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
Subscribe to:
Post Comments (Atom)
773 comments:
1 – 200 of 773 Newer› Newest»So, what do you think that this proves?
I've got a question for you Jaska. What's your explanation for these Uralic clines and clusters?
That is, how is it possible for these groups separated by great distances in North Eurasia to share such relationships to the exclusion of most non-Uralic speakers in the region, when the only other plausible link between them is Uralic speech?
This is not a coincidence, so how do you explain it?
Davidski:
“I've got a question for you Jaska. What's your explanation for these Uralic clines and clusters?
That is, how is it possible for these groups separated by great distances in North Eurasia to share such relationships to the exclusion of most non-Uralic speakers in the region, when the only other plausible link between them is Uralic speech?
This is not a coincidence, so how do you explain it?”
There are many clines in many different figures, based on many different methods and softwares.
We should ask: are all the clines seen in all the figures connected to certain language families? What about the clines that connect populations from different language families? Should we according to your logic always assume that they tell about the spread of some ancient language families?
My point is:
There have been many genetic and archaeological waves. Then there has been the spread of the Uralic language family. How can we reliably tell, which of these many genetic waves is connected to this one language family? The answer is: we must take the linguistic results as a starting point – there is no other scientific method to approach the ancient language stages.
I’m not saying that Kra001 could not in any way be connected to Uralic speakers. I’m only saying that we need evidence and arguments, not mere guessing. We need to explain and reconstruct the situation, an in that, we must take all the possible options and all the possibe evidence into consideration. That reconstructive work cannot be done, if linguistic results are ignored.
No, these are not random clines. You cannot dismiss them as artifacts of the methods.
They connect far flung speakers of languages from the same families, and the Uralic cline stands out as one of the most obvious correlations between genes and languages in Eurasia.
Also, ancient DNA tells us that this cline only started to form after 2,000 BCE.
@Jaska
"I’m not saying that Kra001 could not in any way be connected to Uralic speakers. I’m only saying that we need evidence and arguments, not mere guessing. We need to explain and reconstruct the situation, an in that, we must take all the possible options and all the possibe evidence into consideration. That reconstructive work cannot be done, if linguistic results are ignored."
You are the one who completely ignores linguistics. You have no linguistic results, you are not a linguist, you are a fantasist. Everything you write is just a hoax from someone who knows nothing. You can't explain anything, you can't explain how you are asked how and when Samoyeds migrated to Sayan, you can't explain the Uralic-Yukaghir kinship. You just have one fantasy and unsubstantiated set of empty words, you just deny everything at all.
And the fact that it would be Uralic speaking proves the fact that Samoyedians live there.
kra001 Bronze Age Krasnoyarsk Krai 2336 - 2135 calBC XY N-L392*(xVL29,Z1936) N1c1a1a C4b
Uralic lang tree with dates https://i.ibb.co/dGFTVYM/image.png
cmp
irk007 Early Neolithic Cis-Baikal 7030-6693 calBC XY N-L729*(xL666,M46) N1c2b2
brn008 Trans-Baikal Neolithic 5511-5374 calBC XY N-L708*(xL392) N1c1a1
brn003 Trans-Baikal Neolithic 4690-4519 calBC N-L708*(xL392) N1c1a1
Checkmate.
To make it clearer to Jaska, this distance is a multiple of all modern populations. Let him see that there is no Europe, but there are all Siberian Uralians.
Distance to: RUS_Krasnoyarsk_BA:kra001
0.04467404 Nenets:nenets1
0.04495164 Nganassan:Nganassan8
0.04650398 Selkup:Selkup87
0.04711581 Nenets:nenets11
0.04832763 Nenets:nenets13
0.04846535 Nganassan:Nganassan10
0.04917398 Nenets:nenets9
0.04955038 Nganassan:Nganassan15
0.05007494 Nganassan:Nganassan3
0.05018864 Nganassan:Nganassan2
0.05027305 Selkup:Selkup83
0.05208474 Nenets:nenets2
0.05272580 Nenets:nenets5
0.05279290 Selkup:Selkup3a
0.05309548 Nganassan:Nganassan1
0.05317819 Nenets:nenets14
0.05323655 Selkup:Selkup15a
0.05593773 Nganassan:nganassanVo32_1m
0.05616867 Nganassan:Nganassan12
0.05646796 Nenets:nenets15
0.05775768 Nganassan:Nganassan14
0.05808855 Selkup:selkup7
0.05824697 Nenets:nenets10
0.05883502 Nenets:nenets4
0.06000983 Nganassan:Nganassan7
0.06207560 Selkup:selkup2
0.06279705 Selkup:Selkup4a
0.06493335 Selkup:Selkup82
0.06527059 Yukagir_Tundra:Yuk_021
0.06690830 Khanty:759_R01C01
0.06703245 Evenk:GS000014335
0.06759623 Yukagir_Tundra:Yuk_009
0.06768611 Yukagir_Tundra:Yuk_022
0.06888098 Mansi:Mansi91
0.06916401 Khanty:749_R02C02
0.06949964 Khanty:749_R01C02
0.06971191 Yukagir_Tundra:Yuk_011
0.06974647 Yukagir_Tundra:Ckrd18
0.06985707 Evenk:GS000014336
0.07039133 Khanty:751_R01C02
0.07041236 Khanty:776_R01C01
0.07088660 Khanty:761_R02C02
0.07092679 Selkup:Selkup121
0.07105505 Khanty:751_R01C01
0.07120035 Khanty:748_R01C02
0.07178837 Khanty:751_R02C01
0.07189235 Mansi:Mansi48
0.07195374 Khanty:761_R02C01
0.07243093 Mansi:Mansi76
0.07278839 Yukagir_Tundra:Yuk_023
0.07295074 Khanty:749_R01C01
0.07295457 Khanty:750_R01C01
0.07362622 Yukagir_Tundra:Yuk_019
0.07377425 Khanty:748_R01C01
0.07378957 Khanty:799_R02C01
0.07390169 Mansi:Mansi79
0.07396817 Yukagir_Tundra:Yuk_004
0.07436639 Mansi:Mansi56
0.07455273 Khanty:749_R02C01
0.07479104 Khanty:748_R02C02
0.07493330 Khanty:751_R02C02
0.07614250 Mansi:Mansi94
0.07626382 Khanty:750_R02C02
0.07667542 Yukagir_Tundra:Ckrd9
0.07674138 Khanty:776_R02C01
@Davidski: Could you maybe be more specific about the Uralic source populations in the PCA maps? Now it's just a cloud of "Uralic" all over the maps. I would really like to know where the Samoyeds, Ugrics, Saami etc. are on these PCA maps. Also, a marker with the exact position of kra001 would be nice.
You'll have to activate the interactive version to see all the details. Here...
https://vahaduo.github.io/custompca/
So copy paste the contents of the datasheet linked to in the blog post into the PCA DATA field, then go to PCA PLOT and click on PLOT PCA.
Just to ask the question again, does anyone have any preferred explanations for why the Selkup and Nenets samples are so diverse on the main principal component? Long term structure, recent admixture....?
Now you have produced PC maps that show the Ket-Uralic component in all of them. You have not been able to separate ANE and Arctic KRA001 ancestry in any of them. I think that many representatives of Uralic languages here can agree with you on the result that Proto-Uralics carried a combination of ANE and Arctic ancestry.
PC4 is pretty interesting: https://imgur.com/a/2Y9bgjk
Looks like three clines: Finnic-Saami, one from Mordovians to Ob-Ugrics, and one from Ob-Ugrics to Nganassans.
@Davidski: Thanks for the link.
PCA 1/2 and PCA 1/3 basically show what you want to show. There is an Uralic cline representing mixing between a roughly Paleosiberian kind of population and a European kind of population. There is a different Altaic cline representing mixing between a roughly Neosiberian kind of population and a European kind of population.
PCA 1/4 is very interesting. It shows the substructure of the Uralic populations. There is an Old-Uralic cline running roughly from Nganasan to Saami. Then there is a Central Uralic cline that represents mixing between an Ob-Ugric like population and a European kind of population. There is another West Uralic cline that represents mixing between a Saami like population and a European kind of population.
Annotated PCAs:
https://postimg.cc/H80DVHcW
https://postimg.cc/Z06hfP7k
https://postimg.cc/mtRWKcJP
@Archi: "Uralic lang tree with dates"
Glottochronology is not reliable. You put the modern Germanic languages in one of those algorithms and it will tell you that English was the first to branch off from Proto-Germanic. You put the Indo-Aryan languages in there, and it will tell you that Romani (Gypsy) is the first branch. We know that is not right.
The dates of languages that stay within a dialect continuum get underestimated because there is invisible convergence. The dates of languages outside a dialect continuum get overestimated because there is false divergence. See:
http://www.elisanet.fi/alkupera/Hakkinen2012AfterTheProtolanguage.pdf
(And, yes, that paper by Jaakko Häkkinen)
I don't think pca itself means a lot but when you put all the pieces together it looks very likely that Uralics came from Siberia just like the Turkics, Mongolics, Huns and probably the ancestors of Proto-Indo-Europeans and native Americans.
There is an important counter-argument to be made to "Siberians are the original Proto-Uralic speakers". And for that we need to look at the origins of the Samoyeds.
The Nganasans are held up as the original source population of Uralic in "Ancient Fennoscandian genomes reveal origin and spread of Siberian ancestry in Europe" by Lamnidis e.a. However in the anthropological literature they are clearly described as a very recent mixture of Samoyeds, Dolgan Turks, Tungus and Yukaghirs. See "The Nganasan: Wild Reindeer Hunters of the Taimyr Peninsula" by Chester S. Chard.
"Of the three Nganasan subdivisions, the separate Oko clan stems from a Dolgan family and was founded in the early 19th century. It subsequently became assimilated and acculturated. The Vadeev tribe of Nganasan are Samoyedized Tungus- the process taking place in the second half of the 18th century. The Avam tribe was originally (i.e. in the early 17th century) composed of two groups of Samoyedized tundra aborigines and three groups of Tungus- one of the latter migrating in from the south, the other two from the east or southeast of the present Nganasan territory. These Tungus groups may actually have represented Tungusized Palaeo-Asiatics- most probably Yukagir, who are thought to have been the ancient population of the tundra from Taimyr to the Chukchi Peninsula"
The Samoyeds themselves are indistinguishable from Yeniseians in the PCAs. The Yeniseian language family is an isolate, and looks nothing like the other languages of Northern Eurasia. It is most often linked to the Na-Dene languages in Northern America. Thus, I find it more likely that the Samoyeds are Yeniseians who switched their language to Uralic than that Uralic speakers took up the Yeniseian language from 'we don't know who'. And if the Samoyeds were originally Yeniseian-speaking, that leaves us no originally Uralic speaking source population that is deeply Siberian.
@Onno Hovers said
"Glottochronology is not reliable. You put the modern Germanic languages in one of those algorithms and it will tell you that English was the first to branch off from Proto-Germanic. You put the Indo-Aryan languages in there, and it will tell you that Romani (Gypsy) is the first branch. We know that is not right."
This is all amateur reasoning. Glottochronology is as unreliable as any dating based on statistics, TMRCA, radiocarbon dating, or indeed any method of dating. Glottochronology is just lexicostatistics, you have to know how to do it, and not to take up with profane hands what you do not know how to do. This kind of thing is done by people who know nothing and don't know how to do it because they make beginner's mistakes, professionals get it right. There is no other method of dating in linguistics.
It is much more accurate than Jaakko Häkkinen's free flight of unproven fantasy.
"(And, yes, that paper by Jaakko Häkkinen)"
So everything it says is just nonsense. I will not read it, this man just does not know anything.
" The Samoyeds themselves are indistinguishable from Yeniseians in the PCAs. "
It is not so.
"Thus, I find it more likely that the Samoyeds are Yeniseians who switched their language to Uralic than that Uralic speakers took up the Yeniseian language from 'we don't know who'."
This statement is absolutely nothing and completely useless.
@Onno Hovers
There's a serious flaw with your argument in regards to the Nganasans.
You're assuming that just because they are recently admixed, then they're not informative about where the proto-Uralic population lived.
However, note that the Nganasans lack western admixture, so even if they're mixed, then they're only a mixture of far eastern populations.
Indeed, unless Nganasans are just language shifters, then they do have some proto-Uralic ancestry, and thus proto-Uralics must have been a far eastern population.
And the fact that Nganasans are part of the Uralic cline means that the Uralic cline formed as a result of gene flow from the far east right up to Fennoscandia, rather than, say, gene flow from somewhere in between both to the east and west.
Of course, this is also now what ancient DNA tells us, with kra001 also being part of the Uralic cline, showing no western admixture, and belonging to a Y-chromosome haplogroup that is closely related to the main Uralic lineage N-L1026.
Cpk doesn't have the IQ points for these discussions.
He visits here occasionally and writes a bunch of crap every time.
@Davidski
Jaakko Häkkinen's delusions are connected with the fact that he invented that there was a certain mythical Ugro-Samoyedic branch of the Uralic languages instead of the Finno-Ugric languages, hence he has contradictions and mistakes everywhere, everything does not fit.
He is not at all educated in linguistics, he does not at all understand that in glottochronological calculations, borrowed words are excluded from the list of compared words because it is contamination that creates false divergences.
He literally perceives the word "Uraliс" as originating from the Urals, while it is just a very unsuccessful unscientific term, extremely ideologized. Now in science it is proposed to replace it with the neutral strictly scientific term Finno-Samoyedic, as well as the apt term Finno-Ugric and Indo-European.
If the Nganasans have additional east Siberian admixture , then Nenets might be worth a look
The similar lineages are a pointer even if not matching
The alternative scenario of Hg N lineages representing acculturated hunter-gatherers is interesting, but needs further explanation
I’m fact; I don’t think they’re hunter-gatherers by this stage
I loved Nichols article on subject:
https://www.academia.edu/34581649/Vectors_of_language_spread_at_the_central_steppe_periphery_Finno_Ugric_as_catalyst_language
Basically Finno-Ugric tribe as being in right time and right place to be the first forest trade partners to Indo-Iranians which enabled them to expand and assimilate other (fur) hunters “to the left and right” by having that access to II networks.
However according to Napolskich Tocharian also played some role in Seima Turbino (although he allegedly has changed his views on some of Tocharian loanwords), and according to few other scientists Tocharian might have had an Uralic substrate. So who knows who were the true “initiators” of Uralic spread - IndoIranians or Tocharians or both :)
These are good debates on this blog. I'm going to make a video about genetic evidence Uralic languages coming from Asia on my youtube channel. This big news, as people always thought they came from urals.
I'm going to ask you guys for some information because I don't know anything outside of genetics to do with Uralic languages.
This is what it seems to me.....correct me if I'm wrong.
Uralic languages spread in North Eurasia mainly in the 2nd millennium BC.
They would have made it into Europe by that time. Bolshy Oleni, 1500 BC, is example of Uralic speakers in Europe in 2nd millennium BC.
Saami's European ancestry comes from Russian hunter gatherers & Scandinavians (related to Norse but probably pre-Germanic).
Finns' European ancestry originally comes from Baltic Bronze age and other related groups. But in FennoScandinia they mixed with Scandinavians (once again maybe not Germanic speaking).
Udmurts' European ancestry comes from Srubnaya. But of course lots of Russian Uralic(s) have mixed with Slavs.
Mansi in West Siberia then must be....
Proto-Uralic, West Siberian hunter gatherer (mostly ANE), and Srubnaya.
Very interesting. Uralic groups have ancestry from groups we consider to be prehistoric/extinct.
Mansi cluster in the middle of the Uralic cline, and the population that spread N-L1026 came from the eastern end of the Uralic cline where Nganasans cluster.
So if Mansi are the best proxy for proto-Uralics, then N-L1026 isn't a proto-Uralic marker and Nganasans don't have any proto-Uralic ancestry.
I wasn't clear. I wasn't saying Mansi are proto-Uralic, I meant their ancestry is Proto Uralic+West Siberian HG+Srubnaya.
Onno Hovers:
“…And if the Samoyeds were originally Yeniseian-speaking, that leaves us no originally Uralic speaking source population that is deeply Siberian.”
Excellent point!
And thanks for the great excursion on Nganasan history.
Davidski:
“However, note that the Nganasans lack western admixture, so even if they're mixed, then they're only a mixture of far eastern populations. Indeed, unless Nganasans are just language shifters, then they do have some proto-Uralic ancestry, and thus proto-Uralics must have been a far eastern population.
You cannot jump right from Proto-Uralic to Nganasan: Proto-Samoyed is between them. So, the admixture of Siberian populations concerns only Proto-Samoyeds + other Siberians = Nganasans. You can reach Proto-Uralic when you have first reconstructed the genetic history of the 9 Uralic branches’ protopopulations.
Davidski:
“Of course, this is also now what ancient DNA tells us, with kra001 also being part of the Uralic cline, showing no western admixture, and belonging to a Y-chromosome haplogroup that is closely related to the main Uralic lineage N-L1026.”
1. Kra001 could be close to Proto-Samoyedic speakers.
2. How can you compare N-L1026 with the N lineages of Uralic populations? Please present some chronological schedule. Because L1026* is not found in them, and even its subhaplogroups are only found in some of them.
Genos Historia:
“Uralic languages spread in North Eurasia mainly in the 2nd millennium BC.”
Yes.
GH: “They would have made it into Europe by that time. Bolshy Oleni, 1500 BC, is example of Uralic speakers in Europe in 2nd millennium BC.”
No, BOO couldn’t have anything to do with Uralic speakers: neither in its source area nor its target area were Uralic languages anywhere near for 2000 years.
GH: “Saami's European ancestry comes from Russian hunter gatherers & Scandinavians (related to Norse but probably pre-Germanic).”
European ancestry probably was present already in the Proto-Uralic speakers. We don’t know enough yet. You shouldn’t take Davidski’s guesses as truth.
@Jaska
I'm not here to explain the basics to you.
Try reading some recent papers on Uralic specific lineages and learn the different nomenclatures in them.
Here's a good start.
https://eurogenes.blogspot.com/2019/05/more-on-association-between-uralic.html
@Sam
If you're going to make a video about Proto-Uralic, it be good to reconsider where early Uralics first came across Indo-Iranians.
If you think these contacts were around 2000 bc, you can pick all kinds of exciting options. It wouldn't surprise me if we came across an Andronovo burial on the moon, those people were everywhere.
@Copper Axe,
Yep that is definitely something I need to include.
@Jaakko
“ As Carpelan & Parpola 2001 and Petri Kallio 2006 proposed, a step from Volga-Kama region (Proto-Uralic and its eastern dialect) to Altay-Sayan region (Pre-Proto-Samoyedic) correlates nicely with the Seima-Turbino centers in the Kama region and the Altay-Sayan region.”
This is the weakest point of your hypothesis, which nullifies all your other linguistic constructs. Radiocarbon dating clearly points to Siberia as the primary site of ST around 2500 BC. In Europe, the ST appears 500 years later.
About II and Uralic:
“However, some observations can be made. It thus seems that there are Proto-Iranian loanwords which were borrowed simultaneously into several early branches of Uralic, making it likely that Uralic had split into several branches by the time of these contacts. Also the fact that many of the Proto-Indo-Iranian loanwords either show a restricted distribution (such as West Uralic *waćara, *woraći) or irregular correspondences (*asVra, *śasra, *śi̮ta) can point to the conclusion that Proto-Uralic was fragmenting by the time when contacts with Proto-Indo-Iranian took place.
The earlier, Pre-Indo-Iranian loanwords usually show a wider distribution and regular sound correspondences. Although the number of these earliest loans is quite small, based on their distribution and regular correspondences it can be assumed that the Pre-Indo- Iranian stage (after RUKI, *l > *r and the merger of velars and labiovelars but before the
merger of non-high vowels) was concurrent with Proto-Uralic, with the changes leading to Proto-Indo-Iranian happening after the dispersal of Proto-Uralic.
The distribution of loanwords reinforces the old idea that Samoyed is a lexical outlier, as only few convincing Indo-Iranian etymologies for Proto-Uralic words (*saδa- ‘to rain’, *tora- ‘to fight’) have a convincing reflex in Samoyed. However, the fact that such etymologies exist means rather that the situation is due to lexical loss in Samoyed, and that the earliest contact occurred before Samoyed split off from Proto-Uralic.”
INDO-IRANIAN BORROWINGS IN URALIC
CRITICAL OVERVIEW OF THE SOUND SUBSTITUTIONS AND DISTRIBUTION CRITERION
Sampsa Holopainen, 2019
Few notes: at times of Proto-Indo-Iranian Uralic was split already (but not at time of Pre-Proto-II).
Second, also Seima Turbino (according to Napolskich, Tocharian impact) most likely happened after initial spread of Uralic branches. It helped them spread more but was not the proto-Uralic spread.
@Jaska
"We don’t know enough yet. You shouldn’t take Davidski’s guesses as truth."
It is you and you alone who do not know anything. These are not Davidsky's guesses, these are the conclusions of all science, without exception. There are no other opinions.
Sampsa Holopainen, 2019
"However, the fact that such etymologies exist means rather that the situation is due to lexical loss in Samoyed, and that the earliest contact occurred before Samoyed split off from Proto-Uralic.”
An unsubstantiated opinion based on the typical mistakes of the Finns. For example, *tora- ‘to fight’ is not a borrowing, this word is in the Altai languages, but in the Indo-European languages it has a completely different meaning, that time, as in the Iranian languages, does not exist at all.
@Archi,
That is an interesting remark, because she herself wrote on the curious fact that some of presumably Indo-Iranian loanwords into Uralic were not native PIE words but from an unknown substrate language into Indo-Iranian and then Uralic.
But I missed a part where it was proved that it was
"unknown substrate -> PII -> Uralic", not independently
"unknown substrate -> pre PII" and "unknown substrate -> Uralic".
p.s.
What about rain word *saδa- ‘to rain’? Is that also potential substrate word or is that native PIE?
@Parastais
"What about rain word *saδa- ‘to rain’? Is that also potential substrate word or is that native PIE?"
There is no such word in the Indo-European language at all. The derivation of such a banal word belonging to the general vocabulary as rain directly from the Sanskrit śada "falling" is delusional in consonance.
@Onno , interesting point. On that, on genetics Tambets 2018 (https://genomebiology.biomedcentral.com/articles/10.1186/s13059-018-1522-1#MOESM2), took various new Uralic speaking samples, and their Nganasan samples do have some variation in their Fig 3 ADMIXTURE proportions. (Two individuals look recent European admixed, while others combine a low level of Khanty component with a higher level of Siberian East Asian component than is typical for Mongolic and Turkic Siberian populations).
I'm not sure if all the Uralic samples from Tambets 2018 are on Global 25. I was trying to work out from the Human Origins which regions of Russia the Selkup samples on G25 were from (unsuccessfully), and I think there were some Selkups from Tambets listed in the Human Origins file who are not yet on G25. May have a look at their data table later to compare when I have my PC.
@ Davidski@AG
As for the lack of N-L1026 in modern Nganasans (...) one or two anomalies like that are to be expected.
It's not an anomaly. It just means that Nganasans were a part of Proto-Uralic population that did not take part in the founder effect and subsequent migration to the West.
N-L1026 in Uralics can be like I2-Din in South Slavs. The fact that Bosnians have some crazy high level of it doesn't mean that Proto-Slavs in the Slavic homeland had similar levels of it.
Lack of N-L1026 in modern Nganasans actually goes against their supposed origin in the Volga-Kama region or even in the Urals, as Uralic populations that are found there already show the founder effect (so a back-migration is expected to be N-L1026-rich). On the other hand greater Y-DNA diversity is expected in the original homeland.
Re: PCA
Isn't this a linkage disequilibrium on PC4/PC5 (Saami, Selkups)? It looks so. One related to admixture from Kets, other related to either an admixture from some highly drifted source or a bottleneck in Levanluhta or similar pop.
@ Onno
“ And if the Samoyeds were originally Yeniseian-speaking, that leaves us no originally Uralic speaking source population that is deeply Siberian.”
That’s not true. Khanty, Mansi, Nenets are all deeply Siberian
I agree that caution is warranted with regards to using the Nganasan as an indicator about proto Uralic.
I think there really is enough evidence that they are recent language shifters, and thus don't have too much to say about proto Uralic directly--but wait! As I mentioned before the source population of the Nganasan may be genetically close to the source population for the proto Uralic ancestor language.
I was rightly corrected in the other thread that there is no chance Uralic and Yukaghir are closely related language families. Rather, they may share a distant genetic relationship and may both derive from a single Siberian source population. Predominant Yukaghir ancestry in Nganasan may be showing up as the Siberian source of Uralic (Y haplogroups for example) So I think there is a proto Uralic connection possible...but we have to be cautious with the analysis.
@Synome
"I was rightly corrected in the other thread that there is no chance Uralic and Yukaghir are closely related language families."
You were corrected absolutely wrong. The Uralic (Finno-Samoyedic) language has the Yukaghir language as its closest relative, and it is proven. In general, they form the Finno-Yukaghiric superfamily of a higher order. And genes show it too.
@Davidski Do you believe that Armenian language, unlike European ones who stem from Corded Ware/SGC/Bell Beaker, derived from Poltava/Catacomb/Yamnaya instead? Does it explain its variance compared to Euro and South Asian ones?
What’s your opinion on my theory that the very original PIE speech was a language isolate, completely unhinged from either EHG or CHG dialects?
Presumably Early Proto-Uralic consonant gradation, no later than N1a era, is very productive in Nganassan and a source of fascination for scholars. They cannot possibly have developed this if they were originally speaking something else just recently. I do think it likely that their close resemblance to kra001 is partly coincidental, with their being a great chance they've picked up excess Paleosiberian themselves, wherever it's from. I've noticed Pribislav has associated it with Glazkovo and other surges in Q1a2 deriving from Belkachi and related. In any event some of it was probably everywhere for a while. The Yeniseians were obviously around too.
No calling Nganassan consonant gradation spurious. It's quite uncommon in languages and was retained also in Uralic's far northwest. Modern Karelian features a little and it was once more extensive across Finnic and Lapp.
I always doubted Samoyeds as being “true Uralics”; there was something more “Hunter Gathery” about them, can’t really pin my fingers on.
Yukaghir are a true mystery: I suspect that all these Paleo-Siberian groups in the Northeast Asia - Ainu, Nivkh, Itelmam, Kachdals - have strong or even predominant ANE derived AG1-like component that os shared at least partially with Na-Dene and Native Americans.
Yukaghir could’ve spoken a Nivkh-related language originally.
If according to Davidski’s current entry, Proto-Uralics share a cline with the WSHG Ket Yenisseyan speakers, could it mean that it’s deep language and not just its deep ancestry were joined at the hip?
I remember reading Vajda’s paper regarding some strange unknown etymology of up to 25%-30% in the Hungarian language that couldn’t be sorted out. He claims that 20%-25% of Hungarian comes for IE, 25% from Turkic, *only 20% from Uralic*, and the rest from a language Vajda suspected might’ve been Yenisseyan WSHG one. There is a strong possibility to that, however, because the Huns, Xiongnu, Jie and Avars had an overwhelmingly Q1a uniparental y Hap markers for its conquering elite.
On the same vein, I assume that Botai and Okunevo spoke a para-Yenisseyan language, because all these groups share Q1a and WSHG ancestry AG1-offspring sources.
@Onno Hovers
According to Vajda, Yeniseian toponyms indicate an origin in south-central Siberia. Potentially, something like Baikal LNBA could be Proto-Yeniseian. Kets might have migrated northward relatively recently, into formerly Uralic territory.
I'm going to partially agree with Häkkinen on one point here though. Nganassans do not speak a Para-Uralic language, but a Samoyedic one. So if they got their language from admixture with a kra001-type population, then not only Proto-Uralic but also Proto-Samoyedic speakers had to be like that. So why do even Nenets appear to have Steppe MLBA admixture? It could easily be from southern Siberia, but that doesn't matter for Nganassans.
Of course, Steppe MLBA is also a probable source of Indo-Iranian loans in Proto-Uralic, so admixture from them in Proto-Uralians might not be that surprising.
Davidski:
“So if Mansi are the best proxy for proto-Uralics, then N-L1026 isn't a proto-Uralic marker and Nganasans don't have any proto-Uralic ancestry.”
Why do you now ignore the Nganasan-component, which Mansi also have?
Vladimir:
“This is the weakest point of your hypothesis, which nullifies all your other linguistic constructs. Radiocarbon dating clearly points to Siberia as the primary site of ST around 2500 BC. In Europe, the ST appears 500 years later.”
How is that relevant? Of course there was bronze manufacturing in the two centres, in Altay and in Europe, before they unified to one network. If anything, Proto-Uralic matches the Turbino centre on the Kama, and subsequently the unified wide network, which of course could have been multiethnic.
Parastais:
“Few notes: at times of Proto-Indo-Iranian Uralic was split already (but not at time of Pre-Proto-II).
This is possible yet far from proven. The evidence from different Aryan loanword layers is prone to interpretations: what is the weight of regular sound correspondences and what is the weight of wide distribution?
Parastais:
“Second, also Seima Turbino (according to Napolskich, Tocharian impact) most likely happened after initial spread of Uralic branches. It helped them spread more but was not the proto-Uralic spread.”
See my previous comment to Vladimir above.
Parastais:
“Lack of N-L1026 in modern Nganasans actually goes against their supposed origin in the Volga-Kama region or even in the Urals, as Uralic populations that are found there already show the founder effect (so a back-migration is expected to be N-L1026-rich).”
As we know, we cannot derive Nganasan straight from the Volga-Kama region. Its immediate ancestor was Proto-Samoyedic, spoken in the Altay-Sayan region. N3-lineages are not common in Samoyeds, except in the westernmost ones: Nenets. N2-lineages correlate much better with Proto-Samoyedic.
Davidski:
"I'm not here to explain the basics to you.
Try reading some recent papers on Uralic specific lineages and learn the different nomenclatures in them."
Thanks, but I'm well aware about these things: I started following genetic studies during the time when N-tat was still called Hg16 and all other N's were Hg12.
If you have any real arguments, you should probably learn how to write them down. Nobody can read your mind, therefore writing is the only way to communicate your point.
@Rob
"That’s not true. Khanty, Mansi, Nenets are all deeply Siberian"
That is not so simple, at least for the Mansi. If you look at Davidsky's PCA you will see that the Mansi have two distinct subgroups. One that plots near Kets and Nenets and another that plots near to European groups. This second group is much closer to Udmurts than to the other Mansi group.
Actually their genetics might reflect their history in this case. Some Mansi "clans" have horse cult and they share some horse and riding related vocabulary with Hungarians. But they live in an area that is completely outside of the "horse-zone". It seems pretty likely that the Mansi have partial ancestry from former forest-steppe people who were pushed northward. And it possibly happened relatively recently, hence the genetic disparity.
I'd be interested to know if the genotypes underlying Davidski's PCAs were filtered to remove degree 3 (and maybe 4) relatives. Does Davidski describe a general methodology in any post?
@Rob
More about the Mansi. I run some quick nMontes, only moderns. I picked one genetically very Siberan Mansi (Mansi85) and one less Siberian (Mansi25) as targets. I used the dataset Davidski made for the current post.
Target Distance Bashkir Khanty Komi
Mansi:Mansi85 0.00569254 3.8 88.8 7.4
Mansi:Mansi25 0.00936770 33.0 27.4 39.6
The Mansi main group (main by the number of samples) can be well modelled as a more or less even mixture of Komi, Bashkir and Khanti. The Komi are much more European. The Bashkir have IBD sharing with the Mansi according to earlies studies (probably because they absorbed the Ugric population of the South Ural).
The second group is basically Khanty with some admixture from the first group. My guess is that they are indeed descendants of assimilated Khanty.
@Archi
Uralic connects to Yukaghir, and both connect to Altaic, which in turn all connect to the various Paleo-Siberian languages. Linguistic models can only get you so far back in the past, so the best bet is using genomic ancestry to see how they developed via diverging ancestral lines united by a distant common ancestor. The Uralic question is just scratching the surface of real deep interrelations.
Samoyedic Nenets
Nenets
RUS_Krasnoyarsk_BA:kra001 66.9%
WSGH 14.6%
Bell_Beaker_Bavaria:E09569 8.8%
KAZ_Katon_Karagay_LBA 7.1%
Baltic_EST_BA 2.6%
AmurBasin_N 0%
RUS_Shamanka_EBA:DA334 0%
Baltic_CCC 0%
RUS_Kolyma_Meso:Kolyma_River 0%
USA_Anzick:Anzick 0%
Ugric Khanty
Khanty
RUS_Krasnoyarsk_BA:kra001 46.8%
WSGH 25.2%
KAZ_Katon_Karagay_LBA 13.5%
Bell_Beaker_Bavaria:E09569 8.4%
Baltic_EST_BA 6.1%
AmurBasin_N 0%
RUS_Shamanka_EBA:DA334 0%
Baltic_CCC 0%
RUS_Kolyma_Meso:Kolyma_River 0%
USA_Anzick:Anzick 0%
Distance 4.8749%
It seems a lot of Finns are butthurt with the recent uralic genome :)))
@Davidsky: The arguments you use for "Uralic from Siberia" remind me of the arguments used for "IE from Anatolia". How is "No western admixture in Nganasans" different from "No steppe admixture in Hittites"? How is "Siberian admixture in Uralic speakers" different from "Caucasian/Iranian admixture in Indo-Europeans"?
@KM
I'd be interested to know if the genotypes underlying Davidski's PCAs were filtered to remove degree 3 (and maybe 4) relatives. Does Davidski describe a general methodology in any post?
I posted the results for the first ten dimensions. If there were any duplicates or obvious relatives in the dataset I used then you would see them creating isolated pairs and/or groups in most of the higher dimensions.
So if you're suggesting that any of the key features of my PCA, like the Uralic clines, are artifacts of my methods, then you're out of luck.
@Onno Hovers
How is "No western admixture in Nganasans" different from "No steppe admixture in Hittites"? How is "Siberian admixture in Uralic speakers" different from "Caucasian/Iranian admixture in Indo-Europeans"?
In fact, there was gene flow from Eastern Europe into Bronze Age Anatolia, so this can plausibly explain the appearance of Hittites and other Anatolian speakers there. The problem was a lack of the relevant ancient samples, but they're on the way now and should be published this year.
And there is no Indo-European cline that starts in the Caucasus or Iran, like the Uralic cline that starts in Siberia.
Also, the two most obvious Indo-European paternal markers, R1a-M417 and R1b-M269 are both from Eastern Europe, while the obvious Uralic paternal marker, N-L1026, is from Siberia.
@ Andrze
RE : language isolates
I think PIE, like FU, is most likely to have been a “hunter gatherer language”’, with PIE from Europe rather than from Anatolian farmers or Majkop-related civilizations
Labels like CHG, WhG, EHG are almost irrelevant here , because there would not have been a confluent cline of EHG or CHG related languages
They would have been disjointed and often unrelated. This is what the Amerindian and Australian aborigine languages show
That is why concepts like Indo-Uralic & Nostratic are problematic; they’re constructs of linguists who have little concept about sociology and anthropology; trying to build a castle from twigs
This also invalidates the argument that Uralic can’t be from Siberia .. because Yukaghir
@Onno Hovers
"The arguments you use for "Uralic from Siberia" remind me of the arguments used for "IE from Anatolia". How is "No western admixture in Nganasans" different from "No steppe admixture in Hittites"?"
The fact that no single Hittite has been tested at all.
"How is "Siberian admixture in Uralic speakers" different from "Caucasian/Iranian admixture in Indo-Europeans"?"
The fact that the Eastern European steppe inhabitants are in no way connected with the Caucasus and Iran either by haplogroups, or archaeologically, or anthropologically, or culturally, linguistically, or lexically. And the Finno-Ugrians are very strongly connected by all these parameters with Siberia.
I do not understand this at all, what kind of Eurocentrism and Siberia hating, as if Siberia is some kind of monkey on the branches. Siberia is home to the earliest Neolithic, so be proud that you belong to it. There is no need to feel an inferiority complex in front of the Indo-Europeans, they are also from Siberia from the same places as the Finno-Samoyedians, only they left it not in the late Neolithic, but in the early Mesolithic, that's the whole difference.
@Onno Hovers,
You haven't stated an opinion, it seems you're trying to be the reasonable skeptic.
Y DNA N-L0126 is really the nail in the coffin.
Pre-Historic Europe is covered really well in ancient DNA. We can confidently say Y DNA N1c didn't exist in Europe before 2nd millennium BC.
Then the fact, the oldest N-L0126 is in an individual with 100% Asian ancestry basically proves N-L0126 in European Uralics originated in a population deep in Asia with no West Siberian or European ancestry.
Assuming N-L0126 and proto-Uralic are the same, then the only option is to say Uralic is from a purely Asian population.
The only way to say otherwise would be to say, pre-Uralic was Asian but proto-Uralic migrated to Urals and was part West Siberian & European.
@ SLumberry
Yes fine detailed understanding of the ethngrphic histories of these groups is required.
Can you clarify further what you infer about the relationship between Khanty & Mansi ?
E. Donovan:
“Presumably Early Proto-Uralic consonant gradation, no later than N1a era, is very productive in Nganassan and a source of fascination for scholars. They cannot possibly have developed this if they were originally speaking something else just recently.”
Consonant gradation is not considered as Proto-Uralic. Even Saami and Finnic have it in only some of the languages, and it cannot have old roots because its conditions are different in Saami and Finnic. Such a phenomenon can occur independently, especially in languages with strong, fixed initial stress (like in Uralic languages).
Language shift in North Siberia does not make gradation impossible, because in any case the language has equally deep roots: in Proto-Uralic. Spread of every language happens partly by population expansion and partly by language shift of original inhabitants. Gradation in Nganasan developed after Proto-Samoyedic.
Erik Andersson:
“According to Vajda, Yeniseian toponyms indicate an origin in south-central Siberia. Potentially, something like Baikal LNBA could be Proto-Yeniseian. Kets might have migrated northward relatively recently, into formerly Uralic territory.”
Well, it is usually seen happening in another order: Samoyedic languages have replaced Yeniseian languages (seen in linguistic substrate), and in the process acquired Yeniseian genes, like Y-haplogroup Q in Selkups. Before the Samoyedic expansion around 2000 years ago there are no traces of Uralic-related languages in North Siberia. That is the reason why BOO could not be connected to the Uralic languages: the migration left from the area with no Uralic presence yet until 2000 years later.
Also Yeniseian, Turkic, Tungusic and Mongolic expansions are dated around or less than 2000 years in past. Before that most of Siberia – especially in the north – was occupied by various Palaeo-Siberian language families, mostly disappeared since then. Yukaghir, nivkh and Eskimo-Aleut can be remains of these older variation.
Slumbery:
“The second group is basically Khanty with some admixture from the first group. My guess is that they are indeed descendants of assimilated Khanty.”
This is interesting. Tambets et al. 2018 also found the great diversity in the Mansi, one group even being close to the Mari in the Middle Volga.
Mansi placenames are found in North Russia quite west, around Mezen river. Spreading to Siberia the Mansi probably assimilated the westernmost Khanty, adding their diversity. There are two exogamous fratrias within Mansi and Khanty: mos’ and por. The former is a possible Khanty cognate for the ethnonym Mansi ~ Magyar/Megyer ‘Hungarian’, while the latter is a possible assimilated Palaeo-Siberian “clan”.
Leron:
“Uralic connects to Yukaghir, and both connect to Altaic, which in turn all connect to the various Paleo-Siberian languages. Linguistic models can only get you so far back in the past, so the best bet is using genomic ancestry to see how they developed via diverging ancestral lines united by a distant common ancestor. The Uralic question is just scratching the surface of real deep interrelations.”
It is true that linguistics has its temporal limitations (but see the Dene-Yeniseian hypothesis). Still, there is no scientific method to trace language from the genes. If you take all the different maternal and paternal haplogroups of certain ancient population and you follow them back in time, mutation by mutation, the whole population disperses and lacks coherent history - most of the lineages originate in very different directions. So, how can you guess which of these lineages is connected to certain language? You can’t.
JohnnyOla:
“It seems a lot of Finns are butthurt with the recent uralic genome :)))”
You have misunderstood something. It is a matter of whether we stick with the scientific methods or we let the unscientific guesses and fantasies overtake our reasoning. In the possible connection itself there is nothing to reject – only in the method on which it is based.
Onno Hovers:
“@Davidsky: The arguments you use for "Uralic from Siberia" remind me of the arguments used for "IE from Anatolia". How is "No western admixture in Nganasans" different from "No steppe admixture in Hittites"? How is "Siberian admixture in Uralic speakers" different from "Caucasian/Iranian admixture in Indo-Europeans"?”
Once more an excellent point!
Genos Historia:
“Then the fact, the oldest N-L0126 is in an individual with 100% Asian ancestry basically proves N-L0126 in European Uralics originated in a population deep in Asia with no West Siberian or European ancestry.”
It is L1026.
You must remember that there can be no true connection between uniparental lineages and autosomal components: they are totally independent levels. All correlations are only momentaneous. Autosomal DNA changes in every generation – even strongly, if admixed with very different population.
Genos Historia:
“Assuming N-L0126 and proto-Uralic are the same, then the only option is to say Uralic is from a purely Asian population.”
How could we assume that? L1026* is only found rarely here and there, and none of its subhaplogroups has total coverage in Uralic populations. And even if L1026 was connected to the Uralic language, there is no way to tie it to any autosomal component: all correlations are only momentaneous. In different places and different times L1026 connects to different autosomal components – as you can see already by looking at the autosomal composition of modern populations with this lineage.
All instances of N-L1026 and closely related lineages will be found in East Eurasian samples like kra001 before 2,000 BCE.
That's what Genos Historia was referring to.
@ Archi
''Siberia is home to the earliest Neolithic, so be proud that you belong to it. There is no need to feel an inferiority complex in front of the Indo-Europeans, they are also from Siberia from the same places as the Finno-Samoyedians, only they left it not in the late Neolithic, but in the early Mesolithic, that's the whole difference.''
it is a beautiful place, but i think you are over enthusiastic about it
E.g. the Neolithic, as the rest of the world understands it came from the Fertile Crescent, not Siberia. You have to be aware that Russian schoalrs made up their own definitiion of Neolithic - hunter-gatherer pottery amongst fishers & hunters.
Moreover, your claim that PIE came from Mesolithic Siberia is rather funny, given that when PIE was developing in Europe, nothing remotely Indo-European existed in Siberia. Y-hg R1 might be from West Siberia in the distant Paleolithic, but even that is not clear. In any case Mal'ta has nothing directly to do with PIE
@Matt
I'm not sure where the Selkup samples used in Tambets came from, but I can tell you after looking into it that the big new batch of Selkups from Karafet (converted to G25 by David not too long ago) are all from Yamalo-Nenets, so they are clearly Northern Selkup (Taz variety). In fact, pretty much all the Uralic groups in that study were sampled in Yamalo-Nenets, including the Komi (which is probably why you get some Og-Ugric and Nenets profile outliers among them).
Karafet data:
https://www.ncbi.nlm.nih.gov/geo/browse/?view=samples&display=500&series=73996&zsort=date
And here are the G25 coords:
https://drive.google.com/file/d/15cmDDfKUW24rD0lm7YuLMqopeDhqOftV/view
@Jaakko
“ Before the Samoyedic expansion around 2000 years ago there are no traces of Uralic-related languages in North Siberia.”
The Samoyed expansion into the taiga and Arctic zone was indeed late. But what was in the forest-steppe and southern taiga zone, we can not yet say. Personally, I think this area was populated by an N1a population with an MBA, at least. And if one of the N1a subclades is found in the Neolithic Baraba culture and the Eneolithic Ust-Tartas culture, then even earlier.
@Rob
“ Moreover, your claim that PIE came from Mesolithic Siberia is rather funny, given that when PIE was developing in Europe, nothing remotely Indo-European existed in Siberia. Y-hg R1 might be from West Siberia in the distant Paleolithic, but even that is not clear. In any case Mal'ta has nothing directly to do with PIE”
Moreover, your claim that PIE came from Mesolithic Siberia is rather funny, given that when PIE was developing in Europe, nothing remotely Indo-European existed in Siberia. Y-hg R1 might be from West Siberia in the distant Paleolithic, but even that is not clear. In any case Mal'ta has nothing directly to do with PIE
“ it is a beautiful place, but i think you are over enthusiastic about it
E.g. the Neolithic, as the rest of the world understands it came from the Fertile Crescent, not Siberia. You have to be aware that Russian schoalrs made up their own definitiion of Neolithic - hunter-gatherer pottery amongst fishers & hunters.”
Whether the Neolithic is determined by the producing farm or by the pottery is immaterial, an academic dispute. But the fact that ceramics on the Amur River appeared earlier than anywhere else west of the Altai is obvious. And it is quite possible that it appeared on the Amur River in the N1a-TAT population.
@Jaska,
“ This is possible yet far from proven. The evidence from different Aryan loanword layers is prone to interpretations: what is the weight of regular sound correspondences and what is the weight of wide distribution? ”
Banana vs Wheel. One has wide distribution (all IE branches) but irregular correspondences. One has regular correspondence but not present in all branches (i.e. Baltic). Which one comes from PIE stage?
@ Vlad
''Whether the Neolithic is determined by the producing farm or by the pottery is immaterial, an academic dispute. But the fact that ceramics on the Amur River appeared earlier than anywhere else west of the Altai is obvious. And it is quite possible that it appeared on the Amur River in the N1a-TAT population.''
I agree although the point is specification is required. Even this is not 100%; as some have suggested the earliest pottery is in Southern China
@Rob
"You have to be aware that Russian schoalrs made up their own definitiion of Neolithic - hunter-gatherer pottery amongst fishers & hunters."
This is not a Russian definition, this is the original definition of the Neolithic, which existed until the second half of the 20th century everywhere and for everyone. It is in Europe in the second half of the 20th century that the definition was changed. The concept of farmers was not related to the concept of the Neolithic, the definition of archaeological eras is not associated with economic formations, this is the concept of a technocomplex.
Parastais:
“Banana vs Wheel. One has wide distribution (all IE branches) but irregular correspondences. One has regular correspondence but not present in all branches (i.e. Baltic). Which one comes from PIE stage?”
Exactly.
When considering intra-family cognates, historical phonology (regular sound changes) is the most important argument. A word can be reconstructed to the protolanguage, if the cognates cannot be explained by later spread.
With inter-family cognates it gets a bit more complicated. There are words which from phonological view could represent different stages: they include no “marker sounds” matching the daughter branches of the protolanguage. With these it is tempting to think that wider distribution = older word, but it may be illusory – just like it often is with the intra-family words.
@Rob
"I agree although the point is specification is required. Even this is not 100%; as some have suggested the earliest pottery is in Southern China"
Yes, and it is quite possible that the dates of the collapse of N1 into subclades N1a and N1b of 18000 BP and the appearance of ceramics on the Amur River with 16000 BP and the discovery of the oldest N1b in China, and the oldest N1a near the Amur River are not accidental.
@Jaska,
“With inter-family cognates it gets a bit more complicated. There are words which from phonological view could represent different stages: they include no “marker sounds” matching the daughter branches of the protolanguage. With these it is tempting to think that wider distribution = older word, but it may be illusory – just like it often is with the intra-family words.”
We are not in disagreement here. Apparently pre-Proto-II loanwords have regular sound changes (or say there are loanwords from this stage that do not contradict what regular sound changes might look like), but proto-II loanwords are already irregular, so entered PU already after split.
@ Vlad
I think so. On the other side, the subsequent diminution of Hg N from northern China then the Baikal region, with subsequent finds to the north and northwest implies an onward trajectory.
Theis study from 2006 weren't too far off
https://www.nature.com/articles/5201748
@ Archi
''This is not a Russian definition, this is the original definition of the Neolithic, which existed until the second half of the 20th century everywhere and for everyone''
Im not sure which was first or original definition. But it reflects the different methologies - one socio-cultural the other on the mode of production
But I agree with what you say, it might appear that models expressed here might be 'sidelining' Uralic prehistory but in fact is empowering what have become ethnic minorities within the former USSR. It shows they had pretty complex and expansive histories.
Vladimir wrote,
"Yes, and it is quite possible that the dates of the collapse of N1 into subclades N1a and N1b of 18000 BP and the appearance of ceramics on the Amur River with 16000 BP and the discovery of the oldest N1b in China, and the oldest N1a near the Amur River are not accidental."
Which specimen from an area near the Amur River has produced the oldest evidence of Y-DNA haplogroup N1a?
Specimen brn003 from the Dvorcy-Dacha burial (Kadalinka river) has been dated to 6640-6469 Cal BP and belongs to mtDNA haplogroup D4.
Specimen M54A from Houtaomuga (lower Nen River) has been dated to 7430-7320 ybp and belongs to mtDNA haplogroup B4c1a2.
Specimen brn008 from the Izvestkovaja-1 site, burial 1 (Kuenga river, Sretensky District) has been dated to 7461-7324 Cal BP and belongs to mtDNA haplogroup D4.
Specimen irk007/Popovskij-1 from the Popovskij Lug site (upper Lena River) has been dated to 8980-8643 Cal BP and belongs to mtDNA haplogroup F1b.
Every one of the aforementioned specimens belongs to Y-DNA haplogroup N1a-L729.
@Rob
"Im not sure which was first or original definition. But it reflects the different methologies - one socio-cultural the other on the mode of production"
Well, because you are uneducated in archeology. Neolithic is a concept of the 19th century, and for the first time that at this time there was the emergence of a producing economy, that is, farmers appeared, Gordon Childe expressed, and being a Marxist he called it the "Neolithic revolution" by analogy with the "October Revolution" by the time when these revolutions occurred leading to a change in economic formations. And after him the word "revolution" was dropped, so a new concept of the Neolithic was created.
And only later it was found out that the emergence of a producing economy has nothing to do with the time of the Neolithic, it appeared everywhere in different eras with different technocomplexes.
-lithic this is concept of technocomplexes, as well as the Copper Age, Bronze Age, Iron Age. The economic-productive relationship has nothing to do with it.
@ Archie
“ Neolithic is a concept of the 19th century, and for the first time that at this time there was the emergence of a producing economy, that is, farmers appeared, Gordon Childe”
Right, so the earliest definition was that (of an Australian) within a European framework. Thanks for clearing that up
@Jaakko Häkkinen
"Well, it is usually seen happening in another order: Samoyedic languages have replaced Yeniseian languages (seen in linguistic substrate), and in the process acquired Yeniseian genes, like Y-haplogroup Q in Selkups."
Selkups do appear to have genuine Ket admixture, but I doubt that the ancestry shared between Kets, Nenets and Ob-Ugrians is Yeniseian. Whilst Samoyedic may well have replaced Yeniseian in parts of southern Siberia, the northward migrations of Kets seem to post-date the Samoyedic expansion by a lot.
Georg, 2003: "While the northern boundary of Ket roughly coincided with the river Yeloguy and the settlement of Verchneimbatsk in the 17th century (where the Ket dialect spoken today is called Southern Ket), Ket speakers advanced as far as the river Kurejka (a right tributary of the Yenissey, north of Turuchansk), some 300 km north of Verchneimbatsk and well beyond the polar circle, during the following centuries." ( https://www.academia.edu/1629984/THE_GRADUAL_DISAPPEARANCE_OF_A_EURASIAN_LANGUAGE_FAMILY )
@Archi: "There is no need to feel an inferiority complex in front of the Indo-Europeans..."
Where do you think I'm from?
@Genos: "You haven't stated an opinion, it seems you're trying to be the reasonable skeptic."
Questioning motives are we? The ad-hominem is strong on this forum.
I don't doubt that there was a massive westward migration from Siberia that introduced N Y-DNA and deep Siberian autosomal DNA into West-Siberia and Northern Europe. The part that I am skeptic about is the conclusion that "The Uralic languages come from Siberia".
There are deep linguistic affinities that tie Uralic to Indo-European. I don't think it is possible to explain this with either 'late loans' or with the very early ANE ancestry present in EHG. But given the quality of the discourse about linguistics here, I don't think this is an appropriate forum to discuss all of that.
@Genos Historia
Saami people have some old SHG/EHG-shifted European ancestry but also more recent (likely Viking Age) European admixture from Scandinavians (Norse). We know that typically Germanic haplogroups entered the Saami gene pool relatively late. Their Norse admixture is not pre-Germanic, as Norse is Germanic.
Among Finns, there is ancestry from the Estonian Iron Age as well as from Baltic Bronze Age. In addition to that there is Germanic admixture in Finns, probably dating to the Iron Age. A good proxy for this kind of ancestry is SWE_IA:RISE174, which is a great overall proxy for Germanic ancestry. In the Finnish Iron Age, in Western Finland, Germanic (likely) stone circles known as Käräjäkivet kan be found. They are very similar to the Germanic stone circles from Iron Age Scandinavia and Iron Age Poland (Wielbark culture for example).
@Davidski
You're right, that was the reason I asked about relatives. Obvious things like first-degree kin do tend to appear in a PCA as very obvious floating pairs in my experience, but I myself have been caught out a few times by networks of more distant kinship. What's happened (I'm pretty sure) is that even after I've removed all the 1st and 2nd degree kin, there are a few clan-like groups in the data which have a lot of distant, 3rd+ degree connections, and this very recent structure is pointlessly pulled out and drags a few groups along with it. The most obvious example I can recall are the PCAs I ran on a couple thousand Iranians. Good chance that I'm teaching my grandma to suck eggs, so apologies if that's the case. The pops here are probably distant enough that any clan structure would be basically orthogonal to the rest of the variation anyhow. Not trying to come up with excuses to prop up a pet theory!
Another part of the reason I asked is that your PCAs look really good and I wanted to get some tips! They're well resolved, and clines in the lower PCs are surprisingly interpretable compared to ones I typically generate. But I totally get if you want to keep it a trade secret and part of the blog's USP.
@ Davidski
Genotypes from Harney et al. 2021 (Füzesabony, LaTene, Glăvănești etc.) are published now on Reich's Lab page.
@Rob PIE can not be a WHG language period! All dialects spoken by WSH ancestry people across the Eastern Euro Steppes were para-PIE, which intrinsically contradicts your theory.
@Rob
"Can you clarify further what you infer about the relationship between Khanty & Mansi ?"
I am not entirely sure what you are asking from me here. I have some speculation. Basically the Khanty are the older group in the currently shared region and a part of them got assimilated into the Mansi when the Mansi moved into that territory. Things happened (Tatars, Russians) and now their remnants only live in areas that once were peripheric for them.
@Jaakko Häkkinen
What I read about the topic is pretty much the same in Hungarian. The Hungarian article also adds that "mos" have horse cult and the "por" have bear cult. Note however that the Khanty also have the same two phratries, but in their case there do not seem to be such a genetic variety.
However I think that the Mansi are originally (in a bigger time depth) from the south and not from the west. (By south I do not mean super south, just south from where they live now.) They have genetic links (IBD sharing) with Bashkirs and common paternal lineages (more specific that just N-TAT). It is because the Bashkirs assimilated the Uralic population of their current homeland and that population was related to the Mansi.
The common horse vocabulary of Ugric languages also direct us toward this. The proto language probably formed in the forest-steppe. (And note that this vocabulary is not shared with Samoyedic groups.)
@Heyerdahl Saami people have some old SHG/EHG-shifted European ancestry but also more recent (likely Viking Age) European admixture from Scandinavians (Norse). We know that typically Germanic haplogroups entered the Saami gene pool relatively late. Their Norse admixture is not pre-Germanic, as Norse is Germanic.
Among Finns, there is ancestry from the Estonian Iron Age as well as from Baltic Bronze Age.”
Finns have 50% WSH. The are “Indo-Europeans” in all but language.
@Onno Hovers
"The part that I am skeptic about is the conclusion that "The Uralic languages come from Siberia"."
This conclusion was reached by many different scientists in different sciences in the decades before the advent of genetic research. This conclusion was made independently by each of the natural sciences, such as linguistics, archeology, anthropology, cultural studies, genetics only independently confirmed it and nothing else. Generally speaking, there are no sciences left that would not confirm it.
@Arza
Thanks, I'll be out all day today but I'll run those samples in the evening.
@Andrzejewski
Clearly. Estonian Iron Age and EST Bronze Age has a lot of WSH. So does SWE_IA.
@ Slumberry
“ I am not entirely sure what you are asking from me here. I have some speculation. Basically the Khanty are the older group in the currently shared region “
That’s certainly interesting but I guess it doesn’t change that it is not only Nganasans which have deep Siberian ancestry
It would be strange if all the I1 in Finns was Germanic considering they don't have the corresponding amounts of R1a/b you would expect from a Norse group.
Parastais:
“We are not in disagreement here. Apparently pre-Proto-II loanwords have regular sound changes (or say there are loanwords from this stage that do not contradict what regular sound changes might look like), but proto-II loanwords are already irregular, so entered PU already after split.”
There are actually both kinds in both groups. That’s why I disagree with views that there would be clear difference between these layers.
Sampsa Holopainen 2019, page 343:
“It thus seems that there are Proto-Iranian loanwords which were borrowed simultaneously into several early branches of Uralic, making it likely that Uralic had split into several branches by the time of these contacts. Also the fact that many of the Proto-Indo-Iranian loanwords either show a restricted distribution (such as West Uralic *waćara, *woraći) or irregular correspondences (*asVra, *śasra, *śi̮ta) can point to the conclusion that Proto-Uralic was fragmenting by the time when contacts with Proto-Indo-Iranian took place.”
So, some of the Proto-Aryan loanwords are still regular.
“The earlier, Pre-Indo-Iranian loanwords usually show a wider distribution and regular
sound correspondences.”
So, not all of them are regular. Actually only few of even the Proto-Uralic inherited words are such that no irregularities appeared in any of the nine Uralic branches. Therefore the irregularity in some branches cannot be seen as a conclusive argument against the Proto-Uralic status of the word.
What we can say is: during the Proto-Iranian stage Proto-Uralic was already disintegrated, but during the Late Proto-Aryan stage it was still mostly a coherent protolanguage, as it was during the pre-stages. It is possible that at the very end of the Late Proto-Aryan stage Proto-Uralic was developing in to different dialects.
@Jaakko Häkkinen
Why does there need to be several migration waves of Siberians to Europe from deep in Siberia to explain BOO? Seems excessively complicated.
Erik Andersson:
“Selkups do appear to have genuine Ket admixture, but I doubt that the ancestry shared between Kets, Nenets and Ob-Ugrians is Yeniseian. Whilst Samoyedic may well have replaced Yeniseian in parts of southern Siberia, the northward migrations of Kets seem to post-date the Samoyedic expansion by a lot.”
Yes, good point: there was also secondary expansions like that of Ket to the north.
Slumbery:
“However I think that the Mansi are originally (in a bigger time depth) from the south and not from the west. (By south I do not mean super south, just south from where they live now.) They have genetic links (IBD sharing) with Bashkirs and common paternal lineages (more specific that just N-TAT). It is because the Bashkirs assimilated the Uralic population of their current homeland and that population was related to the Mansi.
The common horse vocabulary of Ugric languages also direct us toward this. The proto language probably formed in the forest-steppe. (And note that this vocabulary is not shared with Samoyedic groups.)”
Yes, this view still stands.
On Non-Uralic substrate in Sami: https://www.academia.edu/4811770/An_Essay_on_Substrate_Studies_and_the_Origin_of_Saami
This is what a WHG or an EHG must’ve sounded like:
Kildin Sami Northern Sami English
kut’t’k kut’t’k heart
nirr nierru cheek
čad’z’ čáhci water
vuntas sand
poav’n hummock
k’ed’d’k geađgi stone
abbr’ arvi rain
piŋŋk biegga wind
ket’t’k’ geatki wolverine
nigkeš pike (fish)
murr muorra tree
cigk mist
mun
@ Andrzejewski
''PIE can not be a WHG language period! All dialects spoken by WSH ancestry people across the Eastern Euro Steppes were para-PIE, which intrinsically contradicts your theory.''
Well nobody said PIE emerged in France but it might very well have developed in WHG-enriched hunter gatherer societies
But that wasn’t even my point, which is - ethnographic data shows that pre-Neolithic societies often had divergent languages. So we can observe genetic clines where adjacent population were genetically related, but the linguistic diversity was not similarly clinal. Hence my point was notions like EHG & CHG languages- which you seem to obsess about, are dubious
But over your head
Expansions, which we can say spread Indo Europeans, so far have been only linked to populations near identical to Yamnaya.
They were the Indo Europeans as far as we can tell.
@andrzejewski
That substrate was provided by a population in north eastern Europe that was strongly Afontonova Gora shifted. They had a lot of WSHG and less WHG. Without knowing you made a perfect case for WHG as the main vector of archaic PIE since the HG in the southern part of eastern Europe ( pontic steppe) were more WHG than Afontonova Gora. CHG and EEF in the steppe are female mediated so they are out of the PIE picture.
What is left if not a WHG rich population?
Arza, are there more samples than the ones we already know?
The following samples have been tabulated on the YFull mtree for haplogroup C4b1, the clade to which YFull has assigned the mtDNA of the kra001 specimen:
C4b1 formed 6900 (95% CI 5500 <-> 8600) ybp, TMRCA 2400 (95% CI 1800 <-> 3200) ybp
C4b1* Kamenka 2 burial (Late Neolithic, Kolyma river, Yakutia) x3, burial 1 at the Nefteprovod-2 site (kra001, Bronze Age, Krasnoyarsk Krai) x1, Mansi x1, Evenki/Selkup (Krasnoselkup) x1, Kyrgyz x5 (four from Tashkurgan), Sarikoli (Tashkurgan) x1, Yakut x1, Buryat x1, Udege x4, Evenk (Stony Tunguska x4, Taimyr x4, Nyukzha x3, Iengra x1), Even (Berezovka x1, Tompo x1)
C4b1a TMRCA 1400 (95% CI 700 <-> 2600) ybp
C4b1a* Buryat (Russia x1, Inner Mongolia x1), Yakut (Northeast) x1, "TUVLI" x1, Oroqen x1, Avam Nganasan (Taimyr) x1
C4b1a1 TMRCA 500 (95% CI 75 <-> 2600) ybp Evenk (Nyukzha x1), Bargut x1
C4b1b TMRCA 1200 (95% CI 550 <-> 2300) ybp
C4b1b* Evenki (Nelkan/Dzhigda x1, Stony Tunguska x1), Yakut (Northeast x2, Central x2), Buryat x1
C4b1b1 TMRCA 275 (95% CI 50 <-> 850) ybp Buryat x3
C4b1c TMRCA 650 (95% CI 50 <-> 2300) ybp Yakut (Northeast x1), Yukaghir x1
C4b1d TMRCA 350 (95% CI 50 <-> 1100) ybp Even (Kamchatka x6, Tompo x1, Sakkyryyr x1, Severo-Evensk district x1), Yakut (Northeast x1)
C4b1e TMRCA 1250 (95% CI 100 <-> 6600) ybp Buryat x1, Yakut (HGDP) x1
C4b1f TMRCA 400 (95% CI 50 <-> 1300) ybp Udege x3
C4b1g TMRCA 500 (95% CI 50 <-> 1650) ybp Yakut x2, Evenk x1
C4b1h TMRCA 500 (95% CI 50 <-> 1650) ybp Uyghur x3
The same clade of mtDNA seems to have been found in the yak022, yak023, and yak024 specimens from the Late Neolithic Kamenka 2 burial near the Kolyma River in Yakutia.
As a proportion of the mitochondrial gene pool of a present-day ethnic group, the clade's density of distribution is probably maximized among Northern Tungusic peoples. However, Yakut and Buryat might account for the greatest number of living individuals who belong to this clade.
C4b2 is one of the most typical mtDNA clades among present-day Koryaks.
@ambron
Here is a full list of samples:
https://genome.cshlp.org/content/suppl/2021/02/12/gr.267534.120.DC1/Supplemental_Table_S1.xlsx
Some of them weren't analyzed yet.
@old europe @rob “ That substrate was provided by a population in north eastern Europe that was strongly Afontonova Gora shifted. They had a lot of WSHG and less WHG. Without knowing you made a perfect case for WHG as the main vector of archaic PIE since the HG in the southern part of eastern Europe ( pontic steppe) were more WHG than Afontonova Gora. CHG and EEF in the steppe are female mediated so they are out of the PIE picture.
What is left if not a WHG rich population?”
You made an interesting but flawed argument:
1. How do you KNOW that it was a WSHG language? Did you find any links to Botai language/Ket etc?
2. Genes don’t speak languages, people do, as Davidski likes to say. The example I had just given about the Finns being 50% WSH Yamnaya related makes them Indo-European in all but their speech.
3. Yamnaya and related WSH populations were 50% ANE, since EHG were 75% ANE and CHG were roughly 35% (I’m referencing the original Yamnaya prior to the significant EEF gene flow). Therefore, based on your line of reasoning, PIE *must* be an ANE language?
4. Both Kartvelians and NWC speakers have mostly CHG component, followed by EEF/ANF and then WSH. Alas, both languages are completely unrelated, sans some areal convergence “sprachbund”. Likewise, Basque and Etruscans were predominantly EEF/ANF but their speech is utterly unrelated.
How plausible is it that Botai may’ve spoken a Yenisseyan language, as both Ket and Botai were WSHG mostly?
“ This is what a WHG or an EHG must’ve sounded like:
Kildin Sami Northern Sami English
kut’t’k kut’t’k heart
nirr nierru cheek
čad’z’ čáhci water
vuntas sand
poav’n hummock
k’ed’d’k geađgi stone
abbr’ arvi rain
piŋŋk biegga wind
ket’t’k’ geatki wolverine
nigkeš pike (fish)
murr muorra tree
cigk mist
mun”
Not really, this is how they sound after going through different Saami sound shifts. Could be quite different sound initially. For example, Baltic word “segt(i)” (to cover) turned into “čiehkat” (to hide) in Saami. You can’t look at “Čiehkat” and say this is how Baltic sounds :))
Arza, I understand, thank you!
Thanks Michalis; although actually I was wrong and Tambets had no new Selkup or Nganasan samples (other Uralic speaking populations though like Mansi), so perhaps those Selkup you mention are from Flegentov 2019 or some other paper.
...
(On a broader thing, if the Nganasan were an unadmixed representative of the first proto-Uralic speakers (and I don't exclude the possibility), that would be an unusual case in which the first speakers of an expanding language family exist in an unadmixed form. That hasn't really happened in Indo-European, Semitic, Turkic, Mongolic, Austroasiatic? In Austronesian it does, through being preserved on the source island, while Sino-Tibetan gets close through sheer population size...
It seems like the Nganasan ethnography is pretty crucial anyway. Analogously, if we were to say "Well we have an extant cline of EEF ancestry through Europe, and it reaches into South Asia too via Steppe_MLBA (who are about 1/3 EEF). Moreover, we have Indo-European speakers in Europe with almost entirely EEF ancestry - the Sardinians. With possibly zero Steppe in the most isolated villages. Therefore the original Indo-Europeans were probably EEF". Disregarding the specific history of Sardinia and their language switch to Latin. That would be pretty wrong, but we'd need to know the ethnography and history to say why, and if we'd just looked at ancestry changes over time, and took it on face value that a present day population with some pure ancestry in a cline indicated the source, it might be pretty plausible).
@old europe
That substrate was provided by a population in north eastern Europe that was strongly Afontonova Gora shifted. They had a lot of WSHG and less WHG. Without knowing you made a perfect case for WHG as the main vector of archaic PIE since the HG in the southern part of eastern Europe ( pontic steppe) were more WHG than Afontonova Gora.
I do not want to touch the topic of paleo-languages, because I believe that it is beyond speculative, it is inevitably fairy tale even with the best intention. But I would like to add some nuance to the above statement.
1. If we use the least AG3 shifted samples (Grotta Continenza) as WHG, then Ukraine Mesolithic is about equally WHG and AG3 (give or take). More WHG if we use a somewhat AG3 shifted sample, like even Villabruna. (Nevertheless, their language is anybody's guess.)
2. This might be true on the Pontic region, but in the Volga-Caucasus region the population was about as AG3 heavy as in the North, in fact the data from Progress suggest that there was even a population that had less WHG than the ones in the North. (Again, what this means for languages is anybody's pet fairy tale.)
3. Scandinavia is not exactly NE Europe, even from Karelia there is a still a lot of distance eastward before you are in Asia. So far the uncovered Scandinavian HG-s are actually more western than Ukraine Mesolithic. In fact significantly so. Even with the plausible assumption that SHG-s in Finland were more eastern than Motala_HG, I'd say they were not significantly more eastern than Ukraine Mesolithic. I'd except them to be close to some Pontic steppe groups in the extent of their WHG:AG3 ratio. So the language they had spoke and influenced the ancient Saami might be very well from an even more ancient WHG one. (But again, anybody's guess. And at the time of the Uralic arrival to historical Saami territory, SHG-s already had many millennia of divergence from the Pontic population on both of their ancestry side. So as far as we know they could be both WHG or both EHG and still be very different.)
@Matt
On a broader thing, if the Nganasan were an unadmixed representative of the first proto-Uralic speakers (and I don't exclude the possibility), that would be an unusual case in which the first speakers of an expanding language family exist in an unadmixed form.
No one's claiming this.
And if it were true, it would be highly unusual, but it's extremely unlikely to be true.
@Andrzejewski
How plausible is it that Botai may’ve spoken a Yenisseyan language, as both Ket and Botai were WSHG mostly?
Ket are mostly not WSHG. When modelled in G25 nMontes as kra001 + Sintashta + Tyumen HG they pick up something like 20% Tyumen HG. Botai + Sintashta + Baikal Neolithic gives something like 35% Botai and that is about the maximum you can get out of them. Their biggest ancestry proportion is what here sometimes called "deep Siberian", that is, in the context of West Eruasian - ANE - East Asian, they are mostly East Asian.
@ arza
Where did you download the Helladic genomes from ?
@ Rob
https://www.ebi.ac.uk/ena/browser/view/PRJEB37782
@Slumberry “ Ket are mostly not WSHG. When modelled in G25 nMontes as kra001 + Sintashta + Tyumen HG they pick up something like 20% Tyumen HG. Botai + Sintashta + Baikal Neolithic gives something like 35% Botai and that is about the maximum you can get out of them. Their biggest ancestry proportion is what here sometimes called "deep Siberian", that is, in the context of West Eruasian - ANE - East Asian, they are mostly East Asian.”
Makes sense. Huns and Xiongnu were described as having East Asian appearance but they likely spoke Yenisseyan. WSH tribes like Yamnaya were Europoid but were 50% ANE AG3 descendants. Going by Rob’s line of thinking, maybe indeed Yenisseyan languages, Okunevo and Native American/Yupik/Na Dene had NOTHING to do with Yamnaya on deep ancestry level, and these languages came from some Paleo-Siberian East Eurasian ethnogenesis.
@Andrzejewski
"Both Kartvelians and NWC speakers have mostly CHG component, followed by EEF/ANF and then WSH. Alas, both languages are completely unrelated, sans some areal convergence “sprachbund”. Likewise, Basque and Etruscans were predominantly EEF/ANF but their speech is utterly unrelated."
Linguistics has it's time and place, but if you keep relying on it beyond it's limitations it's like tying a hand behind your back. Besides Na-Dene and Eskimo, there's dozens of language isolates in the Americas that linguistics will never resolve into one cohesive language family. Yet we know from genetics that the majority of First Americans originate from a common ancestor group. And this group would have obviously spoken one language that through time and isolation caused it's daughter languages to diverge so significantly that it's impossible to classify them into one family. However, genetics tells you that ultimately they all came from a single ANE derived language.
Proto-Uralic is also on a similar position. There's almost 100% pure proto-Uralic genome and ydna with ancestry from eastern Siberia. On it's way to Europe it interacted with different languages that are extinct and which we will never know about, in addition to influence from Indo-Iranian languages and Tocharian. So it's expected for Uralic to seem unrelated compared to languages of its fatherland in Siberia, but genetics clarifies where the root of that language was located.
My line of thinking is more about permitting a diversity of langauge forms within broadly similar populations rather than a priori excluding deep affinities (although the likelihood of PIE being similar to Amerindian is not plausible on a chronological and cultural level). Hence the point was, even if Nganasan are language-shifted groups who originially spoke a different language family, it does not mean proto-Uralic speakers cannot have been nearby
@ Arza
Thanks
Gabriel:
“Why does there need to be several migration waves of Siberians to Europe from deep in Siberia to explain BOO? Seems excessively complicated.”
Hmm… Did I say something like that? I think BOO represents only one wave from Siberia, but I think that Siberian ancestry could have spread to the west also by Volga route, not only the Arctic Sea route. Some Volga populations have more Siberian ancestry than the Saami.
Andrzejewski:
“This is what a WHG or an EHG must’ve sounded like:”
Some corrections: ‘rain’ is an old IE loanword, and ‘water’ has an old cognate at least in Khanty, so it seems that the IE loanword PU *weti ‘water’ largely replaced this older Uralic word.
Parastais:
“Not really, this is how they sound after going through different Saami sound shifts. Could be quite different sound initially. For example, Baltic word “segt(i)” (to cover) turned into “čiehkat” (to hide) in Saami. You can’t look at “Čiehkat” and say this is how Baltic sounds :) ”
Good point. Most of those words postdate the Proto-Saami sound changes, but still probably different sound systems would have meant changes in the substitution of these words into Saami.
Andrzejewski:
“2. Genes don’t speak languages, people do, as Davidski likes to say. The example I had just given about the Finns being 50% WSH Yamnaya related makes them Indo-European in all but their speech.”
Well, in Tambets et al. 2018 in figure 6 even the Saami have 38-51 % Corded Ware ancestry, around 3/4 of it coming from Yamnaya.
Leron:
“Proto-Uralic is also on a similar position. There's almost 100% pure proto-Uralic genome and ydna with ancestry from eastern Siberia.”
How do you know that there is a Proto-Uralic genome? You are only guessing, like Davidski.
Linguistic results show that Proto-Uralic was spoken around Volga-Kama area about 2000 BC. If there is a genome fulfilling this criteria, only then we have a match: only then we could assume that it possibly was a Proto-Uralic speaker.
Leron:
“On it's way to Europe it interacted with different languages that are extinct and which we will never know about, in addition to influence from Indo-Iranian languages and Tocharian. So it's expected for Uralic to seem unrelated compared to languages of its fatherland in Siberia, but genetics clarifies where the root of that language was located.”
It is probable that Pre-Proto-Uralic came from Siberia. Still, you cannot see language from the DNA.
Interesting is the study Karafet et al. 2018: "Siberian genetic diversity reveals complex origins of the Samoyedic-speaking populations."
It shows in Admixture already in K=7 components, which almost totally cover the ancestry of their nominal population:
- Ket-Selkup component
- Nganasan component
- Forest Nenets component
Khantys and Tundra Nenets have all these three, and a little European ancestry in addition.
"ADMIXTURE analysis demonstrates that the “Siberian” component that is highly pronounced in Nganasans, is also frequent in northern Siberian populations (Figure 3, K = 4). Positive f3 values do not support a history of admixture between these arctic and subarctic populations (Supporting Information Table S6). At the same time, this analysis implicated Nganasans and/or Evenks as a mixing source population for the other Siberian groups. These results are consistent with our finding that the Nganasans and Evenks exhibit the highest level of smaller-IBD-segment sharing with nearly all other Siberian populations (Supporting Information Table S7)."
"In sum, our data are consistent with the following hypotheses about the origin of modern Samoyedic populations:
(1) The Nganasans are very likely direct descendants of the ancient sedentary Arctic Paleolithic hunters of northern Asia and Neolithic hunters of wild deer who were culturally assimilated by various Samoyedic peoples. The Nganasan language and material culture suggest prehistoric contacts with proto Tungus and Yukagir populations (Anikin & Helimsky, 2007; Simchenko, 1968).
(2) Our data are compatible with the hypothesis that Tundra and Forest Nentsi are descended from the same aboriginal tribe that gave rise to the Nganasans who were assimilated by ancient Samoyeds (Prokof'ev, 1940). PCA, ADMIXTURE analyses, and treemix tree reveal close genetic ties between the Tundra and Forest Nentsi, while ROH and IBD analyses indicate that these two populations share more short IBD segments (<6 cM) with other populations than with themselves—most likely as a result of a common origin and long-term gene flow."
@Jaska
It's impossible to take you seriously because of the copious amounts of garbage that you're generating here.
You think it's impossible to tell whether Fatyanovo has any Volosovo admix because Fatyanovo has a lot of EHG ancestry from the steppe?
Are you out of your fucking mind?
Learn the basics and stop embarrassing yourself.
@Jaska
It shows in Admixture already in K=7 components, which almost totally cover the ancestry of their nominal population:
- Ket-Selkup component
- Nganasan component
- Forest Nenets component
That's just recent drift.
Old pre-aDNA genetic papers are full of this sort of crap, where the authors mistake recent drift and founder effects for ancient structure.
Grow a brain.
@Jaska
"‘rain’ is an old IE loanword"
This is sheer nonsense.
"it seems that the IE loanword PU *weti ‘water’ largely replaced this older Uralic word."
Don't talk nonsense, this is not loanword, water is an old Nostratic word - Tungus-Manchu: *udu-n, Korean: *òrán-, Dravidian: *jēd_-. No more ancient word exists in the Finno-Samoyedic languages.
PS. From now on, I will not use the word Uralic, but I will write Finno-Samoyedic so as not to give anyone false illusions. FS is like FU.
" Linguistic results show that Proto-Uralic was spoken around Volga-Kama area about 2000 BC."
It's a lie. Scientific data does not show anything like that, these are only your personal fantasies based on your anti-science.
@Jaakko:
"Some corrections: ‘rain’ is an old IE loanword..."
North Saami arvi is not an Indo-European loan, but perhaps an Indo-Iranian one from II *abʰra- 'rain' < PIE *n̥bʰros 'rain (cloud)'.
"... and ‘water’ has an old cognate at least in Khanty, so it seems that the IE loanword PU *weti ‘water’ largely replaced this older Uralic word."
I have heard of this theory. But the Khanty word means 'flood'. And IE *wodr/ud(e)n- 'water' does not have any forms with a front vowel in IE (except perhaps Anatolian, although according to Kloekhorst this vowel is a schwa and the oblique form is still *ud-). So this can't have been borrowed as Uralic *weti. There are other forms that derive directly from the root *wed, like Germanic *wētaz 'wet', Armenian get 'river'. But none of those mean 'water'.
In other words, this theory is wishful thinking. And it is clear what the wish is, a Proto-Uralic that does not look so much like Indo-European.
@Andrzejewski:
"On Non-Uralic substrate in Sami"
And all of this is also a good example why any theory about substrate from an unattested language should be taken with lots of grains of salt. Which is something that has already been pointed out on this forum.
@Onno Hovers
"Indo-Iranian one from II *abʰra- 'rain' < PIE *n̥bʰros 'rain (cloud)'."
rain - There is no such meaning in either Indo-Iranian or Proto-Indo-European. This word only means cloud. In Indo-Iranian, it can mean bad weather < cloudy weather.
@Jaakko Häkkinen
"Linguistic results show that Proto-Uralic was spoken around Volga-Kama area about 2000 BC."
Now, setting aside the entire genetic argument, you must be aware that this is not a consensus view of all historical linguist who study this topic. You have a point saying that the connection between spread of genes and spread of languages is hardly a 100% predictable thing (to say the least), and it is impossible to ascertain/prove some assumptions. But you should not imply that in contrast your theory about the Volga-Kama PU is an immutable scientific reference point, when even fellow linguists apparently disagree. And also, purely linguistic considerations about the localization of proto-languages are no less uncertain and no less impossible to prove that genetic speculations.
BTW, I am familiar with some considerations of why the PU homeland must be in Europe as they are written in some school textbooks in Hungary. As there can be alternative explanations for genetic phenomena, there are alternative explanations for those too.
@Davidski
I get bewildering results for Kets with G25 nMontes (scaled) when I try Botai + Globular Amphora + Afanasievo + kra001 (+ Veretye HG, but it is not of much influence).
Target Distance KAZ_Botai POL_Globular_Amphora RUS_Afanasievo RUS_Krasnoyarsk_BA RUS_Veretye_Meso
Mansi 0.04536930 21.6 12.4 14.8 44.4 6.8
Ket 0.07018073 35.8 16.6 1.0 43.0 3.6
Nganas. 0.07260624 0.0 1.8 0.0 98.2 0.0
Khanty 0.04480577 26.4 11.0 12.8 45.0 4.8
The other populations I included make sense. The Nganassan come out as pretty much pure kra001. The Mansi and the Khanty pack some WSHG and Steppe_MLBA ancestry on it. But the Ket: significant Globular Amphora with noise level Yamnaya. I can't imagine the source of such an admixture. Even modern Russian admixture would not result in this. Do I overlook something obvious?
Makes sense. Huns and Xiongnu were described as having East Asian appearance but they likely spoke Yenisseyan
Dude, you have got to stop speaking bullshit.
Most historical descriptions of the Xiongnu describe them as looking European and the archaeological evidence largely confirms this as well.
See also: https://www.hindawi.com/journals/bmri/2020/2585324/
Their conquerors also weren't majority Q1a, either. Q1a seems to have been a subordinate clade under Iranian Q1b, and the most frequent is R1a.
Davidski:
“You think it's impossible to tell whether Fatyanovo has any Volosovo admix because Fatyanovo has a lot of EHG ancestry from the steppe?
Are you out of your fucking mind?
Learn the basics and stop embarrassing yourself.”
You are the only one embarrassing yourself here: your black-and-white mind cannot understand right even a single sentence. We can continue discussion when you stop building your strawmen and start to read and understand what is said to you. But nice try! Unfortunately bluffing that you know better doesn’t actually mean that you know better.
Davidski:
“That's just recent drift.
Old pre-aDNA genetic papers are full of this sort of crap, where the authors mistake recent drift and founder effects for ancient structure.”
You confuse your beliefs with knowledge, which is sad and unscientific.
The point is, that Forest Nenets are not usually considered in these analyses: therefore they can reveal something more than studies without them.
Besides, usually the components which occur on the lower K-values don’t seem to be late drift and/or founder effect. It tells once again about your unscientific, religious beliefs that you are so eager to ignore everything which may cause a threat to your fantasies.
Onno Hovers:
“North Saami arvi is not an Indo-European loan, but perhaps an Indo-Iranian one from II *abʰra- 'rain' < PIE *n̥bʰros 'rain (cloud)'.”
Yes, but “old IE” contains all the oldest layers from Indo-European branches.
Onno Hovers:
“So this can't have been borrowed as Uralic *weti. There are other forms that derive directly from the root *wed, like Germanic *wētaz 'wet', Armenian get 'river'. But none of those mean 'water'.”
Still considering the phonological and semantic similarity it is best seen as an old loanword. More far-fetched is a view that this would be inherited from the shared Indo-Uralic protolanguage, which remains unproven.
Onno Hovers:
“In other words, this theory is wishful thinking. And it is clear what the wish is, a Proto-Uralic that does not look so much like Indo-European.”
I didn’t think you would be a conspiracy theorist.
Onno Hovers:
“And all of this is also a good example why any theory about substrate from an unattested language should be taken with lots of grains of salt. Which is something that has already been pointed out on this forum.”
It is of course common knowledge. But in Saami there are hundreds of words which have no known cognates anywhere, and great portion of them are phonotactically necessarily Post-Proto-Saamic. So there really is no other credible option than a substrate from an unattested language – especially when we know that Lapland only became Saami-speaking within the last 2000 years.
@Jaakko
Fatyanovo can't have any significant EHG from Volosovo because this would mean that it also acquired extra CHG and EEF from pure CHG and/or EEF sources at the same time, which is impossible.
Grown a brain you hopeless newbie.
Slumbery:
“Now, setting aside the entire genetic argument, you must be aware that this is not a consensus view of all historical linguist who study this topic.“
It is a consensus view of those Uralists who are capable of assessing the arguments. Those who disagree, refuse to understand or accept the supporting arguments, without even trying to give any convincing counter-arguments. Stubbornness is not a scientific argument. :)
Slumbery:
“You have a point saying that the connection between spread of genes and spread of languages is hardly a 100% predictable thing (to say the least), and it is impossible to ascertain/prove some assumptions. But you should not imply that in contrast your theory about the Volga-Kama PU is an immutable scientific reference point, when even fellow linguists apparently disagree.“
Please name one Uralist who has presented valid counter-arguments against the Volga-Kama homeland since 2009. Then we can see how valid those arguments actually are.
Slumbery:
“And also, purely linguistic considerations about the localization of proto-languages are no less uncertain and no less impossible to prove that genetic speculations.”
Wrong: genetics like archaeology alone have zero value for locating languages, because neither of these disciplines cannot even reach a language, even less study it. Their only value is, if there is a match concerning time, place and the direction of spread between the linguistic results and genetic/archaeological result.
Scientific method:
1. Let’s look at the genetic results.
2. Let’s look at the linguistic results.
3. Let’s look if there are in both of these any phenomenons which match concerning time, place and direction of spread.
Unscientific method of Davidski:
1. Let’s look at the genetic results.
2. Let’s ignore the linguistic results.
3. Let’s decide ad hoc that this genetic wave in certain place at certain time represents certain language.
Slumbery:
“BTW, I am familiar with some considerations of why the PU homeland must be in Europe as they are written in some school textbooks in Hungary. As there can be alternative explanations for genetic phenomena, there are alternative explanations for those too.”
Please feel free to try to explain alternatively all the linguistic evidence supporting the Volga-Kama homeland around 2000 BC.
@Jaakko
There's no NWIE layer in Proto-Uralic.
This is just your assumption and indeed wild speculation.
From what I gather, you two (Jaakko and Davidski) both agree that Pre-Proto-Uralic was spoken in Siberia by a population that was likely rich in N-L1026/kra001-like ancestry.
The question is, did Proto-Uralic form in Siberia, or after migrating to the Volga-Kama?
Hofully Jaakko will agree with me here, but more aDNA from the correct space and time should make things clear.
Was N-L1026 present west of the Urals circa 2000BC? Do ancient East-Uralics (Cherkaskul culture?) carry admixture representative of an origin in the Volga-Kama?
Either way, this seems to be coming down to semantics.
@Anthony
I wouldn't call this an argument about semantics.
Jaakko is claiming that he's taking the scientific approach here when in fact all he's doing is pushing some idiotic, outdated ideas that aren't based on anything except wishful thinking and lousy assumptions.
Do you really think there's a chance that Fatyanovo was Northwest Indo-European speaking, and that Proto- or even Late Proto-Uralics were in contact with these Fatyanovo or even Fatyanovo-derived Northwest Indo-European speakers?
It's already obvious that this won't work out. It's nothing more than a fantasy. Anyone who claims that this is science must be mentally unstable.
@Davidski “ Do you really think there's a chance that Fatyanovo was Northwest Indo-European speaking, and that Proto- or even Late Proto-Uralics were in contact with these Fatyanovo or even Fatyanovo-derived Northwest Indo-European speakers?”
Everyone knows that it’s a CWC offspring. PIE was formed in Eastern Europe, PU was in Western Siberia(Asia).
David, have you processed the samples from the new article that Arza wrote about? It would be interesting to know if the sample from Volosovo-Danilovo has admixture Kra001? Formally speaking, this is a culture of fatyanoid ceramics, although individual burials in this cemetery are attributed to both Balanovo and Pozdnyakovo and to the culture of mesh ceramics.
I'm getting this error.
OOPS indiv file has changed since genotype file was created
Aborted (core dumped)
Davidski:
"There's no NWIE layer in Proto-Uralic.
This is just your assumption and indeed wild speculation."
Stop trolling, you know nothing about linguistic results. Read here Koivulehto's article:
https://www.sgr.fi/sust/SUST242.pdf
Anthony Hanken:
“From what I gather, you two (Jaakko and Davidski) both agree that Pre-Proto-Uralic was spoken in Siberia by a population that was likely rich in N-L1026/kra001-like ancestry.”
-- I see it possible, yet far from proven.
AH: “The question is, did Proto-Uralic form in Siberia, or after migrating to the Volga-Kama?”
-- That is difficult to determine, and it also depends on the definition of “forming”. It is easier to say that Late Proto-Uralic centre of expansion was in the Volga-Kama region. At some point before that pre-stage seems to have arrived from Siberia.
AH: “Hofully Jaakko will agree with me here, but more aDNA from the correct space and time should make things clear.”
-- Indeed! A match with the linguistic results would be nice: ancient DNA near the Volga-Kama fork around 2000 BC, and also from both sides from that time. Then we could scientifically suppose that a probable Proto-Uralic speaker is at our hands, and comparing that to the Siberian samples we could get information about the genetic roots of these people – which still cannot prove about the origin of the language, but it would be a good start. Then linguistic search could again be focused on that certain area.
@Jaakko
Read here Koivulehto's article: https://www.sgr.fi/sust/SUST242.pdf
There's no real evidence there that Proto-Uralic came into direct contact with NWIE or anything like it.
Davidski:
“Do you really think there's a chance that Fatyanovo was Northwest Indo-European speaking, and that Proto- or even Late Proto-Uralics were in contact with these Fatyanovo or even Fatyanovo-derived Northwest Indo-European speakers?”
-- Everybody who understands science sees that your method, by which you can see a language from the DNA, is pure fantasy fiction. But hey, feel free to dive deep into your hallusinative bubble! I think your readers don’t like to be mislead, so I keep telling them what is scientific, so they wouldn’t end up hating your for luring them into your pseudo-science. :)
D: “It's already obvious that this won't work out. It's nothing more than a fantasy. Anyone who claims that this is science must be mentally unstable.”
-- Says a mentally unstable man who has no clue about the multidisciplinary science.
Egomanic bluffing may be successfull at the children’s sandbox, but in science it doesn’t take you anywhere.
Andrzejewski:
“Everyone knows that it’s a CWC offspring. PIE was formed in Eastern Europe, PU was in Western Siberia(Asia).”
Late Proto-Uralic spread from the Volga-Kama region, that is the best-argued consensus view in Uralic studies. Please present counter-arguments, if you have any. Knowing without supporting arguments is not actually knowing but believing...
@Davidski
Given the amount of times I've seen Jaakko cited, I'm willing to consider what he's saying, at least in his own field. Even if that may be an appeal to authority.
The connection between NW-IE and Fatyanovo is a tough one. Thats something I'm not going to defend. Saag et al. (2021) is too recent, it will take some time for the linguists to catch up.
What I'm getting at though, is that without a lot more data, it will be hard to determine the exact area Pre-Proto-Uralic became Proto-Uralic.
And in all honesty, geography really doesn't matter. The same populations will need to be involved, whether these contacts occured on the western or eastern side of the Urals. I think you and Jaakko both know that.
@Jaakko
I've got some predictions for you.
Over the next couple of years a scientific consensus will emerge that Fatyanovo was Proto-Indo-Iranian, or at least pre-Proto-Indo-Iranian.
At the same time, the idea that Proto-Uralic had direct contacts with Northwest Indo-European will be completely forgotten.
The idea that Late Proto-Uralic was spoken in the Volga-Kama region around 2,000 BCE will also go the way of the dodo.
@Anthony
I think you know as well as I do that without N-L1026 and at least some kra001-like admix west of the Urals in ~2,000 BCE the chances that the Proto-Uralic homeland was located there at this time are zero.
It looks like it'll take Jaakko many years, or perhaps forever, to adjust to this reality, but that's his problem not mine.
@Davidski
So there are no N-L1026 samples west of the Urals from around 2000BC? Not even the Trans-Ural area?
That would be the nail in the coffin for me.
Not a huge fan of CWC=Uralic.
No, of course not.
That's when Sintashta is in the area, and we've got a very good idea about the substructures in Sintashta and its Y-haplogroups.
There are no plausible Proto-Uralic populations in that area at that time.
Seima, Reshnoe, Yurino, Turbino, Kaninskaya cave, Satyga, etc?
Ancient DNA from just one of these sites would rule out a lot.
@Jaakko
"Still considering the phonological and semantic similarity it is best seen as an old loanword. More far-fetched is a view that this would be inherited from the shared Indo-Uralic protolanguage, which remains unproven."
Full nonsense, issuing your personal anti-scientific fantasies for science. "IT IS BEST SEEN AS" gives you an anti-scientific deceiver in you, this is not an argument at all, your personal preferences have nothing to do with the scientific method. Your denial of science and the issuance of its own anti-scientific nonsense for science shamefully. Linguistics are engaged in smart and honest people to whom you do not relate.
All relations between the ancient linguistic families are proved, because the Finno-Samoyedic language did not occur with airless from air 4 thousand years ago. Гt has ancestral language.
"It is a consensus view of those Uralists who are capable of assessing the arguments. Those who disagree, refuse to understand or accept the supporting arguments, without even trying to give any convincing counter-arguments. Stubbornness is not a scientific argument."
This is a pure and uncovered lie. You are Mr. Liar writing a lie.
All that you write a long time is reliably refuted, you just just do not want to notice because you are not a scientist. As you are certainly not a scientist had never heard of linguistic palaeontology.
"Please name one Uralist who has presented valid counter-arguments against the Volga-Kama homeland since 2009. Then we can see how valid those arguments actually are."
Such scholars as the JANHUNEN, Napolskich, Helimski, Hajdú, and unlike you, who is not a scientist. They are smart people and know much more you.
"Late Proto-Uralic spread from the Volga-Kama region, that is the best-argued consensus view in Uralic studies."
This is a pure and uncovered lie. All that you write a long time is reliably refuted, you don't want to be stuffed to be stuffed because you are not a scientist.
@ Anthony
The only authority should be evidence. Historical linguistics is an anthropological question, not simply one of comparative linguistics. Holistic models require an understanding of all lines of evidence, at least to the extent that one understands why people say what they do, the methodology and influences behind it, limitations and problems. To break down all component lines of evidence then build a new model from scratch - if needed - rather than defending those which already exist. But this is a pretty big demand. At least a couple of commentators here claim to be authorities but they fall well short of those requisites.
@ Davidski
You need to add "hashcheck: NO" in the parfile of convertf.
@Jaska “ Well, in Tambets et al. 2018 in figure 6 even the Saami have 38-51 % Corded Ware ancestry, around 3/4 of it coming from Yamnaya.”
Exactly! Lapps are mostly or close to 50% Indo-European; Basques are at least 25% and so were the Etruscans. Old-fashioned pet theories about these and other populations being relics of some Mesolithic “Paleo-Europeans” are out of window by now. It’s a WSH more than any ANF, WHG, CHG, EHG or else, that formed modern (namely Post-Brone-Age) Pan-Europeans they were they are today, not only linguistically and culturally but largely genetically.
(I don’t doubt that Euros harbor a very significant ANF and in some cases WHG ancestry, in particular among Basques and descendants of Baltic BA populations, and that for most Europeans the EEF and WHG components were mediated through GAC and somewhat Cucuteni Tripolye, but I believe that WSH signal is being underrated and in many cases such as uniparental markers y Hap I2a2a and mtDNA U4, U5 and perhaps even U2) are mistaken for WHG.
@Davidski “ The idea that Late Proto-Uralic was spoken in the Volga-Kama region around 2,000 BCE will also go the way of the dodo.”
Bye bye to adherents of now-obsolete Indo-Uralic theory. Carlos Quiles will be eating his hat in not too long.
I’m sorry to disappoint @Kristiina and other followers of this doctrine (dogma?), but PIE and PU share ZERO cognates; even words like name and water are PIE (or even PIIr—> PU) loan words.
NWIE languages came with the SGC/Bell Beakers, and they are Centum languages. Balto-Slavic and Indo-Iranian are Satem ones and they all are offsprings of a common CWC, which I personally believe was a direct descendant of some EEF-rich Sredny Stog population nearby the Dnieper.
The jury is still out where Tocharian fits in all this mix. We do know now (Holland 2018) that Tarim Basin Mummies were much closer genetically to Sintashta daughter Andronovo than to Yamnaya or Afanasievo, and the presence of Anatolian-derived mtDNA T, K, H and others holds testimony to the historical fact that CWC admixed with GAC (and some CTC) mostly in today’s Poland and West Ukraine and fathered Fatyonovo and then Abashevo etc. is a known fact. I wouldn’t conflate “Tocharians” with “Tarim Base Mummies”, in spite of the fact that the latter ones have R1b rather than the Corded R1a1.
@Rob
I agree, at least thats how it should be. I can'recall the last time I've seen genetics brought up in a linguist's paper though. That wouldn't jive well with their methodology which, I think gets bogged down by minutiae, at the expense of the bigger picture.
Besides, "all lines of evidence" have not yet completely ruled out N-L1026, and thus Proto-Uralic being found west of the Urals circa 2000BC. Even if you don't think this is very likely.
If it was anywhere, it should be found in S-T necropoli like those I mentioned above. I'm not sure if there is aDNA coming from these sites or not.
@Andrzejewski
"Bye bye to adherents of now-obsolete Indo-Uralic theory.
I’m sorry to disappoint @Kristiina and other followers of this doctrine (dogma?), but PIE and PU share ZERO cognates; even words like name and water are PIE (or even PIIr—> PU) loan words."
This is just a downright anti-scientific lie from the first to the last word. Everything that you write is a militant anti-science mixed with outright lies. You absolutely do not understand what you write.
Davidski:
“I've got some predictions for you.”
None of these will happen, unless the whole scientific world gets crazy and dumb enough to adopt unscientific methods. :D
Anthony Hanken:
“The connection between NW-IE and Fatyanovo is a tough one. Thats something I'm not going to defend. Saag et al. (2021) is too recent, it will take some time for the linguists to catch up.”
Here we must separate facts from interpretations:
- Autosomally, all we know is that the ancestry was homogeneous in a huge area from Corded Ware Cultures to the European and Central Asian steppes.
- Y-chromosomally, there is no direct link between Fatyanovo and Sintashta. In the former there are R1a-Z93, in the latter there are its subclades. A missing link between these two can still appear in the future.
As you see, there is still many possible options: you shouldn’t stick with only one of them, that is, supposing that Fatyanovo Culture was Aryan. That is only building guesses upon guesses.
Genetics can never disprove linguistic results, just like linguistic can never disprove genetic results. There can only be either match or mismatch between their results. And if there is a mismatch, then you must find a better genetic match for the linguistic results, because there is no way you can see language from the DNA. Genetic continuity cannot testify for linguistic continuity. Do you accept this?
Then to NwIE loanwords in Uralic. We have these layers:
1. Archaic IE loanwords (formerly known as Proto-Indo-European loanwords), wide distribution in Uralic.
2. (Proper) Northwest IE loanwords (equally archaic), distribution "Finno-Permic" in the areal sense = West and Central Uralic.
3. Proto-Balto-Slavic and Pre-Germanic loanwords in the West Uralic branches Saamic, Finnic and Mordvinic since the late 2nd millennium BC.
As Permic is spoken northeast from the Volga bend, the proper NwIE still must have been spoken close to the Volga Bend. And because East Uralic branches still must be derived from the Late Proto-Uralic from around the Volga Bend, the same location goes with the Archaic IE. So, the difference here seems to be not the location but the time of the respective IE languages.
So the question here is: which culture can be connected to the source language of the Archaic and NwIE loanwords before and after 2000 BC in the Volga Bend region? Fatyanovo-Balanovo happens to arrive there at the right time to match with the linguistic results.
Then to the Aryan branch.
There are loanwords from different stages of Aryan in Uralic.
1. Pre-Proto-Aryan (or Early/Middle Proto-Aryan) loanwords with wide distribution in Uralic.
2. Late Proto-Aryan loanwords with wide distribution in Uralic.
3. Proto-Iranian loanwords with narrow distribution in different Uralic branches.
Again we know the location of Late Proto-Uralic being in the Volga Bend area. So the question is, which culture can be connected to the source language of the Pre- and Proto-Aryan loanwords before and up to 2000 BC in the Volga Bend region?
From the west there comes Fatyanovo-Balanovo Culture, but how probable it is that two different languages were spoken within the same culture? Abashevo Culture has both Corded Ware and steppe roots, so wouldn’t it be the best match for the oldest Aryan loanword layers? Instead, could we connect Abashevo to Archaic/NwIE and and Fatyanovo to Aryan?
Anthony Hanken:
“So there are no N-L1026 samples west of the Urals from around 2000BC? Not even the Trans-Ural area?
That would be the nail in the coffin for me. Not a huge fan of CWC=Uralic.”
RED ALERT!!! RED ALERT!!! RED ALERT!!!
Can’t you see the insanity of your method?
You just decide (ignoring all linguistic results) that some genetic or archaeological phenomenon is connected to Uralic language, and then you circularly deduce that its presence in certain place at certain time testifies for the presence of Uralic speakers, and its absence testifies for the absence of Uralic speakers.
I don't use linguistics to predict anyone’s genes or culture.
I don't use archaeology to predict anyone’s genes or language.
I don't use genetics to predict anyone’s language or culture.
Why do you do so? It is totally unscientific.
Rob:
“The only authority should be evidence.
Yes: the relevant, valid evidence. The only relevant evidence concerning language is the linguistic results. Genetics cannot tell anything about language – it cannot even study language. There are much more population movements than there are linguistic spreads. So, how can you tell which migration is connected to a certain language?
You either guess or you take the linguistic results and try to find match for them in the genetic results. The former is your method, the latter is the scientific method.
Rob:
“Historical linguistics is an anthropological question, not simply one of comparative linguistics. Holistic models require an understanding of all lines of evidence, at least to the extent that one understands why people say what they do, the methodology and influences behind it, limitations and problems. To break down all component lines of evidence then build a new model from scratch - if needed - rather than defending those which already exist. But this is a pretty big demand. At least a couple of commentators here claim to be authorities but they fall well short of those requisites.”
Yes, most people here and other foras have fallen short to the trap of the unscientific method.
Holistic model can be formed either with scientific or unscientific method – which one do you choose?
Scientific method:
1. Let’s look at the genetic results.
2. Let’s look at the linguistic results.
3. Let’s look if there are in both of these any phenomenons which match concerning time, place and direction of spread.
Unscientific method of Davidski:
1. Let’s look at the genetic results.
2. Let’s ignore the linguistic results.
3. Let’s decide ad hoc that this genetic wave in certain place at certain time represents certain language.
@Archi “
This is just a downright anti-scientific lie from the first to the last word. Everything that you write is a militant anti-science mixed with outright lies. You absolutely do not understand what you write.”
I’m inviting you to quote any legitimate scientist who touts the Indo-Uralic theory. And don’t say “Bomhard”.
NEw info seem to prove WHG did not have Black skin.
Skin color genes in Africa.
https://science.sciencemag.org/content/358/6365/eaan8433
They found SNPs in Africa which have more of an affect on skin color than rs1426654. Some East AFricans have rs14626654, but these two SNPs had more of an affect on their skin color.
rs1800404 and rs1868333. Each one explains 20 melanin index difference in Africans. That's a huge difference. Melanin difference on average between Indians & Europeans is 30.
The light skin versions are rare in Africa, more common in India, and much more common Europe than in India. They are as common in San Africans as in Europeans. San have light skin.
The light skin versions are as common in Mesolithic Europe as modern EUrope, according to allele frequencies from Mathieson 2017. Which is mostly madeup of SCandinavian HGs who we already knew had light skin.
If more data on WHG proves they have the light skin version of these SNPs, this excludes the possibility they had African-style Black skin. It even suggests they had lighter skin than dark brown, maybe just brown.
BBC lied to us about CHeddar man being Black.
These study found many SNPs which originated before out of Africa that explain a lot of skin color variation.
India often has same dark versions as Africa which makes sense. San Africans have light versions.
These can help understand skin color variation in prehistoric Eurasia.
@Jaakoo Hääkinen I read your article about the Uralic-Yukhaghir connection. You wrote something in it about the non-IE non-Uralic substrate in Sámi and in Baltic languages. You’re the done who gave me the idea about that Paleo-European languages spoken by pre-Sami and by Baltic HG. Now if you can elaborate more on the topic I’d be glad to read and understand.
I’m starting to get lost meandering through the genetic history of Siberia and who begat whom and when. I have read studies claiming that Ainu people in Japan aren’t descendants of the Jomon people but the are typically Nivkh + Samotsan/Obkhosk Culture, but that the overwhelming majority of their language and culture spin off Samotsan and not from Nivkh. Where exactly do ANE, Beringians, Tungus, Xiongnu, Altaic and Uralics mesh into all this, and what’s the affiliation of any of these aforementioned groups with Ulchi-like Devil’s Gate Baikal HG?
It might be a good idea to also have a detailed look at the Western ancestry in Uralic populations. It looks to me like Uralic populations have extra EHG beyond what they would get from Steppe. They don't seem to have the commensurate amount of CHG that comes Steppe ancestry. But the weird part is this: Somehow the Uralic populations do seem to have EEF ancestry.
Uralic populations (note the low ratio CHG/EHG):
Target: Khanty
Distance: 4.3390% / 0.04338954
44.6 RUS_Krasnoyarsk_BA
23.6 KAZ_Botai
17.8 RUS_Karelia_HG
7.8 DEU_LBK_KD
5.2 TUR_Barcin_C
1.0 GEO_CHG
Target: Selkup
Distance: 4.9769% / 0.04976929
51.6 RUS_Krasnoyarsk_BA
14.2 RUS_Karelia_HG
13.4 RUS_Ust_Belaya
12.2 KAZ_Botai
5.4 TUR_Barcin_C
3.0 DEU_LBK_KD
0.2 GEO_CHG
Target: Saami
Distance: 3.5215% / 0.03521548
38.4 RUS_Karelia_HG
22.8 RUS_Krasnoyarsk_BA
17.6 DEU_LBK_KD
11.4 SRB_Iron_Gates_HG
9.4 GEO_CHG
0.4 RUS_Yakutia_Meso
For comparison (note the increased ratio CHG/EHG):
Target: RUS_Sintashta_MLBA
Distance: 4.9184% / 0.04918431
44.0 RUS_Karelia_HG
29.0 DEU_LBK_KD
18.8 GEO_CHG
4.4 SRB_Iron_Gates_HG
3.8 IRN_Tepe_Hissar_C
Target: Yamnaya_RUS_Samara
Distance: 5.7262% / 0.05726211
56.0 RUS_Karelia_HG
32.2 GEO_CHG
6.0 DEU_LBK_KD
5.8 IRN_Tepe_Hissar_C
Target: Corded_Ware_Baltic
Distance: 3.8882% / 0.03888247
38.2 RUS_Karelia_HG
25.6 DEU_LBK_KD
21.4 GEO_CHG
14.8 SRB_Iron_Gates_HG
@Andrzejewski
We have Corded Ware genomes with R1a and a Yamnaya-like genome, so there’s not much reason to think that they were late CWC-like.
The best evidence for the unreliability of the unscientific guessing method is of course that it produces contradicting results. Compare for example the views of Eurogenes blogger Davidski and Indo-European.eu blogger Carlos Quiles: both of them use the same unscientific methods - they believe that they can see language from the genes.
1. Davidski sees Proto-Uralic in the so called South Siberian (Kra001) ancestry.
2. Quiles sees Proto-Uralic in the so called Corded Ware ancestry.
Is either of these guys even close to the truth? Time will show. At the moment all we can say is, we cannot trust any results achieved with such an untrustworthy method.
@ Anthony
I’m just reflecting about general process & absolutist perspectives sometimes seen (whilst most people are happy to sway with evidence); not exactly where & when certain lineages were
@Onno Hovers
CHG is being eaten up by TUR_Barcin_C because the models are overfitted (too many reference pops, and some with very strong overlapping).
You have to be careful with highly drifted populations, like many of the Siberian groups, because the algorithm will do whatever it can to reduce the distances.
@Jaakko
I'm going to remember these insults and I'll make you pay for this.
@Onno_Hovers,
Please continue digging into this. This (extra) EEF in Uralic is something I noticed for Baltic region too - early Estonian IA (the incoming N rich folk) had increased levels of farmer-ish when compared to local LVA_BA/EST_BA. Nothing crazy, but more than locals.
Where did they get it?
A) further East. Post Fatyanovo? Some Balto-Slavic? Such as Golden Horde Euro sample..
B) option that they assimilated local Germanic substrate near coast already in Baltic (post BA but pre Uralic). That is more theoretical.
I vaguely remember some other instances of increased EEF further East.
Davidski:
"I'm going to remember these insults and I'll make you pay for this."
What insults? It is you who always starts to insult people who disagree with you. Every now and then I play with you in your sandbox and return back to you the words you used.
Or do you mean my comparison of your method to that of Quiles? But it is the very same unscientific method! The longer it takes for you to understand this, the bigger your shame will ultimately become.
You should understand, that YOU win if you desert the unscientific guessing method. So, the countdown has begun. :) See you on this side of the reason.
@Jaakko
I understand that connecting a Y-chromasome to a language family wouldn't fly within your circle at the linguistics department.
This is primarily a genetics blog though, and the connection between Y-hg N and Uralic languages, is something that has been noted since the early 2000s, in many studies.
Trying to decouple N1c from Proto-Uralic requires, multiple founder effects/bottlenecks and language shifts. All of which, must have effected Uralic speakers specifically, across space and time.
It just doesn't seem feasible in the slightest.
@Jaasko
"Scientific method:
1. Let’s look at the genetic results.
2. Let’s look at the linguistic results.
3. Let’s look if there are in both of these any phenomenons which match concerning time, place and direction of spread.
Unscientific method of Davidski:
1. Let’s look at the genetic results.
2. Let’s ignore the linguistic results.
3. Let’s decide ad hoc that this genetic wave in certain place at certain time represents certain language."
1. This is not a linguistic forum. Here linguistics are not discussing. I personally try in every way to avoid linguistic themes because to discuss them from the opponent should be the basic preparation in linguistics, which you do not have because you are not a linguist and he should be adequate.
2. You are not familiar with linguistics, you are not a linguist, and completely ignore linguistics data. You did not even hear what Linguistic Palaeontology is and what results it gave for Finno-Samoyedic languages. A person is not familiar with Linguistic Palaeontology can not even talk about the ancestral homes of Finno-Samoyeds. It is a fact that you ignore linguistics.
3. There is no layer of borrowing some of your hypothetical NWIE into PFS, linguistics denies this, there are only Indo-Iranian borrowing into PFU and FU, there are borrowings from the Balto-Finnish CWC, (Proto-)Germanic, Baltic and Slavic circles into FU languages.
You are an enemy of the scientific method, so you ignore these linguistics and all that you write on them. No one ignores linguistics but you.
PЗ. You and your reaction are very easy to understand, you have been wrong all your life, you have written only mistakes, your life's work has gone down the drain, you are disproved on all counts, your views are dead.
Anthony Hanken:
“I understand that connecting a Y-chromasome to a language family wouldn't fly within your circle at the linguistics department.
This is primarily a genetics blog though, and the connection between Y-hg N and Uralic languages, is something that has been noted since the early 2000s, in many studies.”
Haplogroup N is a genetic object, and Uralic language is a linguistic object.
You cannot see hg N with linguistic methods, and you cannot see Uralic language with genetic methods.
You people here are not stupid, so surely you could understand this?
There is a step where genetic science ends and unscientific speculations begin. It’s a pity that this step is so difficult for people to see, but I’m here to help. True geneticists observe correlations between DNA and language, but they do not claim to see language from DNA - they know that it would be unscientific.
Anthony Hanken:
“Trying to decouple N1c from Proto-Uralic requires, multiple founder effects/bottlenecks and language shifts. All of which, must have effected Uralic speakers specifically, across space and time.
It just doesn't seem feasible in the slightest.”
Please feel free to present a reconstruction of the spread of N1c and Uralic language. If you can present a structure which matches considerably with the spread of Uralic languages on time, space, direction of spread and taxonomic units of the language family, I am ready to accept it. But if it doesn’t match, then the scientific method requires to find a better match.
I still remind you that we cannot exclude other haplogroups like N2 = N1b or R1a.
Archi, we all hope you will find a right medication for you.
Hugs and kisses!
@All
OK, so who wants to label these properly?
Harney_Cheronet_2021_G25.txt
https://drive.google.com/file/d/1E2KusMn1duD-kbtBBLla2JrgQ1jOjiGZ/view?usp=sharing
@Jaasko
Medication are useless for you.
“A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it.” (Max Planck)
Distance to: ROU_Glavanesti_BA:I11910
0.01993113 RUS_Sintashta_MLBA:I1053
0.02135486 Corded_Ware_CHE:MX188
0.02184399 Andronovo_RUS_Krasnoyarsk_LBA:I3395
0.02225511 Corded_Ware_CHE:MX194
0.02251155 Corded_Ware_CZE:I6696
0.02478447 Andronovo_RUS_Krasnoyarsk_LBA:I3394
0.02524044 RUS_Sintashta_MLBA:I1065
0.02528379 RUS_Sintashta_MLBA:I0989
0.02539094 Corded_Ware_CHE:MX199
0.02551431 Corded_Ware_CHE:MX195
0.02560215 RUS_Sintashta_MLBA:I1089
0.02571906 RUS_Sintashta_MLBA:I1024
0.02629277 Corded_Ware_CHE:MX190
0.02667414 Corded_Ware_CHE:MX198
0.02703627 Corded_Ware_CZE:I7280
0.02716505 Andronovo_RUS_Krasnoyarsk_LBA:I3392
0.02769856 Corded_Ware_CHE:MX191
0.02803890 Corded_Ware_Baltic:Plinkaigalis241
0.02827384 Corded_Ware_CZE:I6695
0.02833161 Corded_Ware_CHE:MX189
0.02835384 RUS_Sintashta_MLBA:I0984
0.02838767 Corded_Ware_CHE:MX192
0.02861189 RUS_Sintashta_MLBA:I1011
0.02867176 RUS_Sintashta_MLBA:I1086
0.02929078 BGR_MLBA:I2163
0.03051819 Yamnaya_UKR_Ozera_o:I1917
0.03068632 RUS_Sintashta_MLBA:I0987
0.03069528 Corded_Ware_Baltic:Kunila2
0.03074817 RUS_Karasuk:RISE499
0.03079075 RUS_Sintashta_MLBA:I1027
0.03092184 Corded_Ware_CZE:I7209
.....
0.04934227 RUS_Yana_UP:Yana2
0.04967062 Corded_Ware_DEU:I1544
0.04991713 Yamnaya_RUS_Samara:I0231
0.05053613 RUS_Sintashta_MLBA_o2:I1020
0.05179575 Andronovo_RUS_Krasnoyarsk_LBA_o:I6717
0.05194882 RUS_Kubano-Tersk:RK1003
0.05201500 RUS_Kubano-Tersk:PG2002
0.05213502 RUS_Kubano-Tersk:MK5009
0.05293109 RUS_Sintashta_MLBA:I1008
0.05302358 Yamnaya_RUS_Kalmykia:RISE240
0.05324584 Yamnaya_RUS_Kalmykia:RISE550
0.05424380 Corded_Ware_POL_early:poz81
0.05466013 RUS_Kostenki14_UP:Kostenki14
Distance to: ROU_Glavanesti_BA:I11913
0.02019752 Andronovo_RUS_Krasnoyarsk_LBA:I3395
0.02086768 Corded_Ware_CZE:I7279
0.02112676 RUS_Sintashta_MLBA:I0984
0.02127134 Andronovo_RUS_Krasnoyarsk_LBA:I6716
0.02166403 RUS_Sintashta_MLBA:I1024
0.02183323 Corded_Ware_CZE:I7207
0.02252643 Corded_Ware_RUS_Fatyanovo:NIK008AB
0.02255549 RUS_Sintashta_MLBA:I1086
0.02320840 RUS_Sintashta_MLBA:I1027
0.02340662 RUS_Sintashta_MLBA:I1089
0.02349064 Corded_Ware_CZE:I6696
0.02365312 RUS_Sintashta_MLBA:I1090
0.02366685 RUS_Sintashta_MLBA:I1065
0.02369451 Andronovo_RUS_Minusinsk_LBA:I1852
0.02402374 Corded_Ware_Baltic:Plinkaigalis241
0.02424830 RUS_Sintashta_MLBA:I1064
0.02427715 Corded_Ware_POL:N45
0.02451795 RUS_Sintashta_MLBA:I1053
0.02458272 RUS_Sintashta_MLBA:I0987
....
0.02814623 BGR_MLBA:I2163
......
0.03316489 Corded_Ware_RUS_Fatyanovo:NAU001
0.03321235 Corded_Ware_POL:N44
0.03361741 RUS_Sintashta_MLBA:I1088
0.03367670 Yamnaya_BGR:Bul4
0.03380104 RUS_Sintashta_MLBA_o2:I1057
0.03438052 Corded_Ware_Proto-Unetice_POL:RISE431
0.03442252 Corded_Ware_RUS_Fatyanovo:HAL001
0.03458237 Corded_Ware_RUS_Fatyanovo:HAN002
....
0.04434648 Corded_Ware_POL:N47
0.04442556 Yamnaya_RUS_Kalmykia:RISE547
0.04442837 Yamnaya_RUS_Samara:I0231
0.04480870 RUS_Sintashta_MLBA_o2:I1020
0.04525207 Corded_Ware_POL_early:poz81
0.04541222 Corded_Ware_DEU:I1534
0.04616135 Corded_Ware_DEU:I1544
0.04628542 RUS_Kubano-Tersk:PG2002
0.04650312 Yamnaya_UKR:MJ06
0.04729419 Yamnaya_RUS_Kalmykia:RISE550
0.04730053 Yamnaya_RUS_Kalmykia:RISE240
0.04786679 Yamnaya_RUS_Samara:I0439
0.04787055 Yamnaya_RUS_Caucasus:RK1001
0.04822510 Andronovo_RUS_Krasnoyarsk_LBA_o:I6717
Target: ROU_Glavanesti_BA:I11913
Distance: 0.8248% / 0.00824826 | R5P
46.2 RUS_Sintashta_MLBA
21.4 Yamnaya_UKR_Ozera_o
13.8 RUS_Okunevo_BA
12.8 Corded_Ware_DEU_o
5.8 BGR_Late_C
Target: ROU_Glavanesti_BA:I11910
Distance: 0.6064% / 0.00606378 | R5P
28.8 BGR_EBA
26.4 Corded_Ware_CHE
22.2 Yamnaya_UKR_Ozera_o
17.4 RUS_Sintashta_MLBA_o1
5.2 RUS_LyalovoVolosovo_N
ROU_Glavanesti_BA == ROU_Babino_MLBA (ROU_KMK_MLBA)
@ Davidski
OK, so who wants to label these properly?
https://genome.cshlp.org/content/early/2021/02/12/gr.267534.120/suppl/DC1
Supplemental_Table_S1.xlsx
About 1/3 of the Proto-Indo-European and Proto-Finno-Samoyedic vocabulary coincide, that is, they are cognates. These are not loan words, they are all listed.
Here is an entire publication of Indo-Uralic
https://brill.com/view/title/55752
The arguements are all very weak.
Its an ode to Kortlandt's ''Uralo-Sberian'' which includes PIE
''This puts the dissolution of the Uralo-Siberian language family in the 7th millennium. It now becomes attractive to identify the latter with the abrupt climate
change of 8200 BP or 6200 BC''
It would mean that PIE arrived with haplogrup Q1-rich populations
@Davidski
“@Jaakko
I'm going to remember these insults and I'll make you pay for this.”
It is too late. He has done a lot of damage already. It is irreparable.
Just imagine what interesting discussions we could have had at Biodiversityforum with Constantine Borissoff after Narasimhan et al. 2019 paper. It is all lost. There is no point in discussing with closed minds like Jaska or Carlos, it is a waste of time. Borissoff was right but didn’t fight and left, after that we lost interest in discussing clearly wrong Western European and American point of view on the history of languages and people. Many of these idiots still believe that Slavs came with Huns and the lack of East Asian component in our DNA is a big problem for them. At that time I already knew what papers like Narasimhan et al. 2019 would show.
1/3 ?
Hmm, a/p Nichols
''Out of a pre-set wordlist of 200 items, 8 matches with identical semanJcs are required to exceed chance; out of 1000 items (an esJmate of the total number of roots likely to be reconstructed for a welldescribed protolanguage in the best cases), 28; (out of 2000 items (the typical number of elementary roots found in a well-described non-reconstructed language), 51; with similar but not necessary idenJcal semanJcs, the numbers required are 19, 73, and 121 (for these calculaJons see Nichols 2010).17 We use
the Swadesh wordlist as a convenient pre-set wordlist, though of course it is a beeer comparison for hypotheses of shared descent than for loan vocabulary. Table 3 shows these frequencies and the frequencies needed to exceed chance. Koivulehto's words, though excellent matches taken individualy, are not numerous enough to exceed chance.''
Thousands of yers of language contact can induce not only borrowings, but borrowings of 'fundamental' words and deep morpho-synctactic changes in the target language
If only family-tree method is applied, strange results can be offered. E.g. Warnow's 'Albano-Germanic' branch within I.E.
But population history negates this node. In fact, this is clearly due to a ~ 150 year period of language contact, when Germanic merceneries were known to have been stationed in Dacia ripensis, Dardania. etc
@Eastpole: well well, isn't it yet another ignorant troll from the sewers of internet.
Eastpole:
“It is too late. He has done a lot of damage already. It is irreparable.
Just imagine what interesting discussions we could have had at Biodiversityforum with Constantine Borissoff after Narasimhan et al. 2019 paper. It is all lost.”
Who has done what damage? I had nothing to do with your comment. Your paranoia is not an argument.
Eastpole:
“There is no point in discussing with closed minds like Jaska or Carlos, it is a waste of time. Borissoff was right but didn’t fight and left, after that we lost interest in discussing clearly wrong Western European and American point of view on the history of languages and people.”
Again, paranoia is not an argument.
Eastpole:
“Many of these idiots still believe that Slavs came with Huns and the lack of East Asian component in our DNA is a big problem for them. At that time I already knew what papers like Narasimhan et al. 2019 would show.”
Nice strawman, I personally have never seen anybody present views like this.
Nichols and Koivulehto are not authorities. Swadesh WordList has nothing to do with the general vocabulary. It does not establish relatives.
Indo-Uralic theory is strictly proved. The time of the disintegration of these two languages is the question of relative, on such antiquities, the GLOTOHronology works badly, too few words and too large systematic error, so the decay time is estimated between 10000 and 7000BС with different calculation algorithms. It must be borne in mind that for a long time after PIE penetration to Europe between Europe and Western Siberia continued continuum dialects. I am sure that at this time R1A live there.
------
MOL_Ciumai:I11922-I11928 are Sarmatians.
I11902 Urziceni, Romania, Grave 27 6,300-6,050
I11903 Urziceni, Romania, Grave 27 6,300-6,050
I11904 Urziceni, Romania, Grave 27 6,300-6,050
I11905 Urziceni, Romania, Grave 27 6,300-6,050
I11906 Urziceni, Romania, Grave 56 6,300-6,050
I11907 Urziceni, Romania, Grave 56 6,300-6,050
I11910 Glăvăneşti, Romania, Grave 1, M1 5,450-3,050
I11913 Glăvăneşti, Romania, Grave 1(49), M11 5,450-3,050
I11914 Glăvăneşti, Romania, Grave 1(49), M11 5,450-3,050
I11915 Glăvăneşti, Romania, Grave 1(49), M11 5,450-3,050
I11922 Cimișlia, Rep. of Moldova, Tumulus 8, Grave 7 2,050-1,850
I11923 Cimișlia, Rep. of Moldova, Tumulus 8, Grave 7 2,050-1,850
I11924 Cimișlia, Rep. of Moldova, Tumulus 8, Grave 7 2,050-1,850
I11925 Cimișlia, Rep. of Moldova, Tumulus 8, Grave 7 2,050-1,850
I11926 Ciumai, Rep. of Moldova, Tumulus 1, Grave 10 4,000-1,000
I11927 Ciumai, Rep. of Moldova, Tumulus 1, Grave 10 4,000-1,000
I11928 Ciumai, Rep. of Moldova, Tumulus 1, Grave 10 4,000-1,000
I11929 Polgár-Ferenci-hát, Hungary, 717/1021 7,280-7,035
I11931 Polgár-Ferenci-hát, Hungary, 717/1021 7,280-7,035
I11932 Polgár-Ferenci-hát, Hungary, 717/1021 7,280-7,035
I11933 Polgár-Ferenci-hát, Hungary, 715/1022 7,280-7,035
I11934 Polgár-Ferenci-hát, Hungary, 715/1022 7,280-7,035
I11935 Polgár-Ferenci-hát, Hungary, 715/1022 7,280-7,035
I20743 Kesznyéten-Szérűskert, Hungary 2,600-2,400
I20744 Kesznyéten-Szérűskert, Hungary 2,600-2,400
I20745 Kesznyéten-Szérűskert, Hungary 2,600-2,400
I20746 Kesznyéten-Szérűskert, Hungary 2,600-2,400
I20749 Hajdúdorog-Szállásfold, Hungary 3,700-2,800
I20750 Polgár Kenderföld, Hungary 4,300-3,600
I20752 Besenyszög Berek-ér partja, Hungary 2,250-2,150
I20753 Dereivka, Ukraine 8,392-7,927
I20755 Ekven, Russia 1,400-900
I20756 Ekven, Russia 1,030-790
I20757 Ekven, Russia 1,380-1,010
I20758 Uelen, Russia 1,100-750
I20759 Ekven, Russia 1,310-930
I20760 Ekven, Russia 6,350-6,260
I20761 Ust Belaya, Russia 4,840-4,490
I20762 Volosovo-Danilovo, Russia 4,000-2,000
I20765 Kesznyéten-Szérűskert, Hungary 2,600-2,400
I20766 Kesznyéten-Szérűskert, Hungary 2,600-2,400
I20768 Kesznyéten-Szérűskert, Hungary 2,600-2,400
I20771 Hajdúdorog-Szállásfold, Hungary 3,700-2,800
I20772 Polgár Kenderföld, Hungary 4,300-3,600
I20774 Besenyszög Berek-ér partja, Hungary 2,250-2,150
I20775 Dereivka, Ukraine 8,392-7,927
I20777 Ekven, Russia 1,400-900
I20778 Ekven, Russia 1,030-790
I20779 Ekven, Russia 1,380-1,010
I20780 Uelen, Russia 1,100-750
I20781 Ekven, Russia 1,310-930
I20782 Ekven, Russia 6,350-6,260
I20783 Ust Belaya, Russia 4,840-4,490
I20784 Volosovo-Danilovo, Russia 4,000-2,000
@Onno Hovers “
Target: Khanty
Distance: 4.3390% / 0.04338954
44.6 RUS_Krasnoyarsk_BA
23.6 KAZ_Botai
17.8 RUS_Karelia_HG
7.8 DEU_LBK_KD
5.2 TUR_Barcin_C
1.0 GEO_CHG”
RUS_Kranoyarsk_BA is WSH/Yamnaya
@Onno Hovers “ Target: RUS_Sintashta_MLBA
Distance: 4.9184% / 0.04918431
44.0 RUS_Karelia_HG
29.0 DEU_LBK_KD
18.8 GEO_CHG
4.4 SRB_Iron_Gates_HG
3.8 IRN_Tepe_Hissar_C
Target: Yamnaya_RUS_Samara
Distance: 5.7262% / 0.05726211
56.0 RUS_Karelia_HG
32.2 GEO_CHG
6.0 DEU_LBK_KD
5.8 IRN_Tepe_Hissar_C
Target: Corded_Ware_Baltic
Distance: 3.8882% / 0.03888247
38.2 RUS_Karelia_HG
25.6 DEU_LBK_KD
21.4 GEO_CHG
14.8 SRB_Iron_Gates_HG”
Hogwash!!!
You have just contradicted yourself, by quoting WSH IE Yamnaya, CWC and Sintashta with an EHG ratio almost TWICE that of CHG!!!!
The Sintashta one (outlier?) has twice as much EEF > CHG, and almost 3 times EHG > CHG.
Very strange indeed.
@Parastasis “ Please continue digging into this. This (extra) EEF in Uralic is something I noticed for Baltic region too - early Estonian IA (the incoming N rich folk) had increased levels of farmer-ish when compared to local LVA_BA/EST_BA. Nothing crazy, but more than locals.
Where did they get it?”
Of course they would have EEF admixture!
First, contacts with CWC/Sintashta/Indo-Iranians during BA and later. Iron Age Scythians. Throughout historical period. And East Slavs during Middle Ages.
Therefore, Mordvins, Mari El and other Finno-Permic Eastern European dwellers of modern day Russia have lots of EEF.
By the same token, many Turkic speakers carry R1a (and to a lesser extent R1b) and they have aDNA of assimilated Iranic tribes such as Sarmatians. They are Turkified IE, hence cruet carry significant chunk of ANF/EEF. Like the modern Tatars.
Andrzejewski said...
"RUS_Kranoyarsk_BA is WSH/Yamnaya"
You are again carrying the lies and delirium of a madman. Stop trolling.
Target: RUS_Krasnoyarsk_BA:kra001
Distance: 3.6196% / 0.03619559 | R5P
94.0 RUS_Yakutia_LN
2.6 RUS_AfontovaGora3_UP
2.6 RUS_Yana_MA
0.8 FIN_Levanluhta_IA
NO WSH/Yamnaya, it is far distance to WSH/Yamnaya
I have been waiting for @Archi to do us all a favor, and quote a legitimate study by a well-known scientist to prove his Nostrastic linguistic theory. But so far, he couldn’t even bring up Kortlandt.
@Andrzejewski
RUS_Kranoyarsk_BA is WSH/Yamnaya
How can you write it right after Davidski just wrote two posts featuring this sample and showed that it is almost completely East Asian and in fact particularly close to Ngannassan. Have you not paid attention?
In a WHG + AG3 + BarcinN + CHG + Devils Gate G25 nMontes Krasnoyarsk BA is something like 9:1 East Asian: ANE, with zero WHG or CHG ancestry.
@ Andrzejewski
I do not want to discuss any linguistics here with people who do not understand anything about it. You can't understand anything, you can't even go to Wikipedia https://en.wikipedia.org/wiki/Nostratic_languages#Bibliography and look at online dictionaries (Indo-European - Uralic).
@Archi
I think Andrzejewski just confused Kra001 for the Krasnoyarsk steppe_mlba samples there.
I hope so at least.
@Davidsky: "CHG is being eaten up by TUR_Barcin_C because the models are overfitted (too many reference pops, and some with very strong overlapping). You have to be careful with highly drifted populations, like many of the Siberian groups, because the algorithm will do whatever it can to reduce the distances."
Even when I remove Anatolian ancestry and limit myself to 5 source populations, I still see EHG+EFF and not enough CHG:
Target: Khanty
Distance: 4.4585% / 0.04458489
44.4 RUS_Krasnoyarsk_BA
27.4 KAZ_Botai
14.0 RUS_Karelia_HG
11.2 DEU_LBK_KD
3.0 GEO_CHG
Target: FIN_Levanluhta_IA
Distance: 4.1239% / 0.04123882
48.6 RUS_Karelia_HG
23.0 RUS_Krasnoyarsk_BA
22.2 DEU_LBK_KD
6.2 GEO_CHG
@Parastais: "Please continue digging into this. This (extra) EEF in Uralic is something I noticed for Baltic region too - early Estonian IA (the incoming N rich folk) had increased levels of farmer-ish when compared to local LVA_BA/EST_BA. Nothing crazy, but more than locals.
Where did they get it? ... I vaguely remember some other instances of increased EEF further East."
Thanks! I'm not sure if I am right about this. But I am glad that I'm not the only person who hears this bell ringing. Davidsky does have a point in that it is hard to distinguish between EHG+EEF versus Corded Ware.
@All “
It might be a good idea to also have a detailed look at the Western ancestry in Uralic populations. It looks to me like Uralic populations have extra EHG beyond what they would get from Steppe.”
Bullscraps!
Why are so many people wedded to the idea that Uralics have ANYTHING to do with EHG?
I recall reading on Europedia years back that PIE was allegedly created as a “creole” language because of the Uralic-like elements admixed with Kartvelian-related ones....
Even Carlos Quiles couldn’t make this stupidity up!
So for starters, here are the facts:
1. Basal Hap Y-N is originally from Northern China, a subclades of O-N.
If anything, Uralic’s prehistoric forebears may have had something to do with the East Eurasian side of the Native Americans (“Devil’s Gate) than with anything resembling MA1 or AG3. The WSH had 50% ANE.
2. I understand that Nganassan-like populations have expanded from Western Siberia across the Urals into Eastern Europe, but many of these populations (Bolosovo, BOO, etc) were expelled when Abashevo and its offsprings spread eastbound into Asia from the Forest Steppe.
3. The EHG components were either instrumental in forming WSH, got expelled, wiped out or even (slightly) subsumed by IE.
4. I would put more trust in the Turanic (“Ural-Altaic”) theory than in anything remotely resembling “Indo-Uralic”.
5. Even Anthony (2019) thinks that the Proto-IE were somewhat of a 4-way melange of WHG, ANF, EHG and CHG, with different proportions and genetic drifts vis-a-vis most modern Europeans.
5. I now strongly believe that Pre-Proto-whatever-PIE or whichever speech the Vanyuchka and Progress chose to speak was a language isolate that had nothing to do with either EHG, CHG, ANF or WHG per-se. And every demos with carrying the WSH across the Steppe since 5500BCE spoke a language directly descended from it. Just like Elamite was a language isolate and could not be satisfactorily explained away as an Iran_Neo/Chl one; Basque having same ancestry components with Villanovan Etruscans but speaking mutually unrelated languages, and the list goes on.
@Dadiski
CHG is being eaten up by TUR_Barcin_C because the models are overfitted (too many reference pops, and some with very strong overlapping).
nMontes does not show any CHG there even if you use Barin_N instead of Barcin_C. The populations being extremely drifted is a good point however.
There are a few G25 scaled nMontes runs on Vahaduos's. The fit is horrible for moderns and especially so for Krasnoyarsk_BA. Some of the results are hard to believe. But the problem is _not_ overfitting. This is something else.
Target: RUS_Tyumen_HG:I1960
Distance: 0.04270952
GEO_CHG: 0.0
ITA_Grotta_Continenza_Meso: 1.0
RUS_AfontovaGora3: 89.8
RUS_Devils_Gate_Cave_N: 9.2
TUR_Barcin_N: 0.0
Target: KAZ_Botai
Distance: 0.03814016
GEO_CHG: 0.4
ITA_Grotta_Continenza_Meso: 1.0
RUS_AfontovaGora3: 82.6
RUS_Devils_Gate_Cave_N: 15.4
TUR_Barcin_N: 0.6
Target: RUS_Krasnoyarsk_BA
Distance: 0.20789474
GEO_CHG: 0.0
ITA_Grotta_Continenza_Meso: 0.0
RUS_AfontovaGora3: 13.0
RUS_Devils_Gate_Cave_N: 87.0
TUR_Barcin_N: 0.0
Target: Khanty
Distance: 0.11762305
GEO_CHG: 0.0
ITA_Grotta_Continenza_Meso: 0.0
RUS_AfontovaGora3: 46.2
RUS_Devils_Gate_Cave_N: 42.6
TUR_Barcin_N: 11.2
Target: Ket
Distance: 0.12845948
GEO_CHG: 0.0
ITA_Grotta_Continenza_Meso: 0.6
RUS_AfontovaGora3: 44.2
RUS_Devils_Gate_Cave_N: 43.0
TUR_Barcin_N: 12.2
@ Davidski
I can take care of naming them, but it'll be tricky as 2-3 samples correspond to a single individual, and some of them came out quite differently.
We can discard redundant samples basing on the supplement data about coverage/contamination or using "modelling" with full spreadsheet and picking up the ones with a lower distance to the whole G25 dataset (or ancients only, as contaminated ones should have a pull towards moderns).
It would be easier to decide if you'd publish how much overlap is with G25 SNPs set (it can be rounded to e.g. 5% if you don't want to reveal too much).
But some hard choices will stay as with I20744 and I20766 which have different coordinates but nearly identical coverage according to the supplement.
In such case we can make an experiment with merging genotypes by "calling" a random allele at each position from the redundant samples.
It's your PCA, so you need to decide what you want to keep in it. I can do everything else.
The main differences between Eastern and Western Europeans mostly stem from a couple of factors: Eastern Europeans overwhelmingly have R1a1 and they speak Satem languages, an offshoot of Corded Ware. And their Anatolian part is pre-eminently LBK-descended, largely GAC with sprinkle of CTC.
West Euro’s ancestors come from the SGC within the Corded Ware periphery, which became the Bell Beakers. Their uniparental markers are predominantly R1b in lieu of R1a1. Their language subfamilies are Centum, not Satem; and they have an *additional* significant Anatolian portion, which mostly comes from Atlantic Farmers population and thus is rich in or even majority thereof being Cardial Pottery instead of LBK sourced.
Post a Comment