanation.

Davidski said...

@All

Wolfgang Haak and Johannes Krause are responsible for this sort of naive bullshit. From their paper...

Genetically, the ancestry of Indo-Iranic speakers also derives much more heavily from south of the Caucasus and from Neolithic Iran than from the Bronze Age steppe (16) (see Box 2).

Right, because a set of shared "Caucasus/Iranian" alleles in an admixture analysis that entered India on ~10 separate occasions over a span of several thousand years actually represents the same language group.

On the other hand, one major pulse of Eastern European ancestry along with R1a (which dominates Indian upper castes) apparently left no linguistic traces in India.

Haha.

Romulus the I2a L233+ Proto Balto-Slav, layer of Corded Ware Women said...

We don't agree often but here you are spot on

Vladimir said...


AInterdisciplinary analyses of Bronze Age communities from Western Hungary reveal complex population histories
https://academic.oup.com/mbe/advance-article/doi/10.1093/molbev/msad182/7240678?login=false

vAsiSTha said...

@Virgin said
"Everybody knows that Iranian Plateau Had a Elamite-IVC cultural continuum"

There was no cultural, nor genetic continuum. Elamite locus was SW Iran (Susa), which borders Mesopotamia. IVC, East Iran and SC Asia have minimal Elamite connection till early bronze age. For example, there are at most a couple of Elamite tablets found from EBA Shahr-Sokhta in East Iran, and none from IVC. Proper Elamite trade connection is found only at the BMAC stage in SC Asia.

Wrt genetics, West Iran genetics is completely different from East Iran, SC Asia and India by neolithic.

Hajji_Firuz is the closest location from elamite territory. By 6000bce, 75% of its ancestry was already from IraqPPN (9000bce) or Anatolia (Barcin_N) as it replaced the Zagros neolithic IranN ancestry.

Target: IRN_Hajji_Firuz_C
Distance: 2.3453% / 0.02345310
51.6 IRQ_Nemrik9_PPN
24.2 TUR_Marmara_Barcin_N
15.2 IRN_Ganj_Dareh_N
9.0 GEO_CHG
0.0 RUS_AfontovaGora3

Similarly, 1800bce DinkhaTepe_A is completely under Mesopotamian (9000bce IraqPPN) and Anatolian influence. No wonder DinkhaTepe BA people didnt speak IE.

Target: IRN_DinkhaTepe_BIA_A
Distance: 3.0493% / 0.03049279
65.2 IRQ_Nemrik9_PPN
26.2 TUR_Marmara_Barcin_N
8.6 IRN_Tepe_Hissar_C
0.0 GEO_CHG
0.0 IND_Great_Andamanese_100BP
0.0 IRN_Ganj_Dareh_N
0.0 RUS_AfontovaGora3
0.0 TJK_Sarazm_En


Compare this to SC Asia. Only 6% ancestry is Mesopotamian/Anatolian.

Target: TKM_Geoksyur_En
Distance: 2.9121% / 0.02912072
65.4 IRN_Ganj_Dareh_N
18.6 GEO_CHG
10.2 RUS_AfontovaGora3
5.8 TUR_Marmara_Barcin_N
0.0 IRQ_Nemrik9_PPN

Or IVC, which has no direct Mesopotamian/Anatolian influence, but a common source with SC Asia (distance is high since drifted Andamanese is a bad proxy for AASI.)

Target: IRN_Shahr_I_Sokhta_BA2
Distance: 9.7885% / 0.09788471
41.6 TJK_Sarazm_En
36.4 IND_Great_Andamanese_100BP
22.0 IRN_Ganj_Dareh_N
0.0 GEO_CHG
0.0 IRN_Tepe_Hissar_C
0.0 IRQ_Nemrik9_PPN
0.0 RUS_AfontovaGora3
0.0 TUR_Marmara_Barcin_N

In short genetics of West Asia & SC Asia/India is miles apart.

@Davidski
"Right, because a set of shared "Caucasus/Iranian" alleles in an admixture analysis that entered India on ~10 separate occasions over a span of several thousand years actually represents the same language group."

If this is what you think, you know even less of the region that what I gave you credit for, which wasn't much to begin with.

Davidski said...

@vAsiSTha

I know a lot more than you. And I can tell you that you have no hope with your little agenda.

Not even Heggarty's scribbles in Science or Nature will help you when all is said and done.

Rob said...

@ Vasistha

So according to your (correct) observations of hterogeneity, it seems that there is no common vector or thread in western and northern- South Asia apart from the arrival of groups from Eastern Europe (Via various paths)

vAsiSTha said...

@Davidski

"I know a lot more than you. And I can tell you that you have no hope with your little agenda."

Then why don't you prove 10 separate admixtures from Iran into India. Or show a single reference for that claim.

"Not even Heggarty's scribbles in Science or Nature will help you when all is said and done."

You have no idea what's coming lol. Be ready for more 'Dear ____' posts.

@Rob
I have explained the 'common thread' on my blog post as a comment.

Wee e said...

I was looking at their database, and was surprised at a few of their vocabulary choices in Scottish Gaelic.

I am not a Gaelic speaker, but sometimes it seems they have chosen whatever alternative, local dialect word or very recent borrowing distinguishes Scottish Gaelic from modern or old Irish, even though another word is most accurate according to their semantic description. Or they have preferred one common word over another equally common for no clear reason.

For “dig”, “cladhaich” and “ruamhair” have both wide and very overlapping collocational sets, words for spades and diggers and such, and ruamhair is the more etymologically tricky, but you’d have to say they’re equally appropriate to most mentions of digging, or cladhaich is the one used a bit more as a default. I see that with other languages they have sometimes included two lexemes, so why leave cladhaich out?

“Ruamhair” and Manx “ruerey” are given on the same cognates map as the Old Irish word “claidid” and the modern Irish word is given there as “tochail” (Also a “dig” word in Scottish Gaelic, though leaning more to quarrying and mining usages.) But when you look up the information for the SG and Manx words, it is given as “rúam, Old Irish” and nothing more. There are no other members given for cognate set 6434. I don't really understand why the other word was excluded, as in other languages an alternate “dig” word is included.

Why they have preferred “crog” for claw over alternatives like spuir or spog, that match their description much more accurately is also a bit of a mystery. (crog’s main meaning is the whole paw or set, expressly excluded in the description.)

So this is only one language, but it worries me a bit that their database is a collection of varying quality drawn from sources of varying purpose & quality. It has the potential to be an extremely useful resource in its own right, but these things give me reservations.

It would also be very useful if more other lists could be included besides swadesh, such as number and kinship.

Matt said...

@Wee E, they did list the separate authors for each language, and for Scottish Gaelic, this is - https://www.ed.ac.uk/profile/dr-peadar-o-muircheartaigh. So you could potentially drop him an email about those specific choices even. When it comes to identifying the cognates, although the central team would set the methodology, the actual choices are made by the authors who are different by each language. Although in the case of some families, all my covered by a single specialist (while in Gaelic, the author's responsible for Scottish and Irish Gaelic are separate)...

Rob said...

Objectivey, I would say that the tree by Kassinian & Starostin is the closest one to reality
https://www.degruyter.com/document/doi/10.1515/ling-2020-0060/html?lang=en

Just need to push back a couple of the branchings back by 3-500 years. (e.g. Greek)

Wee e said...

@Matt Thanks.

MaxT said...

Heggarty 2023...are they suggesting PIE emerged as linguistic isolate in Caucasus-West Asia region and then branched out? I'm bit confused. I think only Sumerian language and Elamite language are considered isolates in western Asia, unrelate to Semitic languages.

"This is a deeper root date than previously thought, and it fits with an initial origin south of the Caucasus followed by a branch northward into the Steppe region." Heggarty 2023

Who were living further south of the Caucasus from them if PIE originated there? doesn't make sense.

Samuel Andrews said...

@Davidski,

What do you think about this?

A problem I have always noticed is they think "CHG" is an ethnic group whom they can trace a language back to.

It is a super old ancestry component not an ethnic group.

It's 13,000 years old which is way older than PIE.

Their descendants would have spoke many languages.

Samuel Andrews said...

"Steppe" ancestry however does come from an "ethnic group." An actual population who lived recently enough to have spoken IE.

They treat "Steppe" as the same thing "CHG." It isn't.

Samuel Andrews said...

@Davidski,

Btw, brother, I am almost done with a video on Khvalynsk where I will touch into some of these issues.

I never stopped reading your blog posts because you always share new arguments I can learn from.

Davidski said...

@Samuel Andrews

That's correct.

In fact, I would say that this CHG/Iranian signal is barely a genetic component.

It's actually just a bunch of shared alleles that look like a component because of the piss poor resolution in most of the analyses in these papers.

JR said...

@samuel Andrews Issac
Well,you could say the same thing about all groups.

EHG is 14,000 years old and it's not something u can trace PIE back to.

Same with uralic langauges.
The krasnoyarsk_BA_kra001 profile is very old,probably since 7000-5000 BC,yet proto uralic can only be reconstructed back to 2,500 BC.

Matt said...

I definitely agree that there is a significant weak point in Heggarty's version of the genetic evidence in the lack of reality of CHG-IranN as a single cluster dated to anything like the timeframes he talks about (other linguistic weak points I discussed on the other thread, and people engaged with them about as much as they could or wanted to, so it doesn't do much good to discuss them again here).

It's distinct from what Harvard/Planck proposal which involves a via-the-Steppe impulse to explain most pIE including South Asian, involving with a single specific CHG group, dated to either Mesolithic or Eneolithic, which branched to steppe and Anatolia. (On balance still think this seems less likely than Anatolian explained by an early branch from another direction).

Would also suggest that purely from the perspective of argument, it seems tactically potentially counter-effective to say that India received multiple migrations from different IranN-CHG groups, because then Heggarty could simply say "OK; well, perhaps one of those was the same that gave rise to the Steppe population, and the others are responsible for various other Near Eastern languages. Thanks for helping.".

Though, when the Steppe group formed from CHG or EHG like groups is pretty difficult to say, and whether it was from a late Mesolithic like CHG impulse or an earlier one.

People on here have been pretty critical of DATES, but looking again at the details of the method recently it is pretty hard for me to see how the methodology can be that wrong; when applied to Yamnaya it's looking at the correlation at locations across the genomes between probably CHG and probably EHG alleles, and how far apart those are, i.e. generally if they are present at longer distances, like 20cM, then the admixture time will have been more recent, whereas if they are only correlated hyper-locally like at <1cM, then the time is much further in the past. (Effectively this is based on applying a similar principle to local ancestry inference, but in a approximated form given the nature of ancient dna, and that they're not actually concerned with totally trying to call specific blocks for biomed purposes, just the summary stats). Hard to see how what they find could be the case under any different scenarios where it is much, much more ancient to the time, and there is not at least long term substructure in more extreme proportions than shown by Steppe_En. Likewise for notional trifurcation models between EHG-CHG-Steppe. (What is keeping that correlation together?) Perhaps they need to do more simulations, and maybe there are some simulated tree structures that do undermine the efficacy of DATES.

Davidski said...

@Matt

DATES won't tell you when EHG met CHG to form Yamnaya.

It'll just spit out a rough date of the last major admixture event in the genomes of the Yamnaya samples that was broadly similar to something like EHG and CHG mixture.

In that paper by Nick Patterson the authors used Steppe Maykop and Khvalynsk in the EHG reference pool, which probably tells us how precise DATES is, since these groups have a lot of CHG ancestry.

Steppe Maykop even has West Siberian ancestry.

Davidski said...

Also...

Heggarty could simply say "OK; well, perhaps one of those was the same that gave rise to the Steppe population, and the others are responsible for various other Near Eastern languages. Thanks for helping."

Well, he can't just keep talking shit, he actually has to prove this sort of stuff with genetics. And he can't, because it's not true.

Matt said...

@Davidski, my understanding is that the groups have to be correlated enough with the presumed true population, rather than necessarily unadmixed, for the method to work. (Any admixture uncorrelated with the populations simply reduces the efficacy of finding the signal, rather than biases it in any direction). So you would supposedly still get the same signal and time for EEF in Europe out of doing the European cline with Anatolian and Pitted Ware samples that have some level EHG and EEF in them. But they could certainly test this. Also the case that they have some large unpublished sets of EHG (at least there should be such between Volosovo and what is in Genomics of Stone Age Eurasia). So that could provide another check.

@EastPole: I cannot comment on the method. Let’s assume it is OK. Then the problem is with the data. 170 meanings is small. I suspect that with the set of different meanings or with a much larger set of meanings they would get different results.

Well, so, just to touch on this point, the point of their database is to find generally "basic" terms which reflect features of the environment and human experience which are common to human experience generally (at most, animals found in multiple biomes etc), and not to reflect things which could be recently acquired via contact or change frequently, particularly for prestige dialect reasons (so, e.g. "vomit" or "two", "short" "long", "man" "woman", but not "king"). The idea is that these would reflect deeper relationships, because they would generally lack motivated changes that reflect horizontal links, and they are also the ones which have generally been shown to have most persistence within Indo-European and across different language families.

Of course, yes, people can argue that a larger list of basic lexicon could be used, but it would need to be basic and it certainly couldn't really take a route of going "Now, here's all these things that could be loanwords or wanderworts, so please include those" (words like "horse", "gold", "wine" etc).

At some level it then becomes a statistical question of whether it is possible, within basic lexicon to keep expanding the list in a way where it's even possible to flip the structure of the tree, or if the statistical weight eventually becomes so strong it can be overturned.

I can't really say what this is, and it comes back again to me thinking that a problem with this literature, and a challenge that should be made, is that it is not really much backed by simulated data to set limits on possible it is for models to be uninformative about real rates of change variance and topologies.

Matt said...

@Davidski, yeah, it's true; at some point he would have to put some real genetic substance on what he is claiming. But I just thought that saying that IranN-CHG related ancestry in South Asia is lots of things rather than explained by a largely a couple waves (like say largely an Early Neolithic or Epipaleolithic one mix and then some more via Sintashta), that seems to give some opportunity to confuse matters and then push the responsibility onto geneticists to unpick all the supposed waves before he can be held to task to talk about it.

Davidski said...

@Matt

Well, like I say, there's not much substance in the genetic analysis in Heggarty's paper.

It's a fantasy world where the CHG/Iranian ancestry in Dravidian and tribal speakers is the same thing as the CHG/Iranian ancestry that arrived during the Persian era from the Iranian Plateau, because it lumps together well enough to look more impressive than the Sintashta ancestry in India.

Rob said...

@ Matt

“ Though, when the Steppe group formed from CHG or EHG like groups is pretty difficult to say, and whether it was from a late Mesolithic like CHG impulse or an earlier one. ”

But that’s not what steppe is all about
You’re probably referring to Piedmont, and even that is a 3- or 4-way admixture. A DATES bimodal estimate of CHG/ EHG wouldn’t necessarily even Piedmont steppe types ; let alone the more western variants have yet additional layers of ancestry all mixing since~ 5500 bc

Copper Axe said...

@Matt

You get wildly different dates then you get for Yamnaya on Khvalynsk and Steppe Eneolithic samples, which already shows the reliability of EHG/CHG mixing dates in Yamnaya because quite frankly these should be very similar to progress or Khvapynsk and they arent. If you get a 5000 bc sample from southern Russia you'd get different results too, because ongoing mixtures affect the dates put out by the program.

Ps I never really got an answer from Patterson or Reich why samples older than Yamnaya, particularly 5th millenium BC ones were not used when trying to calculate when "EHG" and "CHG" mixing events and I asked one on my blog (commented) and the other in real life.

Copper Axe said...

Also, I want to point something out about this CHG migration thing. It seems a lot of people are under the impression, academics included, that the russian steppe mirrors the more northern forest-steppe border in terms of ancient samples. We have neolithic EHGs, and then eneolithic CHG mixed samples. Conclusion? CHG must have migrated in Europe during or after the neolithic.

Shouldnt we try and prove this first before running DATES on chalcolithic/bronze agepopulations to calculate when this "migration" occured?

I think there should also be an explanation for the consistent finds of caucasus related lithic technologies in neolithic southern Russia or in some cases even the final stages of the UP if this is supposed to be pure EHG land until 5000 bc. Could it be trade? Perhaps, thats why we should investigate it rather than just leaving it to the side and form opinions on obviously limited data.

Matt said...

@Copper Axe, that's interesting; well, they should have a nice large set of Khvalynsk samples going on that's more robust than the three, so perhaps that will be something that they'll have to confront in published form. (Can't imagine they'll put out that Khvalynsk extended autosomal data without DATES!).

Rob said...

@ Copper Axe

“ I think there should also be an explanation for the consistent finds of caucasus related lithic technologies in neolithic southern Russia or in some cases even the final stages of the UP if this is supposed to be pure EHG land until 5000 bc. Could it be trade? Perhaps, thats why we should investigate it rather than just leaving it to the side and form opinions on obviously limited data.”

That would require reading, learning and other such efforts
Its easier to just invent some “stats”, appeal to “big data” and create a story

Rob said...

Although there’s nothing wrong with probative experimentation, it just shouldnt substitute a more ‘analogue’ synthesis

Davidski said...

@All

New Srubnaya samples...

https://drive.google.com/file/d/14JIMcvT_mUAraFYzByv0GsuD2VOC2Mqu/view?usp=sharing

From here:

https://www.ebi.ac.uk/ena/browser/view/PRJEB63318

Apparently, one of the samples belongs to Y-hg R1a-Y3. You know, the ancestor of tens of millions of Indians.

https://www.yfull.com/live/tree/R-Y3/

Yo, vAsiSTha, you OK? You little rat.

Matt said...

@Davidski, useful to see those so early.

Judging by my eyeball of the results, b10-2 and b28-2 along with tentatively b30-1 are outlying.

Vahaduo plots: https://i.imgur.com/SMFvPtI.png

New ones are labelled Nepljuevski after the burial ground.

There is data on the craniometry of these guys here - https://www.researchgate.net/publication/343945704_Late_Bronze_Age_anthropological_materials_from_the_Nepljuevski_kurgan_cemetery . It would be interesting if the outlying status was mirrored in any way in cranial shape.

Supposedly 25 of the samples are linked by relationships; we'll have to see if these outliers are linked in.

You would guess they could be some mothers from outside the main family sampled, but if that's the case I would expect some intermediate samples between them and the main cluster (unless they had no surviving children), but that would be surprising given the sample information. Alternatively if they were kids, their mothers aren't sampled.

Can teepean or someone establish the sexes of the individuals? In theory all those from the main family should all be from the same y-dna group, although there may not be enough markers in all samples to resolve this.

Matt said...

Although possibly b10-2 is an offspring of b28-2? b10-2 didn't seem intermediate enough to me, but maybe with some projection noise?

Coldmountains said...

B8-2 is the R1a-Y3 sample also one of the most Steppe Mlba like samples. Beside of another sample all other samples seem to be Q-L939
Target: b8-2:b8-2
Distance: 2.6184% / 0.02618355
97.8 Steppe_MLBA
1.8 Neolithic_Iran
0.4 Neolithic_Mongolia

StP said...

Wee , alias SKRiBHa said
I didn’t ask about your post. I just thought it was generally incoherent: oddly literal about centum/satem as a lexical item, &/or casting the classification itself as some nefarious project.
Ebizur answered nothing that I asked: he prefers his “explain ceud” comment to remain gnomic. Or incoherent.


In linguistics, we don't have 100 percent theories and data, because people, despite the patrilineality prevailing in the past, mix and groups rub against each other.
If you know that Ebizur answered incorrectly, write it yourself; how should he write!

vAsiSTha said...

@Davidski

"In fact, I would say that this CHG/Iranian signal is barely a genetic component.
It's actually just a bunch of shared alleles that look like a component because of the piss poor resolution in most of the analyses in these papers."
Bahahahahahaha. What a cope.

Btw, I am still waiting for the proof of your claim '10 migrations from Iran into India.' Maybe next time be more careful about fibbing about a region which you know nothing about.

Ygor C.S. said...

I am not knowledgeable enough to question the methodology and calculations, but I can certainly test the credibility of the results by counterpointing them with the evidences provided by decades or centuries of knowledge amassed by other sciences. An abstract mathematical method coupled with computational calculations can't and shouldn't trump all other forms of scientific evidence. For instance, did Heggarty et al. present a specific culture that would have spread to both the steppe, Turan, South Asia and Anatolia? Did they pinpoint a specific proximal source of ancestry as opposed to a generic and wide-spanning "CHG/Iranian" ancestry that by 5000 BCE was most probably linked to several different languages and even distinct families? Did they manage to explain why the huge CHG/Iran_N spread to the Levant, Mesopotamia, Arabia and Egypt failed to bring IE languages there? Did thy enlighten how exactly the IVC, BMAC, PC Steppe, South Caucasian and Anatolian cultures could be linked archaeologically? Did they present a plausible explanation to NW IE divergence predating the PC steppe peoples' expansion by several millennia? Can they really guarantee that the CHG/Iranian ancestry in Turan and South Asia came from as far as Armenia/Eastern Turkey/NW Iran? Can they demonstrate that by 5000 BC the South Caucasian genetic landscape was still basically pure CHG/Iran_N-like, not yet substantially mixed with ANF e Levant_N? Did they consider the possibility that a pre-historic language family spread mainly by heavy assimilation of foreigners following migration may have evolved (especially in lexicon) more rapidly and intensely than languages spread mainly by post-farming population booms with not as much admixture in most places, such as Sino-Tibetan, Bantu and Austronesian, or languages spread more recently in literate times and under fully organized and centralized states (Turkic)? There are many useful and plausible questions to be made.

Davidski said...

@Ygor C.S.

Yeah, good points. Similar to what I was saying.

I don't know how it's possible for a paper like this to be published in Science.

It's a fuck up, and all of the "big names" in ancient DNA, not just Wolfgang Haak and Johannes Krause, are to blame for allowing this to happen.

Where are David Reich and Nick Patterson? Shouldn't they be trying to enforce some standards?

Kouros said...

@ Matt the outliers are interesting


distance: 1.52
sample: b10-2
Sintashta MLBA: 66.4
Dali EBA: 19.2
Dzharkutan1 BA: 14.4

sample: Custom:b28-2
distance: 1.6
Sintashta_MLBA: 53.2
Dali_EBA: 26.8
Dzharkutan1_BA: 20

sample: Custom:b30-1
distance: 1.860
RUS_Afanasievo: 61.4
Sintashta_MLBA: 38.6

sample: Custom:b30-1
distance: 2.2961
Yamnaya_RUS_Samara: 90.4
Globular_Amphora: 9.6


The Y3 has the Standard Steppe MLBA form

sample: Custom:b8-2
distance: 1.973
RUS_Afanasievo: 67.2
Globular_Amphora: 32.8

JR said...

@davidski

You shouldn't get excited so fast right now.
This isn't the first time we have got something like that,and which didn't turn out to be wrong.

We had a supposed "Y3" from fedrovo too but later after months it turned out it has false-positive calls and wasn't UDG treated so the corrupted strands were giving false positive calls.

(Pic rel)
https://files.catbox.moe.v7y6xm.png

And remeber the supposed 'Y3' from 3500 BC Sredny stog,Alexandria which you guys were using for 2 years to prove that PIE and IR originated in stedny stog? And when 2 years later,the Reich lab corrected the dates and calls and it Turned out to be a 1700 BC fatyanovo sample,you all went silent?

So,I will wait if I was you before dancing on my toes by finding '1' sample which could turn out to be very well, contaminated,low coverage,false-positive calls,Non UDG treated or misdated.

"millions of men belong to this"
Irrelevant since millions of men,in areas where it's concentrated belong to sintashta/mlba mtdna Like
T1,T2,U4,U5,U2e.

Sintashta mtdna is like 15-20% in North West and North india so I don't know why this matters,you should read some studies.

Many balts belong to n1c yet I don't see them speaking a finnic language,same with basques and r1b.

And I don't think it matters anyway since indians get their steppe Ancestry around 600 BC(from DATES and ALDER by even narasimhan 2019) through 900 BC TKM_IA profile which was 50% Andronovo 50% BMAC east iranic and couldn't have brought indo Aryan langauges to south asia,and that in 6000 BC.

Also, Andronovo was 100% Homogeneous for z2124 which is found below 3% in N.India so I dont know what the "supposed" Y3 in srubnaya means here(which could very well turn out to be contaminated,low cov,Non UDG treated,etc).

You will have to show Y3 or something like that in andronovo, since indians get their steppe Ancestry and mtdna from 50% Andronovo derived tkm_ia not from srubnaya or some other group.

There's only 1 wave of steppe Ancestry in india,this is confirmed by runing dates and looking at decay curve and can be confirmed by other means too.
And that wave comes around 600 BC from tkm_ia,at a time when pure unmixed andronovo didn't exist anymore.

And you have to prove that Y3 was mediated by steppe and not born in india from some minor z93 that reached there and got broken down in south asia and increased in frequency later due to population turnovers and bottlenecks from climate change, droughts,mauryan military expansion,etc.

Poznik 2015 dated the expansion of L657 in south asia around 2600-2300 BC at this time no srubnaya existed.

The only L657 from ancient dna in south asia from roopkund lake belonged to a tribal-pallan like profile and had no sintashta/steppe so this proves my point that L657 was associated with a south asian profile early on and didn't differ autosomally from others like L1a,J2b,etc.

The high steppe samples from that study belong to J2,no r1a.

Long Post,I end here.

Looking forward to some ad hominems and misrepresentations from you.



JR said...

@Ygor Coelho/C.S

They will elaborate on archaeological connections and Proximal sources Later,This is Just the beginning.

Vasistha already has shown archaeo records of arrival of Wheat varities and some other crops to Turan and IVC region around 4000-3500 Bc alongwith Levant_n and Anatolian Ancestry.

There's Levant_ppn and anatolia_n Ancestry in a 1:1 ratio in steppe and that's the reason of placing south Caucasus as a source since such a ratio only existed in the Neolithic of Armenia and
North West Iran.

So, I don't know why they should prove 'Pure CHG' existed in south Caucasus since they are actually emphasizing the South Caucasus homeland because of the levant_ppn Ancestry and ratio in steppes.

If you are honest and without any sides,you would give more time, since kurganists were given 30-40 years and they came up with hoaxes like the dyadhanyc horse Head Hoax and placing Andronovo/mlba as a source of Aryans in south asia despite no andronovo pottery found there,despite 40 years of research and quests.

And a failure in genetics too since qpAdm rejects sintashta_mlba direct as a source in rotating models and only accepts 850 BC tkm_ia.
Yet, nobody cared about this and everybody was singing about "muh sintashta Aryans"

Neither do the dates of steppe admixture match as a source of "Aryans from andronovo" since andronovo Ancestry reaches india proper only 600 BC as found from DATES / ALDER on various indian groups and this is too late for the Vedas and Aryans.

Yet,this didn't matter to narasimhan 2019 and Never even saw you either mentioning this, despite this being an obvious contradiction.

Kouros said...

@ Jerome alias Vashishte

Is it required for you to make additional account to spew same nonsense ordure because you have made enough of a buffoon of yourself as Vashiste? If I recall you always stated Y3 has some origin in India , but this sample is death knell for this logic , David has got you by the balls this time.

Davidski said...

@Jerome

This isn't the first time we have got something like that,and which didn't turn out to be wrong.

There's more Y3/2 on the way, including from Abashevo.

Of course, this makes perfect sense, considering all of the upstream markers are in closely related Eastern European populations.

So quit fooling yourself.

Poznik 2015 dated the expansion of L657 in south asia around 2600-2300 BC at this time no srubnaya existed.

Ah, you're a total newbie.

Their estimate was based on modern data, so obviously they couldn't date the expansion of L657 in South Asia.

Their date, which was very rough because it was based on modern data, was just for an expansion somewhere. And obviously we now know that this expansion started in Eastern Europe.

Davidski said...

@Jerome

EHG is 14,000 years old and it's not something u can trace PIE back to.

No one here said anything about EHG dumbo.

The classic Yamnaya genotype formed around 4,000 BCE in Ukraine.

Matt said...

Surprising that the main family at Nepljuevski seems to have Q-L939 (per ColdMountains). Do you guys think this is more likely to have introgressed into the generally Sintashta like ancestry within the Southern Ural region (patrilocality)?

Matt said...

The alternative is that Q haplogroup was pretty common back where these groups migrated in from, or along the path? Or it was present but at low frequency in those areas, and you get some expansion linked founder effect?

Rob said...

A simple enquiry into whether we can distinguish betweewn CHG & Iran_N

1) Turkey_N
Turkey_Epipaleolithic
Georgia_Satsurblia.SG
Jordan_PPNB

best coefficients: 0.645 0.164 0.191
tail prob 0.004

2) Turkey_N
Turkey_Epipaleolithic
Iran_GanjDareh_N
Jordan_PPNB

best coefficients: 0.727 0.107 0.167
tail prob 0.09

right pops:
Cameroon_SMA
Morocco_Iberomaurusian
China_Tianyuan
Russia_MA1_HG.SG
Russia_DevilsCave_N.SG
Italy_North_Villabruna_HG
Serbia_IronGates_Mesolithic
GEO_Dzudzuana_UP
Israel_Natufian_d
+ rotating CHG/ Iran_N

Davidski said...

It's easy with PCA.

https://eurogenes.blogspot.com/2019/12/a-final-note-for-year.html

They're different populations, just as expected based on where they lived.

It would even be possible to split CHG and Iran_N sub-populations from each other with PCA given enough samples.

Virgin_Quilles_Sucks_R1a_Chadvski said...

It wasnt Something that we can Say only by Genetics, Indeed by archeological Research. IVC was Iran Neo+ Minor AASI(Vedoid+Minor ANE?) and It was well seen by several studies.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6800651/
Elamites, Kassites and IVC also Had similar archeological culture.
https://www.jstor.org/stable/2742643

I suggest One more time more interdisciplinare studies, Not only stats Models(that could been misunderstood)

Virgin_Quilles_Sucks_R1a_Chadvski said...

https://www.researchgate.net/publication/330430547_Elam_a_bridge_between_Ancient_Near_East_and_Dravidian_India_Language_Change_and_Cultural_Transformation

A little bit funny cause i Said that some people ignore non-statistical points when analyzing Historical Facts, genetics should help Uus, but without archeology, History, It could produce abasurdism .
Genetics also suggest that IVC(Pakistan/India Not TKM) came with Iran Neo+AASI, archeology and linguistics suggest that Elamites and IVC we're connected.
Also ZMF and Mesopotamian shouldnt been as Far as much From those Sort of people.
Funny imagine How a Matriarcal and Matrelinear Society Magically Became Patriarcal when they ""invaded "" Steppe, Lost their sedentarism and Been Cucked by Patriarcal EHG R1.
Funny imagine those sort of Things.
No chance of PIE Homeland beyound Black or Caspian Sea. No such Kind of Culture was practice on Iranian Plateau Düring Neolithic, they werent nomads, also should been Elamites-IVC Like , example:Kassites, what WE call Iran Neolithic Cultures.
Probably South Steppe was more CHG Like and It explain why Progress took a Better source for Indo Anatolian(that came Fron Caucasiano Route).
Simple, CHG-Like in Progress was an Early Mix between EHG and Dzudzuana-rich pops , simple to understand.
Science was the Most simple,Not the Most improbabile. Imagine thinking that the Argument of Absent based on stats should refute thousands of Researches on different areas.

Virgin_Quilles_Sucks_R1a_Chadvski said...

Sorry If i m flooding Mr. Davidski, but i Just want to explain why those dudes were comiting a terrible misunderstood.

https://www.academia.edu/63932715/Proto_Elamo_Dravidian_The_evidence_and_its_implications

https://dl.acm.org/doi/abs/10.1145/3331076.3331116

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5882685/

https://scholar.google.com.br/scholar?start=10&q=elamo+dravidian+culture&hl=pt-BR&as_sdt=0,5&as_ylo=2019&as_vis=1#d=gs_qabs&t=1692094330936&u=%23p%3DKPtVb4stRm0J

Eastern Iranian Farmers werent exctaly Like Westerns cause they Shared more Genome with AASI , less Basal and more ANE, bit no doubts they shared same cultural connection, Not a single chance of "PIE Homeland" between both, Elamites were, in Facts, partially Eastern Iranian Farmers(that were the Same south Caspian Basin without extra Basal/natufian like zagros?), So there were no argument for refuse the Elamite-Dradivian Group, imagine that PIE should cane from Farmers of Iranian Plateau or Caucasus Not from Nomads. Funny

Anonymous said...

Elamo-dravidian doesn't form a group....As for Indo aryan evidence pre 2000bce...there are scattered evidences that are often ignored....

Wee e said...

@StP
“Wee , alias SKRiBHa said”
— WTF??? I have one account here, I post under this ID and that’s it. Which I daresay Davidski can confirm, if he were remotely interested.

“In linguistics, we don't have 100 percent theories and data, ”
— Who is this “we”? Certainly you’re no linguist. I’m not sure you even understand what a theory is.

Since you resort to ad hominems and baseless accusations of deception, rather than argue your case on its merits, and since most of what you say is literally incoherent anyway, I see it was a mistake to try to take any of it seriously.

Matt said...

OT: https://www.nature.com/articles/s41467-023-40198-w - "The genomic history of the indigenous people of the Canary Islands"

Rob said...


@ Asuwatama

''Elamo-dravidian doesn't form a group....As for Indo aryan evidence pre 2000bce...there are scattered evidences that are often ignored....''

Which are those ?


@ Jerome

''The krasnoyarsk_BA_kra001 profile is very old,probably since 7000-5000 BC,''

That's not true





Virgin_Quilles_Sucks_R1a_Chadvski said...

I posted 3 articles Here and 3 before, you could Accept or not, IT Is a theory that you cant reject, either If you Accept or not both those cultures werent Indo European and there werent any evidence of Indo European cultures on the Iranian Plateau.
You said but not a single article rejects it, only those with bias made by stats nit archeology, linguistic, etc..

MaxT said...

In their previous linguistic study from earlier this year they plotted Eurasian, African, American language families on PCA based on their similarity.

"We cluster languages by similarity which shows that most clustering is driven by related groups of language families:"
https://twitter.com/SimonJGreenhill/status/1648748620387176449

Indo-European family forms it's own distinct cluster. While Uralic and Turkic plot somewhat close together. It's pretty interesting study.

Virgin_Quilles_Sucks_R1a_Chadvski said...

Elamo-Dradivian Theory had fundamental Not on Genetics but Material Archeology and Linguistics, however Not a single article could reject It with proper Interpretation, Iranian Neolhitic wasnt homogeneous but had a similar profile, shared with Mesopotamian Sumerians , Akkadians, etc.. people that were consider Semitic.
Either It was true or not, The cultures between them couldn't been Indo European. Both sides of Iranian Plateau Had non Indo European Like society, both sedentary (Not nomad or semi nomad Like IE), different Pottery, graves (Not a single evidence of Kurgans), people that trust on Southern Arch simple have never seen a single article about Archeology or Ignore It.
If you read Gibuntas or even some Julius Evola you Will See that several cultural aspectos were exclusively Seen on Indo Europeias societies and wasnt seen before It or in Others at that time. For example Patriarcal Organization, Warrior's Elites (Nobitity ), Religious Organization(Sky Father), Männerbund (concept that could bene Seen Like a viking Männerbund), Wolf's Cult, Kurgans Graves, Horse Domesticazione, NOT A SINGLE ASPECT OF IT COULD BEEN SEEN BEYOND THE CAUCASUS.
If WE Accept Southern Arch It means that the original PIE wasnt patriarcal and they where cucked by R1 EHG, funny at leas. Science , by Empirical Method, Always prefer the simple, so PIE homeland should been at Pontic Steppe

Without any dort of DNA model we can prove It, Charleston Coin made It on 50's years using physical Anthropology, Gibuntas indeed.
It was the Most accepted even before the Haak et al 2015.

Davidski said...

@Assuwatama

Elamo-dravidian doesn't form a group....As for Indo aryan evidence pre 2000bce...there are scattered evidences that are often ignored....

Ah, OK, if you say so, but obviously Elamo-dravidian makes much better sense based on the genetic evidence presented in Heggarty et al.

Haha

Virgin_Quilles_Sucks_R1a_Chadvski said...

If those guys COULD find Kurgan's Graves, Horse Bones, evidences of Patriarcal Organization, etc.. BEYOND the Caucasus or on the Iranian Plateau before the Bronze Age or IE Expansion from the steppes, them they COULD prove Something, either there were a migration From Caucasus/Iranian or not(i think that Not) before PIE expansion, PIE Homeland was on Steppe, cause those migrante werent the PIE(i have no doubts about It).
Imo Progress was a EHG Mixed eith Dzudzuana-Like Pre CHG , similar to CHG, and those People formed a continuum between EHG and CHG. They blend with Eneolithic Steppe cultures frim Samara, Dnieper Donestsk, etc.. and Neolhitic substract and them formed PIE on late Sredny Stog(III- Pre or Early Corded) and Yamnaya. Simple, Like Empirical Method suggest to us.

Virgin_Quilles_Sucks_R1a_Chadvski said...

It Is funny imagine why Neolhitic Farmers from CAUCASUS, Iran or Zagros COULD Return to Nomadic Lifestyle and BEEN literraly cucked by a bunch of Cave's Males from Steppe(been PIE= Patriarcal, probably Southern Arc Defenders suggest that Non Paternal Events happened so R1a and R1b were introduce by cheat lol).

Rob said...

Actually R1 steppe males have very little Iran N female mediated ancestry
Basically zero

The I2a chiefs further west on the other hand have it all; Volga-steppe, EEF and Majkop females all at once
Poor blokes must’ve been tired

Rob said...

But seriously virgin boy, grow up. Just because you think a model is wrong you shouldn’t project that onto ancient populations

Anonymous said...

Archaeologically people like Victor Sarianidi has abundantly shown BMAC to be indo-iranian...no need for steppe....Same with sites in India that show no linkages to steppe cultures in late bronze age...having steppe brides and minor steppe male lineages in swat is no garuntee that IE spoken in those parts in Iron age came from steppe...

Anonymous said...

There were meluhan colonies in southern Mesopotamia since the akkadian times....Some meluhan names are attested...Place and Personal names in sumerian myths and so on....

Anonymous said...

Haplogroup H and AASI rich populations are the most likely candidate for dravidian languages...

Haplogroup J2 R2 L1a2 etc aren't related to dravidians....These haplos got incorporated into dravidian speakers as IVC and IVC+TKM_IA rich people got absorbed into Dravidian groups....which also includes several groups that are rich in R1a-z93

Davidski said...

@Assuwatama

It's not widely accepted that Dravidian is an ASI language. It's more likely that it arrived in South Asia relatively recently.

Indeed, some Dravidian speaking groups have the highest ratios of CHG/Iranian ancestry in South Asia.

They also lack Sintashta ancestry.

Davidski said...

@Assuwatama

Archaeologically people like Victor Sarianidi has abundantly shown BMAC to be indo-iranian...no need for steppe

LOL

Anonymous said...

Some dravidians have steppe ancestry especially those of higher and middle castes....Some ruling families had ties with saka in NW India and migration of Buddhists Jains and Brahmins from Northern India post 600bce to South India....

Being shifted towards certain ancestries means nothing.....we have groups in North India who are AASI shifted with least steppe yet speak indo aryan....

Literacy can change the whole dynamics of the language spread...

Modern spread of languages and ancestries can give wrong conclusions....

Anonymous said...

Some people argue for wheel imprints as a sign of indo iranian yet grand fire temples don't prove Indo iranian nature of BMAC....

And horses are tradable....You just need a better offer....

Davidski said...

So the massive expansions in Abashevo/Srubnaya Z93 lineages in South Asia were associated with which language group?

And keep in mind that in traditional archeological and linguistic models before ancient DNA Abashevo/Srubnaya were seen as Indo-Iranian speakers.

Or is that just a remarkable coincidence, and you have a better explanation?

Anonymous said...

Modern data can be misleading....
This massive expansion needs past and new data as a support....

First find site A with 50 samples from 2000-1500bce with no R1a and then the same site with a spike in R1a in next 100 years...

I am more inclined towards R1a being reproductively more successful than large migration of R1a in late bronze age...though I am pretty much in favor of iron age expansion of R1a males in India....

Abashevo burials, Metals and ceramics are missing from Indian sub continent....

Even if we assume a group of Y3 men migrated from there....it was of insignificant nature....

Davidski said...

Right, so what you're really saying is that you hope Abashevo/Srubnaya R1a isn't associated with Indo-Aryan languages in South Asia.

Just so that you can keep pretending until there's no more hope when more ancient samples roll in.

Anonymous said...

I am fact based man...
If data supporting steppe origin comes in, I will switch camp...As of now the evidence isn't conclusive enough for me.

I would rather wait for more data and I have no beef in it. Large number of Indians and Eastern European men share common decent....We should rather cherish that bond instead of being supremacist/racist about it...

Davidski said...

Well obviously these descendants of Abashevo people in South Asia weren't mutes.

They had to speak something, and it was something Indo-European.

So you really don't have much wiggle room to explain them away, and there's nothing supremacist/racist about pointing that out.

Anonymous said...

Obviously it depends on the nature of migration....and nature of interaction with groups in India...

As I have stated earlier....
Andronovo most likely spoke IE....It's interaction with BMAC like group resulted in a newer group called scythians who spoke a variety of attested and unattested languages....A TKM_IA group who could have spoken Avestanian or proto-bactrian language was half sintastha.....

But in iron age there was one expansion after another of steppe groups and iranics into India....Turkish expansion Mongol expansion brought linguistic changes followed by adoption of Persian by these Turkish groups in steppes....

So the process was pretty dynamic...Who knows maybe some of the deities in Indo-Iranic lore originally belonged to these groups in the steppes....

Coldmountains said...

@Assuwatama
Show me a signal Y-DNA match between BMAC and modern Indians dated with TMRCA around 2000-3000 B.C like R1a-Z93 in Sintashta, Srubnaya, Abashevo and Andronovo. Such matches not even exist yet between modern Iranians and BMAC. Sure BMAC influenced Indo-Iranians and surrounding groups in IAMC and the Iranian Plateau but it already lost it importance when Indo-Iranians arrived. It were rather pastoralist IAMC groups that influenced Indo-Iranians and mixed with them. I am also sure that much of the non-IE looking terms shared between Iranic and Indo-Aryan come rather from these groups then from BMAC.

Rob said...

@ Assuwtama

''Some people argue for wheel imprints as a sign of indo iranian yet grand fire temples don't prove Indo iranian nature of BMAC....''

neither of these arguements-are in themselves convincing. People used reductionist approaches of 'lets track horses', or chariots, for tracking IE.


But what I asked for is the earliest evidence of Indo-Iranian in western & southern Asia, not Siriandi's opinion about BMAC. There are the Mittani names, but that has often been over-interpreted.These were cultural terms used by elites, and names adopted upon accession to the Hurrian throne. Before that, the men had Hurrian names

Rob said...

''There were meluhan colonies in southern Mesopotamia since the akkadian times....''


This is the Middle Asia interaction sphere. However the linked image is overly BMAC-o-centric. It was criss crossing trades and colonies across
I dont think it takes too much imagination to believe some form of Elamite & Dravidian interaction

EastPole said...

@Assuwatama

„Abashevo burials, Metals and ceramics are missing from Indian sub continent....”


Don’t you have bronze spiral 'spectacle' pendants in India?

https://upload.wikimedia.org/wikipedia/commons/9/9c/Kultura_%C5%82u%C5%BCycka%2C_lusatian_culture_16.JPG

They were common in Central Europe and in Fatyanovo, Abashevo, Andronovo, Srubnaya.

JR said...

@davidski

The roopkund lake paper had 13 samples and only 2 of them had r1a.
One L657 derived and one Z645.

This is a frequency of 7-14%.
https://haplotree.info/maps/ancient_dna/slideshow_samples.php?searchcolumn=Country&searchfor=India&ybp=500000,0&orderby=MeanYBP&ascdesc=DES


R1a in south asia is as relevant to indo-aryan langauges as much as n1c in balts and Russians is relevant to balto-slavic langauges.

P-IIr can't be from steppe because IIr does not undergo the common agriculture innovations which affect all Yamnaya derived and Sredni Stog/Tripolye affected European languages (Kroonen et al).

Lack of shared agriculture terms/innovations reject your theories and heggarty's paper showing that balto-slavic and indo-iranian don't form a clade and indo-iranian splits earlier puts a nail in the coffin too.

So,I don't know what evidence do you have that Indo-aryan/iranian came from steppes considering that steppe Ancestry only reaches south asia 600 BC through  eastern Iranic tkm_ia(50% BMAC,50% Andro) and the vedas were written in the bronze age,not iron age.
The z2124 of tkm_ia/andronovo is barely found at 2% in india and Y3 barely makes more than 25% either.

It dosent matter much anyway since Steppe_mlba mtdna like u4+u5+T2,etc are itself 10-20% in south asia,almost proportional to the r1a.
(Source-negi et al 2016,Singh et al 2021)

And the Y3 could be a founder effect through population turnover from climate change,etc because it's barely 10% in the roopkund lake samples
Both swat valley and roopkund lake YDNA support this.
Later intra-population Founder effects and bottlenecks don't show the original frequency at which the Haplogroup was mediated.

You will have to find samples from South Asia to prove that high amounts of r1a was mediated through steppes(which is proven wrong by both swat valley and roopkund lake samples),so I have two evidences to support my side while you have none except insults.

550 BC medes hasanlu samples from iran lack any sintashta autosomal or r1a either,so I don't on what terms and evidence you are even shilling all this.

JR said...

@davidski
"Srubnaya"

There's no evidence of srubnaya migration to south asia or even central asia
Srubnaya was restricted to Ukraine and there's no evidence of srubnaya migrating to central asia,let along south asia,only andronovo was the one which reached asia.
Srubnaya isn't ancestral to anybody neither did it leave any succesors,it was replaced compeletely around 700 BC by incoming scythians/sarmatians from central asia(who were mixed with BMAC,Slab grave,WSHG,etc).

DATES/ALDER results from 3 different sources(narasimhan 2019,vasistha) show that steppe ancestry reached india around 600 BC when srubnaya didn't even exist and qpAdm rejects srubnaya

1100 BC burzahom sample from north india leaked shows no steppe Ancestry,so this confirms that steppe Ancestry must come later 100 BC and since dates/ALDER confirmes at 600 BC,it's sure that steppe Ancestry isn't from andronovo proper but from 50% BMAC 50% Andronovo tkm_ia which only passes the source in rotating model,not pure andronovo or srubnaya.

Where's the evidence of srubnaya pottery,textile and archaeological items in south asia?
What evidence do you have that steppe Ancestry reached south asia from "srubnaya" in 1500 BC?

You have found 1 unconfirmed sample out of 21 In srubanya,rest are z2124.

Do you know the ratio of z2124:Y3(2000% in srubnaya) dosent match with that in south asia?
Do you just shut your logic down when focusing on your ethnosupremacist arguments?

You just keep in repeating the same thing over and over again as it would become true,and to fool laymen.

Andronovo indirect(through 50% BMAC tkm_ia) is the source of steppe Ancestry in south asia.

And andronovo pottery isn't found in south asia either,never.
What appears is around 800-600 BC,some ceramics of the Yaz Culture,to which tkm_ia belongs and qpAdm only accepts tkm_ia too.
Tkm_ia,an Iranic can't be a proto indo-aryan right,in 600 BC?

And if steppe_mlba was indo iranian then why do udmurts,who are 50%+ sintashta/fatyanovo/mlba,
speak a uralic langauge and carry n1c mostly?

Target: Udmurt
Distance:2.5779%/0.025778
54.8 RUS_Sintashta_MLBA
19.0 RUS_Krasnoyarsk_BA
11.4 RUS_B_Oleni_Ostrov
7.6 RUS_Karelia_HG
6.8 Kura-Araxes_ARM_Kaps
0.4 Yamnaya_RUS_Samara

Where's the evidence of iranian or Aryan substrate in their langauge?

"Abashevo"

Abashevo can't be a source either,the abashevo culture meets a sudden end around 1800 BC and is replaced.

Show evidence of migration of abashevo around 2000 BC when ivc was in full swing through anthropoloical,genetical and archaeological records.

Those supposed "y3" samples aren't even out yet,we don't have their BAM files,and you already jumping.

You realise the fact that those could be contaminated or not UDG treated,low coverages with false positive calls?

Its not like this hasn't happened before,it has happened,and 3 samples in the past which were supposedly Y3 turned out to be contaminated,misdated or false positive calls.

1 from Fedorovo & others from fatyanovo and Ukraine Alexandria MBA as seen below.

https://files.catbox.moe/vdqiqe.png
https://files.catbox.moe/v7y6xm.png

Call me again when those Srubanaya and abashevo Y3 Samples get released with BAM files.

You shouldn't use unconfirmed samples dishonestly to prove ur BS.
Please wait until they are released before making arguments based on them,or you will have a
surprise pikachu face similar to as when the supposed z93 of sredny Stog was rectified and your whole theory of PIE sredny stog based on z93 came crumbling down.

Rob said...

What's still hard to reconcile is the EEF-rich Fatyanova folk in the East and EEF-poor 'early Corded' groups in central-northern Europe.
How did the criss-cross occur ?

Davidski said...

CWC mixed with GAC in Ukraine.

Some groups went north and west before this happened, but not the ancestors of Fatyanovo.

Copper Axe said...

"And if steppe_mlba was indo iranian then why do udmurts,who are 50%+ sintashta/fatyanovo/mlba,
speak a uralic langauge and carry n1c mostly?

Target: Udmurt
Distance:2.5779%/0.025778
54.8 RUS_Sintashta_MLBA
19.0 RUS_Krasnoyarsk_BA
11.4 RUS_B_Oleni_Ostrov
7.6 RUS_Karelia_HG
6.8 Kura-Araxes_ARM_Kaps
0.4 Yamnaya_RUS_Samara

Where's the evidence of iranian or Aryan substrate in their langauge?"

Terrible model aside, are you not aware of the hundreds of Indo-Iranian loanwords in Uralic languages, including those of the Udmurts?

Here is a good overview:
https://helda.helsinki.fi/items/e5900ba0-5f7e-460f-aec6-3d0db63414e4

As to why they have so much steppe_mlba yet retain their N1c and Uralic languages, female mediated geneflow into a patrilocal society often leads to a retention of the paternal language. Arya being a root word for a term forenemy/slave/servant/southener in several Uralic languages might have something to do with it...

William Anderson said...

@Davidski

"The classic Yamnaya genotype formed around 4,000 BCE in Ukraine"

If this is the case, then why are all the samples so distant from Yamnaya?

Distance to: Yamnaya_RUS_Samara
0.05268483 Yamnaya_RUS_Caucasus:RK1007__BC_3155
0.05271505 RUS_Progress_En:PG2004__BC_4138
0.05912030 RUS_Progress_En:PG2001__BC_4900
0.06072636 RUS_Khvalynsk_En:I0434__BC_4975
0.06748704 UKR_Deriivka_En:ukr104__BC_3614
0.07013417 UKR_Cernavoda_I_En:KTL001__BC_3423
0.07092262 RUS_Vonyuchka_En:VJ1001__BC_4280
0.08395017 RUS_Steppe_Maykop_o:AY2001__BC_3447
0.09347388 RUS_Khvalynsk_En:I0122__BC_4838

Anonymous said...

There is L1a or L1a2 in a 4600bce pre BMAC culture of Turkmenistan....

This is found in abundance in India both north and south....quite high % among Jats who also score quite high on steppe...No idea when it entered India....could have entered India with iranN like ancestry or later with steppe ancestry....

Orpheus said...

@EthanR The "robust farming economy" is a strawman. Corded Ware is responsible for the agricultural shift in European IE languages, having a full blown agricultural economy with animal husbandry supplementing it is irrelevant, and not required.

"Fatyanovo isn't a late CW culture, and we have samples from 2700-2600BC."
It is due to their autosomal DNA which is present only in late CWC individuals, who had already shifted from their non-agriculture heavy economy to a more agricultural one hence the agricultural shift in European IE languages. Moreover Fatyanovo appeared where they were from the west and is intrusive (Nordqvist & Heyd, 2020) and specifically from Central Europe (Grigoriev 2022). This is basically ground zero for CWC agri-shifted languages, this is simply what the data shows

"As is Indo-Iranian, I'm not sure what your point is."
That it is less agri-shifted and thus not related to CWC. If Fatyanovo did not descend from CWC or at the very least did not have 30%+ EEF admixture and came from mainland Europe, then sure they would be a decent candidate for I-Ir. But they're not, the other options are all better and all satisfy the requirements for I-Ir as mentioned in Kroonen et al 2022
Also from Kroonen et al 2022: "However, Indo-Iranian typically does not participate or only marginally participates in the semantic shifts that characterize the European branches."

"Also, the more accurate way to put it is that all European IE languages are agriculture-shifted, whether they come from Corded Ware or something else (Greek)."
Agreed

@Davidski It's not my problem you're lazy and don't look up things
Bader 1937, Kranyov 1972 have already mentioned Fatyanovo practicing agriculture besides Saag. Their origins makes it completely clear too.
also Mallory & Adams 1997 for Abashevo

"There is no ancient DNA evidence of any early R1b in South or even Central Asia"
There is in ancient Iran, and R1a in Central Asia is far more strongly associated with Tocharian speakers. See Kumar et al 2022 for much more massive impact on autosomal and also haplogroup in 2250 to 0BCE Xinjiang. 34/85 samples are R1a, so unless you want to claim that Sintashta spoke Tocharian, your own logic proves you wrong here, and thus the only one resorting to special pleading is you apparently

"Heggarty's model is experimental and it'll be debunked just like his earlier work was debunked"
Heggarty's model alongside all the traditional linguistics who worked on it seems completely fine and gets pretty much every single traditional date right. Its problem is a couple of groups (but these already had a low p probability anyway, 0.26 for example)
Nothing will actually happen. Several of your opinions are simply proven wrong over time, apparently, because you hold preconceived notions that fail to be supported by actual research. That's your problem

Imagine being so invested in a literally inconsequential for real life hobby. Maybe your actual hobby is getting mad at aDNA research haha

Anonymous said...

R2a is another prominent haplogroup among Brahmins...oldest R2 is 8000bce...but question is when and in which form it reached India...There were R2 in BMAC as well as in swat which has BMAC admixture....so question is who is source and who is sink....

archlingo said...

Saturday, August 12, 2023
We're dealing with a bunch of [insert preferred insult here]:

"The Heggarty et al. paper is pure fluff. It offers nothing useful or even remotely interesting."
< I fully aggree. Heggary is a stubborn denier of Indo-European, and the Court Jester at the MPI Leipzig. Thanks for this and all the other comments! archlingo

Orpheus said...

PS. Sintashta indeed has no evidence for agriculture. Several works support this (Anthony et al 2005, Anthony & Brown 2007, Epimakhov 2010, Ryabogina & Ivanov 2011, Ruhl et al 2014, Hanks et al 2018)
Needless to say what this shows about Sintashta and Indo-Iranian, a language that has 4-5x the agricultural vocabulary Tocharian does (something like 2 IE words with ambiguous agricultural meaning whereas in I-Ir it is guaranteed). I-Ir isn't a pastoralist language, not undergoing the same agri shift European IE languages did and instead being peripheral to them when this happened allows it to have its own agricultural vocabulary as well.
Won't go into how 2000 BCE for Indo-Iranian being unsplit also doesn't work (and is required for a steppe entry).

Ok now to the current blog.
"For instance, the authors derive some Indo-European languages in Europe from Anatolian farmers and others from Caucasus hunter-gatherers"
Not at all. There is no pure CHG or pure ANF involved, ergo no actual Caucasus Hunter Gatherers or Anatolian farmers, and this was clear from Kroonen et al 2022 anyway (PIA is not a HG language or a pure farmer language). Their argument is that the same mixed ancestry as in the Southern Arc paper is the source of proto-Indo-Anatolian. Even their (weak imo) hypothesis that Balkanic languages spread via Anatolia is about such a mixed population.
I'm not really sure what the arguments are against Heggarty's dates (which is the important thing here), I haven't see any actual counter-arguments to the paper's calculation of dating splits, besides a couple of non-sequiturs

@Copper Axe "I think there should also be an explanation for the consistent finds of caucasus related lithic technologies in neolithic southern Russia or in some cases even the final stages of the UP if this is supposed to be pure EHG land until 5000 bc. Could it be trade?"
Lazaridis et al 2022 already pointed at an early migration of "pure" CHGs which would have happened probably before 6000 BCE. Probably just migration to look for resources, I don't think that at that date there were any significant trade networks or compelling trade goods that would facilitate migration due to trade, in that specific region

@Rob See my previous comment. It looks like Fatyanovo is a CWC offshoot that expanded eastward from a mainland location initially, and wasn't isolated.
Their autosomal and haplo profile was an indication since day one tbh, if they were some early (and thus possibly isolated) CWC they'd be mostly R1b and without 30% EEF admixture.

Orpheus said...

Forgot
@Rob Isn't Catacomb a dead-end wrt any Balkanic language? They didn't have agriculture, to the point where Kroonen et al 2022 identifies them as possible Tocharian speakers because of this, unrelated to any Euro IE language
Chronologically it's a no-go too. Not "difficult to happen" but completely improbable. For example Greek is a pretty late Balkanic language, separates from Greco-Phrygian which itself split from Greco-Albanian, which itself separated from Greco-Armenian/Balkanic. The commonly accepted split date for Greek from Greco-Phrygian is 2000 BCE, some even say 2200-2100 BCE. Heggarty correctly calculates it at 2000 BCE, and Catacomb is 2500 BCE at earliest. So the argument becomes that within 500 years between proto-Balkanic and Greek with several languages splitting and becoming unintelligble with each other inbetween of those 500 years assuming Armenian was the first to split at ~2500 BCE. Pretty amusing proposition, and shows the value of interdisciplinary research since if we went blindly by DNA only then it would erroneously seem feasible.

RRN said...

Ötzi genome has been revisited (paper released today August 16), and it seems that previous genome was contaminated. Will you redo his G25 coords and add them to the G25 datasheet?

Rob said...

@ Orpheus

''Isn't Catacomb a dead-end wrt any Balkanic language? ''


Dont understand your point. Random people have tried to link Greek with Cataocmb, but I have disagreed with proposition because Catacomb never enetered south of the Danube
proto-Greeks draw dual ancestry from Usatavo & Yamnaya, mixed in Bulgaria / Romania. By 2500 bc they left Thrace and entered northern Greece. So the Greco-Phrygian split began d c. 2500 but the populations might have remained in contact

''Heggarty correctly calculates it at 2000 BCE''

It's not correct. Linguists do not belong in the world of science, they need to stay in their lane- provide supplementary clues via traditional linguistic decipherment.
The key is archaeogenetic reconstruction of populations, by competent people




''See my previous comment. It looks like Fatyanovo is a CWC offshoot that expanded eastward from a mainland location initially,''

Yes we know that, my point was its high EEF
- might warrant a revision of EEF-less early CWC, suggesting heterogeneity within early CW-profile
- provides clues about migratory paths of early CWC

Rob said...

The Greco-Armenian link & split can be placed near the Moldovan part of the steppe, deriving from the younger (Yamnaya) layer of steppe ancestry in proto-Greek and Catacomb (which derives from Yamnaya) origins of proto-Armenia. Specifically, proto-Armenians seem to link with a western variant of Catacomb or post-Catacomb, due to the prevalence of I2c.

What 'word statistics' can ever negate an empirically sober analysis ?

EthanR said...

I'm not sure what the point of bad faith deflecting is. There is nothing about Fatyanovo's economy that makes it incompatible with it being a source on the way to Indo-Iranic.
You seem to have several odd hangups about this. Corded Ware continues into the late third millennium BC, Fatyanovo begins in the early third millennium. Nordqvist & Heyd:
"The beginnings of Fatyanovo date to the period marked by a plateau in the calibration curve (c. 2900–2600 cal BC )... the datings imply that the development of Fatyanovo was not significantly later than the CWC further west. Its end has previously often been placed somewhere in the 2nd millennium BC, but termination over the final centuries of the 3rd millennium is a more realistic option"

Nordqvist & Heyd also distinguishes in figure 11 Fatyanovo from central European Corded Ware groups, who we know went deep into Europe (Germany, Czechia and descendant cultures eventually beyond).
Grigoriev links Fatyanovo to southern Poland.
You seem to be fixated with the idea that the more admixture with GAC (a profile blanketing a large territory as David implies) follows necessarily more agricultural activity and lexicon. I don't see any reason to take this for granted, particularly given that we have little evidence for it in the subsequent Fatyanovo, Abashevo, Sintashta etc.


"Corded Ware is responsible for the agricultural shift in European IE languages"
Greek is a corded ware language?

@William Anderson
Distance to:RUS_Yamnaya_Samara
0.04534211 UKR_Sredni_Stog

Davidski said...

@Orpheus

There is in ancient Iran, and R1a in Central Asia is far more strongly associated with Tocharian speakers. See Kumar et al 2022 for much more massive impact on autosomal and also haplogroup in 2250 to 0BCE Xinjiang. 34/85 samples are R1a, so unless you want to claim that Sintashta spoke Tocharian, your own logic proves you wrong here, and thus the only one resorting to special pleading is you apparently.

Well, obviously, the R1b in northwestern Iran is not linked to Indo-Iranian.

Indo-Iranians entered Iran and India via Central Asia.

See that's why there's also a lot of evidence of Indo-Iranian presence in Xinjiang and surrounds, so I don't know why you even brought that up?

Ever heard of the Wusun, Saka, Kangju, or Tocharoi?

https://www.britannica.com/topic/Tocharoi

Haha.

Davidski said...

@Jerome

And if steppe_mlba was indo iranian then why do udmurts,who are 50%+ sintashta/fatyanovo/mlba, speak a uralic langauge and carry n1c mostly?

Where's the evidence of iranian or Aryan substrate in their langauge?


Congratulations, you've just discovered why Udmurts are Uralic speakers and that there's a lot of Indo-Iranian and even Indo-Aryan influence in Uralic.

By the way, no one ever claimed that Srubnaya migrated to India.

The point I was making was that Y3 was found in a Bronze Age Eastern European who was said to be early Indo-Iranian speaking even before ancient DNA.

That, and the fact that Z93 and Z94 also don't show up in Asia until Srubnaya and related groups move there, means that tens of millions of Indians have ancestry from Bronze Age Eastern Europeans who were likely early Indo-Iranian speakers.

Haha.

Copper Axe said...

@Davidski

To be fair I think the linguistic proposals for Indo-Aryan influence in Uralic are kind of weak, in Holopainen's article most proposed IA etymologies are explained through Proto-Indo-Iranian or Iranic. What seems secure is Proto/early Indo-Iranian > Proto-Iranian > East Iranian languages in terms of linguistic influence on Uralic.

My take on this is that the soundchanges that fully define Indo-Aryan and separate it from Iranian/Nuristani didnt occur in eastern europe or west siberia, but rather somewhere in SC Asia, probably on the eastern (IAMC) side. This would be a result of Proto-Indo-Iranian tribes spreading rapidly across the regions, and dialects forming after.

Proto-Iranian then developed somewhere in the vicinity of the Urals which explains why this layer is significantly present in European and Siberian Uralic languages (even Finnic), and we see migrations from this area towards SC Asia as well as the Altai in the archaeological record explaining the spread of Iranian languages in the later bronze age.

"Haha."

I've been here for years but this still gets me every single time.

Rob said...

Do any of the samples we have approach anything related to variants: the Tazabagyab, Zerafshan, Vakhsh/Bishkent, the Tahirbay and Molali cultures ?

archlingo said...

Orpheus,
"separates from Greco-Phrygian which itself split from Greco-Albanian, which itself separated from Greco-Armenian/Balkanic. The commonly accepted split date for Greek from Greco-Phrygian is 2000 BCE, some even say 2200-2100 BCE."
< NOTHING is "commonly accepted" here. Genes and pots cannot Speak, and Bayes computations are guesses with a VERY WIDE range of scatter/error. For the split of the hypothetical Balkan languages, Bouckaert et al. (2013!) got around 4000 BC, Holm around 3000 BC. You have to give your reference. "Some say" is nothing. BTW, Kroonen is a good linguist, not more. And Heggarty is a good phonologist, and nothing more.
archlingo

Copper Axe said...

@Rob

The Kokcha samples are from a Tazabagyab cemetery. No Vaksh-Bishkent samples as far as I am aware. I might be wrong but the same for the Molali and Tahirbay groups/complexes.

Rob said...

@ Copper Axe

Thanks



@ archlingo

“Genes and pots cannot Speak”

This is a vapid cliche

Coldmountains said...

@Assuwatama

And? L1a is something like 15.000 years old and L1a2 something like 8000 years old. Other said R1a in CWC Eulau or CWC Sweden is closer to R1a in Sri Lanka than L1a2 in BMAC or Hasanlu to L1a2 in Jatts, Kalash and so on. I always like how you guys mention that there is no Y3+ in Sintashta yet but on the other side think L1a2, R2, J2 and so in BMAC, Shar-I-Sokhta or Hasanlu which split from Indic around 5000-15.000 years ago is a close Y-DNA match. Z2124 in Sintashta which split around 2500-3000 B.C from Indo-Aryan L657 is still a closer Y-DNA match to modern South Asians than anything found in Hasanlu, BMAC and even Swat_IA (excluding communities in the northwest near ancient Gandhara)

Coldmountains said...

@Jerome

B8-2 is positive for AMOO483 (Y26/M780 C to T) and AMOO480 (F2597/M727/Y3 T to C). So positive for both snps on Y3 level but negative for R-F1417 and R-M605 downstream of Y3. He has no contradicting calls afaik and is positive for M417. So quite reliable results and his coverage is relatively good (around 1-1.5x)

Survive the Jive said...

mumu Heggarty, mumu Max Planck Institute, mumu Laziridis etc.

*perkʷūnos thunder strike their whole village

Matt said...

Re; Fatyanovo, I can't see anything too much in the Ringbauer's IBD data that would suggest that they had to have forked before other CWC linked groups forked from each other, or any signal of Ukraine_GAC vs Poland_GAC - https://imgur.com/a/hqqnaMZ

It doesn't seem like they couldn't just be groups that forked after admixing in Central Europe at an early time, but then perhaps the group had slightly more admixture than average for their time and then went east.

Perhaps this would be contradicted in some subtle way though.

William Anderson said...

@EthanR

"Distance to:RUS_Yamnaya_Samara
0.04534211 UKR_Sredni_Stog"

What samples make up that average?

@Davidski

Could you please answer my question?

Coldmountains said...

@Jerome
Also B8-2 is positive for Z94 and negative for Z2124. His Y3+ calls also have multiple reads so quite surely Y3+ and hopefully he will be added to Yfull. His coverage should be enough and he surely will be added to FTDNA and theYtree too.

Matt said...

OT: https://www.nature.com/articles/s41559-023-02143-4 - "A genetic history of continuity and mobility in the Iron Age central Mediterranean"

(long time preprint now published, but as far as I know the genotypes were already available?).

Andrzejewski said...

According to a new research paper by Krause, it appears that Ötzi was darker skinned than Cheddar Man:

https://amp.cnn.com/cnn/2023/08/16/europe/otzi-the-iceman-physical-traits-scn/index.html

Matt said...

Andrzejewski, that doesn't seem quite correct; the paper is here (supplement - https://ars.els-cdn.com/content/image/1-s2.0-S2666979X2300174X-mmc7.pdf) and they did not calculate a PRS for skin pigmentation for Cheddar Man, but only for Iceman, Loschbour, LBK Stuttgart and a present day Sardinian person.
Darker is Loschbour -> LBK Stuttgart -> Iceman -> Present day Sardinian. But the EEF group are all closer together than any of them are to the WHG. (Table S11 - https://i.imgur.com/ZaKQnZ3.png)

"The weighted genetic score of dark pigmentation in the Iceman is estimated to be 0.591, higher than the score of present-day southern European populations taking Sardinians as an example (Table S11), which the Iceman shares closest genetic affinity to (Figure S1) and which represent the highest level of pigmentation among modern-day European groups, although it is lower than the score of ancient LBK farmers and the Luxembourg_Loschbour.DG hunter-gatherer (Table S11)."

Comparable PRS scores from Ju and Mathieson's 170 SNP panel from their paper (used by Krause and collaborators for Oetzi) can be seen here in Figure 1 B and C (https://pubmed.ncbi.nlm.nih.gov/33443182/#&gid=article-figures&pid=fig-1-uid-0).

Oetzi fits reasonably well into the EF set in Fig 1C there, who are lighter than HG, but not so much as following Bronze and Iron Age populations.

There is no real revelation about Otzi or EF groups in this paper, it's the same kind of findings that are already out there.

Rob said...

@ Assuwatama

“Haplogroup H and AASI rich populations are the most likely candidate for dravidian languages...”


Yhg H is from the Middle East and has little to do with AASI other than a later founder effect into AASI-rich populations from South Asia

Davidski said...

@William Anderson

None of the populations in your distance calculations are expected to show a close genetic distance to Yamnaya, so I'm not sure why you asked that question?

Virgin_Quilles_Sucks_R1a_Chadvski said...

Androvo Horizon clearly came from Sintasha, that came From Fatyanovo-Balantovo Groups of Corded Ware.LoL It was knew since the XX century, also Vedas and Archeology always pointed to Bronze Age emergence from Indo-Aryan populations. IVC was clearly Dravidian by Archeology, also went further south after Aryans invaded from north(Android Horizon). Every people that said the oppose was literally bias-based

Virgin_Quilles_Sucks_R1a_Chadvski said...

Dr. David another two questions, what do you think about the Tripolli incursions to Steppe just before R1b-M269 went there? As I noticed only another branches of R1b, R1a and I2 were found on steppe before it. We known that Balkan HG and Neo was heavily R1b influenced, so could r1b-m269 being related to Neolithic Balkans? I m thinking about a possibility, it would crush quilles feelings.
Imo it could explains why Mesolithic was R1a and Early Neo R1b-M269 , with R1a surviving at Dom River(with other r1b branches) and North Pontic Caspian.
Other hypothesis that I consider is Rus Progress ( partially CHG) people went south via Caucasus and brought Indo Anatolian cultures.
Indo Anatolian was the most distant group of linguistic Indo europeans, it could easily being a para-Indo European branch. Considering EHG was the first Indo European speakers , it could being brought to any steppe source or even earlier.
Why can't we assume that CHG wasn't one Dzudzuana Rich population+Minor EHG elements? Saying that , how they could say that Anatolia has no EHG DNA at the bronze age? It is Very unlikely that Iran Meso was the proxy for CHG or Iran Neo Zagros, both could easily being Dzudzuana Rich(Para-Epigravettian basal mix) with some EHG strain . I doubt any ANE-rich went Caucasus before Eastern Euro considering mammoths steppe Horizon.

Mononokee said...

This paper is one of the most insolent atrocious garbage i saw and ithe absolute proof there is a bias.
It literally places the divergence date for celtic and germanic languages arround 5000 BC lmao. It goes against every single piece of archeological data, genetic evidence and Linguistic research we have

Davidski said...

@Virgin

Both M269 and M417 are probably from Sredny Stog in Ukraine.

It's hard to say where they were before that. I'm waiting for the samples from the new Allentoft paper and a new paper on the Neolithic in the Russian forest zone for some clues.

Apparently, there's pre-M269 in some of those forest samples. But we need to see the raw data.

Mononokee said...

Btw david, is DCP1 (ANE dna recovered from a pendant on the denisova cave) avaliable for g25?

Davidski said...

I posted the preliminary coords fr DCP1 somewhere in the comments here, but I don't know where.

Someone at Anthrogenica should have them, maybe Teemu?

The official DCP1 coords will end up in the G25 datasheets when the genotypes are included in the AADR dataset at the David Reich Lab.

Davidski said...

I don't know what SA6010 represents exactly. This sample might be wrongly dated.

If not, then it would seem that Progress-like people survived on the North Caucasus steppe until the Yamnaya period, and may have been incorporated into Yamnaya groups there.

Yamnaya and Afanasievo are definitely from much further west than the North Caucasus steppe, and probably from the North Pontic steppe in Ukraine.

Rob said...

Piedmont steppe ancestry obviously made it all the way to Armenia, and there is R1b in Arslantepe. but Im getting major Balkan signals in western-central Anatolia. its on a basis of persistent Ukr_N signal in the F stats which made me test it out.



Isparta_EBA
Bulgaria_C
Azerbaijan_Caucasus_lowlands_LN


best coefficients: 0.354 0.646
std. errors: 0.036 0.036

Tail Prob 0.07


simply using old barcin_N Fails. Using the new Usatavo/Cernavoda genomes do not pass either, nor does Piedement/ Yamnaya.


Yes, Isparta_BA men were j1, but that doesn't matter because Anatolia was mixed and different story to, say, bronze Age Britain. We would need samples from the Citadels

pRight
Cameroon_SMA
Morocco_Iberomaurusian
China_Tianyuan
Russia_MA1_HG.SG
Italy_North_Villabruna_HG
Serbia_IronGates_Mesolithic
Turkey_Epipaleolithic
Jordan_PPNB
Georgia_Satsurblia.SG
Iran_GanjDareh_N
Ukraine_N
Hungary_MN_Vinca
Russia_Sidelkino_HG.SG
Turkey_TellKurdu_EC
Turkey_N

Dranoel said...

@Davidski

A bit on a different topic. In recent DNA studies of the Wielbark culture, a guy from Z2103-CTS7822 was registered in Pruszcz Gdański.
Could you take a look at it? In your opinion, does the aDNA indicate his Gothic affiliation?

On a certain forum I met information where it was defined as "half Slavic, half Germanic", which in my opinion is not correct. What's your opinion?

Regards!

Dranoel said...

* the sample I wrote about has the number PCA0475 - sorry, I forgot to write it right away.

Matt said...

@Andrzejewski, to continue my comment from yesterday, if you go with what I linked above from Mathieson and Ju's comparable panel using the 170 SNPs (https://pubmed.ncbi.nlm.nih.gov/33443182/#&gid=article-figures&pid=fig-1-uid-0) that Krause et al used to predict Otzi and Loschbour's pigmentation risk score, then West Africans score 0.76, South Asians 0.67, present day Europeans 0.5. (East Asians are likely not accurately recorded or identified there).

Slot in the estimated PRS from Table 11 from Krause paper then Loschbour (typical for HG) probably about as dark as South Asians (in 1000 Genomes) or Native American groups but not as much as Africans, EEF farmers about the same as Sardinians and intermediate between the present 1000 Genomes Europeans and South Asians.

Also Otzi is about the same (perhaps slightly lighter even) than the average for the EF in Mathieson's, and Loschbour within the same as HG.

Unlikely to be 100% accurate, but probably indicative, at this point of genetic research of this trait.

So overall this paper tells us nothing new about pigmentation in HG or EF, and seems sensationalized by the media a bit. (HG varied from the west of Europe where about South Asian/Native American with blue eyes to the east where dark eyed and a bit darker skinned than present day people but lighter than EF; EF slightly darker than present day Sardinians but not much). They probably found more variants in the rest of Otzi's genome that will contribute to medical genetics, but things about appearance are easy to get the media to write stories about, even if we're not learning anything new...

Matt said...

@Andrzejewski, as one last thing on my tangent about the Otzi paper and pigmentation, even though in Table S11 they give a relatively dark PRS score comparable to the present-day Sardinian sample, note that in their Table S10 they still give find that he has the derived (lighter) variants at SLC45A2 rs16891982 and SLC24A5 rs1426654.

Some quick comparisons of his variants at these against ancient and current world populations - https://imgur.com/a/H042Dcq

So all in all, he has a overall score similar to other EEF, and he has the derived variants at the two sites that most distinguish present day Europeans from ancient and from other populations (and which not all EF and early Steppe people had, and certainly not all HG). So there is no reason that should dramatically change any vision that we would've had that he was simply like a typical EF person, and that EF people were probably slightly darker skinned than closest present day relatives, but not much. (The results seem somewhat hyped, and I doubt that his skin colour in life was actually closer to the mummy's current colour than the previous reconstructions that have been done for him).

Davidski said...

@Dranoel

PCA0475 looks Danish or Dutch.

ph2ter said...

PCA0475 similarity map:
https://i.imgur.com/DylJ1Fq.png

Bernard said...

Interesting reading:

https://www.helsinki.fi/en/researchgroups/the-yamnaya-impact-on-prehistoric-europe/news/just-released-steppe-transmissions-the-fourth-volume-of-the-monograph-series-the-yamnaya-impact-on-prehistoric-europe

https://www.cambridge.org/engage/coe/article-details/6382645cebc1c78cdcd3e4da

https://www.researchgate.net/publication/365805692_Long-distance_exchanges_along_the_Black_Sea_coast_in_the_Eneolithic_and_the_steppe_genetic_ancestry_problem

Gedrosia said...

What is the individual (single) ancient samples that would represent the following component at the optimal admixture proportion?
Western European hunter-gatherer (WEHG)
Eastern European hunter-gatherer (EEHG)
West Siberian hunter-gatherer (WSHG)
East Siberian hunter-gatherer (ESHG)

MaxT said...

New linguistic study by Max Planck again..

Societies of strangers do not speak less complex languages
https://www.science.org/doi/10.1126/sciadv.adf7704

Fig. 3. The scores of fusion and informativity on the global tree.
The scores with a minimum of 0 (absence of all metric features) and a maximum of 1 (presence of all metric features) have been standardized to a mean of 0 and a variance of 1. We detect many patterns of closely related languages scoring similarly, which might indicate the faithful transmission of grammatical complexity from ancestor languages to their descendants rather than large-scale adaptations of grammatical complexity to changes in sociodemographic factors. Similar to geographic distribution, we see that fusion scores follow a more defined pattern of phylogenetic clustering compared to informativity scores.


Matt said...

@Gedrosia, it may depend on the sample quality.

Davidski said...

@Samuel Andrews

You might want to put this in your video.

Alexey Nikitin got it wrong, because the scientists at the David Reich Lab got it wrong.

That's why it's so important for the scientists at the David Reich Lab to get these basics right.

The EHG-CHG genetic admixture was documented in the communities of the Khvalynsk cultural complex of the middle Volga region (Fig. 1) and the Seredny Stig (Sredny Stog) culture complex (SSCC, the Oleksandria site, Figure 1) in the lower Don/Donets basin from the last two thirds of the 5 th – early 4 th millennium BCE (Mathieson et al., 2018, 2015; Wang et al., 2019).

https://www.cambridge.org/engage/coe/article-details/6382645cebc1c78cdcd3e4da

As I explained, in reality, the samples from Khvalynsk show the admixture between two Eastern European groups, one from the steppe and the other from the forest zone, not between EHG and CHG.

https://eurogenes.blogspot.com/2023/01/dear-david-nick-iosiflets-set-record.html

Dranoel said...

@ Davidski and ph2ter

Thank you very much for the info!

music lover said...

The basic truth here which is also completely supported by the genetics is that all of Eastern and Western Europe, owes almost the entirety of their cultural, linguistic and genetic origins to Iranian hunter-gatherers, who were not from Europe.

Gedrosia said...

@Matt I'm guessing you meant by "sample quality" is the Autosomal SNP count, What your thoughts about the following list

EEHG: Eastern European hunter-gatherer
I0061 (Russia) (Coverage 1025148 SNP)
I0124 (Russia) (Coverage 477876 SNP)
I0211 (Russia) (Coverage 146885 SNP)

WEHG: Western European hunter-gatherer
I0001 "Loschbour" (Luxembourg) (Coverage 1139327 SNP)
I0585 (Spain) (Coverage 1079840 SNP)
I1507 (Hungary) (Coverage 796242 SNP)

ESHG: East Siberian hunter-gatherer
irk075 (Russia-Lake Baikal) (Coverage 377023 SNP)
irk071 (Russia-Lake Baikal) (Coverage 372088 SNP)
irk076 (Russia-Lake Baikal) (Coverage 42207 SNP)

WSHG: West Siberian hunter-gatherer
I1960 (Russia-Tyumen Oblast) (Coverage 805830 SNP)
I1958 (Russia-Tyumen Oblast) (Coverage 31796 SNP)

Davidski said...

@music lover

You're projecting your own deep insecurities here.

All of the R1a-M417 and R1b-M269 in Iran is ultimately from Eastern Europe.

It arrived along with the people who would have a profound impact on the language, culture and genetics of the modern Iranian peoples.

Rob said...

Aside from arrival of steppe lineages, there were significant turnover of original Iranian male lineages by distantly related people from the southern Caucasus & eastern Anatolia.

The corresponding genome-wide shifts have been discussed here before

Davidski said...

music lover = frustrated comedian

R1a-M417 and R1b-M269 are Eastern European paternal markers that spread across West, Central and South Asia like a rash during the Bronze and Iron Ages.

On the other hand, there was zero impact on European genetics by Iranian hunter-gatherers.

MyeDin said...

@music lover

You needn't project what is true for india to what is true for europe.

Matt said...

Title revisions following peer review.

Wee e said...

@Copper Axe. “Where's the evidence of iranian or Aryan substrate in their language?"

There is a ton of IE in specific cognate areas in Udmurt and some other Uralic languages. I would have expected it to be balto-slavic, and some seems to be — but a wedge of very basic vocabulary is more Iranic in form — or something more southern or eastern than Russian anyway.

The very word Udmurt is an example — “murt” (which turns up a lot) human/person, rather than “smert” as you’d expect of a more recent slavic influence.

But I don’t know as I’d call it a substrate. Maybe a lexical residue of contact?

IE forms are the majority in basic cognate sets like pronouns, names for humans & family relationships, question-words and here/there, this/that kind of lexemes; and then a good sprinkling across nouns of common things like body parts, animals, sun/moon, and food/trade items like honey and salt that are wanderworts anyway. It covers phrasebooky “His name is... Where is the small dog? She is my wife,” topics.

Ygor C.S. said...

@Jerome and/or @vAsiSTha

--- "They will elaborate on archaeological connections and Proximal sources Later,This is Just the beginning."

Then let's wait. So far not convinced at all. Multiple holes and dubious assumptions.

--- "Vasistha already has shown archaeo records of arrival of Wheat varities and some other crops to Turan and IVC region around 4000-3500 Bc alongwith Levant_n and Anatolian Ancestry."

If you say so, I can only find it intriguing that it's so hard to get basically any non-negligible Anatolian ancestry (that is not clearly linked to and embedded in Steppe_MLBA ancestry) and any (negligible or otherwise) Levantine ancestry in almost all South Asian aDNA and modern DNA samples using any plausible combination of Global25 coordinates. Anatolian ancestry in most South Asians (modern and ancient samples) basically correlates with steppe ancestry (hence something CWC-like), except in northwesternmost groups such as the Balochi and Pashtuns.

Ditto for Levant_PPN ancestry in the steppe samples, even the Eneolithic Steppe Piedmont of Progress and Vonyuchka, much closer to the presumale Transcaucasian PIE source (heck, even Anatolia_N ancestry is very minor in most steppe aDNA samples not derived from CWC or the WESTERN - not southern, let alone Ciscaucasian - Pontic-Caspian zone close to GAC, CT and the Balkanic Neolithic cultures in Romania and Bulgaria). When you find more substantial gene flow from Caucasian, Anatolian and/or Levantine groups, it's rather more in the western Black Sea (Balkans), such as in Usatovo samples, not anywhere close to the Caspian Sea or the Caucasus.

Usually strong and even not so strong signs of gene flow and admixture are captured just fine by Global25 models of ancestry and PCA charts.

It seems then that the PIE language spread with an even much smaller migration than the one assumed in the Aryan Migration hypothesis, especially in South Asia, and without any significant influence on the Y-DNA lineages (come on, one really got to have a lot of blind faith to believe R1a-M17 and R1b-L23 have more to do with Neolithic Armenia than with Neolithic Eastern Europe), despite an obviously very patriarchal and patrilineal society. Interesting, very interesting.

Or then, alternatively, the PIE-speaking population in Transcaucasia would have had a huge dominance of CHG/Iran_N admixture with very little Anatolia_N and especially Levant_N, which should be proven is something that still existed by circa 5000 B.C. CHG-like admixture in the steppe is much more unlikely to have never existed until the arrival of people from south of the Caucasus than to have existed in the big cline from EHG-proper in the forest zone to mostly CHG in the North Caucasus, spreading gradually northward since even before the Neolithic period (which was not driven by CHG folks south of the Caucasus, as is clearly shown by autosomal changes there).

(continues)

Ygor C.S. said...

(continuing)

If PIE was spoken so close to Mesopotamia and the Levant, it should also be reasonably explained why no indications of Anatolian IE words exist before around 1900-2000 B.C., and even much later for Indo-Iranian words (and already after the spread of steppe admixture to parts of West Asia and Central Asia in the latter); and why no IE language ever spread and evolved on its own in the Levant, Mesopotamia and North Africa (rather than being brought there much later, as Persian and Greek), areas that clearly received a lot of gene flow exactly from the NW Iran-Transcaucasia-Eastern Anatolia region, where decidedly non-IE Hurrian, Urartian, Mannean etc. were later found.

The authors of Heggarty et al. also need to explain how the PIE divergence in their Hybrid Model happened millennia before the spread of steppe ancestry to the rest of Europe (note: "hybrid model" means they assume the steppe peoples had a major role in the spread of IE families, except Indo-Iranian, Anatolian and Tocharian). Would we be dealing with a fast and huge level of linguistic divergence (encompassing most of the known branches of IE? They don't make their hypothesis clear enough in that regard, but, again, it would be really surprising, because by 2,500-3,000 B.C. the IE speakers would have to be speaking very divergent languages and even language subgroups.

--- "And a failure in genetics too since qpAdm rejects sintashta_mlba direct as a source in rotating models and only accepts 850 BC tkm_ia.
Yet, nobody cared about this and everybody was singing about "muh sintashta Aryans""

Well, probably because nobody REALLY believes that Sintashta, let alone the specific and obviously partial population sample that is labeled Sintasha_MLBA, is REALLY the DIRECT AND MOST PROXIMATE source of steppe introgression into South Asia. It's even too early to be likely to be a direct source. Sintashta_MLBA is assumed to be a Proto-Indo-Iranian culture, not a specifically Proto-Indo-Aryan group, let alone the specific Proto-Indo-Aryan culture that expanded to South Asia. It's just very closely related and hence a good enough proxy.

TKM_IA is basically just what happened after Sintashta and Srubnaya-derived steppe groups met and mixed extensively with Neolithic Turan people (several BMAC outliers show that was already happening not long after 2000 B.C.).
There is also very extensive archaeological/anthropological evidence that BMAC had a major influence on the material and immaterial (e.g. religion) of the pastoral peoples that moved to Turan, as expected from a populous and much more complex and sophisticated culture...

Just like the Turks and Mongols that moved to Turan millennia later were in many cases almost wholly changed by Persianization and formed profoundly Persianate states, even though they kept speaking Turkic languages, and such materially Persianate and Islamized societies even spread Turkic languages further, BUT with a far more Islamic and Persianized than Proto-Turkic Northeast Asian culture, e.g. Seljuks in Anatolia. Noticed any pattern?

So, again, a strongly steppe-related immigration is proven to have happened. Now, I hope Heggarty et al. can hope a strongly Neolithic Armenia/NW Iran/Azerbaijan-related immigration did really happen to South Asia, South-Central Asia, Southeastern Europe and the Pontic-Caspian Steppe during the Neolithic, with a common ethnogeographic source.

Ygor C.S. said...

(finally ending... lol)

@Jerome and/or @VaSisTha

--- "Neither do the dates of steppe admixture match as a source of "Aryans from andronovo" since andronovo Ancestry reaches india proper only 600 BC as found from DATES / ALDER on various indian groups and this is too late for the Vedas and Aryans. Yet,this didn't matter to narasimhan 2019 and Never even saw you either mentioning this, despite this being an obvious contradiction."

Well, that's likely caused by what I have read even geneticists (not amateur geneticists, profissional ones, indeed) saying they are pretty convinced of: that those estimations are only reliable enough when there was one single pulse of admixture or when there were several pulses of admixture cumulatively, in which case it will only be showing the approximate date of the last significant occurrence of gene flow from that source being analyzed. If there was a succession of admixture events, you won't know when it began, but when it was last relevant.

So what we are actually seeing is probably that the last major entry of steppe ancestry with a long term impact on the South Asian genetic structure happened around 600 BC. That was not necessarily the only nor the earliest pulse.

A similar scenario was clearly seen in those estimates of Levant-related introgression in the peoples from the Horn of Africa, estimated to have occurred about 3,000 years ago. Oops, but aDNA samples dating to millennia before 1000 B.C. and from even more to the south of the Horn of Africa, clearly involving Horn African-like people from the same major north-south expansion of pastoral peoples, already had a big chunk of Levant_N/Natufian admixture and were even surprisingly very close to modern Horn Africans in genetic ancestry composition.

Those unexpected things do happen, folks.

music lover said...

The steppe has 0 impact on Anatolia, and therefore EHG ancestry or any ancestry seen in any part of Europe cannot be any source of IE languages or culture and of course genetics there. On the other hand Iranian HG ancestry has a significant impact on the genetic ancestry in Greece, the earliest classical civilization of Europe upon which the entirety of western civilization is based. Moreover this Iranian HG ancestry is seen in the formation of the Yamnaya - the major civilizing force - that brings poor pastoralists of the steppe together and gives them a culture. This ancestry then permeates and dominates all of Northern and Western Europe.

The contrast with India and Iran couldn't be more stark. The BMAC are hardly even interrupted by the Andronovo pastoralists. There is hardly a sample over a 800 year period where they are neighbors with any more than a whisker of steppe ancestry. When the ancestry finally arrives, it is a minority ancestry in the SPGT, seen arriving at the earliest times as a female mediated process.

The high proportion of Y chromosomal ancestry is easily explained either by (a) drift (b) events that occur thousands of years later due to the latter success of an indic kingdom.

If you claim otherwise - I challenge you in your next blog post to:
a) find a single sample in Ukraine before 4000 BCE that has the correct ancestry components to fit as the Iranian component in the Yamnaya.
b) find a single sample in Anatolia that has significant proportions of steppe ancestry prior to 3000 BCE. In fact I challenge you to show that Anatolia has any ancestry from progress2/eneolithic piedmont
b) find a single BMAC sample with an R1a/b Y chromosome of the Yamnaya/Andronovo type
d) find evidence of steppe ancestry proportion larger than seen in Italy or Greece in any region of Iran or India.

Sometimes it's silly how terrible @davidski and the bloggers on this forum are. It explains why most of them aren't professionals - they wouldn't get a single paper past peer review.

Here is a challenge: either make a new blogpost answering any of the claims above or stop blogging.

MyeDin said...

@music lover

You are insolent, and are simply parroting various southern arc talking points and claims from cranks which even the researchers you cite with disagree

The idea that the more advanced region must project it's influence into less advanced areas is foolish when the middle east and south asia were themselves subjugated by countless originally poorer or less advanced peoples looking for riches or glory, it's unjustified chauvinism motivated by oikophobic western researchers like Heggarty. I'm going to enjoy seeing the smackdown put on you here.

Davidski said...

@music lover

Right, so you actually noticed that in many parts of Asia, like India, Iran and Armenia, Y-haplogroups from Eastern Europe are often at very high frequencies, while the autosomal impact from Eastern Europe is relatively much lower.

Well done, because this is very important, but obviously you don't have a clue as to why this is so.

It's because the demographic impact from Eastern Europe was male mediated.

This is the classic signature of the Indo-European expansions, because the early Indo-Europeans were patriarchal and patrilineal, and both culture and language in their societies were transmitted by fathers.

It's really a scandal that this hasn't yet been discussed in any papers from the David Reich Lab or Max Planck, because it's the essence of why Indo-European languages are now spoken in India, Iran and Armenia. And it's also a good pointer about how to find the proto-Anatolian signal in Anatolia.

Instead, your professional geneticist role models have been obsessively counting percentage points of autosomal components, and making themselves look like a bunch of amateurs.

By the way, the cat is already out of the bag in regards to Yamnaya being derived from Sredny Stog in Ukraine.

The sampled Sredni Stog populations included individuals who autosomally resembled Yamnaya a millennium before the Yamnaya culture appeared.

https://reich.hms.harvard.edu/sites/reich.hms.harvard.edu/files/inline-files/10.1515_pz-2022-2034.pdf

The fact that you're still not aware of this isn't surprising, since you're not the sharpest tool in the shed.

Tryormaster said...

Different groups are different in terms of advancement in different fields. People who spent too much of time and resources on temple building and literary works will get subjugated by war mongering nomads who has comparatively advanced warfare tactics.
Simply using the term "advanced" for generalisation is not good.

Dranoel said...

@Davidski

One more question about Z2103. We currently know the Z2103 from Central Europe from the following data:
a. CWC Czech Republic,
b. BB Southern Poland,
c. Iron Age - Wielbarska culture, northern Poland,
d. Middle Ages - Merovingian cemetery Alt Inden; Poland Groszowice,
e. early modern - Ellwangen cemetery.

We see the presence of Z2103 in Germanic populations - both in the Iron Age and in the Middle Ages and modern times. How do you think Z2103 got into this group? Since PCA0475 has ADNA indicating Danish/Dutch origin, when could this Y dna get there?

Have a peaceful Sunday! :)

Davidski said...

@Dranoel

I can't give you a definitive answer.

Czech CWC is a likely candidate though, especially since U106 was also found in Czech CWC.

These markers probably made their way to southern Scandinavia at some point, and then expanded back south with Germanic tribes.

Andrzejewski said...

I’m still confused about recent “peer reviewed” studies re: Tarim Basin mummies.

Are they descended from our WSH Eastern European ancestors, from Turanic WSHG or from Paleosiberian admixture of East Eurasians and WSHG?

As far as I’ve always known, Ur David, Chechen Man, Beauty of Lulan and other renowned mummies date to 1,800BCE. Has Reich lab at MIT BROAD challenged this concept too?

Wee e said...

@Copper Axe
Sorry, I got confused there and attributed to you what you were quoting of Jerome’s. Just ignore.
@Jerome
See my misaddressed reply to Copper Axe if you’re interested in a reply about IE in Udmurt.

Samuel Andrews said...

@Andre,

The Tarim mummy DNA samples do not descend from WSH.

They're very similar to ancient western siberians.

But they aren't from western Siberia either.

They are native to central Asia. In other words they were from western China. They were native to the Tarim basin.

Samuel Andrews said...

Indo-Europeans migrated into the Tarim Basin later.

The later Tarim mummies are probably part Indo-European.

But no DNA samples from later Tarim mummies have been sequenced.

Rob said...

@ Sam

If by western China you mean Tarim basin, then that's technically not Central Asia.
Either way, the Tarim basin R1b-PH155 is no way native there. This is yet another screw up by Max Planck.
They do all these fancy tests to write a story, but do the the very basics - actual evidence- very poorly

- little to no pre-Bronze Age habitation in the Basin
- absence of R1b in pre-Bronze Age sample set of Inner Asia

I'd say R1b-PH155 is from further northwest, maybe domino effect by incoming western steppe pastoralists.

https://archaeogeneticsblog.com/2022/04/16/the-tarim-basin-bronze-age-horizon-isolated-paleo-siberians-or-recent-colonists-from-the-west/

Rob said...

There is some kind of R1b in Omsk district, Borvjanka XVII (NEO83) dating ~ 3700 bc
So far, all Altaic & Siberian hunter-gatherers belong to Y-hg C and Q

music lover said...

@davidski It is irrelevant that the Sredni Stog samples resemble the Yamnaya. The point is you have to get this ancestry into Anatolia.

What you don't understand is basic population genetics - drift is a major reason for Y chromosome rise. The only sure evidence we have is the Swat samples and the ancestry is female mediated.

The challenge stands. Show that the difference in ancestry between Anatolia_N > Anatolia_C/BA is mediated by any sample from the Caucuses or Ukraine. The reverse is easy to show. The difference between Steppe_Eneolithic and Steppe_BA is driven by Iranian ancestry.

Wee e said...

@Andrzjewski
The 2021 study in Nature — “The genomic origins of the Bronze Age Tarim Basin mummies”,
(Fan Zhang, Chao Ning, Ashley Scott et al).

As I understand it, they say…

1. The 5 Dzungar/Junggar Basin mummies dated 3k-2.8k BCE had mostly Afanasievo and a little local ancestry. Dzungar Basin is to the north & east of the Tarim Basin region, adjoining it.

2. They think it ”plausible” that the Afanasievo descendants brought an IE language to the Dzungar basin; but there’s no sign of them in the Tarim basin remains.

3. The 13 Tarim Basin mummies dated 2.1-1.7k BCE are the earliest remains of that culture in that region, and all had local ancestry — from a genetically isolated local population at that.

4. They think that the Tarim Basin remains were locals who adopted a pastoral & agricultural lifestyle. The ones from Xiaohe had strong signs of milk proteins in their dental calculus.

(The main-author geneticists are from Jilin Univ & Max Planck. The palaeontologists and archaeologists from the Chinese Academy of Sciences, Beijing and from Xinjiang Institute of Cultural Relics.)

Davidski said...

@music lover

It is irrelevant that the Sredni Stog samples resemble the Yamnaya. The point is you have to get this ancestry into Anatolia.

I just have to demonstrate that European Y-haplogroups entered certain parts of Anatolia at the right time.

You know, like Troy during the Bronze Age.

What you don't understand is basic population genetics - drift is a major reason for Y chromosome rise. The only sure evidence we have is the Swat samples and the ancestry is female mediated.

It wasn't just random drift, because ancient DNA shows Central Asia being suddenly flooded by R1a-M417 during the Bronze Age and Armenia by R1b-M269 also during the Bronze Age.

And steppe ancestry in South Asia isn't female mediated.

See that's why there's a lot of R1a diversity in South Asia, so much so that there are still some idiots who believe that R1a is from South Asia.

The difference between Steppe_Eneolithic and Steppe_BA is driven by Iranian ancestry.

I just explained to you that there are Sredni Stog samples from Ukraine that are like Yamnaya.

Sredni Stog is Steppe_Eneolithic not Steppe_BA.

And there's no Iranian ancestry in Sredni Stog or Yamnaya. This is easy to demonstrate. It's not my problem that the scientists at the David Reich Lab are confused about this.

Rob said...

There are 5 male samples with “European “ Ydna in Swat valley samples - 2 x R1a, 2 x I2a, and a G2a which falls under EEF

Wee e said...

@Rob
I have been wondering what it was that made you sceptical about the local (ie regional) development of a widespread population in paleolithic/mesolithic times that left the Tarim people as an unadmixed relict population. (Not the expression Zhang, Ning & Scott use in their paper, but it’s what they describe.)

You question in your blog why the Tarim people were not exploiting the Tarim Basin before they took up agro/pastoralism. Maybe they were. It’s a pretty huge area, and even though graves are not rare and were literally flagged with pole, and very well preserved, the bronze age remains are an amazingly recent find. Without those graves, would a population for Tarim have been suspected for that period either?

What if burial was new, and previous remains were disposed of otherwise? Or just elsewhere? As an archaeologist, looking for short-lived or seasonal hunting camps or for boat-using foragers, or ancestral Tarim inhabitants transitioning to planting and herding — or even looking for more bronze age graves beyond that south-east region of spectacular preservation… where would you start?

The problem with flatland rivers in a wide, arid basin surrounded by huge mountain ranges is that the watercourses flood, silt and dry, making new courses and meanders all the time. Also, mountain-foot oases and the Basin’s riverine margins would have been the first places transformed by cultivation settlements and cattle grazing. The mountain-foot margins of the basin still have fertile country, though recently degraded. The rivers crisscrossing the Tarim Basin even had linear forests in places. Perfectly good places to live around for nomads or seasonal-nomads. Maybe riversides were the usual burial sites for the mummies’ ancestors — it just wouldn’t preserve.

As the paper remarks, by c2k BCE those in Tarim (unlike the surrounding admixed population) were genetic isolates yet already culturally cosmopolitan: why would that be? Maybe because their way of life had long been mobile but specialised. The barrier to entry for either sex might be skillset.

The Tarim culture burials are noted to be oddly boatlike; it would be logical that if the pre-agro-pastoral lifestyle was tied to rivers, they could have had literal boat funerals wherever they were living at the time — a sendoff down the endless river. No graveyards to find.

As boating people, if not boat-people, potentially the Tarim mummies’ ancestors could accumulate the equivalent of a steppe wagonload of material goods rather than the light kit we usually assume of forager-hunters. Their encounters with western (or admixed) migrants could have been on more equal terms than a lot of HGs or sedentary farmers might manage. If threatened, they could evacuate quickly. Potentially, they themselves could be raiders of migrants’ cattle or crops, appearing and disappearing in the night.

It may also be that neolithic Tarim dwellers seasonally exploited and traded minerals (eg jade) that wash down annually from the mountains. We know the trade existed since the neolithic at least. In this type of life, marrying “out” with another culture means losing specialist skills and giving away trade secrets — in exchange for getting spouses missing vital skills. Not necessarily an attractive deal.

If the Tarim people’s ancestors used boats &/or traded valuable specialist trade items like gemstones (&/or timber or pelts), why wouldn’t they be able to have a good standard of living — and the literal capacity to accumulate material goods? It’s even possible the ancestors of the Tarim mummies were substantially traders as much as HGs, before steppe migrants arrived, or for a thousand years after they did.

Maybe there are fossil river-meanders in that huge landscape that haven’t been flooded or cultivated for 4000 years, where you could find pre-bronze-age traces — but if nobody is looking, they ain’t gonna find them.

AWood said...

@Rob

You say that there was little to no population pre-Bronze Age Tarim Basin. Wouldn't that in effect make the PH155 native to the region and the earliest indigenous people? What is your definition of this word? The Native Americans are considered the natives of my country but they traveled thousands of miles to get there, but were here first. That makes them native, thus R1b-PH155 are the native people of western China, not the Han, not the Uyghur.

Gedrosia said...

@music lover
“b) find a single BMAC sample with an R1a/b Y chromosome of the Yamnaya/Andronovo type”

I’m not sure what are you referring by “Yamnaya/Andronovo type” but V. Kumar et al.2021 paper mentioned that BMAC admixtures traces has Steppe-related affinities such as Corded-Ware, Andronovo, and Sintashta.

https://academic.oup.com/view-large/figure/309303634/msab216f1.tif

At the current version of Y-Full there are few ancient (R) samples which have the deep SNP hierarchy clade age.
https://www.yfull.com/tree/R/

R* = MA1 (RUS) “Mal'ta–Buret” (formed 0 ybp, TMRCA 28200 ybp)
R1b* = I4315 (UZB) “BMAC” (formed 22800 ybp, TMRCA 20400 ybp)
R-Y3370* = I7492 (UZB) “BMAC” (formed 11200 ybp, TMRCA 10100 ybp)

G25 (scaled) distances affinities:

Distance to: Russia_MA1_HG.SG:MA1_noUDG.SG
0.16701494 Udmurt:udmurd2
0.17000957 Udmurt:154_R02C01
0.17440807 Udmurt:udmurd13
0.17462600 Udmurt:169_R01C02
0.17524986 Udmurt:139_R01C02

Distance to: Uzbekistan_Dzharkutan_BA_1:I4315
0.06081483 Makrani:HGDP00155
0.06319011 Balochi:HGDP00062
0.06402231 Brahui:HGDP00045
0.06457500 Balochi:HGDP00088
0.06577608 Balochi:HGDP00080

Distance to: Uzbekistan_SappaliTepe_BA:I7492
0.06294645 Makrani:HGDP00155
0.06554295 Balochi:HGDP00062
0.06925570 Balochi:HGDP00080
0.06999755 Iranian_Bandari:BanII40
0.07051915 Balochi:HGDP00088

Seems MA1 has a Finnic profile while I4315 & I7492 have Gedrosian profile.

As far as I know the determine of the migration path and the locality origin is based upon the SNP age hierarchy, the currant data is showing MA1 region (Siberia) as the possible origin of Haplogroup R-M207 which been suggested by “M. Raghavan et al.2014” doi:10.1038/nature12736

Davidski said...

@Gedrosia

It makes no difference whether there was Yamnaya/Andronovo admixture in BMAC, since BMAC wasn't Indo-European.

The Indo-Europeans from Andronovo basically bypassed BMAC on their way to South Asia.

We know this, because there are zero proximal links between BMAC and the Indo-European speakers of India and Iran, who actually show a much closer relationship with Sintashta in terms of Y-DNA.

Andrzejewski said...

@Samuel Andrews @Rob @Weee “ Indo-Europeans migrated into the Tarim Basin later.

The later Tarim mummies are probably part Indo-European.

But no DNA samples from later Tarim mummies have been sequenced.”

Then why do the mummies look so Europoid and how come their outfits are similar to Celtic and Germanic garments (Tartan)?

Andrzejewski said...

@Davidski “ The Indo-Europeans from Andronovo basically bypassed BMAC on their way to South Asia.

We know this, because there are zero proximal links between BMAC and the Indo-European speakers of India and Iran, who actually show a much closer relationship with Sintashta in terms of Y-DNA.”

But aren’t at least the Tajiks and Pashtuns partially (actually majority ancestry) descended from BMAC, aside from WSH of Sintashta/Andronovo horizon?

Davidski said...

No, they have some sort of BMAC-like ancestry and/or post-BMAC ancestry, probably from several different sources.

The point I'm making is that BMAC cannot have had the same sort of impact on the Iranian and Indo-Aryan speakers of South Central Asia as even Fatyanovo from the forest zone of Russia, which is actually quite far from being a proximal source for their steppe ancestry.

We know this, because they share R1a-Z93 with Fatyanovo. What do they share with BMAC at that sort of level of genealogical relationship?

Copper Axe said...

When it comes to the tarim basin, we have evidence to suggest a human presence circa 5500-5000 bc:

https://www.cambridge.org/core/journals/quaternary-research/article/abs/earliest-welldated-archeological-site-in-the-hyperarid-tarim-basin-and-its-implications-for-prehistoric-human-migration-and-climatic-change/EAFEADD1A7F465B085D5E30A16AF90E2

The tmrca of r1b-ph155 on yfull falls within that daterange. Matches up fairly well no?

Dranoel said...

@ Davidski

Once again, I would like to ask about this PCA0475 sample from Pruszcz Gdański.

I took the data from the forum I mentioned earlier - I don't think it's good to refer to it, because it's rather "secondary" and there are more "political" arguments than a substantive conversation.

To the point - analyzing PCA0475 (probably in K36?) we have, among others, the results:
a. Central_Euro 11.29
b. East_Central_Euro 3.39
c. Eastern_Euro 13.25

Where did this eastern data come from in a person who autosomally resembles a Dane/Dutch? Is this an erroneous analysis by the author or an actual admixture?

According to that gentleman, this result proves that he was a mixed gothic / slavic individual. Is it possible? If so, I assume it was the father who came from Scandinavia who took a local wife... I rather doubt it's the other way around, i.e. some Gothic woman came over and married a local man.

Or maybe it doesn't prove anything, because most Scandinavians at that time had such eastern accents, which, for example, resulted from older periods?

I will be grateful for your opinion.
Regards!

Andrzejewski said...

If Tarim mummies weren’t IE, were they Okunevo instead?

Okunevo were most likely WSHG + transbaikal but I can’t see them resembling a Caucasian appearance.

Last time I read, 11 out of 12 mummies had R1a. Could they have been erroneously dated instead of to Saka, Scythians or to other Andronovo-derived tribes? Plus, who were the Tocharinan and what happened to Afanasievo? I guess we might never know

Ygor C.S. said...

Honestly, I wonder if some day geneticists will at least consider the not so unrealistic possibility that it is not necessary nor reasonable to throw away the overwhelming evidences for a PC Steppe origin of PIE because of the "Anatolian puzzle", simply because Anatolian may as well be a particularly early (earlier split, longer independent evolution >> higher odds of rarer events of linguistic shift occurring) and strange outlier in the history of PIE divergence and later expansion; just like Hungarian for the Uralic family, Gagauz and many Black Sea Turks (especially from Trabzon) for the Turkic family, a lot of the Romance speakers for the Romance family (or was there really any detectable Central Italian-driven long-term autosomal or even uniparental shift in Galicia, Asturias, North France or Romania?).

We don't know much about Luwian and Palaic (except that all Anatolian languages have very considerable seemingly non-IE substrate/adstrate), but for Hittite the clues are interesting.

The names of their main cities, such as Nesa and Hattusa, were non-IE (and non-Anatolian). The name of the very Hittite state was the name of another people (Hatti) that Hittites themselves acknowledged as the natives of the land (they named their language "Hattili", "in the manner of Hatti"). Their liturgical formulas and narratives were full of non-IE Hattic and, less so, Hurrian texts. Their religion was very influenced by neighboring non-IE peoples. They themselves told how they had conquered and unified multiple smaller kingdoms (mostly Hattic ones, it seems), and their language was full of Hattic influence in the parts of the lexicon that pertain to state, political power, religion and the upper class.

Those facts make them look like relatively recent incomers that did not have a long and most relevant history of participation in the early development of complex states and sedentary farming societies within Anatolia (or at least Central Anatolia), during the Late Neolithic, Chalcolithic and Early Bronze Age.

This also reminds me of how even much later in the MLBA there were Greeks without steppe ancestry in the by then much more populated than average Aegean zone, though it is pretty obvious that Proto-Greek arrived with steppe-admixed people, otherwise the huge non-IE substrate in Greek, non-IE prevalence in the toponymy of Greece, non-IE Eteocretan and Eteocypriot inscriptions, and ancient accounts of non-Hellenic native Pelasgians, make no sense. If, say, 10% of Greeks with 5% steppe admixture caused a language shift in a populous region conquered by them, how likely is it that the steppe signal would remain, say, 1500 years after the language shift if new pulses of admixture occurred in the meantime?

Ygor C.S. said...

All those things seem to scream "small ruling elite of 'barbarian' origin conquering a much more populous, politically complex culturally influential society". The difference seems to be that, unlike with Goths or Franks (partly), they achieved language shift ultimately, even if it may have been a slow process (Hattic still prestigious enough to be a liturgical language by the Late Bronze Age is an important hint), and at the cost of a language pretty changed by adstrate and substrate influences

Moreover, if Proto-Anatolian were so archaic and divergent because of a more premature split somewhere largely severed from the core Early-PIE zone, then it's likely that the Proto-Anatolian people themselves were NOT typical Neolithic/Eneolithic Steppe people, but just steppe-admixed EEF in the Balkans (as already seen in some Varna, Cernavoda, Usatovo, Ezero samples) or steppe-admixed in the Caucasus.

But there wasn't a sudden leap from the first EPIE split and Proto-Anatolian proper and from that stage to 2nd millennium B.C. Hittite, Luwian and Palaic speakers. A lot of things may have happened between the first EPIE split and the period of the Hittite Empire, possibly as many as 3000 years. For Anatolian to lose its correlation with steppe ancestry, all it got is for an instance of language shift by elite dominance to happen in some group in the "Hungarian" fashion, and that group's expansion and gradual dilution through admixture in the later (many) centuries. That is unlikely to have been a usual phenomenon in ancient times, but it did happen, and the odds that it happened are somewhat higher in such a diversified family as IE (multiple branches) and in a populous area.

Davidski said...

@Ygor C.S.

What's your opinion about the Lazaridis et al. approach of counting percentage points of Eastern hunter-gatherer (EHG) ancestry in random ancient samples from Anatolia as a means to test whether Anatolian speakers came from the Pontic Caspian steppe?

Davidski said...

@Dranoel

PCA0475 has a Dutch/Danish ancestry profile in every test, including the K36.

So whoever is claiming that this sample is significantly Balto-Slavic isn't interpreting the K36 results correctly.

It seems to me that they're interpreting the K36 results too literally.

Rob said...

@ AWood


''You say that there was little to no population pre-Bronze Age Tarim Basin. Wouldn't that in effect make the PH155 native to the region and the earliest indigenous people? What is your definition of this word? The Native Americans are considered the natives of my country but they traveled thousands of miles to get there, but were here first. That makes them native, thus R1b-PH155 are the native people of western China, not the Han, not the Uyghur.'


It would, but the paper is discussing events c. 2200 BC. Accoding to the view I had, if theyre recent migrants then theyre not natives
Native Americans are obviously native c.f. Columbus and the Pilgrims, but if an aDNA paper was discussing about events c. 16,000 BC, then they wouldnt be native, would they ? Theyd be recent arrivals. So its all about timeframe

Rob said...

@ Weeee


''It may also be that neolithic Tarim dwellers seasonally exploited and traded minerals (eg jade) that wash down annually from the mountains. We know the trade existed since the neolithic at least. In this type of life, marrying “out” with another culture means losing specialist skills and giving away trade secrets — in exchange for getting spouses missing vital skills. Not necessarily an attractive deal.''

That's possible. But during prehistory, things appear to have happpened in specific 'bursts'- segments of time-frame where a lot of things went down, interspersed by longer periods of less flux.

E.g. the 4.2 kiloyear event, c. 2200 BC , a whole lot of things occurred across Eurasia as populations were adopting to a period of relatively more dynamic climate cange compare to previous centuries/ millenia which were more steady state.

- in Europe, we see the development of El Argar & Unetice systems
- widespread systems collapse in the East Med; collapse of the 'palatial' civilization in Helladic Greece, etc
- the Sveti Kirilovo phase of Ezero culture in Bulgaria
- the Mako-Maros horizon in Hungary and return of Tell villages for the first time in 2000 years
- the dispersal of Srubnaya-Andronovo populations across central Asia
- the collapse of Kura-Araxes & expansion of Catacomb into the Caucasus

You get my drift..

THe Tarim basin 'natives' might well have been genetically isolated, but it doesn;t mean the had always dwelt in the Tarim basin. They could have lived along a river system somewhere in further north Siberia and then moved into a new niche because of the climactic shifts and arrival of new populations (Andronovans) into southern Siberia. Possible responces: kill or be killed, assimilate & intergrate, immigrate.







Rob said...

@ Copper Axe

''https://www.cambridge.org/core/journals/quaternary-research/article/abs/earliest-welldated-archeological-site-in-the-hyperarid-tarim-basin-and-its-implications-for-prehistoric-human-migration-and-climatic-change/EAFEADD1A7F465B085D5E30A16AF90E2''



Very interesting,
And they used microblades ! So this does point to a possible ANE population, but it might also be the northeast Asian variant of microblade populations.
The paper also suggests that they were in transit to Nepal.
So it's possible. But where is all the R1b in Nepal ?




''The tmrca of r1b-ph155 on yfull falls within that daterange. Matches up fairly well no?'''


This points to branching dispersal events c 5500 bc, yes

However, R1b-PH155 shares bottlebecks with other R1 lineages, R1a and all other branches of R1b . Check it out on YFull, they all co-bottle neck c. 28 - 20,000 y BP. Y-hg R2, otoh, does not share this pattern. IThis might mean that all R1 co-wintered the LGM together somewhere. Given the branching of all other R1 in eastern Europe, the refuge might have been in central-western Siberia, which is archaeologically sound.

Then , if you look at the tree of R1b-PH155, the Tarim group all lie within 1 of 3 major branhces, two others fall within Xiongnu-Huns lieneages, modern Europeans and Vimicacium barbarian foederate. This also points to a more northern origin


ALso, lets refer to the flood of new data from Yu et al, Posth, & Allentoft preprint. The TTK individual right in the IAMC is Y-hg Q, the Altai-HG are C & Q, Ust-Kyatkha further east is C,.
The earliest individual derived for R1b is an Eneolithic man from Omsk. Funnnily enough very close to Botai & the WSHG meta-population.



Davidski said...

Hasn't R1b-PH155 also been found in Central Asia, like in Botai? Or am I imagining things?

Rob said...

@ Ygor

''
The names of their main cities, such as Nesa and Hattusa, were non-IE (and non-Anatolian). The name of the very Hittite state was the name of another people (Hatti) that Hittites themselves acknowledged as the natives of the land (they named their language "Hattili", "in the manner of Hatti"). Their liturgical formulas and narratives were full of non-IE Hattic and, less so, Hurrian texts. Their religion was very influenced by neighboring non-IE peoples. They themselves told how they had conquered and unified multiple smaller kingdoms (mostly Hattic ones, it seems), and their language was full of Hattic influence in the parts of the lexicon that pertain to state, political power, religion and the upper class.''

The differences run deeper C.f. nuclear IE, Anatolian languages demosntrate a split ergative system, loss of feminine gender, development of enclitic chains of particles and anaphoric pronouns, and near obligatory use of phrase connectors (Watkins). These all point to Hurrian & Hattian influence. These deep differences in Anatolian have fed views that it split so early, as we know, that it did not experience the same development of the rest of IE. Another view however could be imperfect learning due to language-shifting Hattians & Hurrians, with inheritence of their own morphosyntactic features.



''Those facts make them look like relatively recent incomers that did not have a long and most relevant history of participation in the early development of complex states and sedentary farming societies within Anatolia (or at least Central Anatolia), during the Late Neolithic, Chalcolithic and Early Bronze Age.''

No language Family in Anatolia did. Id say most of the languages are 'new' there. Hence I mistrust paleolingusitic exercises by lingusitiics unversed about Anatolian settlement details
There was a big 're-set'' c. 5500 BC, the initial 'Amatolian Farmer' settlements petered out. There was another population decline after 4000 BC.
There are at least 3 possibilities about IE in Anatlia, but we need more samples.
Look at the time frame / sample set from Skourtianioti. Observse the 'gap' in samples between 5500 bc and 3800 BC.
Something was happening



also, Hungarian can no longer be used as a case of 'oddities & exceptions'. The 900 AD Carpathian basin has 100s of Uralic looking individuals with Y-hg N. This was no elite conquest but mass movement into a sparsely inhabited region (desolate after the Franks & their Slav merceneries pillage of the Avars).
What it did have, however, is assimilated Turks, Alans, Germanics, SLavs from the outset.

EthanR said...

There is a R-P297 (xM269) sample from Botai, on a branch distantly related to baltic hunter gatherers.
FTDNA seems to identify a clade related to R-PH155 in a sample from bronze age Uzbekistan (~1500BC). As far as I can tell anything else relevant dates to the iron age and onward.

dave said...

C'mon,.. we all know that children take their mothers language/tongue. Indo-European comes from the Caucus hunter-gatherer woman, a cro-magnon isolated since 41,000 bc. She mixed with the steppe herder about 7000 bc during a single large admixture period. She began mixing with Anatolians about 16,000 bc by slow drift.
It is somewhat crazy to think that the steppe herder men gave their language to the hybrid children they had with the Caucus woman. It is far more likely that the Caucus woman gave the children her language and the steppe men gave the children the knowledge for war and weapons. Why am I the only person in the world to see the "mother tongue" of the indo-europeans. Why has the possibility never been considered?

Davidski said...

I'm pretty sure the above comment is a parody, just in case anyone wants to launch a rebuttal.

Ygor C.S. said...

@Rob

Of course a large migration did happen. As you correctly wrote, we know this for a fact now through archaeogenetic studies.. but where are those people's ancestry now in the speakers of Hungarian in the 21st century? Less than 1000 years later, Hungarians had pretty little to do with those migrants, but the language remained and even expanded through people who have basically zero Siberian admixture (Nganasan-related/Krasnoyarsk_BA-related) and N1a paternal lineages. Hungarian remains thus a textbook case of a language that was adopted and later promoted and expanded by people with barely no links to the initial population (Proto-Hungarians in the South Urals or close) and even less related to the speakers of the family's proto-language (Proto-Uralic). Analyzing modern Hungarians you would never consider that Uralic probably came from West or Central Siberia, but it probably did. Hittites and Luwians by 1500-2000 BCE were probably 2000-3000 years apart from Pre-Proto-Anatolians just removed from EPIE communities. A lot could have happened in the meantime, including Proto-Anatolian proper being spoken by people totally unlike EPIE speakers.

Ebizur said...

Rob wrote,

"Very interesting,
And they used microblades ! So this does point to a possible ANE population, but it might also be the northeast Asian variant of microblade populations.
The paper also suggests that they were in transit to Nepal.
So it's possible. But where is all the R1b in Nepal ?"

Well, bhu-0984 from Bhutan (which, of course, is not Nepal, but is fairly close to it) belongs to R-PH861 > R-PH2588 > R-PH155 > R-BY14586 > R-PH200 > R-PH1578*(xY32801) according to TheYtree.

Then there is nep-0172, another individual from the same study (Hallast et al. 2014, "The Y-chromosome tree bursts into leaf: 13,000 high-confidence SNPs covering the majority of known clades"), who is in fact from Nepal and whose Y-DNA belongs to haplogroup C-V20, which is not R1b, but has perhaps an even deeper connection to Western Eurasia.

R1b-PH155 also has been found in some individuals from India, including at least one from Ladakh, which could reflect a connection with Nepal via Tibet.

Hasn't most Y-DNA in the Tibetan Plateau and the Himalayas been overwritten by the lineages of migrants who have introduced Bronze Age culture to the region (e.g. O-CTS5308)? The expansion of Tibetan proper D-P47 > D-Y14736 seems to be an even more recent phenomenon; according to 23mofang, the TMRCA of extant D-Y14736 should be only 2850 ybp. I suppose it shouldn't be impossible that some members of R1b-PH155 may have traveled all the way to Nepal before being mostly replaced by later incomers.

Ygor C.S. said...

IIRC Botai and perhaps (not as sure) WSHG had R1b-M73 or something close upstream to it, didn't they?

@Davidski
I think Lazaridis et al. are in principle correct in trying to, first of all, check the presence of "steppe" admixture or at least EHG admixture in the BA Anatolian samples (after all, the usually confirmed expectation is that language shifts accompany genetic shifts, and the EHG>IE does work in many places)... But where I disagree with them is in considering the perceived lack of EHG in BA Anatolia (which is still not totally sure) a "smoking gun" for the actual homeland of Early PIE. Apart from rare occurrences of massive language shift by elite dominance or even by migration followed by later changes (replacement, displacement), as in Hungary or in some parts of the Turkic sphere, they need to take into account that it is likely that, by the time Hitites and Luwians are surely in Anatolia (c. 1900 BCE), perhaps as many as 2500 years had passed since the beginnings of the historical path of the Anatolian IE branch. If it were more surprising and dynamic than one might think, it can potentially be like trying to establish the origin of Proto-Germanic on the basis of 21st. century Jamaicans or Singaporeans.

Thus, the "steppe signal" should NOT be expected to be as certain and as detectable as the LPIE expansion from the EMBA to MLBA migrations, especially if (of course a very big if here, but a plausible one) Proto-Anatolian was so divergent and gramatically unique in comparison with the residual PIE-derived branches at least partly because it came from some very peripheral and right from the inception more mixed area, such as the Eastern Balkans or maybe the (pretty undersampled) North Caucasus, i.e. in the outermost boundaries of the core EPIE cluster.

dave said...

But you can't deny that the Yamnaya is full of Caucus HG mt-dna, whereas the steppe herder has none.
Are you suggesting that there was no language contribution from the Caucus woman? To really understand my comment you need full sapiens cognition, not an emotional wisecrack. C'mon Davidski, get your brain out of neutral and fire off a simple rebuttal. You will discover that you can't refute the comment,.. unless you resort to the emotional cognition of an african ghost population. You just never thought about mother tongue and it irks your ego that a new member brought it to your attention. New ideas must be introduced to you slowly.. such a cro-mag trait...

Rob said...

@ Ethan

'FTDNA seems to identify a clade related to R-PH155 in a sample from bronze age Uzbekistan (~1500BC).''

Which one ? Cant seem to see it


@ Ygor

yes I take your point.

''
Of course a large migration did happen. As you correctly wrote, we know this for a fact now through archaeogenetic studies.. but where are those people's ancestry now in the speakers of Hungarian in the 21st century? Less than 1000 years later, Hungarians had pretty little to do with those migrants, but the language remained and even expanded through people who have basically zero Siberian admixture (Nganasan-related/Krasnoyarsk_BA-related) and N1a paternal lineages. Hungarian remains thus a textbook case of a language that was adopted and later promoted and expanded by people with barely no links to the initial population'

Probably all died off fending against the Mongol migrations. King Bela some 300 years later still broe an 'eastern lneage', somethign under R1a-Z3.

However, by then, they had established a European Christian monarchy with institutions & literacy. So the 'general populace' kept speaking the language of State.


@ Ebizur

I suppose R1b-PH155 could be within Dali-EBA type pops, accounting for its present in southern Asia
But then again, who knows - with all the northern nomad invasions into India (Mughals, etc)

Rob said...

@ Ygor

Only other thing is we lack evidence of structure, heirarchical societie sin Anatolia in the late chalcolithic - Bronze Age before HIttites
We see citadels ruling over a patch of countryside around them
So 'elite domiannce' is hard to see in Anatolia, esp when we are no just dealing with Hittites, but a series of other related IE languages

Ygor C.S. said...

Dave just needs to answer some basic questions (not to us, that is not needed, but to himself): do most Mexicans speak Spanish? Do Brazilians speak Portuguese? Do Icelanders speak Norse-derived Icelandic? Do most Serbians, North Macedonians and Bulgarians speak South Slavic language? We all know the sex bias in those peoples' paternal and maternal lineages, and in which of them the link between language and uniparental markers is most strongly found. Patriarchal societies have little room for "mother tongue" in the longer term.

Never mind the racist drivel of "emotional cognition of an african ghost population"...

Ygor C.S. said...

@Rob

Your hypothesis is plausible and often brought to the table when we discuss that puzzle, but there is an a priori assumption that needs to be met and explained before for it to make sense: did medieval Magyars basically avoid admixing with local Slavs, Germans and Vlachs for centuries, or did they have some hyperdescent practice whereby any Magyar-admixed person was deemed one of them and preferably bred with other Magyars in later generations? Otherwise, without enduring barriers, it would make little sense that most Magyars died in the Mongol and Turkic invasions, but their admixture would not have lingered on in substantial if minor proportions in the remaining populace.

Matt said...

https://www.pnas.org/doi/10.1073/pnas.2303574120 - "Descent, marriage, and residence practices of a 3,800-year-old pastoral community in Central Eurasia"

The main paper is walled, but the supplements tell us that b10-2 (male) is in fact the offspring of most reproductively successful brother b32-1 (male) and b28-2 (female).

I was unsure about this in a comment above ("Although possibly b10-2 is an offspring of b28-2? b10-2 didn't seem intermediate enough to me, but maybe with some projection noise?"); perhaps b10-2 took chromosomes from his father and mother than put his position on G25 slightly more to the mother's side. But it also probably helps that b32-1 male is at the most steppe shifted edge of the main cluster. b28-2 is presumably part of the wider Sintashta related network of individuals given here profile.

The paper's y-dna results in the supplementary table agree that all except two male samples are Q1b (in all likelihood the way the tree is reconstructed the same Q1b2b, even though resolution is limited in some samples).

b24-1 and b8-2 belong to R1a1a1b2 and R1a1a1b2a. b24-1 and b8-2 are unrelated to each other, and are not buried with any offspring or associated wives either.

Not all brothers who existed in the pedigree were found at the site, but this may be unfortunate due to grave robbing or something.

Interestingly to me, the guy who seemed like a Yamnaya-ish outlier is part of the main family, and is Q1b2b. His mother and father though, are not sampled, so possibly his mother was from an autosomally Yamnaya like community living in the Urals. Presuming they have reconstructed his relationship to the others correctly! He's less apparently an outlier in their PCA.

Family tree - https://imgur.com/a/rIHQd10

Previous Vahaduo plots - https://i.imgur.com/SMFvPtI.png

The supplement contains a IBD_Nepluyevsky_inter>12 which some pretty decent IBD links between the Nepluyevsky samples and the wider Steppe_MLBA complex at 12cM level.

The state that when it comes to these IBD connections: "Among the male individuals that shared longer segments of DNA with the males connected through our inferred pedigree, only carriers of the Y-chromosomal R1a/R1b haplogroup were found, which rules out direct descent through the patriline.", so there is no direct link in ancIBD to any Q1b2b elsewhere who are yet sampled.

RoH in these samples are limited compared to Fatyanovo and Sintashta, which supports that they are part of a later, larger population.

Rob said...

@ Ygor

“ but there is an a priori assumption that needs to be met and explained before for it to make sense: did medieval Magyars basically avoid admixing with local Slavs, Germans and Vlachs for centuries, or did they have some hyperdescent practice whereby any Magyar-admixed person was deemed one of them and preferably bred with other Magyars in later generations?”

No they intermixed from the outset , thus they differ from Avars in this regard
One of the highest chiefs in the “conqueror” population was I2a-CTS10228-Y4460, a ‘Slav’ lineage , for ex

With continued attrition of military elites, the Central European lines can be easily replenished from the surrounds, but the Uralic& nomadic ones could not be because they were disconnected from the parental region

Gaska said...

@Ygor C (1)

You seem to speak very confidently about the origin of Indo-European, however, the conclusions that Harvard and Max Planck have reached about what you call the “Anatolian puzzle” is simply reasoning based on common sense.

1-If the entire international scientific community has accepted that genetics (using uniparental and autosomal markers) can solve the mystery of the origin of the common Indo-European by tracing the paths followed for hundreds and thousands of years by these markers, then it is evident that the homeland of IE cannot be north of the Caucasus because there is no trace of steppe markers in Anatolia neither in the Neolithic, nor in the Chalcolithic nor in the Bronze Age (except for some isolated cases of male markers and a very small percentage of steppe signal in some ancient samples).

2-However, you may not believe that genetics is the determining factor in establishing the origin and dispersion of a language. Then it is normal to use the same absurd arguments that Kurganists have been using for years when things go wrong, i.e when it is genetically impossible to demonstrate a genetic continuity (male, female or even autosomal) that serves to conclusively connect different cultures linguistically. So, since we are all very intelligent, we use the well-known arguments, some of which you have used in your posts, like inaccuracy of linguistics to clearly establish the relationship between the different branches of a language, inability to agree on the date when that language or branch originated, possibility that languages disperse as a lingua franca, role of women in this business, migration followed by later changes, imposition of a language by a dominant warrior or commercial class, etc. To me all these arguments are loser excuses & desperate attempts to save a collapsing building among other things because they are scientifically unprovable. So it is normal that you agree with Lazaridis in using the steppe signal or EHG to clarify the genetic relationship between the steppes and Anatolia and then say that you do not agree with his conclusions due to the supposed antiquity of the Anatolian branch of IE. Come on Ygor are you serious?

3-Comedy and mockery are superlative when there are people who defend that a 1-3% steppe signal is indisputable proof of when and how an Indo-European language entered a certain region. For God's sake, we're never going to be able to prove what language the Yamnaya culture spoke, have we all gone crazy? In any case, even in the absence of genetic markers that can link different cultures, you can always resort to alternative theories to justify what doesn't make any sense

4-How can you say with such certainty that it is pretty obvious that Proto-Greek arrived with steppe-admixed people?- I suppose it will not be necessary to remind you that the Mycenaean only has written records in the Peloponnese, that there is no trace of a steppe signal in that region until the mid-Bronze age, when there are many Greek samples that have zero steppe ancestry, when the rest of the samples have very small percentages, when even the elites are local (Griffin Warrior), when the genetic relationship of Greece with Anatolia and Iran is overwhelmingly superior than with the steppes. Who are you trying to convince? ignorant people ? or you really think your arguments make scientific sense.

5-We all agree that Anatolian samples should continue to be analyzed, but it is common for geneticists to launch grand theories with insufficient or even absurd evidence, you only have to reread Haak, Mathieson, Allentoft papers (2015) to understand what I am saying. So Lazaridis, Krause and company could have waited? Of course they could, but many times political, racial, social or economic agendas are so powerful that no one escapes their influence. 67 old samples were enough for Haak to state flatly that R1a and R1b brought Indo-European to mainland europe, funny right? at least Lazaridis' latest paper has hundreds of samples.


Gaska said...

Ygor C- (2)

But let's assume that you're right (and this is a big assumption) about PC Steppe origin of PIE, maybe you could answer a few simple questions to clarify what is the current position of those who think like you?

1-Do you agree that the common homeland has to be the origin of absolutely ALL the branches of a certain language? Do you agree that any solution to determine the IE common homeland has to adapt to the temporal limits of the lexical-cultural evidence that indicates that the Proto-IE vocabulary can NOT predate the neolithic, that is, the adoption of a system of life with animals and domestic plants and the technology associated with this form of life i.e 4.500-2.500 BC?

2-Do you agree that it is difficult to accept a common Indo-European homeland that lacks archaeological evidence of the dispersion of their culture in other regions?

3- Do you agree with the overwhelmingly patriarchal character and widespread practice of exogamy in all (not just Yamnya) prehistoric societies for millennia?

4-Do you agree that genetic continuity in male markers is the best test to demonstrate migration and the genetic relationship between different cultures? Do you think it is better to use autosomal signals? What percentages do you find reasonable? what was the role of women?

5-What is the culture or cultures that you propose as the origin and homeland of IE?

6-What were the genetic markers that produced the spread of IE and what branches spread each of them?

a said...

https://www.pnas.org/doi/10.1073/pnas.2303574120 - "Descent, marriage, and residence practices of a 3,800-year-old pastoral community in Central Eurasia"

The state that when it comes to these IBD connections: "Among the male individuals that shared longer segments of DNA with the males connected through our inferred pedigree, only carriers of the Y-chromosomal R1a/R1b haplogroup were found, which rules out direct descent through the patriline.", so there is no direct link in ancIBD to any Q1b2b elsewhere who are yet sampled.

The supplement contains a IBD_Nepluyevsky_inter>12 which some pretty decent IBD links between the Nepluyevsky samples and the wider Steppe_MLBA complex at 12cM level.


When compared to the earlier R1b-Z2109+ Yamnaya culture(ochre-wagon-Dom2horse- copper/bronze/iron burials), we see that
1)R1b-Z2109 Kurgan chieftain burials seemed to be fewer in number, and perhaps slightly more ydna homogenous+ spread out over greater distance, when compared to above ydna Srubnya samples(all the way from Bell Beaker France, Netherlands,Poland, Czech Corded Ware to Altai under Afanasievo culture).
2)Connected to Corded Ware Czech, and Swedish Battle Axe cultures-ancIBD
3)Connected to Abashevo-Pepkino-Pepkinsky shared burial mound(3 men whith R1b-Z2103) results that have not been officially released yet.

4)Fewer children burials, for whatever reason.

Deriivka/Bell Beaker/Sintashta connecting samples,
I5884 Deriivka(Sredny Stog) late phase.
I13467 Czech Bohemia Corded Ware.
Bell Beaker samples-I2787,I7044-Hungary,I4253 Poland,CBV-95 France(2570-2450),13026 Netherlands.
Sintashta and Potopovka related samples,
I7670,I0980,I01020,I01028


LGK said...

@ Gaska
"4-Do you agree that genetic continuity in male markers is the best test to demonstrate migration and the genetic relationship between different cultures? "

Agreed, and since M269 came from Latvia-Bulgaria this confirms that Proto-Iberian and Vasconic were intruders to chalcolithic Iberia - So it turns out you have more in common with Yamna than you believe. Welcome Gaska to the invaders of western Europe club!

Gaska said...

@LGK

How did I not realize before that you are a true genius? Now it turns out that you know the origin of M269, that you know the language that was spoken in Bulgaria and Latvia in the mesolithic and neolithic periods, and that you also know that that language was different from the one spoken in Iberia.

Dude, don't doubt it, you have the ability to write a good article in Science

Do you think that mesolithic WHGs societies spoke different languages based on their different male markers?
Do you think that the different European neolithic cultures spoke different languages?
Do you think the WHGs spoke a different language than the EEFs?
Do you think that the language is transmitted through the male line?
Do you know how many lineages of WHGs were incorporated into neolithic societies?

You don't know what you're talking about, please find another hobby or dedicate yourself to trying to prove the connection between Yamnaya and the Peloponnese, at least that way you will be entertained.

«Oldest ‹Older   1 – 200 of 409   Newer› Newest»