search this blog

Saturday, February 13, 2021

The Uralic cline with kra001 - no projection this time


A whole lot of nonsense was posted online, often by people who should've known better, after I claimed that kra001 was a solid proxy for a proto-Uralic genome (see here).

For those of you who still don't get it, below are three Principal Component Analysis (PCA) plots featuring Uralic speakers and other present-day Eurasians. Kra001 is also there. These graphs are based on genotype data not reprocessed Global25 data. The relevant datasheet is available here.

Compared to my previous PCA with kra001, here I included a bigger range of East Eurasian populations to help mitigate the effects of extreme genetic drift in some of the Siberian groups, at least on the first few Principal Components (PCs). Moreover, kra001 wasn't projected onto PCs computed with modern-day samples, so he was free to influence the outcome of the PCA.


Note the east to west clines made up largely of Uralic speaking groups on the first two plots. These plots are based on PCs 1/2 and 1 /3, respectively. The third plot, based on PCs 1/4, is more complex and thus more difficult to interpret, but it also manages to isolate many of the Uralic populations from the others.

The Uralic-specific clines do intersect with the clines and clusters formed by the other linguistic groups. However, based on the three plots, the Yeniseian-speaking Kets are the only Asian group that can plausibly be confused for Uralic speakers.

Importantly, apart from the Kets, kra001 is the only Asian individual who shifts his position on all three plots as if he were a Uralic speaker. This might well be a coincidence, and we'll never know what language was spoken by kra001, but it does suggest to me that his genome is a solid proxy for a proto-Uralic genome.

See also...

First taste of Early Medieval DNA from the Ural region

The BOO people: earliest Uralic speakers in the ancient DNA record?

Fresh off the sledge

772 comments:

«Oldest   ‹Older   601 – 772 of 772
Simon_W said...

@Mary

"I only know one author who dedicated himself to explaining the spread of this phenotypic trait, Peter Frost. According to him, blue eyes have become common due to sexual selection."

That's also the view endorsed by Johannes Krause in the interview I recently mentioned. Well, personally I don't believe that blue eyes are really more beautiful than brown eyes - and the same can be said about blond hair - but both tends to be rare on a global level, and may have been even rarer in the distant past, so they may have been rather unusual in the beginning, and thus interesting to fascinating.

Simon_W said...

BTW, I see pretty well in the dark, in spite of dark brown irises; but from the outside it's hard to tell if I see as well as Archi. :D

Davidski said...

@Norfern-Ostrobothnian

Jaakko Could we get a quick rundown on why proto-Uralic was spoken in Volga-Ural at 2000 BCE?

He can't give us a quick rundown because it didn't happen.

Anonymous said...

@Dmytro

Wikipedia is written in a wrong. You can't believe everything it says.

The most archaic Slavic hydronyms between the Bug and the Vistula, but the area that you described there Slavic hydronyms on top of the Baltic, Iranian and other languages ​​are quite obviously associated with Korchak culture. Komarov culture even hypothetically can not be Slavic, there is nothing in common with the Slavic elements at all, neither by ceramics, or on burials, nor in a lifestyle, nor contacts, is even impossible to imagine to the Slavs from it.

About the sea and mountains there is written clear not true, the sea of ​​sea is the sea and nothing but the sea, the mountains and rocks are the mountains and rocks, all this is the general Slavic word/concepts. The rest of the concepts about the lakes and swamps do not prove in any way as these concepts are everywhere, the Slavs of the words for the designation of the swamps as other languages, one general, and a pair of three more specific, it is not like the Balts where something about 200 words for designations of marshes.

Anthony Hanken said...

@Jaakko

I've tried to respond to your previous comments twice, but for some reason it's not being posted. So, I'll try once more, with this point.

"You should understand that there is no predicting power of the momentary crossroads between different levels (which I called correlations). Think about it:
1. In certain population speaking language X there are different uniparental lineages.
2. Now, if we go back or forth in time, the situation is such, that we cannot see from DNA if a carrier of certain lineage spoke the language X or not."

What happens when there are multiple "momentary crossroads" with the same correlations? So much so, that migrations can be traced.

Y-hg N has been found at elavated levels, in multiple ancient populations, previously established to be Uralic speaking. The same is true for modern populations, whatever that's worth.

Aleph said...

@Davisdski

Why was my most recent reply to Genos Historia not published? It has been well over a day.

ambron said...

Dmytro, I will find more citations from serious scientific papers at my leisure, but for starters:

The utilization of geographical names yields the following findings:— The centre of Old Slavic names is situated on the northern slope of the Carpathian Mountains, approximately between Bukovina and Krakow; it is based on a substrate of older, Indo-European hydronyms.— The expansion of the East Slavic tribes bypasses the Pripyat Marshes and extends further through Central Russia and especially to the North and the East. — West Slavic settlers reach their new settlement areas through migration from Bohemia and further on to Saxonia and Thuringia, and also through Western Poland to Brandenburg and Mecklenburg-Western Pomerania.— The migration of the South Slavs takes place in two big, yet separate flows, on the one hand through the Moravian Gate to Slovenia, Hungary and Croatia, and on the other hand on the Eastern edge of the Carpathian Mountains to Serbia and Bulgaria.

https://onomastica.ijp.pan.pl/index.php/ONOM/article/view/214/151

Jaakko Häkkinen said...


Norfern-Ostrobothnian:
“VL29 is most definitely a proto-Uralic, or at least European Uralic clade.”

You cannot claim that. How you could see the language from the genes? There are many possible interpretations for the genetic results:

1. It is not a perfect match with the Uralic peoples: it is not found within the Hungarians, Udmurts, Northern Komis, Mansis, Khantys and Nganasans. Outside the Uralic language family it is found within the Russians, Belarussians, Ukrainians, Poles, Latvians, Lithuanians, Crimean Tatars, Volga Tatars, Bashkirs, Chuvashes, Siberian Tatars, Dolgans and Karanogays.

2. Even if it was a perfect match with the Uralic peoples, you still couldn’t know whether it spread earlier, later or simultaneously with the Uralic language.

Do you understand? There are many possible interpretations, so you cannot just decide that it is connected to the spread of Proto-Uralic. That is unscientific.

NO: “@Jaakko
Could we get a quick rundown on why proto-Uralic was spoken in Volga-Ural at 2000 BCE?”

Quick and short is not possible, because it is a long story where location, time and the taxonomic structure of the language family intertwine. You can read the main arguments in Finnish here:
https://www.sgr.fi/susa/92/hakkinen.pdf

Jaakko Häkkinen said...

Anthony Hanken:
“What happens when there are multiple "momentary crossroads" with the same correlations? So much so, that migrations can be traced.”

What do you mean? Still there are gene flows from different directions and/or at different times. You cannot just decide that one of them is connected to a certain language. You have to find the one which matches the linguistic evidence.

AH: “Y-hg N has been found at elavated levels, in multiple ancient populations, previously established to be Uralic speaking. The same is true for modern populations, whatever that's worth.”

N is found in many ancient and modern population, many of them being non-Uralic. You cannot just decide that N or any of its subhaplogroups is connected to the Uralic language. You have to find the one which matches the linguistic evidence.

ambron said...

Dmytro

“Irregular” Sound Substitutions in the Substrate Toponymy. New Constribution to the Chronology and Processing of the Linguistic Slavicization of the Historical Slavia Orientalis

The paper deals with the problem of unexpected phonemic sequences encountered sometimes in East Slavic toponyms suspected of being borrowed from the extinct autochthonous Pre-Slavic languages of the Pripet and Upper Dnieper basins (and their vicinities), supposedly related to Balto-Slavic. These instances can be considered as exceptions to several Late Common Slavic sound laws (hushing spirants continuing IE palatovelars, as in Lithuanian, instead of whistling ones; the second palatalization of velars instead of the expected first one; velars not changed to affricates and / or spirants before the reflexes of the vocalisms *e, *ě or *ь), suggesting that the respective names were borrowed into East Slavic from local linguistic substrates after the sound laws in question had been completed. The most controversial issue are the apparently non-monophthongized diphthongs of various types (on, o etc. occurring before consonants), which do posit certain phonotactic problems.

The results obtained in the paper support the view, expressed by some archaeologists, that the linguistic Slavicization of these areas (including Volhynia), earlier considered as the geographical point of departure of Slavic even by some linguists, was a long and continuous process which was not completed before the end of the 1st millennium AD.

https://www.ejournals.eu/PKKS/2019/Vol-XV/art/15244/

Tigran said...

Looks like there is a possibility that Bacho Kiro guy was not P/pre R.

Also Gaska is embarrassing himself on Eupedia with his R1b from Western Europe nonsense.

ANI EXCAVATOR said...

David, repeating the request, could you run the PCA with Bolshy Oleni Ostrov? Could be run without projection, just like you have done for Kra001.

Genos Historia said...

@Simon_W,

I'm working on a video on current Celtic DNA samples. Then one on current Germanic DNA samples.

I'd like to see how it compares to what you see in the samples.

I haven't looked closely at Alemani samples. What have you seen. Are the actual Alemanni all the same or do they have various amounts of Germanic, Celtic ancestry?

Norfern-Ostrobothnian said...

@Jaakko
I'll have a readthrough of that, but many of those mentioned groups with VL29 are in subclades shared with Uralic speakers, meaning that they spread on multiple occasions and were intertwined with Uralic speakers. So there's a connection kind of like a web.

Anonymous said...

@Jaakko

"Quick and short is not possible, because it is a long story where location, time and the taxonomic structure of the language family intertwine."

It is not difficult, it is just a complete absence of any scientific arguments. You do not wonder my messages ignore because you have nothing to say except verbalia.

"...There are many possible interpretations, so you cannot just decide that it is connected to the spread of Proto-Uralic." That is unscientific.

Your delusions of haplogroups instantly refuted by simply looking at it when he emerged a subclade and who had spread the ancestral subclade. Already tired of your senile insanity.


@all

https://i.ibb.co/7g5MtsF/Reich-New-IBD.png

Aleph said...

I'll try posting a shorter version of my reply from a few days ago to see if it goes through.

@Genos Historia

So here are some numbers which compare Iron gates HG (pre-ANF migration/borderline pre-ANF migration) to Motala HG and Loschbour HG (from after the arrival of farmers and hence a bigger Anatolian input into Europe, but before the big wave got to their locations).

Iron Gates without AHG:
Target,Distance,ITA_Grotta_Continenza_Meso,RUS_Sidelkino_HG
ROU_Iron_Gates_HG,0.05940563,61.0,39.0
SRB_Iron_Gates_HG,0.05780647,67.0,33.0
Average,0.05860605,64.0,36.0


vs Iron Gates with AHG:
Target,Distance,ITA_Grotta_Continenza_Meso,RUS_Sidelkino_HG,TUR_Pinarbasi_HG
ROU_Iron_Gates_HG,0.04632143,58.8,31.6,9.6
SRB_Iron_Gates_HG,0.04738076,65.2,26.2,8.6
Average,0.04685109,62.0,28.9,9.1


Loschbour without AHG:
Target,Distance,ITA_Grotta_Continenza_Meso,RUS_Sid elkino_HG
LUX_Loschbour,0.03011222,93.4,6.6
Average,0.03011222,93.4,6.6

vs Loschbour with AHG:
Target,Distance,ITA_Grotta_Continenza_Meso,RUS_Sid elkino_HG,TUR_Pinarbasi_HG
LUX_Loschbour,0.02874591,92.8,4.8,2.4
Average,0.02874591,92.8,4.8,2.4


Motala without AHG:
Target,Distance,ITA_Grotta_Continenza_Meso,RUS_Sidelkino_HG
SWE_Motala_HG,0.03817561,44.6,55.4
Average,0.03817561,44.6,55.4


vs Motala with AHG:
Target,Distance,ITA_Grotta_Continenza_Meso,RUS_Sidelkino_HG,TUR_Pinarbasi_HG
SWE_Motala_HG,0.03568356,43.8,52.8,3.4
Average,0.03568356,43.8,52.8,3.4


The AHG-like ancestry in Motala and Loschbour could be explained by indirect spreads and raids in a time when there was a lot of Anatolian ancestry to go around even before the farming societies got to those specific places.

As you can see the discrepancy between Iron gates with AHG vs Iron gates without AHG is much greater than the discrepancy between Loschbour with AHG vs Loschbour without AHG and between Motala with AHG vs Motala without AHG. This discrepancy is true for both the amount of AHG ancestry (Iron Gates seem to have more of it) and the difference in the least squares of the models (greater accuracy difference in Iron Gates wrt the Anatolian variable). This is in spite of the fact that Loschbour and Motala were living in a time when there was more autosomal ancestry from Anatolia in Europe (post ANF migration) in comparison to the time of the Iron Gates HG (pre-ANF migration). Now, maybe this is just an illusion created by global25 since it isn't perfect and it only checks for approximate vs distant as opposed to maybe other methods. And so, even if K is native to Europe, it must have originated and incubated entirely within the confines of southeastern Europe since it isn't found in any other European HG until the ANF arrive and spread this mtDNA to other parts of Europe.

Anthony Hanken said...

@Jaakko

"What do you mean? Still there are gene flows from different directions and/or at different times. You cannot just decide that one of them is connected to a certain language. You have to find the one which matches the linguistic evidence."

That's what I mean. What happens when a specific genetic marker, matches the linguistic evidence, at multiple "momentary crossroads". For example, if Y-hg N keeps showing up in aDNA, that corresponds to where, in time and space, the linguistic evidence suggests a Uralic dialect was spoken.

"N is found in many ancient and modern population, many of them being non-Uralic. You cannot just decide that N or any of its subhaplogroups is connected to the Uralic language. You have to find the one which matches the linguistic evidence."

That's not what I'm saying.

If Proto-Uralic was spread by people, one or more, specific Y-haplogroups, would have also been spread by these people. Right now, it looks like N-L1026 is the best candidate, although more may have been involved.

This correlation has been pointed out in multiple peer reviewed studies. This is not just my opinion.

George said...

Off Topic

Human ancient DNA analyses reveal the high burden of tuberculosis in Europeans over the last 2,000 years

https://www.cell.com/ajhg/pdfExtended/S0002-9297(21)00051-3

The Summary from the paper:
"Tuberculosis (TB), usually caused by Mycobacterium tuberculosis bacteria, is the first cause of death from an infectious disease at the worldwide scale, yet the mode and tempo of TB pressure on humans remain unknown. The recent discovery that homozygotes for the P1104A polymorphism of TYK2 are at higher risk to develop clinical forms of TB provided the first evidence of a common, monogenic predisposition to TB, offering a unique opportunity to inform on human co-evolution with a deadly pathogen. Here, we investigate the history of human exposure to TB by determining the evolutionary trajectory of the TYK2 P1104A variant in Europe, where TB is considered to be the deadliest documented infectious disease. Leveraging a large dataset of 1,013 ancient human genomes and using an approximate Bayesian computation approach, we find that the P1104A variant originated in the common ancestors of West Eurasians ∼30,000 years ago. Furthermore, we show that, following large-scale population movements of Anatolian Neolithic farmers and Eurasian steppe herders into Europe, P1104A has markedly fluctuated in frequency over the last 10,000 years of European history, with a dramatic decrease in frequency after the Bronze Age. Our analyses indicate that such a frequency drop is attributable to strong negative selection starting ∼2,000 years ago, with a relative fitness reduction on homozygotes of 20%, among the highest in the human genome. Together, our results provide genetic evidence that TB has imposed a heavy burden on European health over the last two millennia."

Erik Andersson said...

My comment seems to have not came through. I will try again...

@Jaakko Häkkinen
"But these correlations disappear whenever we move in time or place."

How does the correlation between genes and languages "disappear whenever we move in time or place"?

"Really? Please show me the arguments which disprove the Scandinavian origin of the first stone cist grave people."

The Estonian stone-cist grave people were not the same people as the first stone-cist grave people.

"Present some concrete mutation which can be connected to Proto-Uralic, and present the evidence supporting your claim."

N-VL29 (or, more properly, N-CTS10760), which includes N-L550, seems to match Proto-Uralic in age. It is common in some Uralic speakers, rarer in others. Frequencies are likely subject to drift in small populations, as well as founder effects like we see in Indo-Europeans. It is about the same age as N-Z1936 (also under N-L1026), N-Y9022 (under N-L708 but not L1026), and also N-Y3195 (under N-P43). It is likely that all of these were present in Proto-Uralians, or otherwise acquired and spread by them. The alternative is that they spread independently from Uralic languages at around the same time and in the same areas, which doesn't strike me as very parsimonious.
But I suspect that you know this already.

Davidski said...

I don't have a clue why some of these posts aren't coming through. They're not in the spam folder.

Genos Historia said...

@Aleph,

Thanks for running those tests.

Ok, those Anatolian scores are significant for Irongate. This could explain the few mtDNA H samples in Irongate samples.

Davidski speaks of Mesolithic Greek samples he heard about who are identical to Anatolian farmers. But they might be miss dated. The published K1 (related to K1c) in Mesolithic Greece belongs to the same extinct K1 clade as do some IronGate HGs. Based on that, we can expect then to be related to IronGate HGs but let's see.

mtDNA K1 technically also appears in a hunter gatherer in Denmark, Lithuania, and Russia. Although of course it is rare outside SE Europe.

Genos Historia said...

New talk by David Reich.

He mention that there was a "rise in farmer ancestry" in England in 1200 BC.

In other words, there was a migration from France into England. I don't know why he has to be ambiguous.

These were the Celts obviously. This marks the arrival of the Celts in Britain to 1200 BC.

Genos Historia said...

My video on Celts is going to be good.

David Reich recent talk
https://www.youtube.com/watch?v=QoGmPJJS3X8

Jaakko Häkkinen said...

Anthony Hanken:
“That's what I mean. What happens when a specific genetic marker, matches the linguistic evidence, at multiple "momentary crossroads". For example, if Y-hg N keeps showing up in aDNA, that corresponds to where, in time and space, the linguistic evidence suggests a Uralic dialect was spoken.”

Then in those points it can be connected to that certain Uralic language. But you still cannot claim that in other points in time and place that same lineage is connected to the same language. It is easy to see how wrong one can go with this “guessing method”: VL29 is found in the Baltic Sea region within Uralic and Balto-Slavic speakers, but in Asia it is found within the Karanogays, for example.

AH: “If Proto-Uralic was spread by people, one or more, specific Y-haplogroups, would have also been spread by these people. Right now, it looks like N-L1026 is the best candidate, although more may have been involved.”

I have showed you many times already, that L1026* is not a good match for Proto-Uralic speakers, and neither is any of its subhaplogroups. So how can you claim that L1026 is the best candidate? Have you studied any other haplogroups?

AH: “This correlation has been pointed out in multiple peer reviewed studies. This is not just my opinion.”

Geneticists know that correlation is not the evidence, it is just a starting point for further research. Laymen should also understand this. And of those correlation values you see that the correlation is not nearly perfect.

I have told you these same things many times, but you don’t seem to understand. Why is that?

Jaakko Häkkinen said...

Erik Andersson:
“How does the correlation between genes and languages "disappear whenever we move in time or place"?”

Because in every spatio-temporal point the genetic composition of a population changes.
Because all populations have multiple genetic roots, and you cannot see from DNA, which root was connected to the language.

All you have is a momentary correlation, if you find a true match between linguistic and genetic results. And if you ignore the linguistic results, you don’t even have that, but just a mere guess.

EA: “The Estonian stone-cist grave people were not the same people as the first stone-cist grave people.”

Evidence for this claim?
But similarly, neither were the later tarand grave people the same people than the earlier tarand grave people: as I said, in every spatio-temporal point the genetic composition of a population changes. You cannot see the continuity or discontinuity of language from the continuity or discontinuity of population or culture.

EA: “N-VL29 (or, more properly, N-CTS10760), which includes N-L550, seems to match Proto-Uralic in age. It is common in some Uralic speakers, rarer in others. Frequencies are likely subject to drift in small populations, as well as founder effects like we see in Indo-Europeans. It is about the same age as N-Z1936 (also under N-L1026), N-Y9022 (under N-L708 but not L1026), and also N-Y3195 (under N-P43). It is likely that all of these were present in Proto-Uralians, or otherwise acquired and spread by them. The alternative is that they spread independently from Uralic languages at around the same time and in the same areas, which doesn't strike me as very parsimonious.
But I suspect that you know this already.”

Yes, these are possibilities – yet far from proven.
1. The datings are still uncertain: Post et al. 2019 get much older date for Z1936, and that affects all the other branches, too. Then there is no temporal match with the spread of Uralic.
2. Even if we accepted the current datings, do the spread of these branches match the spread of the Uralic branches? Can you present a concrete match for the chronology of the Finnic branch, for example?

Steps:
1. Proto-Uralic in the Volga-Kama area around 2000 BC
2. West Uralic in the Upper Volga area around 1500 BC
3. Pre/Early Proto-Finnic via the Daugava river around 1000 BC
4. Middle Proto-Finnic in (Southern?) Estonia around 300 BC
5. Late Proto-Finnic in Northern Estonia around 200 AD
6. North Finnic arrives in Finland only around 500 AD
7. Old West Finnish, Old Karelian and Old Vepsian around 800 AD

Genos Historia said...

@Simon_W, others

What do you think about the idea that British/Irish Celtic languages are basal to European Celtic languages.

Reich's new youtube talk seems to confirm Celtic languages arrived in Britain pretty early. In 1200 BC.

And, the Celts in continental Europe seem to have all shared a more recent common origin with each other than the Celts in Britain & Ireland.

If you think about it. The Celts in Spain, actually called themselves Celts just like the ones in France did. Which I think confirms a recent Hallstatt origin in France. The Celts in Northern Italy, Central-East Europe were all derived from recent La Tene migrations.

Genos Historia said...

The ethnoym Celt, was used for all Celts except for the ones in Britain & Ireland, correct? Well, also Belgica.

Maybe, Belgica also derived from an earlier Urnfield/Bronze age Celtic expansion. SO weren't proper (Hallstatt) Celts.

Genos Historia said...

Early "Celts"=Urnfield. 1200-1000 BC.

Celt ethnyomn=Hallstatt, La Tene. 800-500 BC.

Just a guess, tell me if I'm wrong.

It is interesting, that according to Julius Ceasar, all Celts in France were the same ethnic group/language in 50 BC. It suggests a recent common origin between all of them. If they derived from an expansion in 1200 BC, they wouldn't still speak the same language.

In other words, the "Celtic" expansion from France then into Britain 1200 BC, happened earlier, than the expansion which created the Celts in France who Julius Ceasar knew.

Anonymous said...

@Aleph

"The AHG-like ancestry in Motala and Loschbour could be explained by indirect spreads and raids in a time when there was a lot of Anatolian ancestry to go around even before the farming societies got to those specific places."

No, it's just an indication of kinship, but not with Anatolia. TUR_Pinarbasi_HG arose as a consequence of migration (Post-Aurignacians) from Europe to Anatolia. ANF has nothing to do with Loschbour and Motala.

Palaeolithic Epipaleolithic Turkey Pinarbași [ZBC] 13642–13073 cal BCE M C1a2 K2b

Anonymous said...

@

"Reich's new youtube talk seems to confirm Celtic languages arrived in Britain pretty early. In 1200 BC."

There is no evidence for this.

"Which I think confirms a recent Hallstatt origin in France."

hahahha. What is this anti-science nonsense?


" Early "Celts"=Urnfield. 1200-1000 BC.

Just a guess, tell me if I'm wrong."

You are mistaken. The Urnfield of that time was inhabited by Italics, Lusitanians, some part of the Germanics, if not all.


Rob said...

@ Genos

''In other words, the "Celtic" expansion from France then into Britain 1200 BC, happened earlier, than the expansion which created the Celts in France who Julius Ceasar knew.''

Yes, although there were earlier versions of Celtic in the mainland before the La Tene expansion.

''Central-East Europe were all derived from recent La Tene migrations.''

Seems correct, although there were Eastern Halstatt-related groups as far as Ukraine. Perhaps stubs.

Norfern-Ostrobothnian said...

@Davidski
Could you take a look at these? They're all labeled according to Global 25 standards.
https://advances.sciencemag.org/content/4/12/eaau5064
https://www.mediafire.com/file/4c3ik9c9pylfjt5/Wright_2018.zip/file
https://www.nature.com/articles/s42003-019-0399-1
https://www.mediafire.com/file/5df96j2ul48j8rl/Kashuba_2019.zip/file

Erik Andersson said...

@Jaakko Häkkinen
"Because in every spatio-temporal point the genetic composition of a population changes.
Because all populations have multiple genetic roots, and you cannot see from DNA, which root was connected to the language."

Yes, the genetic composition changes through admixture of the original speakers with people in the areas they migrate to.
That is why, quoting Tambets et al. 2018, "contemporary Uralic speakers are genetically very similar to their local geographical neighbours", but nevertheless "most Uralic speakers share a distinct ancestry component of likely Siberian origin, which suggests that the spread of Uralic languages involved at least some demic component."

"Evidence for this claim?"

I've already told you, the aDNA. Estonian stone-cist grave people have no ancestry from Scandinavia.
On the other hand, modern Estonians do have ancestry from the first Tarand grave people.
Of course, archaeologists have long noted that Estonian stone-cist graves had native elements. Earlier archaeologists thought they were Finnic speakers co-opting Scandinavian grave forms. Even Lang calls them "mixed natives". Thanks to aDNA, we can remove the "mixed" part.

"Post et al. 2019 get much older date for Z1936, and that affects all the other branches, too. Then there is no temporal match with the spread of Uralic."

From Supplementary Table S2:
N3a2/N3a4 calibration point Age: 7113
N3a4-Z1936 Age: 4911 (95% 3740-6335)
N3a4-B539/Y13850 Age: 4286 (95% 3093-5609)
N3a4-B535/Z1934 Age: 3994 (95% 2803-5325)

Doesn't rule out an association with Proto-Uralic. Even if not Z1936 itself, it's subclades remain possible. Granted, they have more restricted distributions.

"Even if we accepted the current datings, do the spread of these branches match the spread of the Uralic branches? Can you present a concrete match for the chronology of the Finnic branch, for example?"

How about you explain how the expansion of both paternal lineages and autosomal ancestry ubiquitous in Uralic peoples happened independently from the contemporaneous expansion of Proto-Uralic from Volga-Kama?

Ric Hern said...

@Genos historia

I heard he said a rise only in Southern Britain...not in Northern Britain.

Anthony Hanken said...

"Then in those points it can be connected to that certain Uralic language. But you still cannot claim that in other points in time and place that same lineage is connected to the same language. It is easy to see how wrong one can go with this “guessing method”: VL29 is found in the Baltic Sea region within Uralic and Balto-Slavic speakers, but in Asia it is found within the Karanogays, for example."

That's obvious. No one is going to argue that all modern Y-hg N carriers, must be Uralic speaking.

"I have showed you many times already, that L1026* is not a good match for Proto-Uralic speakers, and neither is any of its subhaplogroups. So how can you claim that L1026 is the best candidate? Have you studied any other haplogroups?"

You haven't. N-L1026 and its subclades are a good match. Look at them as a whole instead of trying to pick out a single subclade. They were all probably present in the same population, that then split in to different branches. Each branch was then subject to unique drift resulting in modern distribution.

"Geneticists know that correlation is not the evidence, it is just a starting point for further research. Laymen should also understand this. And of those correlation values you see that the correlation is not nearly perfect."

Further research, meaning more aDNA. Those studies also point out a potential link between N-L1026 and Seima-Turbino. The most likely archeological candidate for the spread of Proto-Uralic.

We already know from rumors, that N-L1026 will be found in S-T. Either spread alongside it, or as the main lineage involved in its spread.

So, if aDNA shows us that multiple Uralic languages, including the Proto language, at different crossroads, all share elevated levels of N-L1026. Can we agree that Uralic was spread by men carrying N-L1026?

"I have told you these same things many times, but you don’t seem to understand. Why is that?"

Parroting the same thing, over and over, isn't very convincing.

Aleph said...

@Genos Historia

The K1 in Russia (after 4500 BC), in Denmark (after 4000 BC) is from long after farmers migration to Europe. I am not sure which early Lithuanian K you are alluding to but there is an H in Lithuania from before 4000 BC so the K would have arrived with that package. Even though large populations of farmers may not have lived close to these HGs at those points of time, the fact remains that since these are thousands of years after ANF migrated to Europe, the long time combined with the high, albeit localized ANF ancestry in neolithic locals of Europe means there was more than enough DNA to be passed around over a long distance indirectly among HGs to result in both a little bit of ANF autosomal ancestry and some ANF mtDNA incidences that we see in the HG. Basically, while I could still buy a southeastern European origination and incubation, the regions outside of southeastern Europe only show mtDNA K after the ANF arrived. The only alternative which allows for a pan-European/non-specific-southeastern European origin of K is HGs in Europe knew about this lineage and were hiding it from future researchers until the ANF arrived and they let them be put in spots where they could be found.

@Archi
Possible. But the test was mostly about the Anatolian affinities in southeastern Europe. So even if the low AHG in those northern and western regions was because of distant post-Aurignacian affinity linking the two, there is still elevated AHG affinity in Balkan HGs which really seems to be direct. Your possibility only further strengthens the Balkan HG-AHG interaction hypothesis since only Balkan HGs have actual AHG ancestry while the rest may only have small distant autosomal affinities instead, and this sets the late Balkan HGs apart.

Genos Historia said...

@Aleph,

Not to get too obsessed with mtDNA. But both the hunter gatherer K1s in Northern Europe are of probably of indigenous European origin.

The K1 in Anatolian-derived Neolithic farmers is........K1a, K1b1
The K1 in Iron Gates is.......K1+16362 (K1e, K1d, K1f), pre-K1c.

The K1 in Denmark is K1e. So it is of indigenous European origin.

The K1 in Lithuania hunter gatherer is K1b2.

K1b2 is only also found in "Steppe" people. They are the source of the modern K1b2a subclade. I think it is of Eastern European hunter gatherer origin.

But it may be of CHG origin ultimately. As T2, which totally Near Eastern, is also present in Neolithic Russia. And of course we know Y DNA J is in Mesolithic Russia.

Anonymous said...

@Aleph

"The K1 in Russia (after 4500 BC), in Denmark (after 4000 BC) is from long after farmers migration to Europe."

mt-DNA K from Europe is the alias name for U8b2.

Paleolithic Bulgaria Bacho Kiro Cave [BK-1653] 34690-34380 cal BC (30570±120 BP, AIX-12024) U8

mt-DNA K1 is Iron Gates.

Mesolithic Iron Gates Serbia Vlasac [I4657 / VLSC_1G/3] 9755-9275 calBCE (9942±66 BP, PSUAMS-2294, corrected for Freshwater Reservoir Effect) F K1c
Mesolithic Iron Gates Serbia Padina [I5244 / PADN_18b] 9115-8555 calBCE (9424±55 BP, PSUAMS-2376, corrected for Freshwater Reservoir Effect) F K1f
Mesolithic Iron Gates Romania Ostrovul Corbului [I4582 / OSTCOR_32] 7021-6473 calBCE [7812±69 BP, corrected for Freshwater Reservoir Effect ORAU _15N value for 8302±32 BP which is a weighted average of (8305±50 BP, OxA-31598), (8300±40 BP, PSUAMS-1749), (8335±45 BP, PSUAMS-1904)] F K1


"Balkan HGs have actual AHG ancestry"

No, this AHG is of Post-Aurignac origin from Balkans.

K2b C2217T * G5231A * A14037G formed 15400 ybp, TMRCA 10000 ybp

Distance to: TUR_Pinarbasi_HG:ZBC_IPB001
0.11125188 BEL_GoyetQ116-1_UP:Q116-1
0.12237136 Levant_Natufian:I0861
0.12839560 Levant_Natufian:I1072
0.15646811 BEL_GoyetQ2_LUP:GoyetQ2
0.17616223 ROU_Iron_Gates_N:I5408
0.17668718 ROU_Iron_Gates_HG:I5411
0.17906303 ROU_Iron_Gates_HG:I4081
0.18294119 ROU_Iron_Gates_HG:I4607
0.18388689 ROU_Iron_Gates_HG:I4582

Target: TUR_Pinarbasi_HG:ZBC_IPB001
Distance: 10.0175% / 0.10017469
62.4 BEL_GoyetQ116-1_UP
37.6 Levant_Natufian

Target: ROU_Iron_Gates_N:I5408
Distance: 5.0194% / 0.05019402
100.0 BEL_GoyetQ2_LUP
0.0 BEL_GoyetQ116-1_UP
0.0 Levant_Natufian
0.0 TUR_Pinarbasi_HG

Target: ROU_Iron_Gates_HG:I5411
Distance: 5.0595% / 0.05059476
100.0 BEL_GoyetQ2_LUP

Target: ROU_Iron_Gates_HG:I4607
Distance: 5.2573% / 0.05257309
100.0 BEL_GoyetQ2_LUP

Target: ROU_Iron_Gates_HG:I4582
Distance: 5.6587% / 0.05658692
100.0 BEL_GoyetQ2_LUP

Target: ROU_Iron_Gates_HG:I4081
Distance: 5.6227% / 0.05622668
100.0 BEL_GoyetQ2_LUP

Aleph said...

@Archi

You are using unscaled coordinates, that too with inputs and targets that are temporally separated by well over 10,000 years in the case of Q116 though you mostly use Q2, which is fine. The unscaled coordinates however prevent me from taking that seriously. Here is the same stuff but with scaled coordinates:

Distance to: TUR_Pinarbasi_HG:ZBC_IPB001
0.22957534 Levant_Natufian:I0861
0.23599305 Levant_Natufian_contam:I1072
0.29164528 BEL_GoyetQ116-1:Q116-1
0.31618474 BEL_GoyetQ2:GoyetQ2
0.37246617 ROU_Iron_Gates_HG:I4081
0.37797128 ROU_Iron_Gates_N:I5408
0.38210014 ROU_Iron_Gates_HG:I5411
0.38476208 ROU_Iron_Gates_HG:I4607
0.38818541 ROU_Iron_Gates_HG:I4582

Target,Distance,BEL_GoyetQ116-1,Levant_Natufian
TUR_Pinarbasi_HG:ZBC_IPB001,0.20650600,32.8,67.2
Average,0.20650600,32.8,67.2

Target,Distance,BEL_GoyetQ2
ROU_Iron_Gates_N:I5408,0.17510035,100.0
Average,0.17510035,100.0
(Pinarbasi doesn't show up, but note how the fit is atrocious, and how it is much worse than the Sidelkino + Pinarbasi + Continenza models which impart some Pinarbasi to it. This model on the other hand is just plain bad)

Target,Distance,BEL_GoyetQ2
ROU_Iron_Gates_HG:I5411,0.17552768,100.0
Average,0.17552768,100.0

Target,Distance,BEL_GoyetQ2
ROU_Iron_Gates_HG:I4607,0.17201340,100.0
Average,0.17201340,100.0

Target,Distance,BEL_GoyetQ2
ROU_Iron_Gates_HG:I4582,0.18084075,100.0
Average,0.18084075,100.0

Target,Distance,BEL_GoyetQ2
ROU_Iron_Gates_HG:I4081,0.17109956,100.0
Average,0.17109956,100.0


All bad models. I could post the ones with Sildekino + Continenza + Pinarbasi, but I fear that if I do so then this comment will not get through like the one from a few days ago. One of the reasons why my response to Genos Historia will be separate. So I will do a brief one instead:

Target,Distance,ITA_Grotta_Continenza_Meso,RUS_Sidelkino_HG,TUR_Pinarbasi_HG
ROU_Iron_Gates_HG,0.04632143,58.8,31.6,9.6
SRB_Iron_Gates_HG,0.04738076,65.2,26.2,8.6
Average,0.04685109,62.0,28.9,9.1

Goyet doesn't show up specifically in Iron gates independent of these inputs. There might actually be something like post-Aurignacian of a type we do not have sampled within Pinarbasi, but it would be a small component or else everyone would already know about it and you wouldn't need to use unscaled coordinates for it.

Aleph said...

@Genos Historia

I notice K2a in EEF as well. U8b1 too, and there is U8b1 from the middle east as well. I haven't checked every single EEF, just glanced at it. U8b1 (outside of K) sub-branches which are limited to European samples have recent TMRCAs on yfull while the ones with mixed European and non-European samples have older TMRCAs. There are however southern European strings like an Italian one with an 11,000+ years TMRCA. It also looks deeply rooted in the middle east as well. K1d apparently has a tmrca of 2800 years, K1e of 7600 years and idk for K1f. But there is a lot of missing history between the formed and tmrca for these branches. If they were generally European then why is there precious little K before the Holocene outside of southeastern Europe? In fact, it seems none at all. There are no such problems with U5 or U8a, or even U4 or post-LGM U2 for that matter. Why K? Not a single pre-holocene K outside of southeastern Europe. So even if it is European, as stated earlier, it must have been restricted to southeastern Europe and it must have spread to other parts of Europe outside of southeastern Europe only after the holocene began. This alternative scenario would also require the Anatolian ancestry in Balkan HGs to be co-incidental or mostly an indication of mtDNA H instead which was only in one Iron Gates sample iirc. I was thinking about another alternative as well- when the proto-WHG that migrated to the near east split from WHG (probably somewhere in eastern Europe), the K could have split both was and spread such that the WHG K ended up somehow being restricted to the Balkans, but that would be one weird distribution for it to be restricted to the Balkans until the Holocene like that.

Aleph said...

@Rob

I am not exactly sure if Archi is talking about the same thing, but the C1a2 in Pinarbasi may have come from some western European HGs moving eastward. Crazy I know, but as of now there is no way to confirm it or rule it out since the precursor genetic pool to Pinarbasi isn't clear. The idea is that something descended from the Aurignacians may have brought C1a2 to Anatolia. BTW if you know more about Pinarbasi then can you tell me about the specific C1a2 subclade that it has. What is it called?

Rob said...

In the maramara/ Bosporus , surveys have found assemblages similar to Bacho Kiro type, but different to anything in further east in west Asia . But none are confirmed with intensive study or C14. If they are confirmed in future study; then it would link NW Anatolia with Europe (no big surprise). The rest of Anatolia hasnt shown much, aside from the far southeast which is part of the Leavtine sequence, and therefore different.

There seems to be nothing during the Gravettian.
Then the Antalya cave sites begin ~ 25,000 BP; which would be post - gravettian / epigravettian; making Pinarbasi related with them, but now exploring the Plains of south central Anatolia & with social contacts with natufians. Thus, the C1/ U8b in Pinarbasi could be from SEE. Similarly, the K3 in Satsurblis could be from E.E.

Aleph said...

@Rob

The post-Aurignacian thing, if something like it did happen, it wouldn't be clear in the distant assemblage since it would have been a more recent thing, post LGM and would only be a small input which would have assimilated with the locals. IDK if Archi has the same view but those are my 2 cents about the post Aurignacian possibility.

As for an origin from SEE, wouldn't that need some kind of population 'nest' for HGs? On anthrogenica some guys posted a map showing no such refugias in the Balkans, which was the thing that made me shift the WHG origin from Balkans to eastern Europe. As for K, I don't think any more can be said aside from what has been said so far- if it is European then it is confined to SEE (x the rest of Europe) until the Holocene, or maybe it is so old that it goes back to the WHG-proto WHG in Dzudzuana split which would explain the EE to CHG possibility. Or it originated in the near east after U8 spread there 30,000+ years ago from Europe (perhaps like U6 on its way to north Africa, but U8 didn't follow the entire U6 route) and then re-entered Europe maybe after the LG in the Balkan and then really re-entered with ANF.

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

It sounds like the Bronze Age collapse is going to turn out to be a really interesting period for aDNA. Can't wait till we start to get a clearer picture of what was going on.

CrM said...

C1a2 in Pinarbasi might be Dzudzuana.

Jaakko Häkkinen said...

Erik Andersson:
“Yes, the genetic composition changes through admixture of the original speakers with people in the areas they migrate to.
That is why, quoting Tambets et al. 2018, "contemporary Uralic speakers are genetically very similar to their local geographical neighbours", but nevertheless "most Uralic speakers share a distinct ancestry component of likely Siberian origin, which suggests that the spread of Uralic languages involved at least some demic component."”

Yes, there are genetic features shared by most but not all Uralic populations.
Still, you cannot see from the DNA, whether this shared autosomal component (or any shared uniparental lineages) truly has spread with the language or if it predates or postdates it.

You see, it is much easier to spread DNA than language to a new population: one childbirth is enough to bring new DNA into population, but it is much rarer that a totally new language is adopted. Therefore there are always more gene flows than language spreads – and you cannot just guess that certain gene flow is connected to certain language.

EA: “I've already told you, the aDNA. Estonian stone-cist grave people have no ancestry from Scandinavia.”

You still didn’t tell me the evidence for your claim. Which study shows that there is no Scandinavian ancestry? Saag et al. 2019 does not show that.

EA: “On the other hand, modern Estonians do have ancestry from the first Tarand grave people.”

…and from other older populations, too.

EA: “Doesn't rule out an association with Proto-Uralic. Even if not Z1936 itself, it's subclades remain possible. Granted, they have more restricted distributions.”

That is exactly what I’ve been telling: there are no perfect matches.

EA: “How about you explain how the expansion of both paternal lineages and autosomal ancestry ubiquitous in Uralic peoples happened independently from the contemporaneous expansion of Proto-Uralic from Volga-Kama?”

It doesn’t require explaining: that is exactly the judgment, if we cannot find any perfect match for the spread of the Uralic languages.
Lack of matches = no straightforward connection between the Uralic languages and these genetic features. Only vague partial correlations. Read my points above in this comment.

Jaakko Häkkinen said...


Anthony Hanken:
“You haven't. N-L1026 and its subclades are a good match. Look at them as a whole instead of trying to pick out a single subclade. They were all probably present in the same population, that then split in to different branches. Each branch was then subject to unique drift resulting in modern distribution.”

Really? You are seriously claiming that a lineage which is found within the Spanish, Norwegian, Swedish, Polish, Latvian, Lithuanian, Belarussian, Russian, Ukrainian, Crimean Tatar, Volga Tatar, Chuvash, Bashkir, Siberian Tatar, Dolgan, Yakut, Evenk, Even, Yukaghir, Eskimo, Chukchi, Koryak, Nanai, Altaian, Tuvan, Khakassian, Shorian, Mongol, Buryat, Kazakh, Uzbek, Karanogay and Karakalpak populations is a good match for the Uralic language family?!?!
Maybe you should think again?

AH: “Further research, meaning more aDNA. Those studies also point out a potential link between N-L1026 and Seima-Turbino. The most likely archeological candidate for the spread of Proto-Uralic.”

Show me the evidence.
Besides, Seima-Turbino could only be connected to the earliest stage of the Uralic spread: from Proto-Uralic to West and East Uralic. It is too early to match the later linguistic expansions from Upper Volga to the west. The ST items are of course more widespread, all the way to Sweden, but the language did not spread with the items.

AH: “We already know from rumors, that N-L1026 will be found in S-T. Either spread alongside it, or as the main lineage involved in its spread.
So, if aDNA shows us that multiple Uralic languages, including the Proto language, at different crossroads, all share elevated levels of N-L1026. Can we agree that Uralic was spread by men carrying N-L1026?”

Hopefully we get more information about the spread of L1026. Still, it is a long way to confirm that it is connected to the spread of Uralic languages, see above.

AH: “Parroting the same thing, over and over, isn't very convincing.”

I have presented arguments. You have presented inability to understand them, but no valid counter-arguments.

Davidski said...

@Jaakko

You keep making really dumb comments.

It makes no difference that N-L1026 is today found at low frequencies in many different populations that aren't Uralic speaking.

Rob said...

@ aleph

“ The post-Aurignacian thing, if something like it did happen, it wouldn't be clear in the distant assemblage since it would have been a more recent thing, post LGM and would only be a small input which would have assimilated with the locals. ”

Can you rephrase that on some way ? Because it’s not clear what you’re saying , sorry
But to clarify; there is an entity called Levantine aurigncian, and archaeologists generally agree that it looks like a movement from Europe to the east Mediterranean Coasts ~ 36000 BP, which would place it at the end of the Aurignacian period in Europe
The Bosopran stuff might or might not relate to this.
Then there is nothing (thus far found) in central - western Anatolia until after the ice age
Where Pinarbasi and Antalya caves came from is therefore an open question . But the cave paintings in Antalya have very similar iconology as in epigravettian Europe.


“ As for an origin from SEE, wouldn't that need some kind of population 'nest' for HGs? On anthrogenica some guys posted a map showing no such refugias in the Balkans, which was the thing that made me shift the WHG origin from Balkans to eastern Europe. ”

They’re definitely wrong
It is common knowledge that northern Bulgarian sites (temneta dupka; Kozarnika), Kadir in northeast Bosnia, etc were occupied during the LGM . Nothing like the dozens and dozens of sites in Iberia but theyre certainly there
Then we also have the Po-Adriatic region and the Adriatic land bridge which was a major refuge during the LGM
Then we have the eastern carpathian regions, incl Molodova etc





Rob said...

“ which was the thing that made me shift the WHG origin from Balkans to eastern Europe”

It would be artificial to split north Balkans from Eastern Europe. In fact, from Liguria to Kostenki, this was a diffuse network, with net flow from east to west

Norfern-Ostrobothnian said...

What was going on in Central Asia during the Upper Paleolithic?

Anonymous said...

@Aleph

"You are using unscaled coordinates, that too with inputs and targets that are temporally separated by well over 10,000 years in the case of Q116 though you mostly use Q2, which is fine. The unscaled coordinates however prevent me from taking that seriously."

The use of scaled coordinates is a categorical error of misunderstanding people. There are no scaled coordinates, the coordinates categorically cannot be scaled to calculate autosomal proportions.

I am the only one here who understands the mathematics of this process. Scaling is the crazy operation of multiplying a PCA coordinate by an eigenvalue. But what is an eigenvalue? It's actually a percentage of the explained variance of the coordinate, which means you're using nonsense values like 0.9%*coordinate. Physically it's like multiplying mass by temperature, what does this quantity mean physically? It means complete nonsense.

Imagine that we have 2 coordinates as a result of PCA calculation, one has 99% explained variance, the other respectively 1%, that is X_scaled = 99%*X, Y_scaled = 1%*Y. Naturally, scaling is an error, in Vahaduo it approximately works only because the eigenvalues in G25 do not change much, from ~10 at the first coordinate (like X_scaled = 1%*X) to ~1 at the 25th (like Y25_scaled = 0.1%*Y25).

I can tell you that the verified method qpAdm works the same way, there are also calculated coordinates, but not through the PCA method, but through twice outgroup F4-statistics, where the source coordinate = f4(0 = target, X = source, -Infinity = right1, Infinity = right2). It gets sets of coordinates for all variants of the right populations (coordinate lines), which makes it possible to calculate autosomal proportions with fine 0% distance always. Now, of course, no one there multiplies the obtained coordinates by their eigenvalues! There is no such thing as scaled coordinates!

So your calculations are wrong and should not be taken as correct. All your results are distorted. All that you write and calculate is ridiculous and not serious simply because you do not understand the essence of the methods used.

Erik Andersson said...

@Jaakko Häkkinen
"Yes, there are genetic features shared by most but not all Uralic populations."

Look at the studies you yourself have quoted.
Post et al. 2019 showed N-Z1936 in Hungarians, uniting them with other Uralic speakers (and Fóthi et al. 2020 showed lots of N3a among Hungarian conquerors).
Saag et al. 2019 showed an influx of Siberian ancestry and N3a into Estonia associated with Tarand graves, and argue (logically) that this is associated with the arrival of Uralic languages. They also show that modern Estonians are closer to Nganassans than Estonian BA individuals are.
So which Uralic speakers don't fit the pattern?

"You still didn’t tell me the evidence for your claim. Which study shows that there is no Scandinavian ancestry? Saag et al. 2019 does not show that."

They don't explicitly test this, but they do show that EstBA (Estonian stone-cist grave people) are basically identical to other Baltic BA people. That's obvious even from the ADMIXTURE analysis (Figure 1).
Quote: "A clear shift toward West Eurasian hunter-gatherers is visible between European LN and BA (including Baltic CWC) and EstBA individuals, the latter clustering together with Latvian and Lithuanian BA individuals [11]."
There's also no I1 or R1b in any of those samples.
The 'second-hand calculator' agrees:

Target: EST_BA
Distance: 1.3767% / 0.01376748
100.0 LVA_BA
0.0 Nordic_LNBA_Average

"That is exactly what I’ve been telling: there are no perfect matches."

Maybe not, but they can still work ;)

Anonymous said...

@Aleph

"If they were generally European then why is there precious little K before the Holocene outside of southeastern Europe? In fact, it seems none at all. There are no such problems with U5 or U8a, or even U4 or post-LGM U2 for that matter. Why K? Not a single pre-holocene K outside of southeastern Europe."

Not a single pre-holocene U8a outside of Magdalenian culture.
Not a single pre-holocene U8c outside of Gravettian culture.
Not pre-holocene U2'3'4'7'8'9 outside of Epigravettian culture (except early Yana-> and ->Late Magdalenian/Azilian).
So why should U8b2 be outside of eastern southern Europe?
What do you think that all cultures were mixed?
No, they lived quite isolated from each other rarely mixing, which by the way autosomes show perfectly.


Jaakko Häkkinen said...

Davidski:
"You keep making really dumb comments.
It makes no difference that N-L1026 is today found at low frequencies in many different populations that aren't Uralic speaking."

No, they are very valid comments to us, who are scientific. It is unfortunate that you are a pseudo-scientist.

You cannot just decide that some lineage is connected to the spread of the Uralic language. You have to explain why the "Uralic" lineage is so widespread; and you have to apply the same criteria also to the "Indo-European" lineages - you cannot have double standards.

Jaakko Häkkinen said...

Erik Andersson:
“Look at the studies you yourself have quoted.
Post et al. 2019 showed N-Z1936 in Hungarians, uniting them with other Uralic speakers (and Fóthi et al. 2020 showed lots of N3a among Hungarian conquerors).”

Yes, but you have to be precise:
- Z1936* is not common in the Uralic populations
- Hungarians and other East Uralic populations have different subhaplogroup than the West Uralic populations.
- Datings of Post et al. 2019 is older than earlier datings, which weakens the match with the Uralic language

Do you understand?

EA: “Saag et al. 2019 showed an influx of Siberian ancestry and N3a into Estonia associated with Tarand graves, and argue (logically) that this is associated with the arrival of Uralic languages.”

That is only a guess. I have told here many times, that the spread of Uralic to Baltia happened via the Daugava route at the Bronze Age, but the Siberian ancestry and N3 spread to Northern Estonia straight from the east at the Iron Age.
THIS GENE FLOW DOES NOT MATCH THE LINGUISTIC RESULTS, SO HOW COULD IT BE CONNECTED TO THE LANGUAGE?

Why you cannot understand this?

EA: “They also show that modern Estonians are closer to Nganassans than Estonian BA individuals are.
So which Uralic speakers don't fit the pattern?”

The spread of Uralic to Baltia happened via the Daugava route at the Bronze Age, but the Siberian ancestry and N3 spread to Northern Estonia straight from the east at the Iron Age.

Can you tell me, why you think that so poor match is a good match?

EA: “They don't explicitly test this, but they do show that EstBA (Estonian stone-cist grave people) are basically identical to other Baltic BA people. That's obvious even from the ADMIXTURE analysis (Figure 1).”

They don’t even compare to the Scandinavian populations, which we know to be very similar to the Baltic populations prior to the spread of Farmer ancestry. So you cannot claim that EstBA could not have any Scandinavian ancestry!

Anthony Hanken said...

@Jaakko

"Really? You are seriously claiming that a lineage which is found within the Spanish, Norwegian, Swedish, Polish, Latvian, Lithuanian, Belarussian, Russian, Ukrainian, Crimean Tatar, Volga Tatar, Chuvash, Bashkir, Siberian Tatar, Dolgan, Yakut, Evenk, Even, Yukaghir, Eskimo, Chukchi, Koryak, Nanai, Altaian, Tuvan, Khakassian, Shorian, Mongol, Buryat, Kazakh, Uzbek, Karanogay and Karakalpak populations is a good match for the Uralic language family?!?!
Maybe you should think again?"

As Davidski said, modern distribution matters very little. Lineages that were once Uralic speaking, may have found success in other populations, after a language shift. We are concerned with what lineage, was initially spread alongside Proto-Uralic. Not what modern populations have.

"Show me the evidence."

Use your scientific method. It's in the right space, at the correct time. Just off the top of my head, there was a recent work by Johanna Nichols, as well as Asko Parapola's 2012 paper. Both connecting the spread of Proto-Uralic to Seima-Turbino. Although, they disagree on where PU and S-T originate.

"Besides, Seima-Turbino could only be connected to the earliest stage of the Uralic spread: from Proto-Uralic to West and East Uralic. It is too early to match the later linguistic expansions from Upper Volga to the west."

I am only connecting it to the initial spread of Proto-Uralic. Later languages, must have had subsequent expansions.

"The ST items are of course more widespread, all the way to Sweden, but the language did not spread with the items."

Seima-Turbino items were also spread by trade. We should be looking at the rare burials, containing human remains, within Seima-Turbino necropoli. Not chance finds in Sweden.

"I have presented arguments. You have presented inability to understand them, but no valid counter-arguments."

Repeating "scientific method" over and over, is not an argument.

Anonymous said...

@Jaakko Häkkinen

"No, they are very valid comments to us, who are scientific."

It is unfortunate that you are a pseudo-scientist.

"That is only a guess. I have told here many times, that the spread of Uralic to Baltia happened via the Daugava route at the Bronze Age, but the Siberian ancestry and N3 spread to Northern Estonia straight from the east at the Iron Age. "

Who told you that your words have anything to do with science? Who told you that you are right about anything? You are wrong about everything and everything you write is a fight against science through your fictions. Your words are empty and no one here will perceive them.

"HIS GENE FLOW DOES NOT MATCH THE LINGUISTIC RESULTS, SO HOW COULD IT BE CONNECTED TO THE LANGUAGE?"

It is not the match for your fabrications. You don't have any linguistic evidence, you're just tricking and trolling everyone here.

Parastais said...

@Jaska,
« The spread of Uralic to Baltia happened via the Daugava route at the Bronze Age, but the Siberian ancestry and N3 spread to Northern Estonia straight from the east at the Iron Age.»
Do you think Uralic speakers arriving Daugava Route (if that was the case, there still is an “If”) did not feature Siberian and N3?
I find that highly unlikely.

Erik Andersson said...

@Jaakko Häkkinen
"- Z1936* is not common in the Uralic populations"

This is irrelevant? Why would basal lineages need to be common in Uralic populations? R1a-M417* isn't common in Indo-Europeans either. It doesn't matter.

"- Hungarians and other East Uralic populations have different subhaplogroup than the West Uralic populations."

Yes, so there are separate founder effects in these populations. So?

"- Datings of Post et al. 2019 is older than earlier datings, which weakens the match with the Uralic language"

Not that much older. Are you confusing the split of N3a4 from N3a2 with the TMRCA of N3a4? I repeat, from Post et al. 2019 Table S2:
N3a4-Z1936 Age: 4911 (95% 3740-6335)
A lot of that is still within the expected age for Proto-Uralic.

"THIS GENE FLOW DOES NOT MATCH THE LINGUISTIC RESULTS, SO HOW COULD IT BE CONNECTED TO THE LANGUAGE?"

Fine, then tell me exactly which 'linguistic results' prevent an association of Proto-Finnic with the Tarand grave people?

"They don’t even compare to the Scandinavian populations, which we know to be very similar to the Baltic populations prior to the spread of Farmer ancestry."

What does this even mean? "Prior to the spread of Farmer ancestry"? Which Scandinavian populations do you mean are similar to which Baltic ones? Certainly not the ones that built stone-cist graves.

Aleph said...

Alright thats it, blogspot sucks at handling comments. This will be my last one here, if it passes through that is.

@Rob

In the map there were a few dots in northern Balkans, my older idea was the southern Balkan origin of WHGs.

@Archi

The unscaled vs scaled comparison that david demonstrated in an old anthrogenica thread looked like the basic fst results. It put Europeans closer to Africans and East Asians than to WHGs. So it looks unreliable right off the bat. Furthermore, modeling Yamnaya unscaled gives it like 12.5% Iran_N ancestry. I am not going to stand by it, but you should since you love unscaldes so much, though I recall you seething about it for a while so IDK if you will be consistent with that.

Lastly, your analogies to other European HGs utterly fail. U8a is not deeply rooted outside of Europe (unlike U8b1 and K) and it would have originated with the Solutreans. So you shouldn't expect to find it outside of the Solutrean-Magdalenian range since it came from a population backed up into Iberia by the Gravettian expansion and then kept there by the LGM.
U8c on the other hand barely exists both in the present and the past. Its presence in Gravettian indicates a minor presence which was then probably wiped out by the LGM. The only other U8c(s) that I am aware off are from Yadavs. This fits well with an early European UP -> Levant Aurignacian -> South Asia route migration for a south Asian U8c, and also for south Asian U2. It was probably accompanied by yDNA H which may have come from Europe or joined in the middle east, doesn't matter either way since it fits. So this fits in the larger scenario where U8 and U6 migrate from Europe to the near east and then U6 goes to north Africa while K remains and develops in the near east. As such the U8c you bring up helps my case more than it helps you. That U'2-'9 from Sicily was a more extreme case of the U8a effect and may have been it's own U like a U10 as per what some people have said over here. Not rooted outside of Europe unlike K so the scenario fails.

For K to be European in origin, it spread to Anatolia before it got to other places in Europe outside of SEEurope, all the while when WHG were spreading around their autosomal cluster and mtDNA U5 very clearly throughout Europe. U8a was limited to Magdalenians for obvious reasons, the Sicilian U10 was a very much local thing, European U8c died a long time ago. So yeah, none of that provides a precedence for what happened with K. One thing which could allow for European K is for Balkans to be more linked to Anatolia than to other parts of Europe or a second alternative to my [U8 (early European UP) -> U8b (Levant Aurignacian) -> K (emerging in Levant)] is K splitting in the WHG and pre-Dzudzuana proto-WHG split.

You can respond if you want, but this is my last comment, so you won't be getting a response from me here. I am on anthrogenica with the same name so you can PM me over there.

Tigran said...

Anybody see Reich's latest youtube video? From what I saw and saw discussed

-Apparently there are over 10,500 unpublished samples. I'd love to know where they are all from.

-Their unpublished data shows the origin of Corded Ware, Bell Beaker and Unetice is in Yamnaya. I would guess that would mean R1b-L51 and R1a-M417 popped up in Yamnaya.

-They also confirmed the steppe ancestry in Tocharians is from Afanasievo (I guess that means they tested mummies in the Tarim Basin and they turned out to be R1b-Z2103 Steppe_EMBA like people).

Also if the Bacho Kiro guy turns out to be something other than P/pre-R how does that change our thinking on the origin of K2b/P? Do we still think that population lives as far east as the Yana RHS site or do we think some sort of K2b/P will pop up between the Urals and Eastern Siberia?

Coldmountains said...

sorry for going off-topic
but can anyone here get access to these samples and convert them to a format usable for Global25 or Gedmatch?. These are 60 new samples from Tajikistan (Tajiks, Sarikoli and Wakhi). Many of these groups are not on Global25 yet.

https://bigd.big.ac.cn/bioproject/browse/PRJCA000278

Davidski said...

@Tigran

-They also confirmed the steppe ancestry in Tocharians is from Afanasievo (I guess that means they tested mummies in the Tarim Basin and they turned out to be R1b-Z2103 Steppe_EMBA like people).

Reich is citing the Ning et al. paper, which claimed that the Shirenzigou nomads belonged to R1b-M269 and had Afanasievo ancestry.

But of course they don't belong to R1b-M269, but rather to a distantly related Central Asian lineage, and one of them actually belongs to R1a-Z93.

What this means is that the Shirenzigou nomads have local Central Asian ancestry, as well as Steppe_MLBA ancestry.

So they do not provide the genetic evidence that Tocharians were derived from Afanasievo.

Rob said...

@ Aleph

The very south of Europe was relatively peripheral for hunter-gatherers, colonised fairly late (might seem counterintuitive; but outlined quite nicely in the Sicily HG papers; which did a great job of combining genetics and archaeology).
But overall, it seems that ‘WHG’ involved accretion of ancestries from a few zones

@ Norfern
I wrote a reply but it didn’t make it through..
In Central Asia there have been sites documented in the western foothills of Inner asian corridor (from ~ 40-10,000 BP); hence would fill the gap between Iran and southern Siberia
Seem to show typological affinities with zagros assemblages

ambron said...

What is also interesting at Reich is the excess of sharing IBD between CWC and Polish GAC. This shows the homeland of the CWC north of the Carpathians.

Jaakko Häkkinen said...

Anthony Hanken:
“As Davidski said, modern distribution matters very little. Lineages that were once Uralic speaking, may have found success in other populations, after a language shift.”

That is exactly the reason why you cannot see the language from the DNA. Nice that you understand it now.

So, how can you claim that this lineage is originally Proto-Uralic and only secondarily spread to the Eskimos and Chukchis? It could equally well be originally Northeast Siberian lineage and only secondarily spread to the Uralic speaking populations.

AH: “We are concerned with what lineage, was initially spread alongside Proto-Uralic. Not what modern populations have.”

And how you could achieve that knowledge? When you cannot see it from the DNA itself?
You have to see the linguistic results: if there is a matching lineage, then you have a strong case.

AH: “Use your scientific method. It's in the right space, at the correct time.”

So you are saying that N-L1026 was present in the Volga-Kama area around 2000 BC? And this is shown in which aDNA study?

AH: ”Just off the top of my head, there was a recent work by Johanna Nichols, as well as Asko Parapola's 2012 paper. Both connecting the spread of Proto-Uralic to Seima-Turbino. Although, they disagree on where PU and S-T originate.”

Nichols is not a Uralicist and does not present any arguments – she just tells what earlier views there have been.
As I said, Seima-Turbino can be connected to only the early stage of the spread. Parpola knows this, too. In 2017 he states that after the Seima-Turbino phase, West Uralic could be connected to the Netted Ware Culture, but not yet Saami, as he had proposed earlier.

AH: “I am only connecting it to the initial spread of Proto-Uralic. Later languages, must have had subsequent expansions.”

Good, here we agree.

Anonymous said...

@ Aleph

You wrote a very rambling text on mtDNA, you think there is logic in it, but I don't see any.

Understand, all your illogical reasoning is broken by the fact that not only U8b2=K and U8b1, but also the typically European U5b2a1a haplogroups migrated to Anatolia in the Late Paleolithic(/Early Neolithic?).

What does scaled coordinates mean, I do not understand and no one does. To calculate in them is like calculating the distance if one coordinate is in meters, and the other coordinate is in feet.

Jaakko Häkkinen said...

Parastais:
“Do you think Uralic speakers arriving Daugava Route (if that was the case, there still is an “If”) did not feature Siberian and N3? I find that highly unlikely.”

Surely the Pre-Proto-Finnic speakers had some Siberian ancestry, as even Mordvins do. But there was less of it in the southern side of Volga than in the northern side of Volga (Saami – Mari – Udmurt continuum). But it diminished in every step of the spread, as the Finnic language carriers assimilated the “Palaeo-Baltian” population on their way.

And very probable they also had some N3 lineages, and it is even possible that there were N3 lineages in Baltia prior to the spread of Finnic. Most of the root branches of N-L550 are found in Sweden, so the spread of this lineage to Northern Estonia could be due to Scandinavian gene flow. As you know, autosomal admixture can change in every generation, so you cannot claim that the autosomal Siberian ancestry could testify for the eastern arrival of the Estonian N-L550. It could very well have come from the south or west.

Jaakko Häkkinen said...

Erik Andersson:
“This is irrelevant? Why would basal lineages need to be common in Uralic populations?”

Because you need to be specific, not vague:
when you talk about N-Z1936, you talk about either Z1936* or some of its subhaplogroup. So, if you don’t talk about Z1936*, which subhaplogroup are you talking about?

EA: “Yes, so there are separate founder effects in these populations. So?”

So, how you connect the eastern subhaplogroups to the western subhaplogroups?

EA: “Not that much older. Are you confusing the split of N3a4 from N3a2 with the TMRCA of N3a4? I repeat, from Post et al. 2019 Table S2:
N3a4-Z1936 Age: 4911 (95% 3740-6335)
A lot of that is still within the expected age for Proto-Uralic.”

Late Proto-Uralic dispersal occurred around 2000 BC. So we can say, that it is possible that Z1936 was present in the Proto-Uralic speaking population, but we cannot claim that it was born in the Proto-Uralic population. It was born in Siberia (as we all agree), and secondarily arrived also to the European side, where it was possibly present in the Proto-Uralic speaking population. Still, it is possible that Pre-Proto-Uralic speakers brought it with them.

You see? A lot of possibilities, no certainties. Only simple minds make black-and-white interpretations, when there is not enough evidence.

EA: “Fine, then tell me exactly which 'linguistic results' prevent an association of Proto-Finnic with the Tarand grave people?”

It depends on the time, of course. The later the stage, the more probable it is that the Tarand Grave people spoke Finnic (like, AD they certainly did). Still, you cannot see linguistic continuity from the archaeological or genetic continuity, so the earliest Tarand Graves people may have had different affiliations – especially if they were locals or came straight from the east, unlike the Finnic language.

EA: “What does this even mean? "Prior to the spread of Farmer ancestry"? Which Scandinavian populations do you mean are similar to which Baltic ones? Certainly not the ones that built stone-cist graves.”

Scandinavian Pitted Ware people were very similar to the Late Neolithic Baltians. (Mittnik et al. 2018.)
Only with the Globular Amphora Culture and Corded Ware Cultures more Farmer ancestry arrived to Scandinavia. During the Bronze Age they were not all very different, either: there were different populations in Baltia and Scandinavia, they were not homogeneous.

Matt said...

@ambron, it seems like the new work is exciting in actually being able to look meaningfully at shared segments from fairly low coverage capture adna, going beyond the already impressive work from last year looking at the simple distribution of RoH classes with a sample's genome (which tells us a fair bit about population size, frequency of consanguinity).

That's obviously interesting wrt understanding the relationships in the 3rd Millennium BCE (between various steppe infused populations, and between pre-steppe populations). Also though, one bit of that which might potentially relevant to our previous discussions is quantifying segment sharing between low coverage adna from the Bronze through Iron Age with present day people.

One of patterns identified from segment sharing between only present day people back in early 2010s was of a high level of sharing among East Europe, and then that was linked in hypothesis to expansion of proto-Slavic group. (See Coop 2013). But! A varying argument in a paper by John Novembre and collaborators a few years later (2018-2019) suggested "high PSC" (segment) "sharing in these regions may be due more to consistently low population densities than to historical expansions (such as the Slavic or Hunnic expansions)". E.g. actually reflects long term low population size then a general increase (because larger populations with more people scramble up segments quicker than small populations where ppl are mating with more third and fourth cousins and keeping segments together).

Directly testing modern and ancient links in segments will allow to compare these hypotheses and see if there was actually a sudden enrichment in ancient segment links with each other, and present day people, after a certain point. That will give us more confidence if there was a sudden expansion of a single people.

I also expect the ongoing high-coverage ancient sampling project to supplement this, providing some sense check (in case there is any unknown inference issue that has been missed).

Anonymous said...

@Tigran

"They also confirmed the steppe ancestry in Tocharians is from Afanasievo (I guess that means they tested mummies in the Tarim Basin and they turned out to be R1b-Z2103 Steppe_EMBA like people)."

No confirmed. It is about Shirenzigou.Everything they wrote is easily refuted. This is one of the tribes from Mongolia, probably related to the Xiongnu, they have Afanasyevo as Mongols and Saka. The Xiongnu fought the Iranian Tokharians.

https://i.ibb.co/nkftP8D/Shirenzigou-PCA1.png
https://i.ibb.co/hZnPBN9/Shirenzigou-PCA1-2.png
https://i.ibb.co/8B8CMD5/Shirenzigou-PCA2.png

No Afanasyevo & R1b-Z2103.

Iron Post-Yanbulake Chine Shirenzigou [M15-1] M R1b1a1a2 R-PH200 I1b
Distance to: CHN_Shirenzigou_IA:M15-1
0.03174476 Andronovo_KAZ_Zevakinskiy_LFBA_Fedorovo:I3976
0.03394378 Saka_Tian_Shan:DA49
0.03613323 Saka_Tian_Shan:DA57
0.03692723 Sarmatian_RUS_Caucasus:MJ38
0.03705712 Saka_Kazakh_steppe:DA11
0.03742513 Saka_Tian_Shan:DA47
0.03762353 Andronovo_KAZ_Alpamsa_LBA_Alakul:I6823
0.03766152 Saka_Tian_Shan:DA51
0.03807545 MNG_Xiongnu_Central_Asian:DA38
0.03811102 Sarmatian_RUS_Caspian_steppe:DA144
0.03811233 Saka_Tian_Shan:DA56
0.03862784 MNG_EIA_5:I13965
0.03874403 MNG_Sagly_EIA_4:I6231
0.03919286 Saka_Tian_Shan:DA55
0.03927098 MNG_Sagly_EIA_4:I6224
0.03937766 MNG_Sagly_EIA_4:I7023
0.03994834 MNG_Munkhkhairkhan_MBA_2:I13173
0.04014648 RUS_Altai_IA:RISE601
0.04015296 RUS_Altai_IA:RISE600
0.04016105 RUS_Altai_IA:RISE602
0.04029801 Saka_Kazakh_steppe_o1:DA17
0.04102840 MNG_Hovsgol_BA_o2:ARS026
0.04117839 Saka_Kazakh_steppe:DA13
0.04120813 Sarmatian_RUS_Pokrovka:I0575
0.04129455 RUS_Tagar:DA2

Target: CHN_Shirenzigou_IA:M15-1
Distance: 2.5211% / 0.02521081 | R4P
35.0 RUS_Sintashta_MLBA_o1
29.8 Saka_Tian_Shan
22.2 MNG_Xianbei_IA
13.0 Sarmatian_RUS_Caucasus

Iron Post-Yanbulake Chine Shirenzigou [M012] 2195±20 BP M R1b1a1a2 R-PH155 U5a2
Iron Post-Yanbulake Chine Shirenzigou [M15-2] M Q1a1a1 Q-M120 U4

Iron Post-Yanbulake Chine Shirenzigou [M820] 285-230 cal BCE (2185±30BP) F U5b2c
Distance to: CHN_Shirenzigou_IA:M820
0.02637309 Saka_Tian_Shan:DA47
0.02904014 Saka_Tian_Shan:DA51
0.02978271 Saka_Tian_Shan:DA56
0.03176508 MNG_Sagly_EIA_4:I6231
0.03185844 MNG_Xiongnu_Central_Asian:DA38
0.03190784 Andronovo_KAZ_Alpamsa_LBA_Alakul:I6823
0.03250462 Sarmatian_RUS_Caspian_steppe:DA143
0.03315796 Saka_Kazakh_steppe:DA11
0.03364313 Sarmatian_RUS_Urals:MJ41
0.03365115 MNG_Sagly_EIA_4:I7023
0.03367210 MNG_Pazyryk_EIA_6:I6263
0.03396115 RUS_Tagar:DA4
0.03398735 Saka_Kazakh_steppe:DA16
0.03419371 Sarmatian_RUS_Urals:chy001
0.03429009 Saka_Tian_Shan:DA57
0.03459032 Saka_Tian_Shan:DA50
0.03459913 RUS_Altai_IA:RISE600
0.03496841 Andronovo_KAZ_Dali_LBA:I1931
0.03498628 Saka_Kazakh_steppe:DA13
0.03508347 Andronovo_KAZ_Zevakinskiy_LFBA_Fedorovo:I3976
0.03511979 Saka_Tian_Shan:DA49
0.03515252 Sarmatian_RUS_Urals:tem002
0.03519077 MNG_Sagly_EIA_4:I6232
0.03523521 MNG_Sagly_EIA_4:I7030
0.03535619 Sarmatian_RUS_Urals:MJ56

Target: CHN_Shirenzigou_IA:M820
Distance: 2.0287% / 0.02028704 | R4P
31.2 RUS_Ust_Ida_EBA
27.4 RUS_Srubnaya_Alakul_LBA
22.0 Sarmatian_KAZ
19.4 RUS_Afanasievo

Anonymous said...

Iron Post-Yanbulake Chine Shirenzigou [M819] F H15b1
Distance to: CHN_Shirenzigou_IA:M819
0.01470034 Saka_Kazakh_steppe:DA11
0.01815544 MNG_Hovsgol_BA_o2:ARS026
0.01828961 MNG_Sagly_EIA_4:I6231
0.01968172 Saka_Kazakh_steppe:DA16
0.01987536 MNG_Sagly_EIA_4:I12970
0.02097713 MNG_EIA_5:I13965
0.02304908 Saka_Tian_Shan:DA50
0.02352679 Saka_Kazakh_steppe:DA13
0.02373942 Saka_Tian_Shan:DA56
0.02384890 RUS_Altai_IA:RISE600
0.02386839 MNG_Sagly_EIA_4:I6230
0.02433783 MNG_Xiongnu_Central_Asian:DA38
0.02450347 Andronovo_KAZ_Alpamsa_LBA_Alakul:I6823
0.02462438 MNG_Sagly_EIA_4:I6232
0.02486302 Saka_Tian_Shan:DA49
0.02517002 MNG_Center_West_LBA_5:I6362
0.02565326 MNG_Sagly_EIA_4:I6226
0.02576781 MNG_Xiongnu_Central_Asian:DA41
0.02588919 MNG_Sagly_EIA_4:I7029
0.02628460 MNG_Pazyryk_EIA_6:I6263

Target: CHN_Shirenzigou_IA:M819
Distance: 0.8838% / 0.00883831 | R4P
49.2 Saka_Tian_Shan
17.6 Corded_Ware_DEU
17.0 CHN_Upper_Yellow_River_LN
16.2 RUS_Ust_Ida_LN

Iron Post-Yanbulake Chine Shirenzigou [M010] 410-350 cal BCE (2290±25BP) F D4j1b
Distance to: CHN_Shirenzigou_IA:M010
0.02272774 MNG_EIA_5:I13965
0.02341047 MNG_Sagly_EIA_4:I6230
0.02355504 Saka_Kazakh_steppe:DA13
0.02437437 MNG_Xiongnu_Central_Asian:DA41
0.02513205 MNG_Sagly_EIA_4:I6356
0.02576199 Saka_Kazakh_steppe:DA16
0.02624119 MNG_Sagly_EIA_4:I6226
0.02636134 MNG_Sagly_EIA_4:I6231
0.02636740 Saka_Kazakh_steppe:DA15
0.02696924 MNG_Sagly_EIA_4:I7022
0.02706418 MNG_Hovsgol_BA_o2:ARS026
0.02712268 MNG_Sagly_EIA_4:I7029
0.02716984 MNG_Sagly_EIA_4:I12970
0.02793278 MNG_Mongun_Taiga_LBA_3:I7033
0.02838785 MNG_Sagly_EIA_4:I6224
0.02850579 Saka_Kazakh_steppe:DA10
0.02864908 Saka_Kazakh_steppe:DA11
0.02867176 Andronovo_KAZ_Alpamsa_LBA_Alakul:I6823
0.02873621 MNG_Pazyryk_EIA_6:I6263
0.02927200 Saka_Tian_Shan:DA56
0.02931245 MNG_Xiongnu_Central_Asian:DA38
0.02931979 MNG_Munkhkhairkhan_MBA_2:I13173
0.02966328 Saka_Kazakh_steppe:DA14
0.03019983 RUS_Altai_IA:RISE602
0.03037417 MNG_Mongun_Taiga_LBA_3:I12976

Target: CHN_Shirenzigou_IA:M010
Distance: 0.9509% / 0.00950856 | R4P
35.0 MNG_Hovsgol_BA
28.0 Saka_Kazakh_steppe_o1
19.8 CHN_Western_Liao_River_LN
17.2 Sarmatian_RUS_Caspian_steppe

Iron Post-Yanbulake Chine Shirenzigou [F004] 370-190 cal BCE (BP) F A17
Distance to: CHN_Shirenzigou_IA:F004
0.03025045 MNG_Xiongnu_Central_Asian:DA41
0.03340763 MNG_Sagly_EIA_4:I6232
0.03376211 MNG_Sagly_EIA_4:I6231
0.03435826 MNG_Munkhkhairkhan_MBA_2:I13173
0.03449710 MNG_Hovsgol_BA_o2:ARS026
0.03455836 MNG_Sagly_EIA_4:I7030
0.03458396 Saka_Kazakh_steppe:DA14
0.03522400 MNG_EIA_2:I6364
0.03525734 MNG_Sagly_EIA_4:I7027
0.03533398 MNG_Center_West_LBA_5:I6367
0.03547619 MNG_Sagly_EIA_4:I7022
0.03551366 MNG_Xiongnu_Central_Asian:DA38
0.03562962 Saka_Kazakh_steppe:DA11
0.03594927 MNG_Pazyryk_EIA_6:I6263
0.03601083 MNG_Center_West_LBA_5:I6362

Target: CHN_Shirenzigou_IA:F004
Distance: 1.4553% / 0.01455308 | R4P
39.2 RUS_Afanasievo
26.0 CHN_Shimao_LN
25.4 MNG_Slab_Grave_EIA_1
9.4 RUS_Ust_Belaya

Distance to: CHN_Shirenzigou_IA:M827
0.02614766 MNG_Sagly_EIA_4:I6231
0.02769982 MNG_Sagly_EIA_4:I7030
0.02823314 Saka_Kazakh_steppe:DA11
0.02824518 MNG_Xiongnu_Central_Asian:DA41
0.02900672 MNG_Hovsgol_BA_o2:ARS026
0.02939047 MNG_Xiongnu_Central_Asian:DA38
0.02944656 Saka_Kazakh_steppe:DA16
0.03008654 Saka_Tian_Shan:DA47
0.03015908 MNG_Sagly_EIA_4:I6232
0.03052589 Saka_Tian_Shan:DA56
0.03060147 Saka_Tian_Shan:DA50
0.03063691 MNG_Sagly_EIA_4:I7022
0.03097338 MNG_Munkhkhairkhan_MBA_2:I13173
0.03176460 MNG_Pazyryk_EIA_6:I6263
0.03203123 MNG_Sagly_EIA_4:I12970
0.03217530 MNG_EIA_5:I13965
0.03252199 Saka_Kazakh_steppe:DA13
0.03300697 MNG_Sagly_EIA_4:I6226
0.03321686 RUS_Altai_IA:RISE600
0.03404747 MNG_Sagly_EIA_4:I6224

Target: CHN_Shirenzigou_IA:M827
Distance: 1.4827% / 0.01482702 | R4P
37.4 Saka_Tian_Shan
34.0 RUS_Okunevo_BA
20.4 CHN_Yellow_River_LN
8.2 MNG_EIA_3

At the end of the paper Chuan-Chao Wang et al. Genomic Insights into the Formation of Human Populations in East Asia // Nature (2021) in general is written complete nonsense, they did not analyze these samples, they just wrote complete nonsense.

Genos Historia said...

@ambron,

mtDNA H1b, U5b2b1a in Globular Amphora is most common in Eastern Europeans today.

Globular Amphora was like 50% French farmer. They came from Funnel Beaker.

This is why Andronovo in Asia has some mtDNA lineages which originated in Western Europe. Very cool to see these connections.

Davidski said...

@Jaakko

That was yet another dumbass comment.

Those N-L1026 lineages in the many non-Uralic speaking populations are irrelevant, because they obviously represent direct and indirect admixture from Uralic populations.

We know this not just thanks to the phylogeny of modern N-L1026, but also due to the records of N-L1026 in ancient DNA.

That is, there is no hint of any N-L1026 in Europe until it arrives there from the east with populations that are especially closely related to modern Uralic speakers.

Ric Hern said...

@ Archi

"Iranian Tokharians" ? Where in Linguistics are Tokharian classified as being Iranian ?

Copper Axe said...

I think those Shirenzigou samples might be related to the Yuezhi, as they aren't really in the right place or the right time to be Xiongnu proper.

In general those samples are odd, they have a lot of Central Asian HG+Pastoralist ancestry, probably some Qiang related stuff as well as Ulaanzuukh type ancestry (probably from the Gobi desert).

Imo most likely option is that they have ancestry from the bronge aze populations in the Tarim and Dzunghar Basins, Scytho-Siberian nomads and ancestry from regiond adjacent to the western Gansu grasslands (Qiang, Ulaanzuukh).

The Yuezhi homeland was described as more or less on the crossroads of those various ancestries.

That being said I don't think these samples were genetically representative of what the average people were like in that region but hey it's not like we have the data to confirm or deny that.

Erik Andersson said...

@Jaakko Häkkinen

The formation of N-Z1936 has nothing to do with this. Surviving subclades of N-Z1936 are closely associated with Uralic speakers and have a common ancestor near the time of Proto-Uralic.

As for Tarand grave people, that's very vague. Can you tell me which linguistic results prevent (Pre-)Proto-Finnic from having came from the east at the time of the first Tarand graves?
Besides, how do you know which route the first Tarand grave people took? Lang 2015 seems to suggest that they are associated with Pre-Proto-Finnic and took the southern route.

"During the Bronze Age they were not all very different, either: there were different populations in Baltia and Scandinavia, they were not homogeneous."

As stated in the quote from Saag et al. and as shown by every analysis of these samples, the Estonian BA individuals cluster with other Baltic BA individuals. And there are no samples like this from Bronze Age Scandinavia. Hunter-gatherers and Corded Ware people are irrelevant to this question.

Andrzejewski said...

@Genos Historia @ambron,

“mtDNA H1b, U5b2b1a in Globular Amphora is most common in Eastern Europeans today.

Globular Amphora was like 50% French farmer. They came from Funnel Beaker.

This is why Andronovo in Asia has some mtDNA lineages which originated in Western Europe. Very cool to see these connections.”

Funnelbeaker = LBK + Erteboelle.

GAC = Funnelbeaker (Balkan Farmers + Erteboelle WHG) + Cardinal Pottery (Med/Atlantic Farmers).

Anthony Hanken said...

"That is exactly the reason why you cannot see the language from the DNA. Nice that you understand it now.

So, how can you claim that this lineage is originally Proto-Uralic and only secondarily spread to the Eskimos and Chukchis? It could equally well be originally Northeast Siberian lineage and only secondarily spread to the Uralic speaking populations."

N-L1026 may or, may not, have spread secondarily to Uralic, at the Pre-Proto/Proto stage. That is irrelevant to whether it was the dominant lineage during the expansion and breaking up of Proto-Uralic. I believe it was.

"And how you could achieve that knowledge? When you cannot see it from the DNA itself?
You have to see the linguistic results: if there is a matching lineage, then you have a strong case."

I've already said, my view is very similar to Parapola's. The main thing I disagree with, is where Seima-Turbino formed. So far, N-L1026 subclades seem to be found, in the archeological cultures, he has designated Uralic.

"So you are saying that N-L1026 was present in the Volga-Kama area around 2000 BC? And this is shown in which aDNA study?"

I don't know if it was. There are plenty of Seima-Turbino necropoli found west of the Urals. IF N-L1026 was in the Volga-Kama region ~2000BC, that would be where.

Angantyr said...

(Blogger gave my first try an error message, so I'll make another attempt...)

@Jaska

"Scandinavian Pitted Ware people were very similar to the Late Neolithic Baltians. (Mittnik et al. 2018.)
Only with the Globular Amphora Culture and Corded Ware Cultures more Farmer ancestry arrived to Scandinavia. During the Bronze Age they were not all very different, either: there were different populations in Baltia and Scandinavia, they were not homogeneous."

What? Where in Mittnik et al. 2018 do you see PWC being similar to Baltic LN? What you do see there is that PWC were quite similar to Narva HGs, but more EEF shifted.

GAC did not exist in Scandinavia. EEF ancestry arrived with TRB farmers, the oldest both dated and genotyped one, Saxtorp, is even in the Mittnik paper: 3945–3647 cal. BCE, that's way earlier than CWC!

Battle Axe culture people were similar to contemporary Baltic CWC, but they did not build stone cists. Late Neolithic Scandinavians did, but those we have the genes of had more EEF ancestry - not an awful lot perhaps, but the difference is visible. Mittnik gathers all Battle Axe, Nordic LN and Nordic EBA samples under the "Scandinavia LNBA" label, which hides these differences.

What are the different populations in LN/EBA (Southern) Scandinavia? (The HG populations of Northern Scandinavia aren't relevant here.) I wouldn't be surprised if there were still people looking exactly like the Ölsund individual, i.e. almost indistinguishable from Baltic LN people, but we haven't found them. Especially not in stone cists.

Onno Hovers said...

@Archi: "What does scaled coordinates mean, I do not understand and no one does. To calculate in them is like calculating the distance if one coordinate is in meters, and the other coordinate is in feet."

The reality according to unscaled G25:

Target: Yamnaya_UKR (unscaled)
Distance: 4.0761% / 0.04076139
29.2 RUS_Karelia_HG (unscaled)
27.2 GEO_CHG (unscaled)
24.4 RUS_Tyumen_HG (unscaled)
9.8 TUR_Barcin_N (unscaled)
9.4 IRN_Ganj_Dareh_N (unscaled)

In this reality, Yamnaya can't be ancestral to later IE populations. For example:

Target: German (unscaled)
Distance: 3.2999% / 0.03299907
48.0 TUR_Barcin_N (unscaled)
38.4 RUS_Karelia_HG (unscaled)
11.8 GEO_CHG (unscaled)
1.2 WHG (unscaled)
0.6 IRN_Ganj_Dareh_N (unscaled)

The reality according to scaled G25:

Target: Yamnaya_UKR
Distance: 6.0527% / 0.06052705
55.4 RUS_Karelia_HG
35.2 GEO_CHG
8.8 TUR_Barcin_N
0.6 IRN_Ganj_Dareh_N

I think it is clear which reality this site is preaching. If scaled G25 is nonsense then so is everything that is published on Eurogenes.

ambron said...

Matt, IBD research tells migrantists to say goodbye to their naive vision of hundreds of thousands of migrating Slavs (such migrations even today would be a difficult logistical undertaking, and at the beginning of the Middle Ages were not possible due to the lack of communication routes and sources of provisionig). Thanks to IBD studies, we know that these were rather small groups movements within a wide area of low density population. So there is a great future for IBD research!

Matt, do you know if Reich has later Bronze Age and Iron Age samples from Poland?

ambron said...

Genos, the admixture of farmers in the CWC is very characteristic because it distinguishes it from Yamna. That is why it is worth knowing what population it comes from.

Ric Hern said...

@ Archi

If Tokharians were Iranian then why were their Language closer to Germanic and Celtic rather than closer to Baltic and Slavic ? Or do you want to tell us that Proto-Indo-Iranian was closer to the Western Branches and by extension Proto-Indo-European as well ?

Davidski said...

@Norfern et al.

Someone please check these coords...

https://drive.google.com/file/d/1uSJS9x8w-tX0ZxXqrpOXKYN8qsMBY4uh/view?usp=sharing

Copper Axe said...

@Ric Hern

The Tocharoi were a Yuezhi-Kushan tribe, they presumeably were East Iranic speaking nomads. The Tocharian language being named after Tokharoi is a misnomer that has produced a century's worth of confusion.

Anonymous said...

@Ric Hern
"@ Archi
"Iranian Tokharians" ? Where in Linguistics are Tokharian classified as being Iranian ?"

These are the true Tocharians, the Chinese called them Yuezhi. The true Tocharians were the ones who spoke Iranian. And about those Tocharians who are called pseudo-Tocharians nothing is known, how they called themselves or were called is unknown, once it is written that they spoke in Kuchanian. They were called Tocharians purely by mistake. A hundred years ago, they were called Tocharians purely by mistake. So, yes, the true Tocharians are Iranian Tocharians, and others are called Pseudo-Tocharians.

Ric Hern said...

@ Copper Axe

Thank you.

@ Archi

So you want to tell me that the Tocharian Language as known by Linguists was spoken by Pseudo-Tocharians ?

Ric Hern said...

@ Archi

So we have to wait for Pseudo-Tocharian DNA to know who spoke the Tocharian Language. Makes perfect sense...

Genos Historia said...

@Andre,

Funnel Beaker was 50% French farmer too. Might have had Ertebelle hunter gatherer. GAC and TRB Sweden both have extra hunter gatherer ancestry from Northern Europe.

Andrzejewski said...

@copper axe @ Archi @Ric Hern

“If Tokharians were Iranian then why were their Language closer to Germanic and Celtic rather than closer to Baltic and Slavic ? Or do you want to tell us that Proto-Indo-Iranian was closer to the Western Branches and by extension Proto-Indo-European as well ?”

Tarim Basin Mummies were found out to be closer to Andronovo than to Yamnaya because of the extra Anatolian ancestry found in them (Hollard 2018). Hollard et al. additionally discovered that 11/12 of the males had R1a1 in lieu of R1b, which likely eliminated their supposed origin from Afanasievo Culture (which like Yamnaya was mostly R1b).

No one can posit any plausible origin for the Tocharian language’s origin and its unique branching on the IE tree, however there’s an approximately 3,000 year gap between Afanasievo culture and the attested Tocharian manuscript.

On top of that, Afanasievo itself was a dead end, likely getting displaced by Okunevo before the latter was swept aside by Andronovo complex.

Therefore, I can more or less agree with Archi that the Tarim Basin Mummies were Bactrian in origin.

Andrzejewski said...

@Genos Historia “
Funnel Beaker was 50% French farmer too. Might have had Ertebelle hunter gatherer. GAC and TRB Sweden both have extra hunter gatherer ancestry from Northern Europe.”

I believe that it was a direct in situ offspring and natural evolution of the LBK, which was a Kurus-Sarcevo-Kris Anatolian Balkanic farmer, and therefore NOT a Cardial (Atlantic farmer) derivative.

PS if Basque Turns out to be a neolithic farmer language relic rather than the Mesolithic hunter gatherer one then it’s probably a Cardial Pottery one And has nothing to do with the Bałkanyk Farmer languages like LBK.

Andrzejewski said...

I am very fascinated with the history of south central and far east Asia of the classical era Because between the Saka, Scythians, Tocharians, Bactrians, Wusun, Greco Bactrians, Yazidi, Ancient northern Indian (ANI), Yuezhi, Kuchans, and other Indo-European nations, most of Asia from southern Siberia to Western Mongolia northern India western China Afghanistan and that’s not even including remnants of Alexander the Great’s Army were all basically Inhabited by European people and lots of the innovations for example like Metallurgy and Horsemanship in China where are due to contact with Europeans steppe people.

Andrzejewski said...

@Davidski Can we attribute the strong Anatolian farmer component for South Eastern Europeans like Bulgarians and Croatians to the Cucuteni Tripolye Culture’s genetic foundations?

If not, then I’m very Curious to know where they came from

Anonymous said...

@Ric Hern
"So you want to tell me that the Tocharian Language as known by Linguists was spoken by Pseudo-Tocharians ?"

That's right. The language spoken and written by the Iranian Tokharians is called Khotanese (Tumshuqese) Saka language by name of the oasis in Tarim.

"So we have to wait for Pseudo-Tocharian DNA to know who spoke the Tocharian Language. Makes perfect sense..."

I have always written about it. All this speculation is useless until the pseudo Tokharians of the 5th-8th century A.D. are tested.

Norfern-Ostrobothnian said...

@Davidski
All but SWE_Meso:ble004 look fine, although it is hard to tell if there's any noise in the Australians. However, by the nature of Australians and Papuans in G25 they model rather fine with Onge and Papuan as sources.

Copper Axe said...

@Andrezjewski

Remember that we only have y-dna data from the Xiaohe cemetery, which is one of the hundreds of bronze age cemeteries in the Tarim Basin.

The Xiaohe mummies having R1a does not confirm or deny that Tocharian came by way of the early steppe migrations to the Altai.

It also doesn't prove that the Afanasievo were a "dead end". The two Chemurchek samples had Afanasievo ancestry and one had M269, and the Chemurchek has been proposed as a missing link of some sorts between the Afanasievo and people of the Tarim Basin.

To understand who the Tocharians were and were they came from ideally we have to work backwards from the historical periods when Tocharian was attested to the bronze age and figure out through which route and material cultures they entered the Tarim Basin.

Jaakko Häkkinen said...


Davidski:
“That was yet another dumbass comment.
Those N-L1026 lineages in the many non-Uralic speaking populations are irrelevant, because they obviously represent direct and indirect admixture from Uralic populations.”

“Obviously”? Present the evidence!
You just cannot decide that it is due to the Uralic admixture, that is unscientific.
And how it could be due to the Uralic admixture, when there never were any Uralic speakers in Eastern Siberia.

Think what you say – too often you make no sense at all.

Davidski:
“We know this not just thanks to the phylogeny of modern N-L1026, but also due to the records of N-L1026 in ancient DNA.
That is, there is no hint of any N-L1026 in Europe until it arrives there from the east with populations that are especially closely related to modern Uralic speakers.”

1. Autosomal composition can change in every generation.
2. You cannot tie Y-DNA to any autosomal ancestry, because they are totally independent levels.
3. What you have, is a mere momentary correlation: you don’t know the autosomal composition of the N-L1026 men in the generations preceding the aDNA sample.
4. There are genetically very different Uralic speakers, so which ones are you referring to?

How many times I must tell you these, until you understand how science works?
If you don’t have any real arguments, you should shut up.

Jaakko Häkkinen said...


Erik Andersson:
“The formation of N-Z1936 has nothing to do with this. Surviving subclades of N-Z1936 are closely associated with Uralic speakers and have a common ancestor near the time of Proto-Uralic.”

Repeating does not make it any more factual.
The truth is, that you haven’t been able to present very convincing correlations so far.

Post et al. 2019 date the split between Z1934 and B539 at 5000 ybp. So, both branches existed during the Late Proto-Uralic dispersal. Still, neither branch is widespread within the Uralic language family: one is western, another is eastern.

Thus we can say that it is improbable that either of them was connected to Late Proto-Uralic. Still, there are too possible options:
1. they spread already before the Late Proto-Uralic, and later they participated correspondingly in the spread of West Uralic and East Uralic dialects;
2. they spread only after the Late Proto-Uralic, with West Uralic and East Uralic dialects.

Do you agree?
There is no evidence whatsoever to connect either of these lineages to Late Proto-Uralic itself.

Erik Andersson:
“As for Tarand grave people, that's very vague. Can you tell me which linguistic results prevent (Pre-)Proto-Finnic from having came from the east at the time of the first Tarand graves?”

Lang's model is not necessarily the final solution for the spread of Proto-Finnic, but it is a synthesis finding archaeological matches for the linguistic results. Some of which I present here:

1. Already in the Upper Volga region there were the northern taiga zone and the southern deciduous/leaf tree zone, northern being original and southern representing the spread of West-Uralic upon the Palaeo-European languages. Saami and Mari derive from the northern zone, having been retained old Uralic words (like *luka '10'), while Finnic and Mordvin derive from the southern zone, sharing loanwords with non-Uralic phonological structures and semantic field of southern forest and agriculture (*kümmin '10', *tammi 'oak', *vešna 'wheat', *lešma 'cow/horse').

It happens that the livelihood-environmental border goes from Upper Volga to Ladoga, and the spread of Pre-Proto-Saami happened all the way in the northern zone to the Onega-Ladoga isthmus.

2. Southern loanword layers like (older) Iranian and Baltic are more numerous in Finnic than in other West Uralic branches. Baltic area started already in the Moscow area, and Upper Volga Pozdnyakovo Culture has for a long time connected to Iranians. This again supports southern route for Finnic. Also the oldest Baltic loanwords seem to be connected to hunting and fishing economy, while in the Germanic loanwords there are more agricultural words. But this does not tell so much about the route.

3. Finnic inner taxonomy shows that the first languages to differentiate were the southernmost ones: first South Estonian, then Livonian. The more northern the area, the later the differentiation of languages. This also supports the southern route for the spread of Finnic.

4. Placenames of Finnic origin are found in Latvia and few even in Northern Lithuania, while no certain Finnic placenames are found east of Estonia (except in Ingria and the Veps region, both however representing quite recent spread of North Finnic from Finland): there are the Meryanic and West Chudic placenames.

So, all the linguistic evidence really seems to exclude the possibility, that Finnic could have spread to Estonia right from the east. That means that any gene flow or cultural wave following that route is impossible to connect to the Finnic speakers.

It is usual for laymen, that when they see a cultural or genetic wave from the east, they think it must represent Uralic language or branch: the one spoken in the area at the present. They are blind to the fact that cultural and genetic waves are numerous, and they keep coming continuously from about every direction. Usually only one of them can be connected to the spread of certain language, and guessing which one it is, is impossible.

Jaakko Häkkinen said...


Erik Andersson:
“Besides, how do you know which route the first Tarand grave people took? Lang 2015 seems to suggest that they are associated with Pre-Proto-Finnic and took the southern route.”

Lang 2020 (Homo fennicus) says that the earliest Tarand graves were born continuing the Stone cist graves, so maybe they were local innovation. Only thing connecting them to the eastern Houses of the dead -tradition is the rectangular shape, which is common all around the world.

Of course the first N-L550 men are found within the Tarand graves, but we all should know by now, that
1. gene flow cannot testify anything about the origin of culture or language; and
2. N-L550 could have predated the arrival of the (Pre-)Proto-Finnic speakers in Baltia, and many basal branches of it are found within the Swedes, so it could spread even from the west.

Erik Andersson:
“As stated in the quote from Saag et al. and as shown by every analysis of these samples, the Estonian BA individuals cluster with other Baltic BA individuals. And there are no samples like this from Bronze Age Scandinavia. Hunter-gatherers and Corded Ware people are irrelevant to this question.”

No, they are very relevant, until we get Bronze Age samples from Scandinavia outside the southernmost parts. The Pitted Ware Culture, people of which were similar to Baltian populations, remained until the beginning of the Bronze Age.

In Baltia, the Mediterranean farmer ancestry increased during the Bronze Age, but again diminished during the Iron Age: different populations seem to have preserved well their genetic compositions. We have no data in Scandinavia which would make it possible to exclude the western gene flow to Baltia during the Bronze Age.

Jaakko Häkkinen said...

Anthony Hanken:
“I've already said, my view is very similar to Parapola's. The main thing I disagree with, is where Seima-Turbino formed. So far, N-L1026 subclades seem to be found, in the archeological cultures, he has designated Uralic.”

Seima-Turbino has two roots: the Northeastern European and Southern Siberian bronze networks united. It is irrelevant which of them made the original proposition.

N-L1026 is also found in Kola BOO, which cannot have anything to do with the Uralic languages: neither in the source area nor in the target area there were Uralic speakers anywhere near for yet 2000 years.

Angantyr:
“What? Where in Mittnik et al. 2018 do you see PWC being similar to Baltic LN? What you do see there is that PWC were quite similar to Narva HGs, but more EEF shifted.”

Chronologically Comb Ceramic Culture was also Late Neolithic, although they don’t use the additional temporal label “LN” with these cultures. Also archaeologically Scandinavian Pitted Ware Culture is connected to the Comb Ceramic Culture. In Sweden, the Pitted Ware Culture remained until the transition to the Bronze Age.

Angantyr:
“GAC did not exist in Scandinavia. EEF ancestry arrived with TRB farmers, GAC did not exist in Scandinavia. EEF ancestry arrived with TRB farmers,”

Yes, sorry, I meant TRB.

Angantyr:
“What are the different populations in LN/EBA (Southern) Scandinavia? (The HG populations of Northern Scandinavia aren't relevant here.) I wouldn't be surprised if there were still people looking exactly like the Ölsund individual, i.e. almost indistinguishable from Baltic LN people, but we haven't found them. Especially not in stone cists.”

So, do we have any data from the stone cist graves in Scandinavia? And even if we had, we couldn’t exclude the Scandinavian gene flow to Baltia, like some people here want to do. That is my point.

Copper Axe said...

@Davidski

I take it that your busy but a new blog entry would he appreciated by me because its hard to keep track of what is being discussed when I see the same twelve paragraphs on Proto-Uralic rambles posted five times a day here.

Cheers.

Davidski said...

I can't do anything this week, but from next week I should have more time for the blog.

Rob said...

Maybe the (pseudo-)Tocharians were buddhists from Western Europe going on a pilgrimage to the East :)

Norfern-Ostrobothnian said...

BOO could have absolutely been an Uralic population. Not every population leaves toponyms, or at least noticeable ones. The ones their small population might have left got Slavified and Saamified, or entirely forgotten. After all, there are hardly any pre-Uralic toponyms left in NE Europe.

Matt said...

@ambron, no I've not heard anything, but Reich says they have about 4x more unpublished samples, and they are rolling out a new process of sequencing by licencing a new type of probe which smaller labs including outside US should be able to use, so between those things I would be surprised if there wasn't something shaping up in a years time or so.

But that's pure guess on my part; I'd say it might be worth emailing him directly and see if he responds in general terms. (That said only time I emailed him, it was about investigating ancestry of some unusual Italian villages who has some potentially Basque like genetics, in terms of adna, but got no response).

Anonymous said...

@Norfern-Ostrobothnian

"BOO could have absolutely been an Uralic population."

There was a version that they were Yukaghirs, the closest relatives of the Finno-Samoyeds, but no one can prove anything yet. All versions are possible about them.


Anonymous said...

@Norfern-Ostrobothnian

Distance to: RUS_Bolshoy_Oleni_Ostrov
0.08674569 FIN_Levanluhta_IA:DA238
0.12420658 Yukagir_Forest:Nel16
0.12439574 Yukagir_Forest:Nel19
0.12675561 FIN_Levanluhta_IA:JK1970
0.12755972 Yukagir_Forest:Nel15
0.13093230 Yukagir_Forest:Nel17
0.13298033 Yukagir_Forest:Nel13
0.15529843 FIN_Levanluhta_IA:JK1968
0.16417273 FIN_Levanluhta_IA:DA234
0.16497823 RUS_Yakutia_LUP:Khaiyrgas1
0.16528954 RUS_Sosnoviy_WSHG:I5766
0.17681968 RUS_Tyumen_WSHG:I1960
0.24220512 RUS_Veretye_EHG_Meso:PES001
0.25255650 KAZ_Zevakinskiy_BA:I3770
0.28882721 RUS_Ust_Ishim_IUP:Ust_Ishim
0.30187785 Yukagir_Tundra:Yuk_024
0.30642949 FIN_Levanluhta_IA_o:JK2065
0.31395881 Yukagir_Tundra:Yuk_020
0.31756055 Yukagir_Tundra:Yuk_004
0.31865638 SWE_Motala_HG:I0013
0.32153459 SWE_Motala_HG:I0014
0.32665994 Yukagir_Tundra:Yuk_011
0.32960657 SWE_Motala_HG:I0015
0.32961867 Yukagir_Tundra:Yuk_022
0.32979021 Yukagir_Tundra:Yuk_009

Distance to: RUS_Bolshoy_Oleni_Ostrov_o:BOO006
0.11990597 Yukagir_Forest:Nel15
0.12000373 Yukagir_Forest:Nel17
0.12581197 RUS_Yakutia_LUP:Khaiyrgas1
0.13472432 Yukagir_Forest:Nel13
0.13945361 Yukagir_Forest:Nel16
0.14364051 FIN_Levanluhta_IA:DA238
0.14565790 Yukagir_Forest:Nel19
0.18470004 FIN_Levanluhta_IA:JK1970
0.21644739 FIN_Levanluhta_IA:JK1968
0.22605128 FIN_Levanluhta_IA:DA234
0.22638049 RUS_Sosnoviy_WSHG:I5766
0.23454231 RUS_Tyumen_WSHG:I1960
0.23613781 Yukagir_Tundra:Yuk_024
0.24620598 Yukagir_Tundra:Yuk_020
0.24900709 Yukagir_Tundra:Yuk_004
0.25777945 Yukagir_Tundra:Yuk_011
0.25958039 Yukagir_Tundra:Yuk_022
0.26071738 Yukagir_Tundra:Yuk_009
0.26430693 Yukagir_Tundra:Ckrd9
0.26473999 Yukagir_Tundra:Yuk_025
0.26513468 Yukagir_Tundra:Ckrd8
0.27084079 Yukagir_Tundra:Yuk_019
0.27304446 Yukagir_Tundra:Yuk_023
0.27648772 Yukagir_Tundra:Ckrd13
0.28452964 Yukagir_Tundra:Ckrd18

Target: RUS_Bolshoy_Oleni_Ostrov
Distance: 3.0149% / 0.03014942
30.8 Yukagir_Tundra
28.8 RUS_Veretye_EHG_Meso
16.8 RUS_Sosnoviy_WSHG
10.4 Yukagir_Forest
9.0 RUS_Tyumen_WSHG
3.4 KAZ_Zevakinskiy_BA
0.8 RUS_Ust_Ishim_IUP

Target: RUS_Bolshoy_Oleni_Ostrov_o:BOO006
Distance: 4.4180% / 0.04417970
43.4 Yukagir_Tundra
24.6 RUS_Sosnoviy_WSHG
17.6 RUS_Veretye_EHG_Meso
6.8 Yukagir_Forest
4.4 RUS_Tyumen_WSHG
2.8 RUS_Yakutia_LUP
0.4 RUS_Ust_Ishim_IUP

A said...

Are there going to be any studies on Seima-Turbino DNA in the foreseeable future?

Janko Raven Johnson said...

Jackson and Archi have really brought out the best in Rob. Thank you, Dave!

Erik Andersson said...

@Jaakko Häkkinen
You were talking about N-Z1936 being "born in the Proto-Uralic population", which is not a claim I made, so it seemed necessary to set the record straight.
The 95% range for the estimate given by Post et al. overlaps the expected age for Proto-Uralic, and other estimates are lower. Even if the split is older than Late Proto-Uralic, there could be separate founder effects in East and West Uralics. So I still think it's very probable that N-Z1936 is linked with Proto-Uralic.

Thanks for the summary. I am willing to consider the arguments in favour of the southern route, but I'm not buying that "gene flow cannot testify anything about the origin of culture or language", as it is a direct attestation of migration and descent. Whether Tarand graves were entirely a local innovation or not, the people buried in the earliest ones were migrants, and their DNA connects them to present and ancient Uralic speakers near and far. That is not something that is easily explained away, at least not in my world.

As for the stone-cist graves, I think you're frankly writing nonsense at this stage. The PWC obviously has nothing to do with stone-cist graves in Scandinavia or Estonia. And the increase in EEF is not tied to stone-cist graves either. The Lithuanian samples have more EEF than Latvian or Estonian ones, so it's more likely from the south.

Andrzejewski said...

@Copper Axe I’m only going by what Hollard 2018 referenced. The other reason he thought that Tarim Mummies were more similar to Corded Ware/Sintashta/Andronovo/Modern Europeans via-a-vis Afanasievo/Yamnaya was because of their rather significant stock of Anatolian mtDNA, which could only come to them by means of GAC assimilated into CWC in the Carpathians.

It is a common knowledge among geneticists and archeologists that Afanasievo was largely replaced by Okunevo, the latter one being indigenous Siberian, and that much later Saka, Cimmerians and Scythians groups (Sarmatians?) would carry a large percentage on their maternal side stemming from East Asian Okunevo.

The Siberian admixture into Tarim Basin Mummies can just as well be attributed to Okunevo.

Anthony Hanken said...

@Jaakko

"Seima-Turbino has two roots: the Northeastern European and Southern Siberian bronze networks united. It is irrelevant which of them made the original proposition."

Roots is probably the wrong word. Seima-Turbino originated squarely in Siberia. The European metallurgical province differed from S-T in conposition, style, and methods of casting.

Seima-Turbino is severely underrepresented in terms of radiocarbon datings. However, the few dates we do have show two trends. They were earlier than expected, and older in the east.

PIIr groups were almost certainly involved. There is strong archeological evidence they may have even lived in mixed communities with Seima-Turbino. It is evident, even in such cases, that the two metallurgical traditions were distinct and treated as such.

For Uralic this makes sense. The earliest metallurgical words seem to be native to the language.

"N-L1026 is also found in Kola BOO, which cannot have anything to do with the Uralic languages: neither in the source area nor in the target area there were Uralic speakers anywhere near for yet 2000 years."

It's possible BOO were pioneering Uralic speakers, although its just a possible they weren't. There is simply not enough data, to make any solid conclusions.

ambron said...

Matt, maybe it's actually a good idea to just ask Reich about it.

Genos Historia said...

@Andre,

The higher farmer ancestry in say Romania compared to Ukraine, yeah is due to Balkan farmers. I don't think we link it directly to Cucuteni Tripolye. But that makes sense and would be cool.

Croatia is a different. It is overwelmingly Slavic anyways. Their South Euro ancestry comes from West Balkans not East Balkans.

The Slavs kind of messed of things up. Europe's variation used to be more cool. They watered down whatever legacy was left of Neolithic Balkans. Good thing they never made it to Western Europe. We are still Western Corded Ware, "unchanged for 4,000 years."

Genos Historia said...

@Andre, I'm joking of course.

There does seem to have been an ancient difference between North of Danube and South of Danube, that the Slavs watered down.

But maybe not. Bronze, Iron age Hungary and Serbia show Welzin ("Slavic-like") ancestry. Suggesting geneflow north and south of Danube before Slavs.

E. Donovan said...

Z1936-influenced cultures should include Itkul (Ugrian), Northern Ananino/Ananyino (some relation to Para-Karelian-Vepsian), Bjarmaland (Para-Karelian). As far as I understand it, Ananyino disintegrated with a lot of it dispersing northwards, although Proto-Permians may also have participated in the culture and the Scythians no doubt absorbed many of its southerners. Z1936, as you all know, may have been among the clades associated with the use of Netted Ware but it cannot have been the only, given its (Netted Ware's) amorphous distribution across northern Eurasia and apparent derivation from the Transbaikal before Z1936 even formed.

E. Donovan said...

I forgot to ask, does anyone know which, within the Ural-Altaic typological complex, of the so-called Altaic families is closest to Uralic? Obviously N is found relatively frequently with their speakers, and Yakut (no surprise) is described to have one, two or a few similarities to Uralic in addition to the overall complex. However, as far as I can tell from the archaeogenetic evidence now it is Proto-Turkic from the U-A complex which will never be given a remotely certain origin. Mongolic and Tungusic are both already easily declared "C with N influence, however much precisely unknowable". Next question is does anyone know what's going on with the Chuvash, the earliest diverged Turkic speakers (followed by the Yakut)? According to early report they don't even have any Q and are principally various clades of N and R-M417. What? Chuvash and Yakut predate any known Turkic empires so the "learned from Turkic Empire" thing won't work here. The only significant influence on Yakut after the 1st Millennium BC is Mongol, followed much later by Russian.

I know people are going to try to say the Xiongnu. Unprovable, whatever they spoke, if anything.

Rob said...

@ Genos
I know you’re joking but it wasn’t actually the slavs - they were moving into often already empty lands
In Central Europe; ‘Germanic’ groups ventured into Roman provinces to try become rulers, or were pushed on by huns
But they were lost amongst a sea of provincials, & often had high mortality rates as they were soldiers . So alls that is left is some Germanic lineages in Italy & Spain

ambron said...

Rob, and where were the empty lands in Central Europe? Palynology has denied this, at least with regard to Poland, the Czech and East Germany.

Rob said...

Ambron- happy to discuss it when the topic comes

Jaakko Häkkinen said...


Norfern-Ostrobothnian:
“BOO could have absolutely been an Uralic population.”

Not a chance. Tell me what of this you cannot understand?
NEITHER IN THE SOURCE AREA (NORTHERN SIBERIA) NOR IN THE TARGET AREA (KOLA PENINSULA) THERE WERE URALIC SPEAKERS ANYWHERE NEAR FOR YET 2000 YEARS.

It is absurd to claim the presence of the Uralic speakers based on DNA - that is circular argumentation at its worst! Wake up now.


Erik Andersson:
“Thanks for the summary. I am willing to consider the arguments in favour of the southern route, but I'm not buying that "gene flow cannot testify anything about the origin of culture or language", as it is a direct attestation of migration and descent.”

Please, try to explain, HOW EVEN IN THEORY gene flow could testify anything about the origin of culture or language? You haven’t read anything I wrote in these last weeks?

EA
“Whether Tarand graves were entirely a local innovation or not, the people buried in the earliest ones were migrants,”

Wrong: only two of them were immigrants, based on the isotopic analysis. You cannot make black-and-white assumptions: they are baseless and unscientific.

EA: “and their DNA connects them to present and ancient Uralic speakers near and far.”

Their DNA connects them also to non-Uralic populations. Why do you ignore this part of the evidence?
You cannot make black-and-white assumptions: they are baseless and unscientific.

EA: “As for the stone-cist graves, I think you're frankly writing nonsense at this stage. The PWC obviously has nothing to do with stone-cist graves in Scandinavia or Estonia.”

Prove it. PWC people lived in Sweden to the beginning of the Bronze Age. How can you say that there is no genetic continuity from them to the Stone cist grave people in Scandinavia? You cannot – you are again making black-and-white assumptions.

EA: “And the increase in EEF is not tied to stone-cist graves either. The Lithuanian samples have more EEF than Latvian or Estonian ones, so it's more likely from the south.”

You cannot claim to know it, because we don’t have enough evidence. There are still many possibilities open.

Jaakko Häkkinen said...

Anthony Hanken:
“It's possible BOO were pioneering Uralic speakers, although its just a possible they weren't. There is simply not enough data, to make any solid conclusions.”

It is not reasonable to suggest linguistic waves of which there are no traces anywhere.
If you claim the presence of the Uralic speakers based on DNA, that is circular argumentation at its worst.

You wouldn’t suggest ghost DNA based on language, would you? So why you would suggest ghost language based on DNA? Both are equally unscientific.

Andrzejewski said...

An interesting article by Lazaridis re: Basal Eurasians https://www.biorxiv.org/content/10.1101/2021.02.24.432678v1.full

Onno Hovers said...

@E. Donovan: "I forgot to ask, does anyone know which, within the Ural-Altaic typological complex, of the so-called Altaic families is closest to Uralic?"

I would say Turkic and Tungusic, because they both have old systems of personal endings on both verbs (like Indo-European and Uralic) and nouns (like Uralic). The rest of 'Altaic' either does not do person marking, or it is clearly a recent innovation (some Mongolic languages).

Anthony Hanken said...

@Jaakko

"It is not reasonable to suggest linguistic waves of which there are no traces anywhere.
If you claim the presence of the Uralic speakers based on DNA, that is circular argumentation at its worst."

I'm not claiming anything. I don't know what language BOO spoke. They were probably isolated being that far north, so whatever language they did speak, may not have left any trace. Whether that was some dead dialect of Uralic, or something completely different.

George said...

OFF COMMENT

Genome-wide analysis of nearly all the victims of a 6200 year old massacre

Mario Novak, Iñigo Olalde, Harald Ringbauer, Nadin Rohland, James Ahern, Jacqueline Balen, Ivor Janković, Hrvoje Potrebica, Ron Pinhasi, David Reich
Published: March 10, https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0247332


From the Abstract:
"Here we provide evidence of killing on a massive scale in prehistory that was not directed to a specific family, based on genome-wide ancient DNA for 38 of the 41 documented victims of a 6,200 year old massacre in Potočani, Croatia and combining our results with bioanthropological data. We highlight three results: (i) the majority of individuals were unrelated and instead were a sample of what was clearly a large farming population, (ii) the ancestry of the individuals was homogenous which makes it unlikely that the massacre was linked to the arrival of new genetic ancestry, and (iii) there were approximately equal numbers of males and females."

Wise dragon said...

It's a bit off-topic, but can someone explain why David Reich now claims that Northern Europeans are a mixture between Ancient Levantine people and Ancient North Eurasians? His wording is pretty misleading since people that haven't read the Lazaridis paper about the ancestral populations of modern Europeans, will think that Northern Europeans are half Natufians/Middle Easterners and half Ancient North Eurasians. He later got more precisely and referred to Anatolian Farmers, WHGs and Steppe too. But again, Reich uses the term Ancient Levantine farmers to describe how the ancestry of Northern Europeans formed, thus as if they were part Natufians or Neolithic Levantines etc. in min. 41-42:22


https://www.youtube.com/watch?v=QoGmPJJS3X8

Anyway, he said nothing new about the Yamanya and the Corded Ware. However, explained their spread into Europe and South-and South East Asia in an understandable way. Reich supports the theory that Steppe people brought Indo-European language and culture into South Asia. He rules based on Ancient DNA the Anatolian Hypothesis for the origin of the IE Language.

Genos Historia said...

@Andre,

Awesome. People with extra basal Eurasian in Eastern Arabia. Interesting.

Andrzejewski said...

Has anyone heard of the Liburnians?

https://en.m.wikipedia.org/wiki/Liburnians

They retained a lot of Cucuteni non-IE substrate.

Erik Andersson said...

@Jaakko Häkkinen
You may have noticed that I have not agreed with what you have written.
Being able to directly see the origin of migrants and how they're related to other peoples tells you a lot about how language and culture might have spread.
And Tarand grave people have a specific relationship with Uralic speakers (in the form of Siberian ancestry and N3a) not shared by other earlier or contemporary peoples of the Baltic (that they also have ancestry in common with).
Since we don't seem to have gotten anywhere, that's the end of this conversation for me.

As to the stone-cist graves, I am not seeing the connection between them, the PWC and Baltic LN HGs. I very much doubt that samples from Scandinavian stone-cist graves would be PWC-like.
The EEF in Baltic BA is not unique to stone-cist grave samples, so it can't be tied to them specifically.

vAsiSTha said...

@andre @copperaxe

Tocharian language has nothing to do with Afanasievo. Andre's points are right about that.

Question remains how a centum language exists there..
Another similar mystery is how Bangani - a centum dialect - remains in what is now Uttarakhand in the himalayas.

Michalis Moriopoulos said...

@Wise dragon

He and others have used the word in that loose way before to refer to ANFs and Natufians as a general Near Eastern farmer group. They are sister groups after all. It's a bit like using Iran-related to refer to CHG, also sister groups. Is it a good practice? No. Is it lazy? Sure. But it's hardly a mystery what he's talking about and it's not what I'd call nefarious. At least no more than the people who doth object to these semantic points a bit too strongly. Whatever you call these people, they WERE Near Eastern and phylogenetically closely related to Natufians. There are more far more interesting things to talk about.

Genos Historia said...

@Michalis,

I can't get away from saying Harvard terminology is a bit inaccurate.

The concept of WHG is outdated. As it is likely the WHG ancestry in Eastern Europe didn't come from Western Europe.

The concept of IranN is outdated.

Now, they're creating a new concept of LevantN which is inaccurate for the same reasons IranN is inaccurate.

Genos Historia said...

AnatoliaN definitely has LevantN ancestry.

But the AnatoliaN which went to Europe had less of it than AnatoliaN in Central-Eastern Anatolia.

And, LevantN itself was already like 50% Anatolian. If anything we are seeing admixture from Anatolia into the Levant during the advent of farming not the other way around.

Yet, in archaeology they see farming go from Levant>Anatolia. SO it seems they're trying to fit DNA in with archaeology.

Genos Historia said...

46:21

David Recih shows family relationship, using 23andme tools, between Corded Ware and Yamnaya.
Also between Corded Ware & Globular Amphora.

Genos Historia said...

Reich's presentation does have new stuff. He mentions new BMAC data, new data from Harrappan civ, new data from Pakistan which is 3,500 years old (and shows Steppe admixture).

Rob said...

@ Andrze


“ Has anyone heard of the Liburnians?

They retained a lot of Cucuteni non-IE substrate.”

Please explain ??

Genos Historia said...

@Wise Dragon,

I don't think Reich was very confusing. I actually like how he chose to describe how the research developed over years instead of saying outright what the results are.

But I'm with you it is annoying that he ignores AnatoliaHG and CHG when describing West Eurasia 10,000 years ago.

And it is annoying he ignores, he doesn't explain that WHG and EHG weren't mutually exclusive. We all here know, they were two opposite ends of a cline.

Andrzejewski said...

Natufians were created when Iberomaurasians speakers of Proto-Afro-Asiatic from North Africa invaded Syria and Israel and imposed their language on the AHG/UHG/Augrinacians indigenous to the environment. So Levant_N and Anatolians were already on same cline.

Later on, during PPNB and PPNC the Levant and Anatolian populations sort of fused. Lazaridis 2016 mentions Peki’in like groups originating from Anatolia, carrying ydna uniparental marker y-T.

The entire ME and NE has Anatolian markers such as G, T, E1b1b, somewhat also K and H, whereas mtDNA associated with LBK - T, N, K, H and others were and are abundant.

It’s fair to assume that Levant_N was 50% Anatolian-like, before a new ethnic makeup similar to Anatolia BA/KAC invaded. Contemporary Lebanese are 52% Levant_N, 48% Iranian/CHG/KAC.

Samaritans, who are descendants of the Kingdom of Ephraim, the only Jewish population which was never exiled, exhibit 26%-27% Anatolian aDNA.

The book of Genesis in the OT, which according to historians was written in the IA period does speak of some “Patriarchs” who migrated from Harran (Tepecik Çiftlik?), and some biblical researchers claim that many customs in that book were Hurrian (i.e. Kura Araxes/Anatolia BA), even some of their name. Could it reflect some faded memory by writers/editors of the book of some “Marianu” ruling elite?

Could it be just Kura Araxes is responsible for the CHG y-DNA’s prevalence among modern and Iron Age onward “Semites”?

Some linguists suppose that the Sumerian language, rather than being an “isolate” was actually a relic of some West Asian language family before being subsumed and assimilated by invading North African Proto-AA speakers. Can it be then that Sumerian may have been related to dialects spoken by LBK?

So it points to Levant and Anatolian populations rapidly merging during the PPNB Neolithic, with a 50% genetic similarity, only to be later on “diluted” - both in Levant and in Anatolia by Iran + Caucasus ancestry from the East.

Andrzejewski said...

@Rob Liburnians were Paleo-Balkan genos living east of the Adriatic, and their language contained a substantial non-IE substrate. I surmised it to be related to the language spoken by CTC

Dave the Slothtopus said...

So Alex W. has identified one of the Dutch Beakers (I5748) as P312>DF19>Z302:
Noord-Holland
aDNA: I5748
MDKA Birth: Netherlands
Bell Beaker, PRJEB23635. 2579-2211 calBCE (3945±55 BP, GrN-6650C). Noord-Holland, Oostwoud, De Tuithoorn Netherlands.
More evidence for the CWC>SGC>BB in the north? This is the first truly ancient DF19 to show up.

Carlos Aramayo said...

@Genos Historia

"Reich's presentation does have new stuff. He mentions new BMAC data, new data from Harrappan civ, new data from Pakistan which is 3,500 years old (and shows Steppe admixture)."

What makes you think so? I watched the lecture and found that David Reich based it only in Narasimhan et al (2019) paper, claiming an arrival between 2000 and 1500 BCE (only slightly "modifying" 1900 to 1500 estimate by Narasimhan). Can you be more specific? Where's the new stuff?

Rob said...

@ Andrze
Yep I know who the Liburnians were; but I’m not sure how you specifically linked a Roman Iron Age Dalmatian group with C-T

Rob said...

Reich seems preoccupied with disproving the Anatolian hypothesis, to the extent that his talks can sound speculative and sometimes wrong ;
Eg he is incorrect about the lack of Anatolian- related ancestry moving into central - South Asia . His inferences about lack of Iranian farmer migration to India is speculative given the absence of relevant adna

Jaakko Häkkinen said...

Anthony Hanken:
“I'm not claiming anything. I don't know what language BOO spoke. They were probably isolated being that far north, so whatever language they did speak, may not have left any trace. Whether that was some dead dialect of Uralic, or something completely different.”

1. How could it be Uralic, when Late Proto-Uralic was still spoken in the narrow homeland in the Volga-Kama region during the time the North Siberians went to Kola Peninsula?
2. How could it be Uralic, when we know that the ancient Palaeo-Laplandic languages were not Uralic?

You could equally well claim that the language of the BOO people was Aryan or Northwest Indo-European – they were spoken equally close to the area than Proto-Uralic at that time.

Jaakko Häkkinen said...

Erik Andersson:
“You may have noticed that I have not agreed with what you have written.
Being able to directly see the origin of migrants and how they're related to other peoples tells you a lot about how language and culture might have spread.”

MIGHT. Indeed. Still, the genetic results cannot testify for language, unless you can present a mechanism how language is inherited in DNA.

Erik Andersson:
“And Tarand grave people have a specific relationship with Uralic speakers (in the form of Siberian ancestry and N3a) not shared by other earlier or contemporary peoples of the Baltic (that they also have ancestry in common with).
Since we don't seem to have gotten anywhere, that's the end of this conversation for me.”

Yes, we should stop, if you still put your beliefs in front of the scientific evidence and cannot understand that correlation is not evidence. It is possible that the spread of some N3 subhaplogroups and Siberian ancestry is connected to the spread of Uralic languages, but it is possible that those are earlier or later, too.

Erik Andersson:
“As to the stone-cist graves, I am not seeing the connection between them, the PWC and Baltic LN HGs. I very much doubt that samples from Scandinavian stone-cist graves would be PWC-like.
The EEF in Baltic BA is not unique to stone-cist grave samples, so it can't be tied to them specifically.”

All I’m saying that at the moment there are many possibilities, so you shouldn’t lock into just one option, when you cannot exclude the other options. You cannot just decide how it was – you have to prove it. You cannot pick up a guess and then build more guesses upon it.

Andrzejewski said...

Has it occurred to anyone that the Cossacks may be the descendants of the Scythians?

The Scythians were offspring of West Iranic Andronovo horseback riders, and their range of occupation mode of less overlaps the Yamnaya one. First, it’s interesting to find out the extent to which the backward migration of CWC/Fatnayovo assimilated instead of dispersed the original Yamnaya/Poltava/Catacomb, and to what extent Scythians retained their Europoid aDNA v. East Asian mtDNA obtained throughout the centuries. Then it’s also intriguing to attempt to seek continuity between Scythians and Cossacks, including assimilated Turkic (Kumans, Pechenegs, etc) and Finno-Ugric - both groups containing a significant amount of Scythians and other IE Steppe nomadic substrate.

Although Ossetian are essentially the only surviving vestige of Scytho-Sarmatians/Alans, Cossacks distinguish themselves from East Slavs and attribute direct heritage and ancestry from Scythians and Sarmatians instead.

Davidski said...

@Jaakko

The remains from the Stone Cist graves are very similar to Bronze Age East Baltic populations and obviously of local origin, with no specific genetic links to Scandinavia whatsoever.

The ancient DNA doesn't allow any other options, so you're having a pointless argument there.

The remains from the later Tarand graves are different, with obvious links to Uralic speakers in Siberia.

Ergo, the Stone Cist grave people weren't Germanic speakers, unless they became Germanic speakers without any gene flow from Scandinavia.

But the Tarand grave people may well have been Uralic speakers, having received their Uralic language along with Siberian DNA from the east.

Anthony Hanken said...

@Jaakko

"1. How could it be Uralic, when Late Proto-Uralic was still spoken in the narrow homeland in the Volga-Kama region during the time the North Siberians went to Kola Peninsula?"

BOO is dated to 3500ybp. This is after Late Proto-Uralic.

2. How could it be Uralic, when we know that the ancient Palaeo-Laplandic languages were not Uralic?

It's not entirely clear where BOO came from. They may have arrived alongside Lozervo ware, this would link them to the Imiyakhtakhskaya culture. The problem is that BOO prefers kra001 over the Imiyakhtakhskaya samples.

"You could equally well claim that the language of the BOO people was Aryan or Northwest Indo-European – they were spoken equally close to the area than Proto-Uralic at that time."

Uralic speakers share a specific Siberian component, also found in BOO. Aryan and NWIE speakers do not.

Jaakko Häkkinen said...

Davidski:
“The remains from the Stone Cist graves are very similar to Bronze Age East Baltic populations and obviously of local origin, with no specific genetic links to Scandinavia whatsoever.
The ancient DNA doesn't allow any other options, so you're having a pointless argument there.”

Show me the evidence that they are very different from the Bronze Age Scandinavian populations. If you cannot do that, you cannot exclude the gene flow, either. As we know, the grave type itself was born in Scandinavia. It cannot have spread to the east without some people moving, too.

Davidski:
“The remains from the later Tarand graves are different, with obvious links to Uralic speakers in Siberia.”

Obvious links? If you mean Y-DNA N3, there is obvious link to Chukchis, Eskimos, and Dolgans etc., all of which are non-Uralic populations.
If you mean the Siberian ancestry, it is also found in non-Uralic populations.

You cannot just look at the Uralic populations and ignore the non-Uralic matches – that would be a serious bias.

Davidski:
“Ergo, the Stone Cist grave people weren't Germanic speakers, unless they became Germanic speakers without any gene flow from Scandinavia.”

If you can prove that there cannot have been any gene flow from Scandinavia. But you cannot, otherwise you would have presented the evidence. Your believing does not make it true.

Rob said...

@ Anfrze
i was just thinking about Coassacks recently ! It seems like they are frontiermen. My guess is they're a mix of East Slavs & Turkic elements . Would be good to research some medieval Cossack aDNA (and there would be at least a couple of sub-groupings)

Davidski said...

@Jaakko

Show me the evidence that they are very different from the Bronze Age Scandinavian populations.

Every relevant genetic analysis shows clearly that they're the descendants of the local Bronze Age population and have no Scandinavian ancestry.

If you insist on not having a clue and making a fool of yourself then that's your choice.

You cannot just look at the Uralic populations and ignore the non-Uralic matches – that would be a serious bias.

The only difference between the Stone Cist grave people and the Tarand grave people is Siberian ancestry that is also shared predominantly with modern Uralic speakers across North Eurasia.

See that's why the authors of the aDNA paper on these graves correctly concluded that the Tarand grave people were probably Uralic speakers.

Like, you know, there was a genetic shift to a part Siberian, Uralic-like genetic profile, so it makes good sense to say that it was accompanied by a language shift, especially since this fits linguistic and archeological data too.

CrM said...

@Andrzejewski

"Cossacks distinguish themselves from East Slavs and attribute direct heritage and ancestry from Scythians and Sarmatians instead."

It's a LARP. They're Slavs with sometimes a bit of Turkic and Caucasian ancestry. Some even go as far as to claim Scythian ancestry to justify their presence in Krasnodar and mark the surviving Northwest Caucasians as invaders who committed a genocide on the Scythians in their "original homeland".

Speaking of which, all Northwest Caucasians harbor a good amount of Cimmerian and Scythian ancestry, especially Kabardins.

Target: Kabardin:303_R01C01
Distance: 1.9187% / 0.01918749
50.0 RUS_North_Caucasus_MBA:KDC001
21.8 HUN_Prescythian_IA:IR1
18.4 Scythian_UKR:MJ34
9.8 KAZ_Kipchak:DA23

Target: Kabardin:307_R01C02
Distance: 2.2854% / 0.02285385
52.2 RUS_North_Caucasus_MBA:KDC001
41.0 HUN_Prescythian_IA:IR1
6.8 KAZ_Kipchak:DA23

Target: Kabardin:Kab-5
Distance: 2.8668% / 0.02866792
55.2 RUS_North_Caucasus_MBA:KDC001
35.2 Scythian_UKR:MJ34
9.6 KAZ_Kipchak:DA23

Jaakko Häkkinen said...

Anthony Hanken:
“BOO is dated to 3500ybp. This is after Late Proto-Uralic.BOO is dated to 3500ybp. This is after Late Proto-Uralic.”

That is the dating of the remains, yes. That is not the dating of the Siberian gene flow, which must be earlier. Around 1500 BC the closest West Uralic speakers were still in the Upper Volga region.

Anthony Hanken:
“It's not entirely clear where BOO came from. They may have arrived alongside Lozervo ware, this would link them to the Imiyakhtakhskaya culture. The problem is that BOO prefers kra001 over the Imiyakhtakhskaya samples.”

How does this answer to my sentence, that Palaeo-Laplandic languages, which Saami replaced, were not Uralic?

Anthony Hanken:
“Uralic speakers share a specific Siberian component, also found in BOO. Aryan and NWIE speakers do not.”

Also some non-Uralic speakers share that Siberian component. Why do you look only at the Uralic speakers? Then you cannot see anything else than Uralic speakers.

Jaakko Häkkinen said...

Davidski:
“Every relevant genetic analysis shows clearly that they're the descendants of the local Bronze Age population and have no Scandinavian ancestry.”

If that was the case, you would have given me a genetic study which shows that. But there is no such study so far. So, are you claiming that genetic studies are not relevant, but only your own results are relevant?

WHEN THERE IS NO EVIDENCE DISPROVING THE OPTION, YOU CANNOT EXCLUDE THE OPTION. That is how science works.

Davidski:
“The only difference between the Stone Cist grave people and the Tarand grave people is Siberian ancestry that is also shared predominantly with modern Uralic speakers across North Eurasia.”

…and non-Uralic speakers, too, as I already told you. Why do you look only at the Uralic speakers? Then you cannot see anything else than Uralic speakers. Your view is biased.

You still cannot disprove the possibilities that the Siberian component either predated or postdated the spread of the Uralic language. You are only deciding and guessing, which is not a scientific method. When will you understand it?

Davidski:
“See that's why the authors of the aDNA paper on these graves correctly concluded that the Tarand grave people were probably Uralic speakers.”

Guessing a language is only guessing, even if it is made by a geneticist.
We can be quite certain that the latest tarand graves during AD, which also spread to Finland, were connected to the Finnic speakers. But we cannot claim that the earlier tarand graves were also connected to the Finnic speakers, because archaeological or genetic continuity cannot testify for linguistic continuity.

I can repeat this as many years as it takes from you to understand scientific principles.

Davidski:
“Like, you know, there was a genetic shift to a part Siberian, Uralic-like genetic profile, so it makes good sense to say that it was accompanied by a language shift, especially since this fits linguistic and archeological data too.”

Does it, really? Show me that it matches the linguistic results. Here is the linguistic frame:
1. Proto-Uralic in the Volga-Kama area around 2000 BC
2. West Uralic in the Upper Volga area around 1500 BC
3. Pre/Early Proto-Finnic via the Daugava river around 1000 BC
4. Middle Proto-Finnic in (Southern?) Estonia around 300 BC
5. Late Proto-Finnic in Northern Estonia around 200 AD
6. North Finnic in Southern Finland only around 500 AD
7. Old West Finnish, Old Karelian and Old Vepsian around 800 AD

Andrzejewski said...

@CrM “It's a LARP. They're Slavs with sometimes a bit of Turkic and Caucasian ancestry. Some even go as far as to claim Scythian ancestry to justify their presence in Krasnodar and mark the surviving Northwest Caucasians as invaders who committed a genocide on the Scythians in their "original homeland".

Speaking of which, all Northwest Caucasians harbor a good amount of Cimmerian and Scythian ancestry, especially Kabardins.”

Aren’t Northwest Caucasians WSH because of Catacomb and Poltava?
Isn’t it because of these Yamnaya-offshoots that Armenians speak an IE language, rather than the latter one being a “Balkan” derived speech?

Recently someone posted here that Georgians have 15% Yamnaya, 55% CHG and 30%, whereas “Northwest Caucasians would’ve had the highest amount of CHG were it not for the high WSH ratio among them”. However, WSH is roughly 40%-50%. IOW, without Yamnaya or the Scythians the CHG in Kabardians and Adyghe would’ve been actually lower.



Andrzejewski said...

Re: Scythians, Sarmatians and Andronovo:

It recently occurred to me that the term “Aryan” might’ve been an endonym that Andronovo Horizon Culture used to refer to themselves, before branching out. Any proof to that?

I also read some ignorant comment on Wikipedia, claiming that Sintashta was 15% Anatolian; shouldn’t it be at least twice of that, ranging between 30% -40%, like modern Polish and Balts?

The Cimmerian’s locus on the IE tree is still controversial: were they Iranic or were they something else completely (Poltava, Catacomb, Strubnaya)?

Matt said...

Re; Narasimhan paper and Reich group on South Central Asia, kind of agree with others in thread that group are going a bit beyond what they have all the data for (somewhat understandably in a way but might be useful to have more cautions).

I have a lot of reservations about the models of Eneolithic and Copper Age (pre-BMAC) Central Asia even as IranN+WSHG+AnatolianN.

The p values in paper for these models are often very low, and the corresponding G25 model distances are quite high, and there is not much correlation between the paper model and G25. Moreover, the low model fit in G25 is correlated with "WSHG" proportion, which suggests this is the reference that is "off".

When I create a G25 ghost to try and better fit than their models, regression models firstly refuse to generate ghosts with exactly the same proportions as in their models (that ratio of BarcinN:IranN) and then the ghost is quite diverged to WSHG.

(Ghost: https://pastebin.com/k6kNcJWg).

Imgur link: https://imgur.com/a/OxRcqsH

Many times this manifests in G25 models as preference for CHG ancestry, but this may represent something different like Iran_N with less "basal Eurasian" ancestry.

It's also not necessarily likely that the WSHG would be closely or recently related to the HG groups living in South Central Asia (even if this works better as a reference than WHG/EHG, etc).

The Narasimhan paper is a good first draft, but I am skeptical they will really understand the situation unless we can get good adna from Kelteminar Culture skeletons.

J.S. said...

@ Dave

I am French. Recently a friend email you my genome file to get my G25 coordinates.
I must say that the G25 results with vahaduo are extraordinarily accurate and match perfectly my genealogy from different places of France. I'm just amazed.

Congratulations for your great work and thanks a lot!

Copper Axe said...

@Andrzejewski

Uralic languages have several words with an etymological root derived from Aryan so you can assume that the ethnonym was in use during Proto-Indo-Iranian times.

Cimmerians are like first/second generation migrants out of Siberia, bringing the deer stone traditions with them.

They either spoke Iranic, or a non-IE language perhaps related to Yeniseian. The second option is very unlikely and Cimmerian names such as Dugdamme and Sandaksatru seem pretty Iranic.

Copper Axe said...

Speaking of Uralic languages...

@Davidski

Your boy is talking shit on AG:

"Davidski banned me from his blog because I'm too scientific, so we can continue here."

Lmao.

Why is your account there listed as suspended anyways if you dont mind me asking?

Andrzejewski said...

If Poles are 55% WSH: 30% EEF : 15% WHG, then we have twice as much Steppe than Anatolian. Previously, I thought the ratio was almost even (CWC males marrying GAC women).

Nevertheless, the fact that Finns are pre-eminently Indo-European autosomally, on top of the tremendous amount of Steppe in Basques and Etruscans/Villanovans and Northwest Caucasian (and the fact that Semitic people carry a CHG dna but speak an iberomorausain language) shows that the correlation between genetics and language is misleading, and they aren’t always synonymous.

CrM said...

@Andrzejewski

"Aren’t Northwest Caucasians WSH because of Catacomb and Poltava?"
That's Northeast, Dags specifically.

"Recently someone posted here that Georgians have 15% Yamnaya, 55% CHG and 30%,"
Who posted that? The currently available West Georgian samples are about 2% Yamnaya max if you model them with bassal components.
Now, I know a Svan who recently purchased his G25 coordinates, he's very high CHG and scores some Steppe and Turkic on top of it, most likely due to the fact that Svans are mixed with Balkars and Abkhaz.

Target: Svan
Distance: 2.8411% / 0.02841127
60.0 GEO_CHG
24.4 TUR_Barcin_N
7.8 Levant_PPNB
4.6 Yamnaya_RUS_Samara
2.0 RUS_Devils_Gate_Cave_N
1.2 IRN_Wezmeh_N

Original Svans (maybe there are some unmixed Svans left) and certain Rachvels should look like this.

Target: Svan_-30%Balkar
Distance: 3.8408% / 0.03840835
62.2 GEO_CHG
28.0 TUR_Barcin_N
7.2 Levant_PPNB
2.6 IRN_Wezmeh_N
0.0 RUS_Devils_Gate_Cave_N
0.0 Yamnaya_RUS_Samara

Little to no Yamnaya, this likely being the original Proto-Georgian profile.

15% Yamnaya should be possible only among Northeast Georgians who are heavily mixed with Nakh, it's even possible that they're Georgianized Nakh themselves.
Below is a G25 model based on K12b. You can see Yamnaya being above 15%, but the CHG is nowhere near 55%.

Target: Jabushanauri
Distance: 2.3919% / 0.02391886
40.6 GEO_CHG
24.4 TUR_Barcin_N
17.2 Yamnaya_RUS_Samara
10.4 Levant_PPNB
7.2 IRN_Wezmeh_N
0.2 RUS_Devils_Gate_Cave_N


"However, WSH is roughly 40%-50%. IOW, without Yamnaya or the Scythians the CHG in Kabardians and Adyghe would’ve been actually lower."

Once again, 40%-50% is for Dags. This is what a Kabardin would have been like without the Scythian/Cimmerian/Turkic. Quite similar to the Svan_-30%Balkar ghost, but without Iran_N, and with less Anatolia but more Levant.

Target: Kabardin_-Steppe
Distance: 3.8788% / 0.03878788
62.6 GEO_CHG
20.6 TUR_Barcin_N
16.4 Levant_PPNB
0.4 RUS_Devils_Gate_Cave_N
0.0 IRN_Wezmeh_N
0.0 Yamnaya_RUS_Samara

Davidski said...

@Copper Axe

For being a little too honest about Jaakko's IQ.

Ric Hern said...

@ Andrzejewski

If going by what you say that Genes do not always equal Culture then what will you say about this question ?

How can all Cultural finds of supposedly Modern Humans be attributed to Modern Humans if there are no bones available from many sites to study the Morphology/Anatomy of their creators ? Not to mention Genes to clarify the picture ? Then adoption of culture could have taken place without any significant genetic contribution.

Michalis Moriopoulos said...

@Davidski

Yeah, the mods are very strict over there; even I've gotten warnings over this kind of thing and I CO-FOUNDED the place. I understand how frustrating it can be to deal with folks who can't or won't grapple with the preponderance of evidence, but I wouldn't let him tempt you into breaking the rules. It just isn't worth getting perma-banned at AG over this guy (considering what an invaluable resource for community engagement it is), so once your suspension is inevitably revoked I'd just flat-out ignore him. The data clearly favors your account here, and I'm quite confident in the years to come no one will be able to deny a Siberian urheimat, not even "the Uralic Elhaik."

Andrzejewski said...

@Ric Hern “ If going by what you say that Genes do not always equal Culture then what will you say about this question ?

How can all Cultural finds of supposedly Modern Humans be attributed to Modern Humans if there are no bones available from many sites to study the Morphology/Anatomy of their creators ? Not to mention Genes to clarify the picture ? Then adoption of culture could have taken place without any significant genetic contribution.”

It just proves that Anthony 2019 is wrong assuming that Yamnaya spoke an EHG language just because of its y-DNA. We communicate in a language that in all likelihood is a Sredny Stog heir, if not a Progress/Vonyuchka one. PIE dates back only 7,000 years, and I tend to increasingly see it as a language isolate and not as a WHG/ANF/EHG/CHG one.

I’m pretty sure that a CHG-rich population from Piedmont north of Caucasus Mountains migrated into the lower Don valley, created Sredny Stog by admixing with GAC and CTC, and that Progress —> Sredny Stog —> CWC. CWC must be the “PIE” reconstructed by linguists.

Wise dragon said...

@Michalis Moriopoulous,

I beg to disagree.To me geneticists too often use pop. genetics terms way too loosely in interviews, hence confusing people who are not familiar with genetic studies. The point is, that ANF was distinct from Natufians/Levantine Neo no matter that they shared some "components" from thousands of years before the Neolithic. Besides, you seriously think people who want precise language when referring to populations/components that leave no room for misleading and confusing interpretation, are nefarious? Uhm, nope. Actually, people who strongly oppose the sloppy use of population reference are the very opposite of being nefarious. I'm afraid, that researchers who keep on being lazy, continuing using pop. genetic terms too sloppy, do it on purpose, thus to a degree with the intent to mislead. They do that perhaps to counter nationalism. The thing, there is no need to say Northern Europeans are part Ancient Levantine and part North Ancient Eurasians when there is a more accurate way to describe their ancestry. North Europeans are around 50% Yamanya/Steppe, and around 40% Anatolian farmers and the rest WHG.

vAsiSTha said...

I also want to add to my comments re the steppe_en qpAdm model pasted above

details: Georgia_Kotias.SG Russia_MA1_HG.SG -0.007532 -10.815322
details: Russia_HG_Karelia Russia_MA1_HG.SG 0.005029 7.266309
dscore: Russia_MA1_HG.SG f4: -0.001730 Z: -3.215880

The other ancestry with significant generated dstats is with regard to MA1. ie. the EHG + CHG model of Steppe_En harbours much less ANE ancestry than the actual samples.

The region which possesses genetic profile of IranN like + ANE like ancestry in the neolithic is the SC asian region east of caspian, which is why vahaduo picks up Sarazm as an input.

All of this will be much clearer with availability of dzudzuana samples (no clue why they arent made available, they can help differentiate between CHG and IranN), and of course neolithic samples from the east of caspian.

Erik Khan said...

So what exactly do this mean? We know already that a group of Neo Siberians migrated westward from Siberia during the bronze age and went as far as Scandinavia(and their language supposedly survived to the 10th century as paleo-laplandic and paleo-lakelandic), and this new sample is clearly one of the migrating neo siberians at the time, but why it has to be Uralic? none of the modern Neo Siberian populations are Uralic other than the samoyeds.

«Oldest ‹Older   601 – 772 of 772   Newer› Newest»