search this blog

Tuesday, April 21, 2020

Aesch25


During the early 3rd Millennium BC much of Central and Northern Europe was being infiltrated by pioneer herders, often young men, from the east associated with the Corded Ware culture (CWC).

In some important ways, this expansion may have been very similar to the European colonization of the more remote parts of the Americas during the 16th and 17th centuries.

For instance, the European newcomers weren't always able to dominate the indigenous peoples, and, sometimes, instead of trying to impose their culture on them, they accepted theirs.

I suspect that Aesch25, an ancient sample from the recent Furtwängler et al. paper on the social and genetic structure of the prehistoric populations of the Swiss Plateau, represents a similar case.

Aesch25 wasn't buried with grave goods so he wasn't given a cultural context in the said paper. However, dated to 2864-2501 calBC, he's the earliest individual in this part of Europe with the originally Eastern European Y-haplogroup R1b-M269 and a CWC-like genome-wide genetic structure.


Indeed, the other fourteen samples from the same burial site, dated to more or less the same period as Aesch25, are overwhelmingly of local Neolithic farmer origin.

In any case, irrespective of his cultural affiliation and life story, Aesch25 represents an important data point in the search for the homeland of the so called Bell Beakers who spread across much of Europe during the Copper Age. That's because most Bell Beaker males belong to R1b-M269 and are very similar to Aesch25 in terms of overall genetic structure, apart from an excess of Neolithic farmer ancestry.

My view is that the Bell Beakers were an offshoot of the Single Grave culture (SGC), the westernmost variant of CWC. Of course, the SGC was centered on what is now northwestern Germany and surrounds, and didn't reach into the Swiss Plateau. However, in all likelihood it was founded by men closely related to Aesch25.

Below is a Principal Component Analysis (PCA) based on Global25 data featuring Aesch25 and several other individuals from the Furtwängler et al. paper. To view an interactive version of the plot, copy paste the data from the text file here into the relevant field here, then press Add to PCA. Also, you should copy paste each population separately to make sure that they don't form one grouping in the PCA key.

Aesch25 can easily pass for a CWC individual from what is now Germany (DEU_CWC_LN). On the other hand, the CWC samples from the Swiss Plateau (CHE_CWC_LN) are clearly shifted "south" relative to the German CWC cluster, which suggests that they harbor more Neolithic farmer ancestry. Indeed, they all belong to Y-haplogroup I2, which is especially closely associated with Middle Neolithic European farmers.

MX265, from Singen in southwest Germany, is the only sample in the Furtwängler et al. dataset that belongs to Y-haplogroup R1a. This is a somewhat unexpected outcome, because R1a is, overall, the most common Y-haplogroup in CWC males (see here).

Another surprise is that this individual is dated to just 763-431 calBC, which is a period that overlaps with the Hallstatt and La Tene cultures in Central Europe. Considering that these cultures are often associated with early Celts, was this person perhaps the speaker of a long lost Celtic language?

See also...

Single Grave > Bell Beakers

Dutch Beakers: like no other Beakers

Hungarian Yamnaya > Bell Beakers?

Hungarian Yamnaya predictions

The Battle Axe people came from the steppe

911 comments:

«Oldest   ‹Older   201 – 400 of 911   Newer›   Newest»
mzp1 said...

Well Yamnayas Southern Ancestry certainly isn't Anatolian.

Turkmenistan is the obvious place where EHG and Near Eastern (CHG/Iran_N) would make contact.

We need to ask ourselves:

1) What is the likelihood of non Anatolian-admixed CHG/Iran existing in the Caucuses to provide the admixture source? CHG in the Caucuses is fully Anatolian admixed, not surprising given the association of Anatolian with the Neolithic and the level and spread of the Neolithic in the Caucuses by the time of Yamna.

2) Is there evidence of Northward gene flow from the Caucuses to the Pontic Steppe?

Wang asked this important question in his paper, referencing an earlier study which concluded the Caucuses was a barrier..

"Yunusbayev and colleagues described the Greater Caucasus region as an asymmetric semipermeable barrier"

Yunusbayev states "Genetic discontinuity between the North Caucasus and the East European Plain contrasts with continuity through Anatolia and the Balkans"
in his paper titled "The Caucasus as an Asymmetric Semipermeable Barrier to Ancient Human Migrations"

Wang replies to Yonusbayev: "Second, our results reveal that the Greater Caucasus Mountains were not an insurmountable barrier to human movement in prehistory."

The problem is he is looking at the Anatolian component in the steppe samples north of the Caucuses and using that to contradict Yunusbayev, but he misses the possibility that that ancestry could have come from Turkmenistan. It's quite sad actually.

So, yes, there is little actual evidence for ancestry crossing the Caucuses barrier from South to North, but if anyone has any it would be worth seeing.

Samuel Andrews said...

@mzp1,

Piedmont Eneolithic who is basically identical to Yamnaya is contemporary to Darkveti-Meshoko Eneolithic. In other words, all of Yamnaya's ancestors already live in Russian Steppe by 4500 BC (by the Eneolithic).

That is why Eneolithic genomes from Caucasus aren't direct ancestor of Yamnaya. And no one here argues they are.

What we argue is Yamnaya's CHG acestors arrived in the Mesolithic. That Yamnaya is derived from a hunter gatherer, Mesolithic population from Southern Russia who had mixed CHG and EHG ancestry.

Anonymous said...

@ mzp1

"CHG in the Caucuses is fully Anatolian admixed"

What is this nonsense? Do you understand a word of the subject you're writing about? CHG has no Anatolian admix.

"It's quite sad actually."

It's quite sad to watch you fantasize about something that can't be. There's never been an EHG in Turkmenistan.


Samuel Andrews said...

"Turkmenistan is the obvious place where EHG and Near Eastern (CHG/Iran_N) would make contact."

EHG didn't live in Central Asia, so Turkmenistan is not the place EHg could mix with a Middle Eastern pop.

The best place for formation of Yamnaya, is North of the Caucasus mountains.

mzp1 said...

There is no evidence of such early CHG HGs crossing the Caucuses. If they had we would see that ancestry somewhere given it would have had enough time to spread out, being HGs and not settled agriculturalists.

Evidence suggests the Caucuses was a barrier. This is also evident by the fact that the Neolithic is early and well-developed South of Caucuses but it does not make the jump across the Caucuses to get to Europe. This is noteworthy because otherwise the Neolithic is quite an expansive force.

Rob said...

@ mzp
It’s all very interesting. But whatever we want to call it, the CHG stuff is just hunter gatherers that had minimal role in the formation of the “kurgan cultures”

Rather it was more westerly paraneolithic societies which were in contact with European farmers which expanded Eastward, incorporating the steppe signal (falsely named by Harvard) within the step and then expand back from out there

mzp1 said...

EHG/ANE is admixed into Iran HGs from the Zagros at 10% and that is much more Southern than the northern foothills of the Kopet Dag.

Samuel Andrews said...

That's a fair argument, it is strange CHg ancestry is basically absent in Ukraine and Russia HGs. But, the fact is Yamnay has CHG ancestry from the Cauasus. The fact is CHG only lived in the Caucasus and never lived in Central Asia, therefore it is impossible for Yamnaya to have Central Asian ancestry.

Ukraine Neolithic HGs dating 5500 BC do have very minor CHG ancestry. Russian HGs in Samara do not. However, I think hunter gatherers in southern Russia did have lots of CHG ancestry.

Anonymous said...

@ mzp1 "There is no evidence of such early CHG HGs crossing the Caucuses. If they had we would see that ancestry somewhere given it would have had enough time to spread out, being HGs and not settled agriculturalists."

Your ignorance of the material is just going off the charts, you know absolutely nothing about the samples studied and therefore you can't see anything, you just fantasize funnily.

"Evidence suggests the Caucuses was a barrier. This is also evident by the fact that the Neolithic"

Evidence suggests that you're having a funny fantasy. Moreover, it is not clear here Neolithic at all, we are talking about genetic components that are already in the North Caucasus from the end of Paleolithic. All the time, people have crossed the mountains freely.

Max said...

@Samuel Andrews
How can you be so sure that CHG was not in Turkmenistan? There are still many gaps in the central Asian fossil record.

Davidski said...

@mpz1

EHG never migrated to Turkmenistan and CHG never moved into the steppe you dumbo.

Yamnaya is a mixture of EHG/Ukraine_N with mostly hunter-fisher populations from the north Caucasus foothills rich in CHG-related ancestry.

You'll see that this mixture took place well north of the Caucasus near the Don River around 4,500 BCE!

mzp1 said...

Not only absent from EHG HGs but also in EEF farmers.

mzp1 said...

I thought it was clear EHG or something close was mixing with Iran_N/CHG in Western Iran?

"The population structure of the ancient Near East was not independent of that of Europe (Supplementary Information, section 4), as evidenced by the highly significant (Z=−8.9) statistic f4(Iran_N, Natufian;WHG, EHG) which suggests gene flow in ‘northeastern’ (Neolithic Iran/EHG) and ‘southwestern’ (Levant/WHG) interaction spheres (Fig. 4d)."


Wang in the more recent paper references the above but explains it slightly differently, instead of it being a 'sphere of influence' it is ANE admixture on CHG/Iran_N

"The extra ENA/ANE ancestry also explains the affinity between Iran/Caucasus and EHG previously proposed as part of a North/East West Eurasian interaction sphere12, which our results suggest was created by admixture of ENA/ANE ancestry on top of the Villabruna→Basal Eurasian cline. In the north, Karelia_HG traces its ancestry to a Villabruna-related source modified by ENA/ANE admixture, while CHG/Iran_N were Dzudzuana+Basal Eurasian (or, equivalently Villabuna+Basal Eurasian) derived populations also modified by ENA/ANE admixture.

We need HG samples from the Kopet Dag but that ANE admixture has to be coming from somewhere and it is reasonable to expect greater ANE going North from the Zagros to the Northern foothills of the Kopet Dag (Jeitun).

This has to be the source of the famous steppe signal because there are too many 'weird' samples like the Greek Neolithic Corded-Ware-like, Haji Firuz Chalcolithic etc which just look random and chronologically and geographically do not tie up with the formation of Yamna and Corded ware.

Samuel Andrews said...

@Mzp,

>Mesolithic genome Eastern Iran has more ANE ancestry than Neolithic Western Iran. Neither has EHG ancestry.

>Haji Firuz Chalcolithic_o is Bronze age, dates 2000s BC, she doesn't predate Yamnaya, she has European Steppe admix.

>Greek Neolithic Corded-Ware-like, once again this is Steppe admix.

There's no evidence of a Yamnaya-like pop indigenous to the Middle East or Central Asia. Mainly because EHG only lived in Eastern Europe.

Anonymous said...

@ mzp1 "Greek Neolithic Corded-Ware-like"

You're talking nonsense, some Greek Neolithic corded ware pottery are made up. In Greece, corded ware pottery appears at the end of the 3rd millennium BC. Previously, it only appeared in northern Greece territory. It's all Eneolithic/Bronze age, not some Neolithic.
Don't you dare ramble on topics in which you understand absolutely nothing.

mzp1 said...

I'm not sure how important the distinction between ANE and EHG really is here, I have seen both used interchangeably in literature regarding Iran_N ancestry.

That could be a valid argument if it means something significant.

Having said that, all the components of the EMBA and MLBA steppe profile can be found in Northern Iran, EHG/ANE, CHG and Anatolian.

It is important to understand the interaction between herders/hgs and farmers because one is greater in CHG/ANE and one in CHG/Anatolian . Both groups exist in Northern Iran in the relevant period. One is naturally more visible in archeology, and one much less so.

I was not happy when Haji-Firuz Chalcolithic was re-dated. All a bit suspicious and they never did give a good reason for the change.

Matt said...

The upthread link by Jatt_Scythian seems to be an abstract book with a slight mis-phrasing compared to another version of the abstract:

e.g. https://onlinelibrary.wiley.com/doi/epdf/10.1002/ajpa.24023 from AAPA 2020 - "Here we target the population-genetic transition processes through genome-wide next-generation sequencing data of 25 Eneolithic to Bronze-Age individuals from seven archaeological sites in southwestern Russia. We observe a consistent signal of the hunter-gatherer -like ancestries, followed by the earliest occurrences of individuals with Iranian Neolithic-related ancestry (previously described as ‘steppe ancestry’) mixed with these."

while a better phrasing same paper from ESEB2019 appears to be: https://app.oxfordabstracts.com/events/653/program-app/submission/123334 -

Here we target the population-genetic transition processes through genome-wide next-generation sequencing data of 30 Eneolithic to Bronze-Age individuals from seven archaeological sites in southwestern Russia. We observe a consistent signal of the Eastern Hunter-Gatherer (EHG) -like ancestry, proposing its status as the erstwhile predominant genetic substrate in the region. In close geographic and temporal vicinity, we also show the earliest detected occurrence of mixed EHG and Iranian Neolithic-related ancestry (previously described as ‘steppe ancestry’).

I am not sure why a newer abstract seems to have a worse phrasing and fewer samples than an older version of same abstract, but there we are. (Leaving aside the CHG vs IranN topic).

Davidski said...

@mzp1

I was not happy when Haji-Firuz Chalcolithic was re-dated. All a bit suspicious and they never did give a good reason for the change.

The reason was that a C14 (radiocarbon) direct dating was obtained, you clown.

Henrique Paes said...



@Samuel Andrews

"He doesn't even include CHG in this story. Basically, Harvard completely ignores CHG. According to them CHG doesn't exist. They prefer a nice and simple story, in which there is just IranN.

*****So, it isn't that they think Yamnaya's Near Eastern ancestry is more related to iranN than to CHG. it is that they are not even aware CHG exists."

CHG is just IranN with an amount of ANE (not much, but enough to differentiate itself from IranN). I believe that the population of the Dzudzuana cave was a source for both IranN and CHG.

Assuming they "don't know about CHG" doesn't make any sense. They just don't make CHG a fundamental reference Because CHG itself is already a mixture

Anonymous said...

@Henrique Paes
" They just don't make CHG a fundamental reference Because CHG itself is already a mixture "

No, they do it purely for ideological reasons. IranN is a mixture, too, it's not pure. It's about who invested in the Steppe component, CHG or IranN. But for that they don't notice CHG excluding it from the analysis purely for ideological reasons. Because elementary computation shows that there is CHG in WSH, but not IranN.

Davidski said...

@Henrique Paes

The scientists at Harvard and Max Planck have been ignoring CHG because they made a mistake.

They make mistakes, like everyone else.

Yamnaya didn't form when early farmers from Iran or Armenia migrated into the steppe, but this is the story that these scientists were hoping for.

What you'll see in the new data is that a population almost exactly like Yamnaya already lived way up north near the Don River at least a thousand years before Yamnaya.

The formation of this population is still a mystery, and I don't think the new data will resolve this problem anytime soon, but it seems to me that ultimately it was a fusion of two or three different hunter-fisher populations living in different parts of the steppe.

Mayuresh Madhav Kelkar said...


"I was not happy when Haji-Firuz Chalcolithic was re-dated. All a bit suspicious and they never did give a good reason for the change.

The reason was that a C14 (radiocarbon) direct dating was obtained, you clown"


http://eurogenes.blogspot.com/2018/12/the-hajji-firuz-fiasco.html


"Are you on drugs or something?

Read carefully what I'm writing: that Hajji Firuz sample now has a radiocarbon (C14) date and it's several thousand years younger than first assumed. Also, it has steppe ancestry.

Wait for the paper and you'll see."

mzp1: Have you seen that paper? Thank you.

TLT said...

@JuanRivera
Have you seen the Lazaridis preprint of the Dzudzuana paper? Almost all EHGs are modeled as part-Dzudzuana. CHG is distinct from Dzudzuana in that it has ANE and additional basal Eurasian (red component) ancestry. However, in the components breakdown of EHGs, you only see ANE, WHG and Dzudzuana, none of the extra red component outside of whatever is present in Dzudzuana. I don't know for sure, but I think that this will mean that the CHG-like ancestry in EHGs might be better represented by Dzudzuana than by CHG proper.

IDK when CHG or Iran HG formed, but the presence of the extra basal ancestry in them and the absence of it in northern populations like EHG (outside of whatever is brought in by Dzudzuana alone) probably means that the origin of CHG is separate from that of EHG and that the CHG-like ancestry in EHG is more like Dzudzuana, therefore that input didn't come from CHG either.

Davidski said...

@TLT

It's been a long time since the Dzudzuana preprint was posted at bioRxiv, and it hasn't been published yet, nor are there any signs that it will ever be published.

So I would say that there were some major issues with it and we're likely to see very different models in the paper that the Dzudzuana samples are eventually published with.

Samuel Andrews said...

A major issue is they gave way too much Dzunda to Europeans in their models. Wish they would make Dzunda public to amateur experts online so they can do lots of study on Dzunda. They're an incredible find.

Michalis Moriopoulos said...

Back in November, Razib asked Lazaridis on Twitter what was holding up the Dzudzuana study. Laz replied, "We have more data now that we're analyzing. They waited 26 thousand years, they can wait a bit more."

Maybe that means they have tested some more samples relevant to the paper.

Samuel Andrews said...

@Michalis, That would be sweet. There's a lot to untangle, about relationship between Epipaleo West Eurasians.

For example, mtDNA H, T2, J1, N1a1 all existed in IranNeo and AnatoliaNeo. All clades are less than 30,000 years old. Yet, David Reich says IranNeo and AnatoliaNeo are as distinct as modern Europeans & East Asians. Obviously, this isn't true.

Simon_W said...

An interesting pattern:

CZE_Hallstatt_Bylany:DA111 = closest to French_Occitanie

DEU_Lech_MBA = closest to Spanish_Camp_de_Tarragona

DEU_Roman = closest to Spanish_Catalunya_Central

CHE_IA = closest to Spanish_Castello

Now we begin to realise the magnitude of the shift caused by Germanic migrations.

DEU_Lech_EBA was still closest to French_Pas-de-Calais. Exogamy and/or population movements have reduced the steppe element. (But as Matt pointed out, the Lech_MBA is only one individual and not the most typical one, so the change has taken longer. On the other hand Lech_MBA is not fully MBA/Tumulus culture.)

BTW, among the ancient samples, the closest match for CHE_IA is DEU_Roman.

Coldmountains said...

@Arza

Slavs not formed in the Bronze Age and models using only Bronze Age pops or even earlier grups don't really say anything about Proto-Slavs. The genetic base of Slavs is Baltic_IA-like but with a shift towards Balkan_IA. Central and North European populations like Hungary_BA, DEU_Welzin, Halstatt_Bylany or North Germanics are not picked up significantly what speaks against Slavs arriving from Central Europe. Rather Proto-Slavs look like a Balto-Slavic population shifted towards the Balkan with DEU_Welzin-like and Central Euro (Halstatt, Germanic) ancestry picked up mostly among West Slavs.

Target: Ukrainian
Distance: 1.7657% / 0.01765737
74.4 Baltic_IA
24.2 Balkan_IA
1.4 Hun_Tian_Shan

Target: Polish
Distance: 1.5177% / 0.01517724
69.4 Baltic_IA
25.4 Balkan_IA
3.0 DEU_Welzin_BA
2.2 Germanic

Target: HUN_Avar_Szolad:Av2
Distance: 2.1661% / 0.02166115
75.4 Baltic_IA
18.2 Balkan_IA
6.0 DEU_Welzin_BA
0.4 Hungary_BA


Also many of the Hungarian Scythians pick a lot of DEU_Welzin/Hungary_BA-ancestry when adding them to the models. Only DA191 and DA197 have significant Baltic_IA-like ancestry the other Scythians from Hungary got their high HG-shifted ancestry from other populations.

Anyways the Scythian samples MJ-14 and Scy009 pick more Balto-Slavic drift than any Hungarians from Scythians

Target: Scythian_UKR:scy009
Distance: 2.0386% / 0.02038588
41.2 Baltic_IA
23.0 DEU_Welzin_BA
21.2 Hungary_BA
8.8 Germanic
5.8 Balkan_IA

Target: Scythian_UKR:MJ14 (generally had bad fits in any model but the Balto-Slavic is always high)
Distance: 4.3402% / 0.04340219
69.8 Baltic_IA
24.2 Balkan_IA
6.0 Hun_Tian_Shan

J.S. said...

@Simon_W

Have you checked the autosomal of the French Iron Age? Someone on AG said they plot between Northern Italy and Northern France.

Simon_W said...

@J.S.

I haven't checked them. So far I only work with Global25 data and they're not there, because they're too low coverage. Where in France are they from?

Simon_W said...

DEU_Singen_EBA and CHE_Wartau_EBA are also closest to French_Pas-de-Calais, precisely like DEU_Lech_EBA. But interestingly CH_Bad-Zurzach_EBA is already closest to French_Occitanie, similar like CZE_Hallstatt_Bylany:DA111. This is because one of the two is Barcelona-like, while the other is Breton-like. So they were not yet a homogenous Occitan-like mix. Also interesting to note is that the Catalan-like DNA of CHE_Wartau_IA cannot be from an old local substrate, as CHE_Wartau_EBA was much more northern, as noted above.

And yes, @Davidski, it's Bad Zurzach, or Zurzach at least, not Zuzach as they misspelled in the Supplemental table 1. In the map in the main text they got it right.

Targamos the Based, son of Kavkasos son of CHG son of said...

@Samuel Andrews
Progress and Vonyuchka pick extra Hotu ancestry no matter how much ADC you put. They have on average around 10 Hotu.

Targamos the Based, son of Kavkasos son of CHG son of said...

Even Yamna picks Hotu
https://cdn.discordapp.com/attachments/576401520648781825/703170424850350090/unknown.png
https://cdn.discordapp.com/attachments/576401520648781825/703170473135046666/unknown.png

zardos said...

Coldmountains: Isn't the most logical assumptions that Proto-Slavs developed along a cline from Baltic to Thracian? This would be reflected in your model if its solid. What about the main difference between Balts and Slavs being exactly that Balkan, Thracian or Thracian-like and possibly also Indo-Iranian influences on them? In most models Thracians are the direct neighbours of Proto-Slavs and like with Celts the Iron Age might have been the time of an expansive movement from the Thracian sphere.

Davidski said...

Hotu is a garbage sample, so it doesn't prove anything. It's just noise.

ambron said...

We probably have only two Thracians from Sikora. One looks like a Sardinian and the other looks like a Slav.

Davidski said...

Those Thracian samples are garbage as well. One is totally contaminated.

ambron said...

David, thanks! I did not know...

EastPole said...

@zardos
“Isn't the most logical assumptions that Proto-Slavs developed along a cline from Baltic to Thracian?”

I don’t see any logic here. Slavic language is more archaic than anything Greek and Thracian and probably Baltic. Only phonology of Baltic languages is considered to be more archaic by some, but others consider phonology to be a pseudoscience as we cannot know PIE sounds with any certainty. Etymology is considered to be amore reliable indicator of how archaic the language is by some.
As far as words, roots, morphology, grammar, syntax etc. Slavic is more archaic and definitely not a mixed language between Baltic and Thracian but most likely the opposite is true.

https://i.postimg.cc/bYxfWzPZ/screenshot-81.png

Mouthful said...

We got a comedian in here.

zardos said...

@EastPole: And what has this to do with the movement of people from the Thracian sphere (including Dacians, from the Carpathians) in the Iron Age to the Proto-Slavs directly North of them? Even if your linguistinc considerations would be correct, what is debatable, this won't make a difference. You see the potentially Thracian-like influences in Slavs being much stronger than in Balts and I say this influence was there from the Proto-Slavic stage in the Iron Age. That might be wrong, but you can't disprove it with linguistic arguments alone, whether they are for that field valid or not.
Balts got influenced comparatively more by Uralics, Slavs by fellow Indo-Europeans from different directions, but in the early phase during the Iron Age mostly from the Thracian/Balkan sphere most likely.

ambron said...

Ralph's study shows a massive gene flow between Poland and the Balkans in the Iron Age. First, the Balkan population fled to Poland from the Romans, and then the Polish population migrated to the Balkans with Germanic tribes.

Henrique Paes said...

@Samuel

"Davidski noticed even before CHG genomes were published, that Yamnaya's Middle Eastern ancestry is from Caucasus. This was back in 2015.

http://eurogenes.blogspot.com/2015/03/modelling-yamnaya-with-qpadm.html

So even looking modern pops, Yamnaya choses the Caucasus. If back in 2015 with only modern genomes, Davidski could see this, Harvard geneticists should be able to see it using the countless ancient genomes published since 2015."

For geographic reasons it has always been obvious that CHG and steppe would be related. You acting like this is an 'incredible discovery' is funny.

Henrique Paes said...

@Archi

"@Henrique Paes
" They just don't make CHG a fundamental reference Because CHG itself is already a mixture "

No, they do it purely for ideological reasons. IranN is a mixture, too, it's not pure. It's about who invested in the Steppe component, CHG or IranN. But for that they don't notice CHG excluding it from the analysis purely for ideological reasons. Because elementary computation shows that there is CHG in WSH, but not IranN."


CHG has ANE while IranN does not. If it is to mention a population in the south related to CHG, but without ANE it makes sense to talk about IranN. But some people prefer conspiracy theories about the 'ideological' nature of scientists.

Anonymous said...

@Henrique Paes
"CHG has ANE while IranN does not. If it is to mention a population in the south related to CHG, but without ANE it makes sense to talk about IranN. But some people prefer conspiracy theories about the 'ideological' nature of scientists."

Nothing like that, CHG has EHG, while IranN has ANE and possibly ASI.

Jatt_Scythian said...

I think Southern ancestry in the North clearly comes from the Caucasus. But I agree with Henrique its hilarious to read about what the scientists were looking for. Clearly the Harvard scientists were dying to find Iranian ancestry in the steppe.

Singh said...

@Henrique Paes

Both CHG and IranN have ANE, they were both formed around the same time - when ANE arrived around ~13 kya. It was around the same time when ANE expanded into Eastern European, which lead to formation of EHG.

Before that, both CHG and IranN were Dzudzuana-like but with extra Basal Eurasian compared to Dzudzuana. CHG and IranN formed when ANE admixture arrived around ~13 kya or else they would just be mix of Dzudzuana and extra Basal, which is not the case.

"The Dzudzuana population clarifies the origin of these populations by showing that European
affinity in the Caucasus decreased between Dzudzuana at ~26 kya and Satsurblia at ~13 kya as additional
ENA/ANE ancestry arrived. Thus, Iran_N/CHG are seen as descendants of populations that existed in the
Villabruna→Basal Eurasian cline alluded to above, but with extra Basal Eurasian ancestry (compared to
Dzudzuana), and also with ENA/ANE ancestry. The extra ENA/ANE ancestry also explains the affinity
between Iran/Caucasus and EHG previously proposed as part of a North/East West Eurasian interaction
sphere12, which our results suggest was created by admixture of ENA/ANE ancestry on top of the
Villabruna→Basal Eurasian cline. In the north, Karelia_HG traces its ancestry to a Villabruna-related
source modified by ENA/ANE admixture, while CHG/Iran_N were Dzudzuana+Basal Eurasian (or,
equivalently Villabruna+Basal Eurasian) derived populations also modified by ENA/ANE admixture."

vAsiSTha said...

@samuel
Ehg+chg model fails for steppe_eneolithic. A source similar to geoksyur/parkhai is required to make the models pass.
Make of that what you will.
Have debated this gazillion times with davidski before, and user @matt came to a similar conclusion to mine independently.

Anonymous said...

@Henrique Paes "But some people prefer conspiracy theories about the 'ideological' nature of scientists."

The Conspiracy theory is a fact, as they specifically exclude CHG from the analysis, deliberately, to which they are not entitled. Moreover, they exclude it without any analysis, without showing that they are entitled to it. They published a postulate about the spread of Indo-Europeans from Iran even before any analysis, Krause actually stated directly that he is not objective, and everywhere he pushes his hypothesis.



vAsiSTha said...

@jatt
"I think Southern ancestry in the North clearly comes from the Caucasus. But I agree with Henrique its hilarious to read about what the scientists were looking for. Clearly the Harvard scientists were dying to find Iranian ancestry in the steppe."

Rightly so because models with only chg totally suck. You should try it out for yourself and give us a passing qpAdm model.

CrM said...

@Henrique Paes

Iran_N, Hotu and CHG all have some ANE ancestry which peaks in Hotu. I tried experimenting with the current available samples on G25, and it shows that there's a strong preference of MA1 over AG3 (You can model MA1 as AG3 + Yana(Kostenki14/related + Tianyuan/related) with a decent fit).
https://i.imgur.com/Jo8hjwg.png

Now obviously the fit isn't ideal due to a lack of samples.
You can however check the models done by Lazaridis et al with includes Dzudzuana. Even there you can see Iran_N having ANE.
https://i.imgur.com/z7ke6cQ.png

But I'm not entirely sure how instrumental Dzudzuana really was in the formation of CHG and Iran_N, because Dzudzuana seems to be really close to Anatolians genetically, and including Anatolia doesn't make the fit ideal either (unless there's a drift that needs to be taken into account, there's around 10k years difference between Dzudzuana and Pinarbasi), however you can see CHG strongly preferring Pinarbasi in G25, something that Hotu and Iran_N do not.
Another thing with Lazaridis et al model is that when Dzudzuana is included, you can see a stronger preference of AG3 over MA1.

Henrique Paes said...

@Archi

'Nothing like that, CHG has EHG, while IranN has ANE and possibly ASI.'



In this case, it would be more meaningful to say that ANE has IranN than the other way around. A relevant part of the ancestral components of ANE has to do with Iran and even populations further south such as 'IRN_Shahr_I_Sokhta' and even 'PAK_Saidu_Sharif'. ANE appears to be connected to South Asia.

Distance to: RUS_MA1:MA1
0.28097828 IRN_Shahr_I_Sokhta_BA2:I11466
0.28832937 PAK_Saidu_Sharif_H_o:I7722
0.30038195 IRN_Ganj_Dareh_Historic:I1955
0.32912995 GEO_CHG:KK1
0.34839628 MAR_Taforalt
0.36736833 Jarawa
0.42181247 WHG
0.42408893 Anatolia_Pinarbasi_HG:ZBC_IPB001
0.43785248 Levant_Natufian:I0861
0.53988371 Han
0.75650487 Yoruba
0.84837526 Khomani_San


Henrique Paes said...

@CrM

You make a lot of sense and explain things without appealing to conspiracies. Thanks.

Anonymous said...

@Henrique Paes
"In this case, it would be more meaningful to say that ANE has IranN than the other way around. A relevant part of the ancestral components of ANE has to do with Iran and even populations further south such as 'IRN_Shahr_I_Sokhta' and even 'PAK_Saidu_Sharif'. ANE appears to be connected to South Asia."

This is nonsense, so to say impossible. MA1 is fifteen thousand years older than IranN. You are misinterpreting. ANE is Siberian component and its appearing in Iran is result coming's R2 from Siberia.

Henrique Paes said...

It appears that all populations further east in relation to Anatolia have remained closer to populations in southern Asia than populations further west such as Anatolians, natufians and WHG. It is not absurd to associate populations of steppe, CHG and even ANE with populations in southern Asia. Although it is a fact that it makes more sense to compare CHG with steppe populations.



Distance to: Jarawa
0.22002135 PAK_Saidu_Sharif_H_o:I7722
0.29888676 IRN_Shahr_I_Sokhta_BA2:I11466
0.32498490 Han
0.36736833 RUS_MA1:MA1
0.42490462 IRN_Ganj_Dareh_Historic:I1955
0.45881026 RUS_Karelia_HG:I0211
0.46213056 GEO_CHG:KK1
0.47667515 Levant_Natufian:I0861
0.50614544 Anatolia_Pinarbasi_HG:ZBC_IPB001
0.57452780 WHG

Distance to: PAK_Saidu_Sharif_H_o:I7722
0.13076110 IRN_Shahr_I_Sokhta_BA2:I11466
0.22002135 Jarawa
0.28832937 RUS_MA1:MA1
0.31529524 IRN_Ganj_Dareh_Historic:I1955
0.34741377 GEO_CHG:KK1
0.38981493 RUS_Karelia_HG:I0211
0.43252868 Levant_Natufian:I0861
0.44615494 Anatolia_Pinarbasi_HG:ZBC_IPB001
0.47727134 Han
0.53633498 WHG
0.73487522 Yoruba
0.83561853 Khomani_San

Samuel Andrews said...

@vAsiSTha, Davidski

Davidski, do you also see qpADM models with CHG fail? Not saying I think it means Kurgan is from Turkmenistan.

@All,

I think I get big implication of Wang 2018 which Davidski referred to here but didn't explain. http://eurogenes.blogspot.com/2020/02/ancient-dna-vs-ex-oriente-lux.html

The big find from Wang 2018, is there is no Eneolithic Middle Eastern ancestry in Steppe.

Why is this such a big blow to the Southern Hypothesis? Because Eneolithic migrations were the cornerstone of the Southern hypothesis.

The idea, was advanced Eneolithic colonist from the NEar East established Kurgan culture. Well, this is no longer possible

The Southern hypothesis can adapt by saying it was Neolithic migrations from the Middle East into the Steppe.

Ok, if so you'll have to accept those migrations only affected the southern tip considering Neolithic Ukraine and Russia basically lack Middle Eastern ancestry.

And you'll have to accept, that the Southern Russia ancestors of Yamnaya were primitive hunter gatherers. Yet were somehow founded by an Neolithic farmer society from the Near East? Where is the archaeological evidence of that?


But, let's say you go with that idea Neolithic migrations from the Middle East brought IE language to the Steppe.

If so, you would have to say Proto-Indo European originated in the Neolithic Caucasus in 6000-7000 BC, migrated in the Southern edge of the Steppe and Anatolia in 5000 BC. That sounds so unrealistic.

Think about it. Anatolia is right next door to Balkan peninsula which has a long history of contact with the Pontic Caspien Steppe. If Anatolian languages are indigenous to the Middle East, where are the other ancient IE languages in the Middle East?

Iranain can linked to Steppe, Armenian to Balkans. We see no evidence of indigenous IE languages anywhere else where in the Middle East. If IE languages originated in the Middle East, why is Anatolian the only very ancient IE language there?

It makes a whole lot more sense to say Anatolian language is from Steppe than to say it is not, considering we know Steppe IE had long been active in Balkan peninsula.

Henrique Paes said...

@Archi

"This is nonsense, so to say impossible. MA1 is fifteen thousand years older than IranN. You are misinterpreting. ANE is Siberian component and its appearing in Iran is result coming's R2 from Siberia."

Your answer makes sense, but ANE has a relationship with other populations in southern Asia that it can be tricky to distinguish what exactly 'returned' to the south or was part of very primitive migrations to the north autosomatically.

Samuel Andrews said...

It is confusing, frustrating that Yamnaya (KUrgan) has Middle Eastern ancestry yet they got it from hunter gatherers. It would make more sense they got it from farmers.

Why, was it the hunter gatherers with Middle East admix who founded Kurgan culture? Why not the Neolithic Ukraine and Russia hunter gatherers? It's confusing. Archaeologist traditional see Kurgan culture as descendants of Ukraine and Russia hunter gatherers, not ones from southern tip of Russia next to Caucasus mountains.

Copper Axe said...

I saw a video once of Krause saying that Germans only have 10-20% ancestry from the Central Asia (the steppes).

mzp1 said...

Samuel,

IE influence would be present in both ANE, for example populations like Botai with their Horses and certain cultural similarities to later Scythians, and also in Iran_N-like farmers from the Near East, taking IE further West into Anatolia and developing the Neolithic there.

ANE and Iran_N are both getting their IE elements from contact with South Asia. Current genetic modelling is missing pre-Bronze Age contacts that would have occurred between ANE-like pops and South Asians directly North from the Southern Himalayas to Badakshan and Sarazm.

Henrique Paes said...

@Samuel Andrews

"It is confusing, frustrating that Yamnaya (KUrgan) has Middle Eastern ancestry yet they got it from hunter gatherers. It would make more sense they got it from farmers.

The ancestry of European farmers is related to Anatolians and natufians, not CHG. Although the relationship between them is probably not very distant, but definitely different.


Distance to: GEO_CHG:KK1
0.17353853 IRN_Ganj_Dareh_Historic:I1955
0.26288000 IRN_Shahr_I_Sokhta_BA2:I11466
0.32912995 RUS_MA1:MA1
0.34706562 Anatolia_Pinarbasi_HG:ZBC_IPB001
0.34741377 PAK_Saidu_Sharif_H_o:I7722
0.36419977 Levant_Natufian:I0861
0.39547366 RUS_Karelia_HG:I0211
0.46213056 Jarawa
0.51932591 WHG
0.61611320 Han
0.76016422 Yoruba
0.85410880 Khomani_San

Anonymous said...

@Copperaxetotheneck
"I saw a video once of Krause saying that Germans only have 10-20% ancestry from the Central Asia (the steppes)."

Central Asia has nothing to do with the steppe. Germans have 50% of steppe.

https://i.ibb.co/hKGCKcb/WSH-EEFmixes.png

Samuel Andrews said...

@Copperaxetotheneck,

Btw, I was wrong about Early Corded Ware having same amount of CHG as Yamnaya. They have several percentage points less. CWC_Early=30% CHG, Yamnaya=35% CHG something like that. CWC_Early=60% Eastern HG, Yamnaya=55% Eastern HG. Small difference.

mzp1 said...

I'm saying both EHG Samara_HG and Iran_N-like southern Herders (not HGs, these need to be herders, HG is a misleading term for pre-Neolithic pops) were IEs, and both likely of the Iranian linguistic sub-group, though there are reasons to believe Iran-N may have been less Iranian linguistically compared to Northern ANE type IEs.

Anonymous said...

mzp1 said...
"IE influence would be present in both ANE"

A fantastically meaningless sentences, written by a child.

mzp1 said...

I meant "both ANE, (like Samara_HG, Botai and Sarazm_Eneolithic etc) and Iran-N populations".

Copper Axe said...

@Archi

I know that, I was just pointing out Krause's obvious biases regarding the kurgan hypothesis.

Anonymous said...

mzp1 said...
I meant "both ANE, (like Samara_HG, Botai and Sarazm_Eneolithic etc) and Iran-N populations".

Once again, you write meaningless phrases using words you don't understand. You've got all these messages, a meaningless set of words.
"Samara_HG, Botai and Sarazm_Eneolithic etc) and Iran-N" are not ANE.

Copper Axe said...

@Samuel Andrews

Thanks! I guess the differences in percentage are purely due to geography. Kind of surprised by that amount of CHG ancestry in the Yamnaya, I've always seen it reported as being around 50%. Are those numbers elevated due to the Dzudzuana component in the EHG or something akin to that?

Ric Hern said...

Caucasus > Kammenaya Balka.

vAsiSTha said...

Yamnaya_samara does not have 30% CHG or 50%CHG. Its more like 22%


Russia_Yamnaya_Samara

EEHG: 46.7 +- 1.6
Georgia_Kotias.SG: 22.2 +-2.7
Ukraine_Globular_Amphora: 12.5 +- 1.4
Shahr_I_Sokhta_BA1: 18.6 +- 2.7
p-value: 0.087
result file: https://pastebin.com/ssgUCG71

All nested 3and 2 way models are a big fail with p-values in e-10 range. SiS is a proxy for Turan. Geoksyur, sarazm, parkhai should also work.

Henrique Paes said...

@Copperaxetotheneck

"I saw a video once of Krause saying that Germans only have 10-20% ancestry from the Central Asia (the steppes)."

Was he not referring to the amount of ANE and not Yamnaya?

Henrique Paes said...

@vAsiSTha

Target: Yamnaya_RUS_Samara
Distance: 9.5010% / 0.09500965
42.6 RUS_MA1
32.4 GEO_CHG
15.6 WHG
7.6 IRN_Ganj_Dareh_Historic
1.8 Anatolia_Pinarbasi_HG

Distance to: Yamnaya_RUS_Samara
0.18077186 RUS_MA1:MA1
0.20505940 IRN_Ganj_Dareh_Historic:I1955
0.22904924 GEO_CHG:KK1
0.29399902 IRN_Shahr_I_Sokhta_BA2:I11466
0.32788848 Anatolia_Pinarbasi_HG:ZBC_IPB001
0.33835538 MAR_Taforalt
0.33876388 PAK_Saidu_Sharif_H_o:I7722
0.37501344 Levant_Natufian:I0861
0.37744914 WHG

Including EHG distorts the values ​​because EHG also already has a lot of ANE.

epoch said...

@Archi

You showed a book snippet assigning Aesch to the BB culture. I reckon that would be a part of the non-steppe, southern Eurpean BB culture, with communal burials? Is that based on the assignment of other megalith burials in the near vicinity?

Samuel Andrews said...

I sent an email to heads at David Reich lab and Max Institute. Instead, David Anthony responded.

What David Anthony said is.....he thinks unadmixed CHG *hunter gatherers* lived in Orlovka culture in South Russia from 6200-4500 BC. And he thinks pure CHG and pure EHG mixed in 4500-4300 BC.

This matches everything he said in a paper 2019, but instead he saying they were hunter gatherers not farmers. So, David Anthony is moving towards idea Kurgan culture was created by Steppe hunter gatherers (some CHG, some EHG).

He said most at the Reich lab isn't totally decided yet....
"Not everyone in the Reich lab agrees on a single view of Eneolithic steppe genetic dynamics, and I have heard spirited discussions on this topic, so there isn't yet a Reich lab orthodox view on this"

"Our interpretation is still evolving as we move toward publishing many new samples, Eneolithic and Yamnaya, a project we are working on now. "

Copper Axe said...

@Henrique Paes

No he specifically said ancestry from cattle herding pastoralists coming out of central Asia, which doesn't sound like ANE to me.

Rob said...

@ epoch

“ would be a part of the non-steppe, southern Eurpean BB culture, with communal burials? Is that based on the assignment of other megalith burials in the near vicinity?”

No such thing

Anonymous said...

You've all made sure that vAsiSTha doesn't understand anything about modelling!! It's got to be an outgroupp to include the ingroups (sources). It's just a matter of thinking about including CHG as the right population and including it as the left population!!

I excluded from the right side of the CHG population and as you would expect I got this combination impossible!

Russia_EBA_Yamnaya_Samara Russia_EHG Georgia_Kotias.SG Ukraine_Globular_Amphora Iran_BA1_ShahrISokhta
0000 0 7 2.737 0.908196 0.472 -0.584 0.266 0.846 infeasible
0001 1 8 8.779 0.36126 0.474 0.375 0.150 0.000
https://pastebin.com/kJztf4xH

He includes in his models those in the original populations that it has in outgroups or exist later modeled populations! Iran_BA1_ShahrISokhta = Iran_GanjDareh_N is + north (steppe) influence. Exclude Iran_GanjDareh_N give correct result.

Russia_EBA_Yamnaya_Samara Russia_EHG Georgia_Kotias.SG Ukraine_Globular_Amphora Iran_BA1_ShahrISokhta
0000 0 6 1.799 0.937261 0.514 -0.146 0.262 0.371 infeasible
0001 1 7 2.561 0.92242 0.531 0.225 0.244 0.000
https://pastebin.com/033Grviw


You made sure that all models of vAsiSTha are wrong. It absolutely does not understand anything in modelling, it does not understand how it is calculated that there is admissible, and that is not.
The man does not understand at all how much modelling depends on the right populations and how much his choice of the right populations is bad. And of course the choice of sources is bad.

@Henrique Paes

Yamnaya_RUS_Samara has nothing to do with ANE. The presence of a small amount of ANE in the EHG has nothing to do with it.

Anonymous said...

@Henrique Paes

9.5010% is absolutely fault. The model failed.

vAsiSTha said...

qpAdm for Steppe_Eneolithic

Russia_Steppe_Eneolithic

Russia_HG_Karelia: 47.5 +- 1.9
Georgia_Kotias.SG: 24 +- 3.1
Geoksyur_EN: 28.5 +- 3.3
p-value: 0.097
Result file: https://pastebin.com/vc9v2kwP

both 2 way nested models have p-values around e-11, so theyre utter failures.

Samuel Andrews said...

@Copperaxetotheneck,
"Kind of surprised by that amount of CHG ancestry in the Yamnaya, I've always seen it reported as being around 50%."

Roughly 35% CHG for Yamnaya is the number I have always seen in this blog. First when we used D-stats. Then when David created G25 PCA.

Overall, Yamnaya according to G25 PCA estimate is 45% Middle Eastern because you have to include 35% CHG and 10% AnatolianEF. Which is close to the original 50% estimated by Harvard in 2015.

I agree, Corded Ware_Early and Yamnaya are apart of the same Cline, with Corded Ware closer to UkraineHGs/EHG and Yamnaay closer to CHG. But they are basically identical.

vAsiSTha said...

@archi

1. I looked at your results pasted here https://pastebin.com/kJztf4xH. it uses an HO database with 133k snps used. Use the 1240k database will allsnps:YES option in param file.

2. Your 4 way model gives -58% for Kotias and 84% for ShahrSokhta, clearly indicating that SiS is preferred over Kotias. No of course this is incorrect because you dont have Satsurblia in the right pops. having satsurblia will leave you with the correct result.

3. Your nested 3 way model for Russia_EBA accepts the model
Russia_EHG + Globular_AMphora + SiS with p-value = 0.82. Its the highest feasible model in your results file.

proportions are russia_ehg: 47.1, Globular_A: 19.6, SiS: 33.2

learn modeling and how to read the results better.

Anonymous said...

Everyone sees that vAsiSTha is frank and brazenly cheating with the results (changes outgroups as he wants, but leaves the wrong ones), straight and uncovered, adjusting them to the desired result. It's a fact.

Anonymous said...

@vAsiSTha "No of course this is incorrect because you dont have Satsurblia in the right pops.
learn modeling and how to read the results better."

Kid, you write nonsense. The right populations are called outgroups, they don't have the right to be like the left ones, and they should be as far away from the source populations as possible.

You don't understand anything, learn, kiddo.

vAsiSTha said...

Lmao, you said "exclude iran_n to give correct result". Lol we see who is cheating. Hahahaha. Pathetic.

Include both iran_n and satsurblia in right pops and see what yamnaya prefers.

vAsiSTha said...

Samuel, I challenge you to model steppe_eneolithic as ehg+chg, or yamnaya as ehg+chg+whatever else in qpAdm.
Do it and I will accept defeat. Just don't leave out iran_n from right pops as suggested by that loser archi.

Hell, even vahaduo fits improve for steppe_eneolithic when geoksyur is added to ehg+chg

vAsiSTha said...

@archi youre modeling with 133k snps, you should just stfu.

As far as right pops are concerned, heres what the invenor team of qpAdm says in the latest preprint https://www.biorxiv.org/content/10.1101/2020.04.09.032664v1.full.pdf

"Previously, ‘right’ populations 77 were referred to as ‘outgroup’ populations, but we avoid this term because it suggests that 78 reference populations should be outgroups in phylogenetic sense (i.e. equally closely related to 79 all 'left' populations)"

stop yapping when you know absolutely nothing. "leave iran_N out from outgroup to get correct result" lmao. hahahahahahaha

Anonymous said...

vAsiSTha said...
" Lmao, you said "exclude iran_n to give correct result". Lol we see who is cheating. Hahahaha. Pathetic.

Include both iran_n and satsurblia in right pops and see what yamnaya prefers."

LOL. Include yamnaya in right pops and see what yamnaya prefers.

Russia_EBA_Yamnaya_Samara Russia_EHG Georgia_Kotias.SG Ukraine_Globular_Amphora
fixed pat wt dof chisq tail prob
0000 0 8 4.227 0.836129 0.562 0.919 0.158 -0.639 infeasible
0001 1 9 15.070 0.0890326 0.495 0.402 0.103 -0.000
0010 1 9 8.506 0.484097 0.572 1.137 0.000 -0.709 infeasible
0100 1 9 37.834 1.86705e-05 0.614 0.000 0.156 0.230
1000 1 9 19.668 0.0200771 0.000 4.173 0.171 -3.345 infeasible
0011 2 10 19.178 0.0380637 0.503 0.497 0.000 -0.000
0101 2 10 42.237 6.803e-06 0.762 0.000 0.238 -0.000
https://pastebin.com/SS6rc31h

You're disgraced at the top as always.


Samuel Andrews said...

@vAsiSTha, I don't have any of the computer stuff for qpADM.

Geographically speaking, it makes sense Steppe Eneolithic only has CHG ancestry. Turkmenistan is too far away. I understand that's how your models work but I don't believe qpADM makes the final word on this issue. I don't have an explanation for why you get the results you get, but I have seen enough evidence over the years to say Yamnaya's Near Eastern anecstry is from Caucasus.

Henrique Paes said...

@Samuel

"Calling Yamnaya's Middle Eastern ancestry IranN-related is like calling Latin American's European ancestry Russian-related.

We have CHG genomes, they need to do basic analysis, they need to talk to Davidski, and finally understand we know where Yamnaya's Midle Eastern ancestry is from so we no long have to call it mysterious IranN-related."

This is crying because of words. When the influx of genes from the 'south' reached the EHG region, perhaps CHG did not even exist in the south of the Caucasians - or even formed in the same period. Probably IranN, CHG and Yamnaya have common ancestors. The only fact that CHG appears closer to Yamnaya is because CHG itself also contains a little more ANE than IranN. It does not mean that an ancestral population shared between Yamnaya and IranN did not exist. Or that EHG has mixed specifically with CHG.

CHG and Yamnaya formed in a nearby region. What they have in common may be the result of common ancestors, not the direct mix between one and the other.

vAsiSTha said...

@samuel Ok then that's surely how it happened.
Why does steppe_eneolithic prefer added geoksyur on both global25 as well as qpAdm? What's your best guess?

Matt said...

@Sam, thanks for sharing Reich's latest thoughts (right or wrong?).

Reich is a close reader of David Anthony I think (unless he has Russian language of his own, he has to be). On the topic of this Orlovka Culture, DA seems to suggest that, at least as of 5200-4800 BCE (perhaps not earlier), their settlements (and they are apparently settlements and not mobile hunters camps) show evidence of domesticated animals, *but* that these domestic animals ultimately came to them from SE Europe, via Ukraine, *not* from the Near East. See - https://imgur.com/a/2fUHPgz

So in a sense, if you synthesize that with Reich's comments, the hypothesis would be that they were "Near Eastern" in a broad genetic sense (unadmixed CHG) and they were agriculturalists in the sense of being settled food producers, but they were not necessarily "Near Eastern agriculturalists" in the sense of a demic impulse from the Near East, and of having many more agricultural (or pastoral) elements than neighbours to north.

Ancient dna from the animal specimens could really sort it out. (Did this actually happen as proposed?). Unfortunately I would bet life will not be that easy...

Henrique Paes said...

@Archi

"Yamnaya_RUS_Samara has nothing to do with ANE. The presence of a small amount of ANE in the EHG has nothing to do with it"


Are you telling a joke? The amount of ANE in Yamnaya is not small. EHG is basically 75% ANE and 25% WHG. Yamnaya has slightly less ANE than EHG, but remains its main autosomal component.


Target: Yamnaya_RUS_Samara
Distance: 9.5010% / 0.09500965
42.6 RUS_MA1
32.4 GEO_CHG
15.6 WHG
7.6 IRN_Ganj_Dareh_Historic
1.8 Anatolia_Pinarbasi_HG

Target: Yamnaya_RUS_Samara
Distance: 6.1586% / 0.06158622
54.0 RUS_Karelia_HG
40.8 GEO_CHG
5.2 Anatolia_Pinarbasi_HG

Anonymous said...

@Henrique Paes
"EHG is basically 75% ANE and 25% WHG."

joke.

No, EHG is not ANE + WHG.

Distance: 9.5010% is a joke.

Target: Yamnaya_RUS_Samara:I0357
Distance: 3.7252% / 0.03725209
63.4 RUS_Karelia_HG
36.6 GEO_CHG

Samuel Andrews said...

@Matt,

I sent the email to David Reich. But, Reich asked Anthony to respond. So, the response explained are the thoughts of Anthony.

"Orlovka Culture, DA seems to suggest that, at least as of 5200-4800 BCE, their settlements show evidence of domesticated animals, *but* that these domestic animals ultimately came to them from SE Europe"

Yeah, that's a key point. Orlovka just seem like one of many of the hunter gatherers transitioning to pastoralism but they probably had lots of CHG ancestry unlike the others. It is a weird coincidence that it was a heavily Near Eastern (CHG) hunter gatherer pop who dominated the Steppe with more advanced pastorlism, yet their rise had nothing to do with migration from the Near East.

It convinced a lot of us for a log time, that Kurgan society was derived from Near Eastern pastorlists. This is what we see in Neolithic Europe, in South Asia, in East Africa. It made so much sense that is what happened in Steppe.

Samuel Andrews said...

@Matt

Anthony, believes Orlovka was pure CHG. And that pure CHG and pure EHG mixed in 4500-4300 BC creating Khvlanksy. He still believes early Indo Europeans were a recently mixed population. Which creates confusing of who the real PIEs were CHG or EHG.

But, I agree with what Davidski says, which is that the CHG/EHG mix happening entirely in North Caucasus foothills long before 4500 BC, probably back in the Mesolithic era.

And that therefore, Early Indo Europeans were not a mixed population but instead derived mainly from one CHG rich population from Lower Don who spread into Ukraine creating Sredny Stog, and into Volga-Samara area creating Khvalnsky.

EastPole said...

@Samuel Andrews
„What David Anthony said is.....he thinks unadmixed CHG *hunter gatherers* lived in Orlovka culture in South Russia from 6200-4500 BC. And he thinks pure CHG and pure EHG mixed in 4500-4300 BC.”

So it probably looked like this:

https://i.postimg.cc/L5bYLYst/Orlovka-Culture-CHG-EHG.png

I used to think that only CHG wives were transported along rivers. But it turns out that unadmixed CHG cultures could develop along river routes:

https://s14.postimg.cc/cpnp160gh/screenshot_391.png

http://eurogenes.blogspot.com/2018/05/on-genetic-prehistory-of-greater.html?showComment=1526536374182#c4945358758153117098

If Khvalynsk developed this way why not Sredny Stog too?

Matt said...

@Sam, my bad with skim reading some bits there, too many Davids...

Davidski said...

@Samuel

When did you have this correspondence with Reich and Anthony? It seems like it was a few months ago. I doubt they'd be saying the same thing now, after they had a good look at their new samples.

Khvalynsk is overrated by Anthony. It formed parallel to Sredny Stog and Yamnaya from similar processes, but it was then replaced, maybe partly absorbed, by Yamnaya.

The CHG-related ancestry in Sredny Stog is definitely from the North Caucasus foothills. That's where it formed a long time ago.

There was some gene flow from the Caucasus proper into the Eneolithic steppe, but it was minimal and after the fact.

Samuel Andrews said...

@Davidski,

I emailed them yesterday, they responded today. These are his current thoughts on PIE origins. He said this about new Eneolithic Steppe genomes....

"....we move toward publishing many new samples, Eneolithic and Yamnaya, a project we are working on now. It does look like the Volga-Don-Caucasus steppes had a series of Eneolithic sites on a north-south transect with almost unadmixed EHG in the north and greatest CHG in the south (Progress-2)"

I guess, he is saying Eneolithci DNA shows is uadmixed EHG in North and heavily CHG pops in the south. No cline, but two distinct regions. He thinks Orlovka will come out 100% CHG, and that CHG-EHG mixed from 4500-4300 BC.

Davidski said...

@Samuel

Thanks.

Well it looks like we'll have some fun here after they publish this stuff.

Anonymous said...

@Davidski

The wait is eternal if they can't publish an article with Welzin for three years. The fact that they have tested Khvalynsk and Mariupol is also known for several years. But when they publish papers, so the information about the samples is just depressing and it becomes unclear why they wrote so long?

Matt said...

@Sam, I suppose it will depend on what the samples show really (either we'll see CHG turning up only in Piedmont_En type ratios with EHG, and a max at not much more than the Piedmont_En samples, or we won't!).

I guess obviously the advantage of an eneolithic / early copper age population that would be almost purely or very high CHG (with perhaps low level EHG such the reverse of how Khvalynsk samples CHG?), if there is such a thing, and which might speak early form of IE, would be... that a population like that would plausibly be able to disperse through Caucasus into Anatolia (Anatolian languages) without dispersing much EHG related ancestry, which they don't apparently see. And provide an explanation for lack the later derived cultural terms which Anatolian is thought not to share with core IE (wheels and such). Wherever the language ultimately derived from. Pure speculation.

Anonymous said...


Naturally, I don't know if Orlovka was CHG, but I have long written that CHG was probably spread to the north of the Caspian Sea. But I know that with the end of the Neolithic Orlovka culture the northern forest population flooded there, bringing new forest pottery and at the same time formed the Eneolithic Pre-Caspian culture.

Davidski said...

@Archi

All of the samples that we need to understand what happened are already publicly available. So let's give them credit for that.

Davidski said...

@Matt

I don't know why Anthony is saying that Orlovka might be CHG.

There's no chance that you'll see anything more CHG on the steppe than Vonyuchka.

Matt said...

Btw, off topic - for anyone interested in Siberian / East Asian type stuff:

https://www.cambridge.org/core/journals/evolutionary-human-sciences/article/homeland-of-prototungusic-inferred-from-contemporary-words-and-ancient-genomes/950020A1799B4D7B71CCC57D04743B3E - "The homeland of Proto-Tungusic inferred from contemporary words and ancient genomes - Chuan-Chao Wang (a1) and Martine Robbeets"

Published online a couple of days ago.

(I believe the abstract of this was present in a ISBA 2018 conference a couple years ago - https://eurogenes.blogspot.com/2018/11/on-spread-of-dairy-pastoralism-to-east.html).

Anonymous said...

@Davidski

Is open access samples of Mariupol and Khvalynsk? Is DZuDZuana publicly available? Is full sources of Welzin with haplogroups in open access? In order to correctly interpret the results we always need complete information about the archaeological context, as complete as possible, rather than a scrap with a message of approximate place and approximate time.

vAsiSTha said...

A pure iran/chg type sample without ehg from the eneolithic caucasus will be great for modeling and will solve most debates.

Davidski said...

There's no such thing as a pure CHG sample in the Eneolithic Caucasus.

That's what is slightly pulling Progress/Vonyuchka from the EHG/CHG cline, and making it look like Geoksyur ancestry.

But it's just minor admix.

Samuel Andrews said...

As you guys already know Harvard has a lot of unpublished Khvalnsky genomes. Anthony said in the email, the samples from the original huge Khvalnsky cemetery vary from 0-50% CHG.

If CHG never gets above 50%, then it must have not been a pure CHG pop migrating into Volga area but instead a pop with 50% CHG.

Davidski said...

@Samuel

If CHG never gets above 50%, then it must have not been a pure CHG pop migrating into Volga area but instead a pop with 50% CHG.

Exactly.

There's no evidence in the new Khvalynsk data that Orlovka was CHG, or that anyone even close to pure CHG made it onto the steppe.

Khvalynsk is just a mix of EHG and something like Progress.

Copper Axe said...

You know what would be nice for the simpletons like me? A chart of the various rates of admixtures in these early populations, on average across time.

So something like (these numbers are made up by me and likely completely off): Samara culture = 90 EHG+ 5% WHG + 5% CHG -> Dnieper Donets =... -> Khvalynsk = 80% EHG 20% CHG, Sredny Stog = 60% EHG, 20% CHG 20% EEF -> Piedmont steppe = 45% EHG + 55% CHG -> Repin(?)/Yamnaya/Afasanievo = 55% EHG,35% CHG, 10% EEF.

Once again these are just b.s numbers, but something akin to that, in a simple comprehensive view. Is there anything like that out there? I don't really trust any other sites outside of this one when it comes to ancient genetics.

Henrique Paes said...

@Archi

"No, EHG is not ANE + WHG.

Distance: 9.5010% is a joke.

Target: Yamnaya_RUS_Samara:I0357
Distance: 3.7252% / 0.03725209
63.4 RUS_Karelia_HG
36.6 GEO_CHG"

The only joke here is that you switch to Karelia and pretend that Karelia is not mostly ANE just to get a few dots in the distances. This kind of fallacy only deceives idiots.

EHG it is mostly ANE and by no means the amount of ANE in steppes is very small as you guessed.


Target: RUS_Karelia_HG:I0211
Distance: 10.4910% / 0.10490963
68.8 RUS_MA1
31.2 WHG

Henrique Paes said...

@Archi

Nor is it necessary to model populations to say obvious things. In all articles, the role of ANE in EHG ancestry is fundamental. I don't even know why I waste time arguing with you.

Anonymous said...

@Henrique Paes This kind of fallacy only deceives idiots.

WHG = 55% Levant_Natufian + 45% RUS_MA1 uhahaha

Target: WHG:I1875
Distance: 18.9652% / 0.18965153
55.4 Levant_Natufian
44.6 RUS_MA1

Anonymous said...

@Henrique Paes

Therefore, you should use Levant_Natufian instead of WHG.

vAsiSTha said...

The iran component of Khvalynsk is itself not 100%CHG

Russia_Khvalynsk_Eneolithic

EEHG: 78.9 +- 2
Georgia_Kotias.SG: 7.3 +- 3.9
Geoksyur_EN: 13.9 +- 4.1
p-value 0.11
result https://pastebin.com/ZnptRjQn

2 way model with eehg + kotias fails with p value 0.004, while that with eehg + Geoksyur passes with 0.0627.

So there would be a chg + geoksyur like population which existed in the caucasus between Kotias & eneolithic ie, between 7000bce & 5000bce.

This pop X made a cline with EHG to give Progress_EN to the south with 50:50 EEHG:X and Khvalynsk up north with 80:20 EEHG:X.

Davidski said...

@vAsiSTha

Geoksyur has nothing to do with this, except that it's derived from distantly related populations that colonized Central Asia from the west, you idiot.

vAsiSTha said...

They did find a J1 at khvalynsk, 2 of which have been found in geoksyur as well, 1 in gonur. the subclades will need to be checked. None of wang's caucasus samples are J1 except 1 from kura araxes vellikent.

Davidski said...

There are two EHG samples with J1, of course.

mzp1 said...

Where do you think the EHG J1 came from?

Davidski said...

The J1 in EHG is probably a very ancient signal of gene flow from the Near East into Eastern Europe.

We're talking the Upper Paleolithic or early Mesolithic here.

Samuel Andrews said...

I sent David Anthony another long-winded email about possibility Proto-Indo European is from a a long established mixed CHG/EHG population from Southern Russia.

I'll see if he responds. Hopefully, this theory passes on to Harvard and they take it into consideration.

Davidski said...

I don't know about PIE coming from the Progress-2/Vonyuchka area (North Caucasus Piedmont steppe), if that's what you mean. That might be a different issue.

But the point I've been making is that there's too much focus on EHG and CHG, when there's no evidence that CHG ever lived on the steppe. It seems to me that the Progress-2/Vonyuchka population is about as CHG as we should expect on the steppe.

PIE may have formed when this Progress-2/Vonyuchka population mixed with other hunter-fishers further north. The fact that PIE wasn't a mixed language doesn't preclude this possibility.

Samuel Andrews said...

I meant Southern Russia in general not a specific part. But in the email to David Anthony I did highlight the North Caucasus Steppe.

If, let's say PIE was mainly from Hunter-Fishers from North Caucasus Steppe but also has significant ancestry Hunter Fishers further north, isn't is most likely the pops from North Caucasus Steppe spoke pre-Proto Indo European?

Maybe that isn't very important if the real Proto-Indo European community originated significantly north of there.

epoch said...

I hear some people stating that EHG isn't WHG and ANE. On what is that conclusion based?

Rob said...

@ Davidski


PIE may have formed when this Progress-2/Vonyuchka population mixed with other hunter-fishers further north. The fact that PIE wasn't a mixed language doesn't preclude this possibility.”

These are separate issues- how PIE emerged vs how Vonuchka-like groups formed

vAsiSTha said...

East of caspian PIE homeland theory of Kozintsev (2019), which is of course my no 1 contender as well

"ABSTRACT. The multivariate analysis of the matrix of pairwise lexical matches between 104 languages representing 16 families from The Global Lexicostatistical Database suggests that the two-dimensional (quasi-spatial) model is informative even when basic vocabulary is concerned. Two Semitic languages intuitively chosen at the initial stage of analysis –– reconstructed Northwest Semitic and Akkadian–– are not merely geographically but also lexically closest to IE. Also, the Semitic family as a whole and Middle Egyptian are significantly closer to IE than are other branches of the Afroasiatic macrofamily whereas the one lexically furthest from IE –– Chadic –– is also geographically furthest. While no agreement with the genealogical classification of Afroasiatic languages is seen, the role of the areal factor is obvious. The Dravidian family cannot be regarded as a sister branch of either Kartvelian or any other. The analysis provides no evidence of genetic or areal ties between Dravidian languages and either of the two sister branches of the Indo-Uralic macrofamily. The totality of linguistic, genetic, and archaeological data suggests that the IE homeland was situated east of the Caspian Sea, close to the presumed Indo-Uralic homeland, which can be associated with the eastern Caspian Mesolithic. Proto-Uralians could correlate with the Kelteminar culture, and Indo-Hittites, with one of the early farming cultures of southern Turkmenia. The first Indo-Hittite migration route evidently passed across northern Iran toward Anatolia. Three secondary northward migration routes can be reconstructed: the first, along the western Caspian coast toward the Volga (Khvalynsk ancestors), the second, related to the Darkveti-Meshoko culture, along the eastern Black Sea coast to northwestern Caucasus (both in the fifth millennium BC), and a later migration, Leilatepe-Maykop, in the early fourth millennium, from Iranian Azerbaijan along the Kura valley to north central Caucasus. While all the three northward migrations contributed to the dispersal of IE dialects to the steppe, only the first one appears to have been accompanied by gene flow."

http://etnografia.kunstkamera.ru/en/archive/2019_3/kozincev_a_g_novye_leksikostatisticheskie_dannye_o_yuzhnom_adstrate_v_praindoevropejskom_yazyke

zardos said...

But in the end a group of CHG colonists must have come North to the Piedmont steppe, even if we won't find a pure CHG population in the Eneolithic time. The exact time and character of the admixture is what is still unknown.

About PIE it still stands that this language was brought by EHG, because that are the dominant paternal lineages and have a heavy autosomal impact as well. There is no other element in PIE of which both could be said: Absolutely dominating the male lineages and having the dominant overall genetic impact.

The adoption of pastoralism is a separate issue and will be connected to the timing of the CHG input and imho agro-pastoralism wasn't adopted by the steppe inhabitants at one time anyway, but in a step by step path. I'd still say that first came ovicaprids from the Transcaucasian area, at a time they Lower Don people were still largely hunters and fishers, with the later introduction of bovines from the West and finally the development of horse breeding from within, possibly with influences from Botai and related Eastern groups.

gamerz_J said...


@Davidski
"Hotu is a garbage sample, so it doesn't prove anything. It's just noise."

The official line from Harvard seems to be that the proximal population of Iranian/CHG-related into Yamnaya is still unknown.
https://twitter.com/iosif_lazaridis/status/1253750847659479040

(not my twitter account asking the question btw- just came across this)

gamerz_J said...

@epoch

I think it is that WHG does not give a very good fit and in some models CHG is also needed.

gamerz_J said...

I missed a lot of interesting comments here, the past few days but (and anyone pls correct me if wrong) does not Iran_N also have some ENA beyond the ANE, BE and Dudzuana related I see mentioned here? Probably more than CHG as well. So, could this not be used to test if populations that descend from it also have this signal?

gamerz_J said...

@TLT

"There is only a trace small of Iran_N related ancestry in Sardinia and that might as well either be a toss up between Iran_N proper and CHG clade. It would go back to the aceramic Anatolians, personally I am kind of leaning towards it being phylogenetically closer to the Iran_N/Zagros group even if it is a third group distinct both the Zagros specific and the (IMO more distant) Kotias specific CHG branches. However the Iran-like ancestry found eastward of Iran is most definitely not even close to CHG."

Could you elaborate a bit on this? Why would it be closer to Iran_N? Based on Global25?

vAsiSTha said...

This is what Laziridis said yesterday
"The proximal source of the Near Eastern ancestry in Yamnaya is not known. We have used Armenian/CHG/Iran over the years to model this as more samples became available."

This is the same position as that of Patterson, and i guess Reich as well.

It is a fact that Steppe_en cannot be modeled as EHG+CHG with any set of outgroups. Wang 2019 shows its possible but the details of only that model are completely missing from the supplement. Using Wang's setup gives p-values in the range of e-10 and do not match whats written in the paper. Even Davidski cannot get EHG +CHG model to pass. I had emailed Wang about this but have received no response. So its clear that we need a new source population different from CHG.

qpAdm cannot differentiate between CHG and Iran_N for yamnaya, khvalynsk & steppe_eneolithic with only Iran_N in the right pops. You need a CHG sample in the right pops as well so the model can differentiate between the 2. However, that will make you end up with CHG in both the right as well as left pops which is against modeling principles. A compromise is to break up CHG into Satsurblia & Kotias and use Satsurblia in right pop and Kotias in left pop. This should work as Kotias is a good 4 millenia later than satsurblia and the probability that there was minor admixture into an ancestor of Kotias sample from external sources is high. Using this, EHG + CHG + Geoksyur_En like model is a pass, as i have shown above.

Given that Patterson sometimes comments here, would love to get his take on this.

Anonymous said...

It is perfectly clear that CHG is much closer to Yamnaya than Iranian farmers. Therefore, the use of Iranian farmers in the text about Yamnaya creates a false impression. You can see that from the D-statistics (D, Z, s).

Mbuti.DG Yamnaya_Samara CHG Iran_N -0.032651761 -9.90778 0.0032956
Mbuti.DG Yamnaya_Samara CHG Iran_LN -0.030051322 -7.87958 0.0038138
Mbuti.DG Yamnaya_Samara CHG Iran_HotuIIIb -0.026203181 -4.69495 0.005581
Mbuti.DG Yamnaya_Samara CHG Iran_ChL -0.017181072 -6.51740 0.0026362
Mbuti.DG Yamnaya_Samara CHG Iran_recent -0.009159683 -2.52353 0.0036297
Mbuti.DG Yamnaya_Kalmykia CHG Iran_N -0.031994772 -8.71378 0.0036717
Mbuti.DG Yamnaya_Kalmykia CHG Iran_LN -0.031571172 -7.70066 0.0040998
Mbuti.DG Yamnaya_Kalmykia CHG Iran_HotuIIIb -0.023053164 -3.97537 0.0057
Mbuti.DG Yamnaya_Kalmykia CHG Iran_ChL -0.018901638 -6.61262 0.0028584
Mbuti.DG Yamnaya_Kalmykia CHG Iran_recent -0.007807957 -2.03964 0.0038281

Mbuti.DG Yamnaya_Samara Iran_ChL Iran_HotuIIIb -0.009323583 -1.86785 0.0049916
Mbuti.DG Yamnaya_Samara Iran_ChL Iran_LN -0.009754644 -2.87556 0.0033923
Mbuti.DG Yamnaya_Samara Iran_ChL Iran_N -0.016487468 -5.59159 0.0029486
Mbuti.DG Yamnaya_Samara Iran_ChL Iran_recent 0.009338622 2.91241 0.0032065
Mbuti.DG Yamnaya_Samara Iran_HotuIIIb Iran_ChL 0.009323583 1.86785 0.0049916
Mbuti.DG Yamnaya_Samara Iran_HotuIIIb Iran_HotuIIIb 0.000000000 0.00000 1.0000
Mbuti.DG Yamnaya_Samara Iran_HotuIIIb Iran_LN 0.001218970 0.16482 0.0073956
Mbuti.DG Yamnaya_Samara Iran_HotuIIIb Iran_N -0.010709445 -1.79306 0.0059727
Mbuti.DG Yamnaya_Samara Iran_HotuIIIb Iran_recent 0.017117877 2.54461 0.006727
Mbuti.DG Yamnaya_Samara Iran_LN Iran_N -0.007067214 -1.67005 0.0042317
Mbuti.DG Yamnaya_Samara Iran_LN Iran_recent 0.017324644 3.73795 0.0046348
Mbuti.DG Yamnaya_Samara Iran_N Iran_recent 0.024059571 6.02428 0.0039938
Mbuti.DG Yamnaya_Samara Iran_recent Iran_N -0.024059571 -6.02428 0.0039938
Mbuti.DG Yamnaya_Kalmykia Iran_ChL Iran_HotuIIIb -0.005512973 -1.05687 0.00521
Mbuti.DG Yamnaya_Kalmykia Iran_ChL Iran_LN -0.011225659 -3.10282 0.0036179
Mbuti.DG Yamnaya_Kalmykia Iran_ChL Iran_N -0.012707461 -3.90331 0.0032556
Mbuti.DG Yamnaya_Kalmykia Iran_ChL Iran_recent 0.012538626 3.79766 0.0033017
Mbuti.DG Yamnaya_Kalmykia Iran_HotuIIIb Iran_LN -0.013180332 -1.72185 0.00765
Mbuti.DG Yamnaya_Kalmykia Iran_HotuIIIb Iran_N -0.014033094 -2.22286 0.00631
Mbuti.DG Yamnaya_Kalmykia Iran_HotuIIIb Iran_recent 0.014916956 2.14297 0.0069
Mbuti.DG Yamnaya_Kalmykia Iran_LN Iran_HotuIIIb 0.013180332 1.72185 0.00765
Mbuti.DG Yamnaya_Kalmykia Iran_LN Iran_N -0.002245723 -0.48906 0.0045919
Mbuti.DG Yamnaya_Kalmykia Iran_LN Iran_recent 0.022585938 4.70698 0.0047984
Mbuti.DG Yamnaya_Kalmykia Iran_N Iran_recent 0.024235606 5.81438 0.0041682

zardos said...

To me a three part mixture could make sense with an original EHG + CHG steppe population and a third, CHG related, but not identical people, arriving in early Neolithic times, introducing new technologies (like waddle and daub constructions) and first domesticated animals (ovicaprids). However, if this third element exists, it won't change the general picture, because it would be the smallest contributor, not dominant, but quickly assimilated, leaving behind only some innovations and primarily autosomal, low level genetic influences.
Their role would be even less impressive overall than the one of the Western Neolithic populations for the ethnogenesis of PIE and their genetic make up. I asked about this possibility in the past, but only new samples might solve this question sufficiently?

vAsiSTha said...

Chg is closer to yamnaya than iran_n because of more shared alleles with EHG. This has been known since long.

Anonymous said...


We only see the genomic influence from the north to Iran, associated with the spread of wheel wagons, but we do not see the influence of Iran to the north to Steppe.

Matt said...

@epoch, it depends on what they mean by that. If they mean that EHG isn't broadly between WHG groups and AfontovaGora3/West SiberiaN, then that seems wrong.

If they mean that EHG may not have been formed by a pulse admixture between ANE and WHG, but may have been due to a trifuraction of ancestries followed constant bi-directional geneflow exchange to east and west. As in the sort of complex scenarios laid out in Harney's recent preprint. That's possible but we simply don't know.

Additionally some models like the ones from Yana paper would allow for EHG to have some ancestry from CHG, as seems more "basal eurasian" than product of EHG+WHG.

There is probably some meaningful genetic drift that separates populations from around Ukraine and Western Russia from simply being West Siberian/AG3+WHG.

E.g. some PCA plots of outgroup f3 statistics, PC1 to PC3: https://imgur.com/a/JSRwsIP

PC1 represents being loaded on f3 towards all Euro+Siberian HG groups.
PC2 splits WHG from AG3, with UkraineN / SHG etc intermediate.
PC3 gives a separate split of UkraineN on one end and AG3+Bichon on the other.

Notice that both the steppe EMBA (green dots) and MN EEF have a bit of scatter on the PC3 dimension. Compared to Yamnaya / CW_Baltic_early, the Catacomb set seem to have more affinity for UkraineN than suggested by their position on WHG vs AG3 dimension. (Some caution about this as low level changes in f3 can be artefactual).

Exploiting UkraineN and other eastern HG groups in pRight would probably add to the ability of pRight to distinguish between Steppe_EMBA streams, just as including more WHG related groups in pRight is used to split apart GAC / Iberia_MN, etc.

gamerz_J said...

@Matt

West Siberian_N refers to AG3?

Matt said...

Not quite, West Siberia N is the slightly later population (or set of samples) sampled in Narasimhan's paper that is broadly the same but seems to have some very low East Asian ancestry (single digits, less than Botai) and may have some admixture back from EHG as well. AG3 is the late Upper Paleolithic samples from Afontovagora.

vAsiSTha said...

@gamerz
refers to Tyumen_HG

gamerz_J said...

@Matt

Thanks, that was my understanding as well. I was just confused because I read your post as AG3 being the same as West_Siberia_N. I recall Narasimhan to have placed EHG on a cline between them and WHG with bi-directional gene flow but I am not sure as to the reason why he did that.

So basically it seems to me that there has always been bidirectional gene flow East-West Eurasia (or East to West mainly, I don't know) because it seems this East Asian/ENA ancestry is present among many ancient populations (EHG,ANE,CHG,Iran_N, Villabruna etc)

vAsiSTha said...

Not only does Steppe_eneolithic, yamnaya & khvalynsk prefer geoksyur+chg, so does steppe_maykop

Target,Distance,GEO_CHG,RUS_Karelia_HG,RUS_Tyumen_HG,TKM_Geoksyur_En
RUS_Steppe_Maykop:AY2003,0.03369183,8.6,24.2,48.4,18.8
RUS_Steppe_Maykop:SA6001,0.02532208,13.0,16.0,53.4,17.6
RUS_Steppe_Maykop:SA6004,0.02463860,15.4,24.4,45.0,15.2
Average,0.02788417,12.3,21.5,48.9,17.2

Significantly worse distance without geoksyur, from 2.7% to 3.4%
Target,Distance,GEO_CHG,RUS_Karelia_HG,RUS_Tyumen_HG
RUS_Steppe_Maykop:AY2003,0.03984472,24.6,19.0,56.4
RUS_Steppe_Maykop:SA6001,0.03236179,27.6,14.6,57.8
RUS_Steppe_Maykop:SA6004,0.03025204,28.6,21.6,49.8
Average,0.03415285,26.9,18.4,54.7

gamerz_J said...

@vAsiSTha

These are much later though aren't they?

Matt said...

@gamerz_J, basically I did that as West Siberia N and AG3 are close enough for the purposes of describing the general clines, even though they're not identical.

Yes, I would say that re bidirectional flow, though it's low. That said there is not really any (not recent) West Eurasian ancestry among many people today from quite north in East Eurasia. I'd expect this is probably because of systematic waves of expansion towards the north probably from China in the last 20kya from people who didn't really get any West Eurasian related ancestry (until recently and still in many cases not very much).

epoch said...

@Matt

Ah, OK. As in EHG being not the result of one clear admixture event.

Best possible fit (4%-5%) I get with Golbal25, for what that's worth, is with Narva and AG3. That is interesting because from that recent Krim Gravettian paper you could see that the site was abandoned at onset of the LGM and only reused somewhere around 10,000 y BC, with a cultural layer assigned to the Swiderian culture.

That is interestingly far eastward. Far enough to be part of a contact zone. Narva is supposedly coming from Kunda, which itself comes from Swiderian.

Maybe I see too much in this though.

[1] "distance%=4.4106"

RUS_Karelia_HG

RUS_AfontovaGora3,56.8
Baltic_EST_Narva,43.2


[1] "distance%=4.9667"

RUS_Samara_HG

RUS_AfontovaGora3,51.2
Baltic_EST_Narva,43.2
GEO_CHG,5.6

Anonymous said...

"Steppe_eneolithic" is a deceptive name given to misleading, it doesn't reflect the essence, it puts a lie in the name misleading.

old europe said...



@epoch

I remember that Narva is the culture where they find a whole bunch of R1b 297. There have been rumours that these folks could be ancestral to the samara R1b hunter gather. I wonder if these samples were ultimetely also from the swiderian culture in Poland. I guess it is a possibility. Because narva was 75% WHG and Swiderians were likely full WHG.
Also some R1b M 269 seem to pop up not far from Narva. I mean the hints about Volosovo samples yet to be published.

Mayuresh Madhav Kelkar said...

http://etnografia.kunstkamera.ru/en/archive/2019_3/kozincev_a_g_novye_leksikostatisticheskie_dannye_o_yuzhnom_adstrate_v_praindoevropejskom_yazyke

Wish Kozintsev cited linguist Igor A. Tonayen-Belayev who could be living across the street from him in St. Petersburg.

old europe said...


Intersestingly enough I came up also with the Dneper Donets foragers that seem to have been linked to a migration from north eastern europe. Some authors link them with the swiderian too. And guess what...they were predominantly of the same Y line of narva and kunda R1b and I2 something.........

maybe Archi could provide more informations...
R1b and I2 .....a combination that sounds too familiar...

old europe said...


I'm referring to this kind of trail:


After the Ice Age, we have Swiderian culture in modern Poland, circa 11,000 B.C.E. — circa 8,200 B.C.E. Swiderians were not newcomers from the east. According to Wikipedia: “Rimantiene (1996) considered the relationship between Swiderian and Solutrean "outstanding, though also indirect", in contrast with the Bromme-Ahrensburg complex (Lyngby culture), for which she introduced the term "Baltic Magdalenian" for generalizing all other North European Late Paleolithic culture groups that have a common origin in Aurignacian.”

Then, we have Kunda Culture, with its roots in Swiderian culture in the Baltic forest zone extending eastwards through Latvia into northern Russia dating to the period 8000–5000 BC.

Later on, we have the Dnieper–Donets culture (ca. 5th—4th millennium BC) in the area north of the Black Sea/Sea of Azov between the Dnieper and Donets River and Samara culture (late 6th and early 5th millennium BC) at the Samara bend region of the middle Volga. Again according to Wikipedia:
“There are parallels with the contemporaneous Samara culture, and a larger horizon from the lower half of Dnieper to the mid-to-lower Volga has been drawn, particularly by the advocates of the Kurgan hypothesis as expounded by Marija Gimbutas. Dmitry Telegin (ru) assigns them to a broad cultural region that spanned the Vistula in Poland southeast to the Dnieper.”

old europe said...


@rob

Yes the definitive refinement of PIE was likely due to the interaction between proto europoid HG of the Dneper Donets culture with the balkan carpathian cultures and globular amphora. Every piece of the puzzle now seems to fit into the picture.

what is CHG is irrelevant in the PIE equation. It was there but the real deal was what happened between the Dneper and the balkans. And the I2 were important in the PIE stage

old europe said...



@juan rivera

it is a fact that r1b is an EHG lineage. problem is if it comes from the WHG side or the ANE side of the equation EHG equation.
also nearly 100% uniparentals from the east but narva folks ended up being 75% WHG....

Anonymous said...

@JuanRivera "The EHGs from Karelia and Samara have extra ANE
including Y-DNA Q1a2, which is clearly from EHGs."

No, Q1a2 is clearly from ANE.

Anonymous said...

Q1a2 most likely formed a WSHG cluster.

Matt said...

@old europe, the thing is that while you can hypothesize that there is a language transmission chain between some autosomally and uniparentally different group towards the steppe, this seems like it will probably never really going to be accepted by historical linguistics. They really work on the basis of attested distribution and paleolexical reconstruction (through written and spoken sources, and histories that can without too many steps be reconstructed uncontroversially from those).

These models that go "Well, I think based on the transmission of archaeological culture and archaeogenetics that these was this linguistic transmission... but I don't actually have any actual linguistic evidence of this, because it was all overwritten by later language expansions" are going to be met with "That's all very fine, but we can never say it that's actually true or not". (I think then one could counter "Transmission demonstrated in archaeology and archaeogenetics should be prima facie considered to be language transmission, with the burden of proof that it's otherwise!" but I don't think linguistics will ever accept this, and any linguist to hand will very likely immediately launch into a long monologue listing all the cases that violate the assumptions).

The actual attested distributions and very immediate inferred distributions really are the final frontier of evidence here.

old europe said...


@juan rivera

r1 is from Siberia sure but I'm referring to the kind of poulation( genetic marker ) that brought the R1b P297 to the Middle Volga. ( which is important for us at least in this part of the world)
Anyway I'm going with ancient dna. If we find R1b P297 older than the Narva ones east of the Volga and overpacked with ANE ancestry I will follow the evidence. I have no ethno agenda really.

old europe said...


I men R is from Siberia for sure

Gaska said...

@Richard Rocca said "Gaska-looks like France, the final place you've been hiding your fantasy scenarios, crumbles this week as well. Good luck in your future endeavors"

Final place?-We just found 3 Neolithic farmers R1b-M269 buried in Swiss dolmens/collective burials that have nothing to do with the BBC or with the CWC-It seems that your informants about the Swiss paper were wrong, but you can check the thread “Is Yamnaya overrated” and you will see that 7/8 months ago we were already commenting that this could happen because the rumors that reached me were very different-You no longer speak of mass migrations but of solitary explorers, nor of the Yamnaya culture but of the CWC and not even about individual burials following steppe customs but of collective burials in dolmens-You and many of the scientists you quoted were wrong about R1b-L51, but don't worry, you can always continue to indoctrinate your smart cultured, anthrogenica friends aand keep using stupid terms like "kurgan bell beakers"-Exogamy and CWC are sufficient to explain the famous steppe ancestry detected in the first quarter of the third millennium BC and Switzerland has shown that R1b-L51 has nothing to do with steppe cultures (just the opposite of what you said), so I wish you luck in your efforts to find that marker in the SGC/Poland/Russia/Ucrania or Mongolia, you will certainly need a lot-Remember this site-Valle de las Higueras (Huecas, Toledo), you might get a surprise, others will surely suffer a stroke


Simon Stevin said...

@gamerz_J There is no East Asian ancestry in Khvalynsk, Piedmont Steppe, Sredni Stog, Samara, Karelia, Mesolithic/Neolithic Baltic, and Yamnaya. Clearly the relationship with WSHG, is based on a common ANE ancestry. Every paper you look into on the subject, concerning Khvalynsk, Yamnaya, Piedmont Steppe, and Sredni Stog, demonstrates this. We also need more data on Neolithic and Mesolithic Siberia before we make anymore definitive statements on ancestry.

gamerz_J said...

@Matt

Re: ENA in EHG
It seems to me to be around 2-3%. But it is not always needed either.

Re: West Eurasian in ENA
It seems to me to be around 5-15% in the northern and western Chinese provinces, (ie Inner Mongolia, Qinhai, Xinjiang) particularly in the nothern minorities. It is of course much bigger in Xinjiang Uyghur and Kazakh, exceeding 40%. Most of it should be relatively recent, at least in Siberians,I have seen ALDER estimates of 4-6 generations.
There is some in Northern Han as well, about 2-4% I think but please correct me if I am wrong.
Generally, there seems to be more ENA gene flow in West Eurasians than the reverse (at least so far).

gamerz_J said...

@Simon Stevin

To be more accurate I don't disagree that most of these population can be modeled without any East Asian ancestry (recent one that is) but some papers have suggested low East Asian in EHG and CHG which contribute to many of these populations.

vAsiSTha said...

Wheat and barley spread to east ukraine from Jeitun/SW asia through the caucasus.
So did livestock as per the paper posted by @mzp1 above

2019 paper https://www.nature.com/articles/s41598-018-35758-w
Intensification in pastoralist cereal use coincides with the expansion of trans-regional networks in the Eurasian Steppe

"After the initial domestication of wheat (Triticum sp.) and barley (Hordeum sp.) in southwest Asia by 8,500 cal BC, these cultivars spread into the Iranian Plateau by 6000 cal BC and into Pakistan by c. 5500 cal BC12,13,14,15. In southern Central Asia, domesticated forms of barley and wheat identified in Jeitun cultural contexts in Turkmenistan suggests these cultivars had spread to western Central Asia around 6100 cal BC16,17. Domesticated wheat and barley were cultivated in the southern Caucasus by c. 6000 cal BC18,19 and were present in the northern Caucasus by c. 4500 cal BC20. Wheat and barley spread into eastern Ukraine by 4000 cal BC and into the Crimean steppe by 3400 cal BC19."

Anonymous said...

vAsiSTha said... "Wheat and barley spread to east ukraine from Jeitun/SW asia through the caucasus."

Nothing like that, it doesn't say that, don't be deceived.

vAsiSTha said...

commenting on Hajji_Firuz_IA as somebody said it has steppe ancestry.

I2327 has absolutely 0 steppe ancestry. it has absolutely 0 WHG.

left pops:
Iran_IA_HajjiFiruz

Iran_C_HajjiFiruz: 76.8 +- 3
Kyrgyzstan_BA_Aygirdjal: 23.2 +- 3
p-value: 0.9359 (extremely high p value)
result file https://pastebin.com/x7WRcYzk

Aygirdjal can be modeled as Sarazm_en + Botai

same is confirmed with vahaduo

Target,Distance,IRN_Hajji_Firuz_C,KGZ_Aigyrzhal_BA,RUS_Sintashta_MLBA
IRN_Hajji_Firuz_IA:I2327,0.02832953,76.0,21.6,2.4
Average,0.02832953,76.0,21.6,2.4

Target,Distance,IRN_Hajji_Firuz_C,KGZ_Aigyrzhal_BA
IRN_Hajji_Firuz_IA:I2327,0.02856426,76.8,23.2
Average,0.02856426,76.8,23.2

Samuel Andrews said...

David Anthony is the led author of the upcoming Harvard paper with new Eneolithic Steppe and Yamnaya DNA. He is the Indo European expert in David Reich's lab.

He is open to the idea, Proto-Indo Europeans were a long-standing population from Southern Russia not a recently mixed CHG-EHG pop.

"Yes, I actually think we agree on almost everything. I had not said that the genetic admixture happening earlier, but I'm very open to it. It fits with the underlying north-south dynamic that I think we agree on. After all, why did the IE gods live on mountains if their homeland was just a flat plain? But a plain near the highest mountains in that part of the world fits pretty well. I've been hired by the Reich lab to bring all of these steppe samples to publication--it's an experiment in which an archaeologist takes the first shot at a draft narrative in a genetics report. It's certainly a challenge for me, but we are moving along slowly.

David A"

Samuel Andrews said...

The "North-South dynamic" he is talking about, I think is that North Eneolithic are pure EHG while South Eneolithic are CHG-EHG (look like Yamnaya).

old europe said...



@samuel andrews

if the northern foothils of the caucasus harbors the PIE population were are the Y lines that provides the expansion from the steppe. Whre are there in the northern caucasus?

R1a and its expansion is from the Dneper
R1b M 269 has been found in the elite burials in the Suvorovo culture (Bulgaria). I2 also not from there
R1a another in north eastern Romania.
Cultural package of single grave culture part native Sredni stog and part old european borrowings.

So the strong patrilocal patrilineal clans of the western steppe were overrun by which male tribes in particular?

Anthony is aware of the presence of the carpathian mountain range?

Mayuresh Madhav Kelkar said...

"Wheat and barley spread to east ukraine from Jeitun/SW asia through the caucasus.
So did livestock as per the paper posted by @mzp1 above"

A related presentation by Michael Frechetti from the 20 min mark


https://www.youtube.com/watch?v=lP16l1PFoLA&t=2854s

Jatt_Scythian said...

@Samuel Andrews

Thanks for the info.

Did he give you a hint to forest steppe y dan vs steppe y dna?

Matt said...

@gamerzj, yes, talking about ENA total, EHG seemed in past papers to likely have some. (When talking about West Siberia vs AG3+EHG, I meant, "East Asian (in excess of that which may have contributed to ANE)").

It's difficult to gauge low level in modern populations; I've read that about Northern Han as well. Adna+qpAdm seemed to reject very low level West Eurasian contributions in modern East Asian folk from Northern China (and likewise the reverse), or be unclear, but given the problems and advice not to model recent populations with ancient dna in Harney's paper, and some of the dispersal of ancients on easily measured f stats (e.g. Native American vs North Eurasian) which would provide a proxy, and how confounded those are by stats, I think it's difficult to really check this at all at very low bounds (1-3% ish).

(To be honest, looking at f3 stats, I think there is room for uncertainty even about how much in Europe Steppe vs EEF contributes to specific recent populations compared to forebears in the same region. Some of the batch effects between modern and different ancients may be too severe.).

Anonymous said...

old europe said... "R1b M 269 has been found in the elite burials in the Suvorovo culture (Bulgaria)."

No, Smyadovo burials do not belong to Suvorovo group. That burial is not an elitist one.


Davidski said...

@Jatt_Scythian

There isn't a huge difference between the Y-haplogroups in the steppe and forest steppe of Eastern Europe during the Eneolithic.

In both regions R1b is the most common Y-haplogroup while R1a shows up sporadically.

Based on what I've seen, the main difference seems to be that Q1a is a lot more common in the forest steppe.

Anonymous said...

@Davidski "Based on what I've seen, the main difference seems to be that Q1a is a lot more common in the forest steppe."

Bronze Steppe Maykop Russia Sharakhalsun 6 [SA6004 / kurgan 2, grave 18, BZNK-003/4] 3336-3105 cal BC (4500±40BP ,GIN-12401) M Q1a2

"In both regions R1b is the most common Y-haplogroup while R1a shows up sporadically."

The Mesolithic population of North-Eastern Europe was R1a, the Neolithic population of Eastern Europe was mainly R1b. The Mesolithic population was replaced by the Neolithic population (including the Volosovo culture and the Dnieper-Donets), while the Neolithic population was completely replaced by the unrelated Eneolithic population. These are all different populations.

gamerz_J said...

@Matt
Apologies, do you mean ENA total here as in via ANE as well? Then yes I also agree, what I think is not certain is whether it has East Asian or not, albeit that has been suggested too but at very low levels (1-3%).

About northern Han I was referring to Chiang et al (2018) as well as the recent paper on East Asia. Chiang et al I think suggested something like 5% which to me seems to much for Han populations. But I could be wrong. In any case, it is localized in the north.

"Adna+qpAdm seemed to reject very low level West Eurasian contributions in modern East Asian folk from Northern China (and likewise the reverse" Could you please elaborate on the Adna and qpAdm rejecting it in northern Han? Are you referring to a study?

Small percentages are always hard to trace, I think that was an issue already dating to STRUCTURE/ADMIXTURE algorithms.

"To be honest, looking at f3 stats, I think there is room for uncertainty even about how much in Europe Steppe vs EEF contributes to specific recent populations compared to forebears in the same region."
As you probably know I am new to this but trying to understand f3 stats it does seem that even they (and most methods apparently, even D stats etc) have limitations etc. Lipson (2020) was a really interesting read but I still don't understand them perfectly.



Rob said...

@ old Europe

“Yes the definitive refinement of PIE was likely due to the interaction between proto europoid HG of the Dneper Donets culture with the balkan carpathian cultures and globular amphora. Every piece of the puzzle now seems to fit into the picture.

what is CHG is irrelevant in the PIE equation. “

Exactly , that makes perfect sense anthropolgically , genomically, archaeologically

Jatt_Scythian said...

@Davidski

Thanks.

What does that say about where R1a was during the Eneolithic? Why do you think R1a was less common?

Rob said...

@ Zardos

“ About PIE it still stands that this language was brought by EHG, because that are the dominant paternal lineages”

Wow; that’s some high level evaluation, Zardos

EastPole said...

@Samuel Andrews

“why did the IE gods live on mountains if their homeland was just a flat plain? But a plain near the highest mountains in that part of the world fits pretty well.

David A"

pcw250 R1a R-Z645
pcw420 R1a R-M417 (xZ645)
pcw430 R1a R-M417 (xL664, FGC9988)

https://anthrogenica.com/showthread.php?p=660299#post660299

In Polish languge dad is ‘ta-ta’ or ‘pa-pa’, mom is ‘ma-ma’
‘*pa-tri’is ‘pa-pa’ who looks after, holds, ‘*ma-tri’ is ‘ma-ma’ who looks after, holds.
‘*ta-tri’ would mean ‘ta-ta’ who looks after, holds, or father.
Interesting that ‘Ta-tri’ is also the name of the highest mountain range in the Carpathian Mountains.

https://i.postimg.cc/SNw7SqBK/Tatry.jpg

If my theory that CWC-R1a were PIE is true then David A’s statement would make a lot of sense from linguistic point of view.
Caucasus mountains names are not IE at all.

Jatt_Scythian said...

@EastPole

Your theory is wrong. Its likely both R1a and R1b spoke PIE. if it originated among one of them then we'll never know.

Rob said...

@ Matt

“ the thing is that while you can hypothesize that there is a language transmission chain between some autosomally and uniparentally different group towards the steppe, this seems like it will probably never really going to be accepted by historical linguistics. They really work on the basis of attested distribution and paleolexical reconstruction (through written and spoken sources, and histories that can without too many steps be reconstructed uncontroversially from those)”

Dunno what your rambling on about , because that’s what the data shows
I don’t think you’re in a position to dictate what’s going to be accepted or not

Davidski said...

@Jatt_Scythian

What does that say about where R1a was during the Eneolithic? Why do you think R1a was less common?

R1a was found in much of Eastern Europe during the Eneolithic, but usually rather thinly spread.

R1b was more common probably because it was around for longer. Overall it's an older haplogroup with a deeper mutation structure.

Anonymous said...


The hypothesis that early CWC had populations with R1a and with R1b is unscientific, it is wrong. Early CWC was one population (with R1a). BBC were another population (with R1b). The BBC was not Indo-European speaking, it is a proven fact. And the mixing happened later.

Matt said...

Rob, I'm not interested here in having another discussion where you swear and insult your way through assertions that historical linguistics should simply accept historical linguistics whatever you assert is true based on your read of archaeological evidence, with no actual attested linguistic evidence, and repeatedly delete your own posts. It's not controversial or an unusual viewpoint that historical linguistics bases its consensus of linguistic history on linguistic evidence which can only be *supported* and *corroborated* by archaeology, not blog comment rants which making linguistic history up whole cloth from an individual's personal interpretation of archaeology. The people you have to persuade on this are the historical linguistics community, so perhaps if you want to make this argument, you would probably be placed to go and make this argument to them on a forum that serves them.

Anonymous said...


As far as I understand Davidsky, he is talking about the Volosovo culture, but it is not Eneolithic, but Neolithic. It is only attributed to the Eneolithic as a synchronization with other cultures of the South.

Rob said...

@ Matt
And I’m not interested in your empty pontificating
What O/ E is correct ; and linguistics supports it
Now if you have some data or model to contravene it, lay it out

EastPole said...

@Jatt_Scythian

“Your theory is wrong. Its likely both R1a and R1b spoke PIE.”

It is just a theory, so it can be wrong. But…
R1a speaks more archaic IE languages then R1b.
R1a links Europe and India, so it was truly Indo-European.

Where is the proof that R1b spoke PIE and how and when and where had it indoeuropeized R1a?

vAsiSTha said...

Q: has R-Y3 or it's subclade been found in any ancient so far?

Jatt_Scythian said...

@EastPole

More branches of IE can be linked to R1b than R1a including the earliest ones , Anatolian and Tocharian (unless the Tarim R1a is really Z93-)

Mayuresh Madhav Kelkar said...

Anthony:

""why did the IE gods live on mountains if their homeland was just a flat plain? But a plain near the highest mountains in that part of the world fits pretty well. "


Ok. Here is the list:


Dyaus Pitar (Vedic), Zeus Pater (Greek), Jupiter (Roman), Dei Patrous (Illyrian), Dievs (Baltic).
Uṣas (Vedic), Eos (Greek), Aurora (Roman), Aushrine (Baltic).
Varuṇa (Vedic), Odinn/Wodan (Germanic), Ouranous (Greek), Velinas (Baltic).
Asura (Vedic), Aesir (Germanic), Ahura (Avestan).
Marut (Vedic), Ares (Greek), Mars (Roman).
Parjanya (Vedic), Perkunas (Baltic), Perunu (Slavic), Fjorgyn (Germanic).
Traitana (Vedic), Thraetaona (Avestan), Triton (Greek).
Aryaman (Vedic), Airyaman (Avestan), Ariomanus/Eremon (Celtic).
Saramā/Sārameya (Vedic), Hermes (Greek).
Pūṣan, Paṇi (Vedic), Pan (Greek), Vanir (Germanic).
Rudra (Vedic), Ruglu (Slavic).
Danu (Vedic), Danu (Irish).
Indra (Vedic), Indra (Avestan), Inara (Hittite).
Śarvara (Vedic), Kerberos (Greek).
Śrī (Vedic), Ceres (Greek), Freyr/Freya (Germanic).
Bhaga (Vedic), Baga (Avestan), Bog (Slavic).
Apām Napāt (Vedic), Apām Napāt (Avestan), Neptunus (Roman), Nechtain (Celtic).
Ṛbhu (Vedic), Elbe (Germanic).
Yama (Vedic), Yima (Avestan), Ymir (Germanic).

Now find the highest mountain with the flattest plains around these.

mzp1 said...

It's not even an advanced topic, just basic research would reveal that to be incorrect. It does seem like he lives in his own bubble. I do fear for the Reich paper if they let his imagination loose on it.

zardos said...

@Rob: In the ancient world of the steppe there were primarily two modes of language transmission: 1st by demic diffusion on a big scale or 2nd by demic diffusion on a limited scale with male elite dominance.
The already attested uniparental and autosomal profiles of the steppe people and their descendents don't allow both from any other people than the EHG hunter fishers. There is no way around this, no other logical explanation left for the early phase.

For the later phase in the Bronze Age there would be a miniscule chance for the 2nd option for many parts of the Indo-European world, but you won't get it straight for the whole sphere of IE speech.

Whereever you go you need specific clades of R1a and R1b. They are the most consistent markers for the PIE people and modern IE speakers.

If there ever was a predominantely CHG population speaking an IE language, chances are high we will find an elite which is comparatively more on the EHG side of things in comparison to the commoners.

As I see it, you don't need to introduce complexity when the simplest models, best supported by the known facts, don't just explain the case the best, but are actually the only ones which can explain it thoroughly and without introducing special pleading.

Your model might actually be worth considering for one group of steppe descendents: For the majority of BB, because a large fraction of BB seem to have actually adopted farmer languages. But the same is not proven and highly unlikely for the root of PIE, for which we have not the slightest proof for 1) or 2) as mentioned above.

Angantyr said...

@old europe:

"it is a fact that r1b is an EHG lineage. problem is if it comes from the WHG side or the ANE side of the equation EHG equation.
also nearly 100% uniparentals from the east but narva folks ended up being 75% WHG...."

Really? When I count all Narva samples published in Jones 2017/Saag 2017/Mathieson 2018/Mittnik 2018 I find (ISOGG 2019 nomenclature):

1 I (undetermined)
2 I2a1a-P37.2/PF4004
1 I2a1a1b1a1-L233/S183
1 I2a1a2-M423
2 I2a1b1a-CTS616
2 I2a1b1a2-CTS10057

1 Q1b-M346
1 R1
1 R1b1a1-P297/PF6398
5 R1b1a1-P297/PF6398(xR1b1a1b-M269/PF6517)

Let's add one more late R1b1a1-P297 with uncertain cultural belonging, then we get 50% each for "eastern" and "western" Y Hgs.

Rob said...

@ Zardos
Trouble is , people don’t really evaluate the data fully. They just see in it what they want to

So when they say PiE is associated with EHG , it actually tells us nothing, because EHG, ukrHG, WHG are just points on a punctuated cline who’s descriptive utility beyond the Stone Age become irrelevant
R1a-M17 and R1b-M269 probably originated in different parts of the Sredni Stog chain , so what use is there in sweeping then under one rubric
So Who lived on the steppe during the Neolithic/ early eneolithic ? An entire gamut of communities with different lineages , ranging from I2a2 in west to Q1 and J1 in south/east; all with distinctive autosomal make up initially
What happened subsequently ? A homogenisation in genomic make up, parallleled by homogenization in burial ritual (exactly what Matt is arguing against)
From a YDNA perspective; this seems to have played out by a marked reduction in diversity, with prominence of 3 major lineages (for reasons which remain to be determined, but obviously territoriality and dominance played a part)

So if we say R1b is the most important or R1a, what do you see this on :
Are they the earliest ? Are they the richest ? Are they the proto-anatolians ? Have you considered relative demography to account for observed patterns ?
Has any alleged paleolexicon shown that PIE comes from the R1b hills ?
Your analysis ignores all these aspects

mzp1 said...

The early R1B in Western Europe, it comes from Cattle Herders who travelled from Asia and reached as far as Western Europe. This is how widespread was early IE. Their relatives in Antolia created Farming and with Yamnya you are seeing the expansion of a late-IE, post-farming society.

Talking about the formation of Yamnaya is just talking of the formation of Yamnaya, not necessarily the formation of PIE.

Yamnaya is IE but not necessarily PIE. PIE from a purely linguistic standpoint, is a predominantly Cattle herding population with the use of Wheels and Horses. Linguistically, and according to the most conservative layers (Indo-Iranian) this is a pre-Neolithic copper age society.

I am looking through the data on Mesolithic Cattle Management and will post a comment here when I can write it up to make a good case.

zardos said...

@Rob: We have now data points from the start to the end of the journey of PIE and in all these data points we find similar clades of R1a and R1b, which first appear in EHG hunter gatherers, which spread with EHG hunter gatherers and which keep up their dominant position throughout all these changes and expansions.

Now you could say, "but hey, in the important phase of the horizon X we have another lineage from the West jumping in, taking over and being the true PIE founders" - but where is this? Where is this proof for a continuous chain? And why and into what did they disppear soon after in too many branches?

Probably new data points can change the picture somewhat, but not completely any more. There are enough data points already for supporting the story of EHG hunter warrior lineages taking over in the North Pontic area and moving on from there. Where are the dominant lineages of Yamnaya, Corded Ware, Bell Beakers and Usatovo supposed to come from? They all came form the steppe region, they all are R1 with the EHG:CHG profile. Its up to anybody who criticises this theory to prove the hole in the logic, to prove a broken chain, not vice versa. If you have a continuous chain from A - B - C - D its the most likely path. No complexity needed. Any alternative needs to break the chain.

The same will be true for the developed culture from the Lower Don (not sure for the earliest phase, but its about continuity from the later phase after all) and Sredny Stog proper.

What kind of additional data do you need to be convinced? What if Cernavoda and Troy show up with steppe and R1 lineages? Not even on a big scale, but sufficient for the transmission, probably with a reduction North to South. Would that satisfy you?
Even if Cernavoda and Troy would not produce the results, this won't disprove what I said before, but would make us require a 2nd thought. But what if we find steppe influences in Cernavoda and Troy already in larger samples? I'd say that's all what is still needed and if its done, you will find traces in Anatolians too, no doubt about that.

"From a YDNA perspective; this seems to have played out by a marked reduction in diversity, with prominence of 3 major lineages (for reasons which remain to be determined, but obviously territoriality and dominance played a part)

So if we say R1b is the most important or R1a, what do you see this on :
Are they the earliest ? Are they the richest ? Are they the proto-anatolians ? Have you considered relative demography to account for observed patterns ? "

This are important questions. We probably will never know for hundred percent, but for 90 plus percent. Because if we have a hunter fisher lineage which is on top of its tribal people throughout all phases, and expands over others, what is what we see so far, in all available data points, who is supposed to have changed their language how? Its not likely that they adopted a foreign language, but only influences from other languages as they seemed useful. Anything else is remotely possible, but highly unlikely.

mzp1 said...

@Zardos

"story of EHG hunter warrior lineages taking over in the North Pontic area "

Do you agree that Linguistics and Literature point to PIE being a Cattle Herding society (not hunter warriors), with Copper, Horses, Wheels but no Bronze or Agriculture? I'm not saying your wrong, but I'm asking how you square the circle here?

R1A and R1B are ANE lineages. Those EHG guys came from the East, where ANE was in contact with IE populations in Central Asia.

If you can actually show the formation of Yamnya in the Pontic-Caspian you may have an argument. At the moment it looks like EHG and Yamna are both arriving West from the East, fully formed. You need a population significantly greater than 50% CHG for EHG to mix with and you will not find the required donor population to be able to say Yamnaya formed in the PC Steppe.

«Oldest ‹Older   201 – 400 of 911   Newer› Newest»