search this blog

Sunday, December 1, 2019

Big deal of 2019: ancient DNA confirms the link between Y-haplogroup N and Uralic expansions


The academic consensus is that Indo-European languages first spread into the Baltic region from the Eastern European steppes along with the Corded Ware culture (CWC) and its people during the Late Neolithic, well before the expansion of Uralic speakers into Fennoscandia and surrounds, probably from somewhere around the Ural Mountains.

On the other hand, the views that the Uralic language family is native to Northern Europe and/or closely associated with the CWC are fringe theories usually espoused by people not familiar with the topic or, unfortunately it has to be said, mentally unstable trolls.

The likely close relationship between the CWC expansion and the early spread of Indo-European languages was discussed in several papers in recent years (for instance, see here). This year, we saw the first ancient DNA paper focusing on the transition from the Bronze Age to the Iron Age in the East Baltic, including the likely first arrival of Uralic speech in what is now Estonia.

Published in Current Biology courtesy of Saag et al., the paper showed that the genetic structure of present-day East Baltic populations largely formed in the Iron Age (see here). It was during this time, the authors revealed, that the region experienced a sudden influx of Y-chromosome haplogroup N, which is today common in many Uralic speaking populations and often referred to as a Proto-Uralic marker. Little wonder then that Saag et al. linked this genetic shift in the East Baltic to the westward migrations of early Uralic speakers.

The table below, based on data from the Saag et al. paper, surely doesn't leave much to the imagination about what happened.


Unfortunately, I have to say that the genome-wide analysis in the paper was less informative than it could have been. The authors focused their attention on rather broad genetic components, and, as a result, missed an interesting fine scale distinction between their Bronze Age and Iron Age samples. The spatial maps below, based on my Global25 data for most of the ancients from Saag et al., show what I mean. The hotter the color the higher the genetic similarity between them and present-day West Eurasian populations.

Note that the Bronze Age (Baltic_EST_BA) samples are most similar to the Baltic-speaking, and thus also Indo-European-speaking, Latvians and Lithuanians, rather than the Uralic-speaking Estonians, even though they're from burial sites in Estonia. On the other hand, the Iron Age (Baltic_EST_IA) samples show strong similarity to a wider range of populations, including Estonians and many other Uralic-speaking groups.




See also...

It was always going to be this way

Fresh off the sledge

More on the association between Uralic expansions and Y-haplogroup N

327 comments:

1 – 200 of 327   Newer›   Newest»
Davidski said...

Interesting map for OLS10, eh? I wrote about this guy here...

https://eurogenes.blogspot.com/2019/05/fresh-off-sledge.html

Sofia Aurora said...

Guys just a question please.

I am trying to enter the Degruyter online site in order to check out news about its journals but it gives out a message

403 Forbidden
request forbidden by administrative rules

What the hell is that?
I was trying to see the latest issue of open archaeology, Baltic-Pontic studies etc. and it sends a message that the site is not safe and to press on a specific sentence if i want to continue!

When i do that it gives out the 403 Forbidden message!!!

It did the same when i visited the Brill site in order to see the last issue of Indo-European issues and generally it does not allow me to access the Brill online and the Degruyter online!

All the other sites work fine!

Does this happen to you too or it's just me?

Mauri said...

I am not sure about the eastern/Volga route of the Baltic Finnic languages, but more northern origin through the Ladoga region is possible. Hopefully Russian researchers in future have success and are able to find something in one way or another. Moving from Volga to Estonia sounds unlikely.

Vinitharya said...

Oh no, Carlos! Alexa, play Despacito. Truth always wins out.

Queequeg said...

@ D and re "The authors focused their attention on rather broad genetic components, and, as a result, missed an interesting fine scale distinction between their Bronze Age and Iron Age samples." Could you kindly elaborate what are those "fine scale distinctions" you're referring to?

Kristiina said...

@ Davidski Thank you for this very interesting post. It is very enlightening.

So, now we are anxiously waiting for the dearly cherished theory of the Volgaic origin of the Uralic languages being possibly debunked?

I will accept the picture as it emerges from the ancient DNA data. In any case, I do not think that the Proto-Uralic language could have been spoken in a culture if it yields next to 0% of haplogroups carried by modern Uralic populations.

Anthony Hanken said...

A rereading of Tambets et al. 2018 may be interesting in light of this paper and rumours. In it they offer two archealogical models for the spread of N3a into northeast Europe, which they link with Uralic languages. Comb-Ceramic/Volosovo favouring an older age estimate of PU and Seima-Turbino favouring a younger sge estimate.

In the picture of unpublished Reich samples it looks like there is a large cluster just east of the Urals in the forest-steppe. If these are ~2000BC they would likely be from the Krotov culture and maybe N-L1026. I know the SUGRIGE project is working with Reich's lab so maybe they have a better understanding of things than we currently do regarding unpublished data.

Either way the Estonian IA samples definitely seem to support a demic diffusion of N-L1026 along with Uralic languages as the Tambets paper suggested, especially considering their relationship to modern Uralic speakers.

Mauri said...

@Kristiina,

you write about Volgaic origin of Uralic languages. I wonder how can it be evaluated by Estonian IA samples. Do you assume a tour via Estonia to Ural?

Vladimir said...

@Anthony Hanken. Second option. In Central Siberia (before the arrival of the Andronov Indo-Iranian culture R1a-Z2124) it is a Krotov culture. After the arrival of the Andronov culture, it is a syncretic Samus culture (Western Siberia 3500-2000 BC). At the same time there were two crossings through the Urals. First, the tribes of Garino-Bor culture (Y9022) passed. Further back in Siberia, the M2019 subclades separated from the second group. Forth transition through Ural apparently makes L1026. In the Volga region, this group is the Chirkov culture it is also the Seimin-Turbin culture. Here one group Z1936 goes North, forming the Ananyin culture, and another group goes West, assimilating the remnants of the Fatyanovo culture R1a-?, Abashev and Srubnaya culture R1a-Z2124, Volosov culture I2a, Pozdnyakov culture R1b-Z2103, Lola culture G2A or J2 ???, as well as R1b-M73, the result is a syncretic culture of mesh ceramics ( CTS10760).

Vladimir said...

By the way, these assimilated groups could not go to the Baltic, and remain in the Volga region. The structure of the gene pool of Mordvins (Fino-Ugric people): N1a-CTS10760-20%, R1a-Z2124-20%, I2a-20%, R1b-Z2103 -15%, R1b-M73 -10%, J2a - -10%

Kristiina said...

@ Do you assume a tour via Estonia to Ural?

No, it would be too convoluted. I only mean that if there are no Uralic yDNA haplos in the Neolithic/Comb Ceramic Volga Bend and the Corded Ware only brings R1a1a1-M417, there is not much justification for Proto-Uralic.

If you take a look at fig. 1 in M.G. Kosmenko, "The culture of Bronze age Net Ware in Karelia", you see that the Net Ware more or less overlaps with David's second and third map which shows the highest sharing of ancestry with Estonian Iron Age genomes. Moreover, the earlist area of expansion of Net Ware south of the Lake Ladoga corresponds to the highest concentration of Estonian Iron Age ancestry on the map. This core area is the historical Veps area: https://www.wikiwand.com/en/Vepsians

"Net Ware formed during the second half of the second millennium BC in an area bounded by the Upper Volga in the south, Lake Onega in the north, the upper reaches of the River Sukhona in the east … Later, during the first half of the first millennium BC, it spread over a wide area, extending to the coasts of the Baltic and the White Sea, along the middle reaches of Volga to the mouth of the River Kama, and to a lesser extent southwards to the basin of the River Kama. From the middle to the second half of the first millennium BC it ceased to exist independently and merged with a new wave of Early Iron Age culture of the so called Ananyino type that spread over the vast territories of the forest zone to the north of the Volga between the Ural Mountains and Baltic Sea."

Ananyino culture is dated to the late 8th to 3rd centuries BCE in present-day Tatarstan. Net Ware starts to spread c. 1500 BC.

In any case, I am not that much expecting to see N-L1026 in Krotovo. IMO, other N clades are more probable. Moreover, from the genetic point-of-view, Krotovo genomes will hardly be a good match with the Iron Age Estonians.

Davidski said...

@Huck Finn

Baltic_EST_BA is similar to Balts and relatively dissimilar from Uralic speakers.

Baltic_EST_IA is also similar to Balts, but also to a wide range of Uralic speakers, especially Estonians.

OLS10 doesn't appear to be exceptionally similar to any extant population, but is relatively similar to a lot of groups from Uralic and former Uralic speaking regions from the Baltic to the Volga.

Mauri said...

@Davidski,

if someone of those ancient Estonian samples spoke Uralic it was sure OLS10. All other N's are questionable. If we look at the new Finnish study about mitochondrial groups in Iron Age Southwestern Finland, it looks like there was a strong connection between them and Estonian Late Bronze Age stone-cist graves, still in Finland during Vendel and Viking eras, but not between them and later Estonian burials. So we don't know if the later burial context between SW Finns and Estonians (Tarand type) was more a cultural adoption. Until now Finnish research has not revealed corresponding ancient N subclades. It is possible that ancient Ingrians were genetically closer Tarand Estonians, but we need more evidences to understand this connection and linguitic probability.

Mauri said...

@Kristiina, thank you. Sounds credible.

Davidski said...

@M. Myllylä

The people from the Estonian stone-cist graves were Balts genetically.

So if their descendants spoke Uralic, then it's not because they spoke it, but because their descendants learned it from someone else. Probably from OLS10 and his relatives from the east who used Tarand graves.

But in any case, I find it extremely unlikely that the unadmixed descendants of the stone-cist grave people still lived in Estonia and surrounds during the Iron Age.

Eastern admixture is seen in the Y-DNA and autosomes of the Iron Age samples from Finland and Ingria, and even in the mtDNA lineages that you speak of.

Where do you think the U1b2 in Hittola came from? The stone-cist people? Nope.

Anthony Hanken said...

@Kristiina

Why don't you dont think there will be N-L1026 in the Krotov culture? If there is no N in Volosovo there likely won't be in Garino-Bor either unless its been mediated through the Urals. Krotov artifacts have even been found in Garino-Bor sites like Turbino.

The oldest Tarands are over a thousand years younger than the beginning of Krotov/S-T materials spreading. I don't really expect the Proto-Uralic population to be significantly like the Esontian Tarands anyway, just related.

Anthony Hanken said...

Pretty much every forest culture from an archealogical perspective west of the Urals, seems to decend from the same sub-neolithic hunter/fishers of the Laylovo ,Upper-Volga and Comb Ceramic complex of cultures. Probably with a mix of Y haplos like R1b, R1a, Q, I2 and maybe J.

This doesn't really change until the spread of Seima-Turbino from the east and cultures infleunced by S-T like Garino-Bor and later Netted-Ware. This IMO is a good case for N-L1026 coming from west Siberia or possibly in a Krotov influenced site like Turbino, but either way closly related to a group east of the Urals.

Mauri said...

@Davidski,

Vendel era SW Finns were sure mixed, but not as much as today.

Hiitola samples are too young to be comparable to Iron Age SW samples. Översti et al makes vaguenesses and I am not only one saying this, many Finnish professionals share this with me. Maybe Översti et others felt too much pressure and had to say something neat.

Kristiina said...

@ Anthony

Why should Indo-European people be genetically significantly like their linguistic ancestors, but Proto-Uralics would be completely different?

Moreover, to argue that "If there is no N in Volosovo, there likely won't be in Garino-Bor either unless its been mediated through the Urals" does not make much sense. IMO, if there is no N in Volosovo, there can very well be N in Garino-Bor. And if there is N in Garino-Bor, there is no need that this N was (recently) mediated through the Urals. Haplogroups fluctuate a lot in time and space.

In any case, my prediction is that Krotovo genomes will be a mixture of Karasuk, WSHG, Glazkovo and Steppe. The Finnic Siberian is not from Altai. It is from a Saami like population in Finland and its northern surroundings.

My prediction is that Krotovo will yield a lot of Q and R1a1 and some N(xL1026).

Rare root types of L1026 have been found in the following populations: Karelians, Vepsans, Arhangelsk Russians, Pinega Russians, Latvians, Lithuanians, Belgorod Russians, Komi Permians, Tatars, Khantys and Dolgans. The origin of L1026 will probably be somewehere in between the areas inhabited by these people.

Davidski said...

@M. Myllylä

Well, we've got the stone-cist grave people from Estonia who look like extreme Latvians and with a frequency of R1a of 100%.

And then we've got all of the later folks from Iron Age Estonia, Ingria and Finland, who always show something from the east in their Y-DNA, mtDNA and/or autosomes.

So the big picture is clear. Uralic languages and DNA came to the Baltic after the stone-cist grave people.

Mauri said...

Tuukkala and Hiitola samples are about from the same era with my oldest known ancestors (gelealogically known). Översti could have dug up thousands samples almost everywhere in Finland.

Anthony Hanken said...

@Kristiina

I don't think a PU genome would be completely different from IA Tarands however they certainly won't be the exact same. The fact that IE groups share a majority of their DNA with their lingustic ancestors is true but it is not 1:1 and the ratio can vary. With Uralic groups we are dealing with smaller populations which can obviously lead to a larger change in overall genome over time.

Garino-Bor is an offshoot of Volosovo, it is still derived from the same sub-neolithic Volga-Kama cultures which I don't think any N will be found in. Any diffrences genetically between the two would have to be because of gene flow from the east.

In regaurds to modern populations S-T spread extremely fast both east and west which could explain its sporadic distribution.

Queequeg said...

Re D's "Where do you think the U1b2 in Hittola came from?": I wouldn't like to play devils advocate here but the location of Hiitola in Översti's PCA makes me also wonder. Why for instance those samples are so close to IA Poland and Viking Age Sweden? Karelian Hiitola after all is located in Ladoga area, which was the starting point of first Rus' state and a real commercial Viking Age hot spot. So, my question is: to what extent these "Neolithic Farmer" mtdna lineages represent the original local Finnic gene pool? Based on loan words those early Karelians possibly assimilated in IA a quite sizeable pool of Slavic speakers. If the nature of these words is of any guidance, they were Slavic speaking women.

Davidski said...

@Huck Finn

U1b2 isn't a Slavic marker. It's from the Caucasus and surrounds. There weren't any Slavs there at that time.

It's also debatable whether the Polish IA samples are Slavs. They're probably Goths or something like that.

Mauri said...

@Davidski

"Well, we've got the stone-cist grave people from Estonia who look like extreme Latvians and with a frequency of R1a of 100%.

And then we've got all of the later folks from Iron Age Estonia, Ingria and Finland, who always show something from the east in their Y-DNA, mtDNA and/or autosomes.

So the big picture is clear. Uralic languages and DNA came to the Baltic after the stone-cist grave people."

Of course, Uralic languages are young in Baltic sea area, probably only 2000 years old, Saami a bit older. But the mitochondrial message tells about older story, and Översti et al. used it, so I made some conclusions too. Making conclusions about genetic history of any areal population needs also to fit it to the present and there are obvious and admitted shortages now.

Davidski said...

@All

Any thoughts about this? Basically, Proto-Uralic as a language of the fur road...

A diachronic linguistic geography for Uralic

Davidski said...

@M. Myllylä

Do you have a PLINK file of the samples from this Icelandic study?

https://science.sciencemag.org/content/360/6392/1028

The versions I have show a lot of aDNA damage and pseudo-African admix.

Samuel Andrews said...

@Davidski, I think you overrate that single U1b2 found in Medieval Finland. Weird haplogroups exist in every population. It could be from Steppe pops considering they were 40% CHG. The main thing about that Medieval Eastern Finland mtDNA is they for sure Finnic and direct ancestors of Karelians.

Rob said...

Nichols outlines a spread of Samoyedic toward the East, but places her putative FU homeland in cis-Altai region on her imaging. Doesnt seem congruent

Davidski said...

@Samuel Andrews

It's not just U1b2 (of which there are two different haplotypes in the Hittola data), but also T1a1+@152 and T2d1b1 that are present in very small sample sets from ancient Northeastern Europe.

The last one is also found in ancient DNA from kurgans in West Siberia associated with Ugric speakers.

Of course, there may be different explanations for each of these unusual mtDNA haplotyes, like, say, Circassian slaves belonging to U1b2 being taken up to Lake Ladoga by Vikings. But one possible explanation is that they're remnants of contacts between early Uralic speakers and late steppe groups near the Urals.

Gabriel said...

Just wondering, Finns are equally or less similar to these samples than they are to Scandinavians, right?

Samuel Andrews said...

You don't have to be an expert on Uralic studies to know: Y DNA N1c and Uralic languages are from Asia (Siberia) not Europe. Seems people avoid acknowledging this.

I'd say both ultimately trace back to a population with no European ancestry at all (meaning nothing with WHG in it). You guys seem to think it is likely the actual proto-Uralics who spread the language had mixed European, Asian ancestry. That makes sense. But, the ancestor of proto-Uralic definitely is from an Asian population.

If Bolysho-Oleni doesn't qualify as Uralic according to expert archeaology and lingustics, they were definitly most derived Asian population who the proto-proto Uralics came from. But you really have to think they may have been Uralic.

Because, what was this mostly Asian population with Y DNa N1c doing so far deep in Erope, if they didn't have something to do with Uralic languages? You have to consider Bolysho-Oleni was proto-Saami.

Davidski said...

@Samuel Andrews

The Bolshoy Oleni Ostrov site is very mysterious. No one knows who these people were and where they came from.

But yeah, looking back, I can't see any convincing arguments why they couldn't have been Uralic speakers.

Vladimir said...

@Samuel Andrews. “ You have to consider Bolysho-Oleni was proto-Saami.” Apparently all so. But the big deer island is R1a-YP1272. Confuses that its sister line R1a-M198 appeared so far, in Mesolithic of Ukraine. Where they parted, that's the question. It is logical that, too, somewhere in Siberia.

Vladimir said...

@Davidski. According to the Russian researcher of Fino-Ugric languages Napolsky, based on ancient toponymy, under the layer of Fino-Ugric languages lies an unknown, but definitely not Fino-Ugric language. He calls it paleo European. One can assume that it is Fino-Ugric only in one case, accepting the theory of Carlos Quilis that the pit-comb pottery culture was the R1A-M459 culture. Then we can assume that haplogroup N1a1, arriving in Europe adopted an autochthonous language. However, this does not happen in life, the conqueror always imposes his language, and hardly N1a1 were an exception. Looking at the modern Finns is not think, but reading the appearance of the Volga tribes Chirkov culture becomes clear that it was a well-armed tribes, assimilated on its way all the tribes that existed in the Volga region before 2000 BC. Not just the same as the first kings of Rus and Lithuania were haplogroup N1a1.

Davidski said...

@Vladimir

The R1a sample from the big deer island isn't related to the samples from Bolshoy Oleni Ostrov.

There's a huge genetic difference between them.

The R1a sample is EHG, with no Siberian ancestry, while the samples from Bolshoy Oleni Ostrov have around 50% ancestry from a population related to Nganasans from Siberia.

So no, it doesn't look like R1a arrived in Eastern Europe from Siberia.

rozenblatt said...

@Vladimir @Davidski Ok, there seems to be some kind of a confusion. "Bolshoy Oleny Ostrov" literally means "Big Deer Island" in Russian.

Davidski said...

Right, well there's no R1a in the samples from Bolshoy Oleny Ostrov.

There is an R1a in a sample from Yuzhnyy Oleni Ostrov (Karelia_HG I0061).

But this sample isn't related to the samples from Bolshoy Oleny Ostrov. They form two completely different populations.

Vladimir said...

@Davidski. I agree. There is another version. So as R1a-YP1272 discovered together with man haplogroup J, on my J2a, then can be suggest, that they where the from Caucasian region. Off topic, but ngasany N1a-L666 diverged with the ancestors of the future Urals N1a-Z1956 about 16,000 years ago. This is a very distant connection, for the base you need to take something more approximate.

RobertN said...

Vladimir, the conqueror does NOT always impose his language. Look at the case of the Bulgars who conquered the Slavic tribes in the Balkans, yet the language spoken to this day is Slavic.

Davidski said...

@Vladimir

The J in the Mesolithic sample from Karelia is J1, but this doesn't mean that R1a arrived in Karelia from the Caucasus.

The oldest R1a is on the steppe in Ukraine (~8,700 BC), and shows no signs of any recent ancestry from outside of this region.

https://eurogenes.blogspot.com/2019/09/y-haplogroup-r1a-and-mental-health.html

Davidski said...

@RobertN

Vladimir, the conqueror does NOT always impose his language. Look at the case of the Bulgars who conquered the Slavic tribes in the Balkans, yet the language spoken to this day is Slavic.

Bulgarians have a lot of Slavic ancestry but very little, if any, ancestry from the Bulgarians from the Volga.

That's probably why they speak Slavic.

Gabriel said...

If the closest population to Finns is Iron Age Finns and Balts, then why do Finns seem to have a weak relationship to Baltic_EST_IA compared to Balts?

Davidski said...

Because Baltic_EST_IA is from the East Baltic where Baltic-related people lived before Uralics moved in, and Finns have more complex recent ancestry than Balts, including Saami admixture.

Kristiina said...

Johanna Nichols paper is a superficial mixture of geographic maps and linguistic graphs that are not explained.

In any case, I don’t buy her idea that Estonian, Hungarian, Finnish or any Uralic language spread on the basis of fur trade.

Her main linguistic conclusion seems to be this: ”Typologically, Proto-Uralic clusters with the greater Pacific Rim population of language: high causativation, inflectional person, fairly high POS flexibility, head final.”

Pacific Rim languages are Ainu, Nivkh and Chukchi-Kamchatkan languages.

In this map you can see where a Western Uralic language (Finnish) and an Eastern Uralic language (Hungarian) cluster typologically among the world languages (fig 1 https://royalsocietypublishing.org/doi/full/10.1098/rspb.2010.0051). Finnish and Hungarian are closer to the core IE than Hindi is. Nivkh and Ainu are together on a completely different direction.

I suspect that Nichols has picked up some features that somehow support her idea and has discarded the rest. This is a good technique to get the result you want to have.

In the paper, causativization is explained with examples from two languages, Spanish and Ingush. I can add that for example in Khanty and Finnish, causative verb is usually made with infix ”tt”, e.g. istua ’to sit’, istuttaa, ’make sit’. In Arabic, the verb ’know’ is ʿalima, and it can be causativized with gemination, ʿallama ’teach’. I cannot see a huge difference between, for example
Finnish: istua ’to sit’, istuttaa, ’make sit’
Arabic, ʿalima ’to know’, ʿallama ’to teach’

As for inflectional person, it is not explained in the paper. An explanation of ”inflectional person” can be found here (https://books.google.fi/books?id=lnbnBQAAQBAJ&pg=PA323&lpg=PA323&dq=%22inflectional+person%22&source=bl&ots=qRcFkJ24oA&sig=ACfU3U2nH76xJ3Ty6v67xq7OnIsRNQo-iQ&hl=en&sa=X&ved=2ahUKEwjs4ZW_6pjmAhWv_CoKHQLzDk4Q6AEwDXoECDEQAQ#v=onepage&q=%22inflectional%20person%22&f=false)

I cannot see much difference in this respect between Uralic and IE languages.

Again, ”POS flexibility” (Part of speech flexibility) is not explained. However, I am wondering how any Uralic language could be more flexible than English in this respect. In all languages, it is possible to make verbs from nouns and nouns from verbs and adjectives from nouns, etc.

As regards, head final languages, English is head-initial and German is head-final. In Uralic language Finnish is head initial and Nenets is head-final. In this map (https://wals.info/feature/81A#2/18.0/153.1) you can see the distribution of SOV and SVO languages and in this map (https://wals.info/feature/87A#2/18.0/152.8) order of adjective and noun. I really cannot see any Uralic-Pacific Rim unity.

Samuel Andrews said...

I made a breakthrough on ancient Roman DNA.

Of the imperial Roman samples, 16 of 49 have significant Greek ancestry. Their European ancestry is from Greece not Italy. They are Greek+Middle East mixed. The Middle East ancestry in most is from Mesoptamia.

This confirms many of the Middle Eastern immigrants into ancient Rome were "Greek Middle Easterners" who came from Greek kingdoms in Asia. A lot probably come from Selecudia kingdom considering they have Greek & Mesoptamia ancestry.

Queequeg said...

The idea of an Uralic speaking Fur (/Bronze etc.) Road is not bad as such however it is probably only one part of the explanation as for instance northern river valleys are also in many cases natural pastures. Even nowadays cattle in Northern Russia grazes in those places, I've seen that. I'd therefore guess that early Uralic speakers were herders (vs. earlier HG's), besides some opportunistic slash and burn type of agriculture. Besides, cattle bones have indeed been found for instance in Ladoga area, connected to early West Uralic speakers of that area.

The roots of that herding institution are for sure in Ural area, as for instance Saami word for reindeer boazu is placed on Indo Aryan/Proto Iranic/smthng < *pasa "tamed". In other words, the idea of herding just reindeers is a new one, whereas the idea of herding as such is much older.

Fur Road might actually explain BOO, too. Northern trade posts are for sure natural places for both trade and some recreational sex.

And D: many thanks for the clarification re mtdna.

Ryan said...

Does this mean Panonian Avars were Uralic?

Craig said...

Johanna Nichols' presentation seems pretty non-controversial. She's summarizing what can be inferred about the history of the Uralic language family based historical linguistics, and current and known historical distribution of the Uralic languages. It seems pretty consistent with what I've read from other historical linguists discussing this language family.

Uralic likely spread as a trade language. Traders married women from local peoples they traded with (as Nichols puts it, "Likely mechanism: local chiefs' daughters marry rich traders (as with with French fur franchises in North America)". The Uralic languages borrowed extensively from other languages, including IE languages as different places and time periods.

Immediate- family kinship terms in Uralic languages are frequently loanwords, but terms referring to extended family are retained from Proto-Uralic. This makes sense if traders' children often grew up in bilingual households, but it was important to maintain extended family contacts across larger areas of the trade network where the local languages were different. (Terms in the Uralic lingua franca would be preferred in this context. Within the immediate family terms from either Uralic or the local language could be used.)

In the case of Estonian and Finnish, the substrate language is IE, presumably a language ancestral to Baltic and Slavic, for instance, the Estonian word for "sister" is sõsar, in Finnish, it's sisar,from PIE *swesōr. (Sami languages and Finnish also have loans from Proto-Germanic that are not shared with Estonian.)

Nichols also discusses loanwords from Indo-Iranian languages in Uralic. (If I remember correctly, the pattern of borrowings here suggests trade contact, but not intermarriage. I might be misremembering.) These loans aren't shared with Samoyedic languages.

Anonymous said...

@ Davidski @ Vladimir
R1a, like R1b, arrived in Eastern Europe from Siberia, but in the Final Palaeolithic/Early Mesolithic.

@RobertN

Bulgarians did not win Slavs, they have entered into the union relations with Seven Slavic tribes against Byzantium and Avar, thus the union with the subsequent creation of the state was formed.



CrM said...

https://imgur.com/a/fhLdv36
Is this model correct? Did BMAC really have Progress ancestry? Is it possible that Progress was also situated further East, closer to Central Asia, and thus explaining why Steppe Maykop had so much of it?
Incidentally, could U1b in Uralics come from some post BMAC/Progress populations?
Many Maykop samples, including the one which carried U1b (I6272), had some Progress component, plus the weird Areni samples from Eneolithic Armenia.
https://imgur.com/a/uZ9nys1

Rob said...

@ Archi

R1b is in Europe by at least 16000 BP; long before FP/ early Mesolithic

“Seven Slavic tribes against Byzantium and Avar, thus the union with the subsequent creation of the state was formed.”

Modern Bulgarians are Vlachs and Slavs from
Macedonia . First Bulgarian empire ceased to exist after Basil destroyed it ; the Rus overan it
It was the settled mostly by Turkics (Patztinaks; Cumans) until the aforesaid reconquista

Davidski said...

@AuckeS

Steppe Maykop probably has Progress ancestry because it lived in former Progress territory.

But I suspect there may have been some sort of a cline of Progress/Sarazm_En ancestry around that area. So if you try adding Sarazm_En to your models that might flesh it out.

The missing piece of the puzzle might be samples from the Kelteminar culture.

CrM said...

@Davidski

Well, I've been thinking that one of the Progress groups were hunter-fishers from the Caspian that eventually reached Central Asia by following the Caspian coast.
I agree on Kelteminar, hopefully samples from this culture will clear things up.

Nomic Belief said...

According to my understanding of recent archeology (the little read from papers by W. Lang), the stone cist graves in Estonia should rather represent coast-dwelling Germanic culture than Baltic. The ppl in those graves being "genetically Baltic" may thus not mean they were Baltic by language or culture, despite shared ancestry?

Anonymous said...

@Kristiina @Craig

Don't pay attention to Johanna Nichols. She's always delirious, she's getting IE out of "India". She always fantasizes without any arguments and without any understanding of the issue.

Rob said...
"R1b is in Europe by at least 16000 BP; long before FP/ early Mesolithic"

LOL. 16000 BP is Final Paleolithic, which you have specifically reduced to FP so that it wouldn't be visible.


gehn"
“Seven Slavic tribes against Byzantium and Avar, thus the union with the subsequent creation of the state was formed.”

Modern Bulgarians are Vlachs and Slavs from
Macedonia . First Bulgarian empire ceased to exist after Basil destroyed it ; the Rus overan it
It was the settled mostly by Turkics (Patztinaks; Cumans) until the aforesaid reconquista"

You don't know what you're writing about and what you're talking about.

Davidski said...

@Nomic Belief

There's absolutely nothing Germanic about the ancestry of the people in the stone-cist graves, so even if they were Germanic speakers they were derived from a Baltic-related population that probably spoke something related to Baltic.

Kristiina said...

@ Nomic Belief

What are the archaeological elements that show that Stone Cist people spoke Germanic?

Craig said...

I agree with Kristiina that Nichols' discussion of typological features is pretty dubious. Historical linguists usually caution against reading to much into these kinds of resemblances unless there's also other evidence of a connection. The rest, though, seems like pretty standard stuff.

The idea that trade networks may have been involved in the spread of Uralic languages isn't new, or something Johanna Nichols came up with. Ante Aikio has also discussed this, and has gone into some detail about how he thinks this may have played out in the spread of the Sami languages. I think Aikio knows his stuff, whatever you think of Nichols. ("Knows his stuff" doesn't mean he's infallible, of course, but I take what he has to say quite seriously.)

Nomic Belief said...

@Kristiina

"What are the archaeological elements that show that Stone Cist people spoke Germanic?"

Well, you should better be asking this from prof. Lang. Anyway, the grave style is probably from Gotland and Scandinavia, as is their advanced coastal agriculture. Also the geography and time frame match the people whose pre-germanic coastal life vocabulary the Uralic immigrants assimilated. That sounds more probable to me than taking them for some kind of pre-Balts, or for whichever of the previous paleo folks that were decimated apparently due to severe climate. So, I rather see them more probably Germanic than Baltic notwithstanding the genetic profile.

Is there some other reason not to see the stone cist people (pre-)Germanic?

Olavi said...

@ Kristiina

Kristiina said:
"What are the archaeological elements that show that Stone Cist people spoke Germanic?"

This is one of V. Lang's theories from his book "Finnic be-comings" ("Läänemeresoome tulemised"). Two paragraphs from his summary:

"At around 1500 BC in western and southern coastal Finland and around 1200 BC in northern coastal Estonia people started to build monumental stone-cist graves above ground. As the grave type (together with some other evidence) refers directly to the west, i.e. Proto-Germanic-speaking Scandinavia, one has to consider the establishment of Germanic settlement on both sides of the Gulf of Finland. One can suppose that this Germanic population came from different parts of Scandinavia because in northern Estonia they were well-established agriculturalists (the closest parallels being with Gotland; Fig. 6.1) but in coastal Finland there is not much evidence on agricultural activity by this population at that time."

"From the Daugava valley the members of the language group Finnic AB shifted northwards to the Estonian, Finnish and central Swedish coasts, where they met Proto-Germanic speaking populations. Unlike the very frst pioneers, the communities of fortifed settlements that reached coastal Estonia and Finland in the 9th century or around 800 BC settled within the core areas inhabited by this population. As a result of the establishment of mixed, bilingual settlements of Finnic and Germanic communities, the language Finnic AB received a strong Germanic influence and gradually became Finnic ABG (the middle PF). This could happen only in the coastal zone of Estonia, Finland, central Sweden, and Courland (Fig. 6.7) inhabited supposedly by Germanic-speaking populations before the arrival of newcomers from the SouthWestern Passage. The earlier Germanic population east of the Baltic Sea was later assimilated by the Finnic; the communities burying their dead in stone-cist graves gradually acquired new ‘eastern’ customs, such as using pottery in funerals and adding new tarand-like structures to graves that had initially built to be circular in design (Fig. 6.8)."

Davidski said...

Actually, some ancient Gotlanders have genetic profiles similar to the stone-cist grave people in Estonia, and also to Balts.

So the migration of the stone-cist people may have gone from the East Baltic to Gotland, rather than the other way around.

There's really no ancient DNA evidence of any Germanic settlements in Estonia during the Bronze Age.

Olavi said...

Link to "Finnic be-comings" pdf format. The English summary is at the end.
https://www.etis.ee/File/DownloadPublic/680cdbdb-88f5-421c-aea0-6a2b12ba3759?name=tulemised_etis.pdf&type=application%2Fpdf

Kristiina said...

This is the description of the same period in Gotland:

In the Early Bronze age around 1800 BC the same burial pattern tied to the coastal areas is continued, and at that time it is sometimes seen that some of the late Neolithic stone cists were covered by a monumental stone cairn, and additional burials were added outside the stone cist (Burenhult, 1986, p. 344–351; Stensköld, 2004, p. 155–157). This act protected the old genealogical burial place and possibly tied these earlier generations to the followers, as well as, the cairns became clearly visible monuments in the landscape.

The bronze finds on Gotland show local character in the ornamentation already in period I (Montelius period system) but with close resemblance with finds from areas to the South West and South East. Finds of so called Mälardal axes show connections to the East but also ties to the East Swedish area. During the late Bronze Age the connections with the East Nordic area are stronger than contacts with South Scandinavia according to Hansson (1927, p. 100p). Eriksson (2010xx) who have studied pottery from the Bronze and Iron Age contexts in East Sweden suggests that Gotland show a mixed find material and contacts from several areas around the Baltic Sea are indicated.

During this time was an extensive farming and herding method used. According to Lindquist (Ibid) the land-use changed into intensification of agriculture with arable meadows and grazing in smaller “privatised” established areas with a fencing system, during the pre-Roman Iron Age. These types of smaller irregular farming units are also found in Estonia. Lang (1996) calls these “Baltic fields” and according to him they reflect the boundaries of clearing of the arable soil and centered on clearing cairns. Thus they diverge from the larger regular Celtic fields, which reflect a conscious land-division and land ownership. Hallin (2002, p. 34) suggest that the Baltic field type is found on Gotland as well.

Source : Prehistoric lifestyles on Gotland – Diachronic and Synchronic perspectives
http://www.diva-portal.org/smash/get/diva2:579478/FULLTEXT01.pdf

Anonymous said...

@Davidski "There's really no ancient DNA evidence of any Germanic settlements in Estonia during the Bronze Age."

For the Bronze Age it is incorrect to speak about the Germanics, we can only talk about links with specific territories of Scandinavia. There were no Germanics at that time. Therefore, nothing contradicts the fact that Estonia and Eastern Scandinavia were connected in the Bronze Age, they also have the haplogroup R1a-Z283. Nowadays, genetics does not show the direction of connections, but archeology says that it was from Scandinavia to the Eastern Baltic.
Of course, there were no Finns there at the time.

Kristiina said...

So the intensified farming means this: "According to Lindquist (Ibid) the land-use changed into intensification of agriculture with arable meadows and grazing in smaller “privatised” established areas with a fencing system, during the pre-Roman Iron Age."

In Scandinavia, Pre-Roman Iron Age is from 5th to 1st centuries BC.

Can anyone find out from Lang's book when the "Baltic fields" appear in Estonia? Pre-Roman Iron Age is already the period of Tarands in Estonia.

Shaikorth said...

@Davidski
Can you try how OLS10 fits in qpAdm as a mixture of Baltic_BA (the local part) and Mereke_MBA (proxying the Uralic part), the latter's N not directly ancestral to it but nothing else available is either and it might be slightly better than the Sintashta outliers.

Kristiina said...

According to Lang’s book
Estonian Archaeology 3 Valter Lang The Bronze and Early Iron Ages in Estonia, p. 101,
the oldest part in the Saha-Loo field complex is from the Middle Bronze Age, ca 14th-11th centuries BC.

This means that these intensive fields are clearly older in Estonia than in Gotland.

Anthony Hanken said...

@Samuel Andrews

In Lamnidis et al. BOO is connected with the spread of as asbestos mixed lozervo ceramics and the Imiyakhtakhskaya culture which originated in Yakutia.

I would bet this populations was 100% Nganassan like and non-Uralic speaking originally. Wether it was originally N-L1026 is something I can not say but by the time this group reached the Kola peninsula it was clearly mixed with something else, maybe something Uralic?

Slumbery said...

@Ryan
Does this mean Panonian Avars were Uralic?

Hungarian scholars considered this possibility for a while (for example he: https://en.wikipedia.org/wiki/Gyula_L%C3%A1szl%C3%B3 ), but the current academic consensus is against it.

Nomic Belief said...

@Davidski

"Actually, some ancient Gotlanders have genetic profiles similar to the stone-cist grave people in Estonia, and also to Balts."

Actually, I almost brought this up but was not sure enough. Thanks for confirming this. That is a telling point about the ancient Gotlanders. It is, however, difficult to identify them as plain Balts based on genetics. Probably these supposedly Pre-Germanic folks or whoever were some kind of mix, at least on the islands and the Eastern shore of the Baltic Sea.

"So the migration of the stone-cist people may have gone from the East Baltic to Gotland, rather than the other way around."

True, it is a possibility. OTOH the population and grave numbers are very few in Estonia/Saaremaa, so which direction is more probable is up for debate.

"There's really no ancient DNA evidence of any Germanic settlements in Estonia during the Bronze Age."

I guess this depends on the ethnic identity of the individuals found in stone cist burials.


Ryan said...

@Slumberry - Hungarian scholars considered this possibility for a while (for example he: https://en.wikipedia.org/wiki/Gyula_L%C3%A1szl%C3%B3 ), but the current academic consensus is against it.

According to this a majority of Avar-era Y-DNA sampled was N1, so it seems pretty damned likely that there were at least some Uralics in that coalition I'd think.

http://dispatchesfromturtleisland.blogspot.com/2019/12/european-barbarians-has-varied-y-dna.html

Mauri said...

@Davidski

"Do you have a PLINK file of the samples from this Icelandic study?

https://science.sciencemag.org/content/360/6392/1028

The versions I have show a lot of aDNA damage and pseudo-African admix."

I processed all Icelandic IA samples already in January and yes, there is a small pseudo-African admixture. Without SSA references all those samples are just where I expect they should be, but after adding SSA samples they move toward South European on plots.

I also tested Bolshoy samples (without outlier) using qp3Pop to see shared drift with Europeans with no or only minor Siberian admixture, Russians, Poles, Scandinavians, Brits, Orcadians and Balts. Balts showed highest drift followed by Russians without Siberian admixture, Brits lowest drift. Hard to say how this turns to reality, but it would be interesting to see where the European part of their ancestry comes from. Of course it can be more like Uralic, but the common Siberian can be older than their common European and it is difficult to see inside it. For background, still around 1000 years ago an unknown group warlike people and merchants lived near Kola Peninsula who disappeared later totally. They left cultural signs, artefacts. We know very little about ancient northern people, because researchers show so little interest on them.

capra internetensis said...

@Ryan

The Avar N1 was mainly N-B197, the branch containing Turko-Mongolic N-F4205 (peaks in Buryats) and Chukotkan N-B202 (Chukchis and Asian Eskimos). None of it looks Uralic AFAICT. On the other hand the Conqueror period has several N-Z1936(xL1034), a largely Uralic branch.

Kristiina said...

@ Ryan

All Avar samples (with enough resolution) belong to N-Y16323/B197, or more specifically to Y16312 (https://www.yfull.com/tree/N-Y16323/). The highest frequency of Y16312 is in Buryats (18.4%). It is also found in Altaians. Y16323 is not found in any Uralic population. IMO, it is quite safe to say that these Avars spoke a Turkic language.

Instead, none of the Hungarians belonged to Y16323/B197. Hungarian yDNA haplotypes are mostly shared with Uralic populations.



Ryan said...

@capra, @Kristiina - Ah, gotcha. They are sister clades though no? Could Buryats have assimilated para-Uralics in the distant past? Because the languages don't seem particularly related, but the Y chromosomes certainly are.

Rob said...

@ Archi

“16000 BP is Final Paleolithic, which you have specifically”

Nope. It’s Bolling Allerod-> late glacial.
FP is 12/11,000; and certainly not Mesolithic, as you claimed .

“You don't know what you're writing about and what you're talking about.”

Aha. Where’s all the Hunno-Bulgar ancestry in modern “Bulgarians”?
Basil the Bulgar slayer ....

Davidski said...

@M. Myllylä

Great, can you post those PLINK files or send them to me? Your versions of these samples might be less noisy than mine.

By the way, here are the modern pops that match BOO best using Global25 data. Veps, Karelians, and Finns are way down the list, but ahead of Balts.

Distance to: RUS_Bolshoy_Oleni_Ostrov
0.08166835 Mansi
0.08186721 Tatar_Siberian_Zabolotniye
0.08837377 Khanty
0.10585811 Ket
0.11593419 Tatar_Siberian
0.12054473 Bashkir
0.12353995 Yukagir_Forest
0.12620272 Tlingit
0.14334612 Saami
0.14481923 Tubalar
0.14523161 Udmurt
0.14635313 Shor_Mountain
0.14700201 Mari
0.14702680 Shor_Khakassia
0.15148961 Shor

Distance to: RUS_Bolshoy_Oleni_Ostrov_o (BOO006)
0.06210746 Khanty
0.06729425 Mansi
0.07708945 Tatar_Siberian_Zabolotniye
0.08453697 Ket
0.10003686 Selkup
0.10958907 Shor_Khakassia
0.11040395 Shor_Mountain
0.11351358 Shor
0.11391265 Nenets
0.11661708 Tubalar
0.12340279 Khakass
0.12595806 Tatar_Siberian
0.12798673 Yukagir_Forest
0.14958261 Tlingit
0.14963160 Bashkir

Vladimir said...

@Ryan In the South-West krotovtsy reached Northern Kazakhstan (monument Vishnovka-1), where they, in our opinion, coexisted with Petrovka culture Z2124, as evidenced by the presence of characteristic features of Petrovka in the ornamentation of ceramics Krotovka monuments of the area. So the subclade of future Avars Y16323 separated from future Balto-Finns CTS10760 about 2700 years BC, i.e. in Siberia. I think that Y16323 were the southernmost tribe, but they were obviously Fino-Ugric language, but apparently were assimilated first by the Scythians, then the Turks and then joined the Avar cocktail.

Gabriel said...

@Rob

What evidence suggest Bulgaria harbored high Turkic populations that were replaced continually by Vlachs and Slavs from Macedonia?

Rob said...

@ Gabriel
Are you telling me that you’re not aware of the Ogur -Turkic origin of Bulgars ?
Or that the newly conquered Bulgarian territories were settled by cumans & Pechenegs who settled in 10-12th cc
The 2md Bulgarian empire was led by Vlachs, who’s origin is a mystery but they first attested in the central balkan region & supported by DNA and by the fact that they spoke romance

Gabriel said...

@Rob

No, I mean, many believe Turks to have been a minority population in Bulgaria. That’s why I was asking.

Rob said...

20,000 horsemen isn’t exactly a minority in 6th century mountain lands. That’s a population of 50,-60,000

Anonymous said...

@Rob gehn
"Archi

“16000 BP is Final Paleolithic, which you have specifically”

Nope. It’s Bolling Allerod-> late glacial.
FP is 12/11,000; and certainly not Mesolithic, as you claimed ."


You are a liar, I did not even write this date, but you're talking nonsense as usual, 12,000 BP is an early Mesolithic. At least don't write your nonsense from an uneducated person, for example, we looked at the dates of the Iron Gate Mesolithic.
Mesolithic England Gough's Cave, Cheddar [Not Cheddar Man] 12700 BC
Mesolithic Natufian Israel Raqefet Cave, Mount Carmel [I1685 / Nat 4] 11840-9760 BCE
Mesolithic Italy Grotta Continenza, Abruzzo [Continenza] 11200-10510 cal BP
Mesolithic Iron Gates Serbia Vlasac [I4657 / VLSC_1G/3] 9755-9275 calBCE (9942±66 BP, PSUAMS-2294, corrected for Freshwater Reservoir Effect)

And in general - R1b 14000BP,
Palaeolithic Epigravettian Italy Villabruna, Sovramonte - Belluno, Veneto 12230-11830 calBCE (12140±70 BP, KIA-27004) M R1b1a

mammoth hunter is as shameful as ever.

Rob said...

@ Archie
You can’t even articulate basic sentences

Anonymous said...

Mr. Rob gehn mamonth hunter ... you're incapable of writing a word of truth at all. You write only one meaningless shout.

Rob said...

@ Archi

''And in general - R1b 14000BP,
Palaeolithic Epigravettian Italy Villabruna, Sovramonte - Belluno, Veneto 12230-11830 calBCE (12140±70 BP, KIA-27004) M R1b1a''

So if it made it to Italy by 14,000, it was certainly present by 16,000
EPigravettian period stretches back to 25,000 BP

And I don't know what Israel or Gough's cave have anything to do with it; there is no ''Mesolithic'' in Isreal, its called an 'Epipaleolithic' . You have no clue

Davidski said...

@Archi and Rob

Can you two get a room?

Anonymous said...

Mr. Rob gehn mamonth hunter ... As a typical embarrassed troll, you're twisting through a lie. I wrote to you to look at the Iron Gates because you were shamefully yelling that in 1200BP there was no Mesolithic. You lied that the R1b you found had a date of 16000BP, and now you're twisting it. You shouted that 16000BP is not the Final Paleolithic, and now you are twisting.
Epipaleolithic Turkey Pinarbași [ZBC] 13642-13073 cal BCE
Epipaleolithic France Rochedane, Villars-sous-Dampjoux 11140-10880 calBCE (11120±50 BP, GrA-41739)

I actually wrote the Final Paleolithic/Early Mesolithic period, you're the one who set the dates. Remember, I'm not going to listen to you scream ignorant, you're disgusting.

Rob said...

@ Archie

No; “Final Paleolithic” has a very specific designation; from 13000 later
https://www.jstor.org/stable/25801180?seq=1

Late glacial and B-A are distinct; and importantly so

Ryan said...

Rob, Archi - can we just chill and move on please?

Samuel Andrews said...

@Anthony Hanken,
"I would bet this populations was 100% Nganassan like and non-Uralic speaking originally. Wether it was originally N-L1026 is something I can not say but by the time this group reached the Kola peninsula it was clearly mixed with something else, maybe something Uralic?"

By, Y DNA N is an Asian haplogroup (originally) so they didn't pick it up in Europe.

Anonymous said...

@Rob

You don't know anything about terminology at all. You think it's the same everywhere and you give some meaningless references. The Final Palaeolithic (Epipalaeolithic) is simply an abbreviation of words of the final Upper Palaeolithic (final stages), any specific meaning is given to it by local archaeologists studying specific cultures. All terminology is relative. Read and learn https://en.wikipedia.org/wiki/Epipalaeolithic.

Davidski said...

Yeah, see that's why there's so much talk about the link between Siberian ancestry and the Uralic language family, even though those early Uralic speakers from the Tarand graves had very little Siberian ancestry on their autosomes.

The general consensus is that proto-Uralics came from somewhere near the Urals, but some linguists propose that the pre-proto-Uralic language came from Siberia.

And from what I've been seeing in recent years, there might be a push coming to do away with the pre-proto-Uralic thing, and just say that the proto-Uralic language came from Siberia. You can see that in the stuff posted here IMHO...

https://sites.utu.fi/paleogenetics/en/recent-news/

Anonymous said...

@Davidski "The general consensus is that proto-Uralics came from somewhere near the Urals, but some linguists propose that the pre-proto-Uralic language came from Siberia."

Napol'skich and other Russian scientists have long argued that Uralic languages existed in Siberia, and only the Finno-Volgian (FU without Ugrian) ones crossed the Uralic mountains.

All the bother in a completely unsuccessful name "Uralic" languages.

Samuel Andrews said...

@Davidski, How do you make those spatial maps? I'd like to learn how to make them.

Davidski said...

@Samuel Andrews

https://anthrogenica.com/showthread.php?18960-G25-Distance-Maps-To-Modern-Europeans

Ric Hern said...

That Hallstatt map is interesting...All those relatedness in Western and Northern Europe could not have come exclusively from Hallstatt but rather much earlier...Does this point to the overall closeness of Beakers, Unetice, Tumulus, Urnfield etc. All of these Cultures show interaction with Britain and Ireland and interestingly the Tumulus Culture which reached the Isles originated not far from Kromsdorf where we know about the Bell Beaker connection...so basically broad continuity stretching over a distance of a 1000 kms...The same people sharing different ideas...

Samuel Andrews said...

@Davidski, He doesn't explain how he makes those spatial maps? Do you know how to?

Kristiina said...

As everything always seems to boil down to the speculation about the area of origin of N N-M2126, formed 7500 ybp, and N-L1026, formed 6300 ybp, I hope that the geneticists start screening the taiga area.

Although I am not at all convinced that N-M2126 spoke Proto-Uralic, it is clear that the place where these y lines will hopefully be one day found is very significant for the understanding of the origin of Uralic languages. I hope that I need not wait the rest of my life until this riddle is solved.

Davidski said...

@Samuel Andrews

There are different ways to make spatial maps. It's not something I ever got into, and I don't know what he's using exactly, but this might help...

https://en.wikipedia.org/wiki/List_of_spatial_analysis_software

Queequeg said...

@ D: how about OLS10 as a qpAdm mixture of Baltic_BA (the local part) and Mereke_MBA (proxying the Uralic part) Shaik was asking for?

Garvan said...

“Samuel Andrews said...
@Davidski, He doesn't explain how he makes those spatial maps? Do you know how to?”

I have made maps like these using QGis, but I have only needed them every few years, so I have to relearn how to do it each time. QGis may be the easiest method, but I have seen most people use R.

QGIS
https://docs.qgis.org/2.18/en/docs/gentle_gis_introduction/spatial_analysis_interpolation.html

R
https://mgimond.github.io/Spatial/interpolation-in-r.html


Anthony Hanken said...

@Samuel Andrews

N being from Asia is completely irrelevant. I am talking about N-L1026, a specific subclade relevant to Uralic speakers. Are all R1b sublades from Asia because R* is? We currently don't have enough data to know where N-L1026 was born 6300ybp.

Davidski said...

@Huck Finn & Shaikorth

Using the same setup as in the qpAdm runs in my "fresh off the sledge" blog post:

0LS10
Baltic_EST_BA 0.848±0.041
KAZ_Mereke_MBA 0.152±0.041
chisq 14.259
tail prob 0.430632
Full output

Again, it's a pretty good fit, but I have a feeling that there's something missing in the outgroups because Global25 fits for the same model are kind of crap. Even adding the Nganassans and BOO doesn't help much.

Target: Baltic_EST_IA:0LS10_1
Distance: 4.7564% / 0.04756373
83.4 Baltic_EST_BA
16.6 KAZ_Mereke_MBA

Target: Baltic_EST_IA:0LS10_1
Distance: 4.3565% / 0.04356516
85.4 Baltic_EST_BA
10.8 KAZ_Mereke_MBA
3.8 Nganassan

Target: Baltic_EST_IA:0LS10_1
Distance: 4.1679% / 0.04167906
83.4 Baltic_EST_BA
11.6 RUS_Bolshoy_Oleni_Ostrov
5.0 KAZ_Mereke_MBA

Queequeg said...

Thanks D, much appreciated. Re "Even adding the Nganassans and BOO doesn't help much." How about Levänluhta, did you try that?

CrM said...

Target: Baltic_EST_IA:0LS10_1
Distance: 3.9810% / 0.03980987
80.0 Baltic_EST_BA
10.0 RUS_Bolshoy_Oleni_Ostrov
10.0 RUS_Lola
0.0 KAZ_Mereke_MBA

Lola seems like a slightly better source than Mereke.

Mauri said...

In my opinion the biggest difference between Bolshoy and Levanluhta is that B is a mixture of Baltic HG and Siberian and the latter one Baltic CWC and Siberian. How did this happen? Bolshoy looks like an older concept of Siberian-European admixture and Levanluhta went through a modernization process somewhere near Ladogan, in Finland or near the Baltic area anyway. But there is nothing Baltic Finnic in both samples. In this area the cock crow of Uralic people was younger and Levanluhta admixture was born in Finland before them, meaning that CWC, or Baltic BA if you wish, was the ruling component in South Finland still around 2ky ago.

Davidski said...

@Huck Finn

qpAdm actually prefers Mereke_MBA to Levanluhta_IA, at least with the outgroups I'm using anyway.

@M. Myllylä

I can't see any direct relationship between Levanluhta_IA and CWC or Baltic BA. Wouldn't it show a lot of Baltic/Slavic drift if it had Baltic BA ancestry? So how come it doesn't?

Levanluhta_IA is clearly a mix between something like BOO and Levanluhta_IA_B (the female Germanic outlier).

Queequeg said...

@ D: thanks again!

Queequeg said...

Regarding Mereke_MBA: is this model of JuanRivera solid?

Mereke_MBA: (Botai+Afanasievo+Steppe_Maykop, 1.979, Sintashta_MLBA 82.5% Botai 17.5% Afanasievo 8.33% Steppe_Maykop 74.17%)

It anyways seems to me that one way or another people end up finding something ANE/WSHG-like behind OLS_10, even if the amount is relatively limited. The fact that OLS_10 prefers Mereke_MBA over Levänluhta is also very interesting, especially considering the geographical distances.

Arza said...

@ Davidski

Wouldn't it show a lot of Baltic/Slavic drift if it had Baltic BA ancestry? So how come it doesn't?

B-b-b-but... it does.

Target: FIN_Levanluhta_IA
Distance: 2.7547% / 0.02754726
48.6 RUS_Bolshoy_Oleni_Ostrov
27.4 Baltic_EST_BA
24.0 FIN_Levanluhta_IA_o

The source of admixture is more Nganassan-shifted than BOO:

Target: FIN_Levanluhta_IA
Distance: 2.4554% / 0.02455416
36.2 RUS_Bolshoy_Oleni_Ostrov
30.8 Baltic_EST_BA
23.4 FIN_Levanluhta_IA_o
9.6 Nganassan

With Polish and Norwegian to allow movement along the clines:

Target: FIN_Levanluhta_IA
Distance: 2.0310% / 0.02031005
38.4 RUS_Bolshoy_Oleni_Ostrov
24.2 Baltic_EST_BA
24.0 Norwegian
9.2 Nganassan
4.2 Polish

With CCC:

Target: FIN_Levanluhta_IA
Distance: 1.9840% / 0.01984032
33.8 RUS_Bolshoy_Oleni_Ostrov
23.0 Norwegian
21.6 Baltic_EST_BA
10.0 Nganassan
6.6 Polish
5.0 Baltic_EST_Comb_Ceramic_low_res

Davidski said...

@Arza

Your model isn't really picking up Baltic drift. It's compensating for the fact that FIN_Levanluhta_IA_o is exceedingly western, even for a Germanic.

I did some qpAdm models, and with the outgroups I'm using Baltic_EST_BA is rejected in favor of a two-way model between BOO and FIN_Levanluhta_IA_o. But the statistical fit is poor.

Arza said...

@ Davidski
Can you run these?

Chimp Levanluhta_IA Levanluhta_IA_o Baltic_EST_BA
Chimp Baltic_EST_BA Levanluhta_IA_o Levanluhta_IA

CrM said...

Another interesting thing is that Armenia MBA Lchashen also had ancestry that is similar to Steppe Maykop/Lola/Mereke. Maybe that component was what spread U1b to Caucasians and Uralics/from Caucasians to Uralics.

Target: ARM_Lchashen_MBA:DA35
Distance: 3.6464% / 0.03646385
82.8 Kura-Araxes_ARM_Kalavan
11.0 RUS_Sintashta_MLBA
6.2 RUS_Steppe_Maykop

Target: ARM_Lchashen_MBA:DA31
Distance: 2.6441% / 0.02644117
79.0 Kura-Araxes_ARM_Kalavan
11.4 RUS_Sintashta_MLBA
9.6 RUS_Steppe_Maykop

Mycenean also provides a better fit, and increases the Steppe_Maykop component. Lchashen, iirc, can be connected to the Trialeti Horizon, which showed many paralleles with Myceneans.

Target: ARM_Lchashen_MBA:DA35
Distance: 2.8196% / 0.02819602
66.4 Kura-Araxes_ARM_Kalavan
19.4 GRC_Mycenaean
14.2 RUS_Steppe_Maykop

Target: ARM_Lchashen_MBA:DA31
Distance: 2.0692% / 0.02069208
66.8 Kura-Araxes_ARM_Kalavan
17.4 RUS_Steppe_Maykop
15.8 GRC_Mycenaean

Mauri said...

@Davidski

"I can't see any direct relationship between Levanluhta_IA and CWC or Baltic BA. Wouldn't it show a lot of Baltic/Slavic drift if it had Baltic BA ancestry? So how come it doesn't?

Levanluhta_IA is clearly a mix between something like BOO and Levanluhta_IA_B (the female Germanic outlier)."

I'll make something more practical, but in a historic point of view the Levanluhta is a lucky spot find, because most of organic remains have destroyed in Finnish acid ground and Levaluhta burials were unusual, more like sacrifices. What we know about the Levanluhta site and the area in the Vendel era is that the area was for a short time populated by Scandinavians without Finnish or any other influences. By archaeological evidences we can say that the Snandinavian era was very short, only around 100-150 years and barely left much genetic spots and such homogeneity to Levaluhta samples, if they were not from same family, but they can't be due to the wider time span of finds. Nothing else has been found, including evidences of Saami or other populations. It is hard to understand how the lack of evidences would prove something. On the other hand we have a lot of evidences about Saami people in Southern Finland BEFORE Finns or any FU people. It is hard to believe that Levanluhta, being a short Scandinavian oddity could show something to generalize or even locally meaningful. Until now we have no finds from more southern and meaningful areas, just because of the acid ground. But I'll do something soon.

Davidski said...

@Arza

Chimp FIN_Levanluhta_IA FIN_Levanluhta_IA_o Baltic_EST_BA 0.0112 1.624 143054
Chimp Baltic_EST_BA FIN_Levanluhta_IA_o FIN_Levanluhta_IA -0.0146 -2.122 143054

@M. Myllylä

Well, Levanluhta_IA isn't a mix of BOO and Baltic BA. You need a more western source to make the model work.

Mauri said...

Here are Icelanders, a new store offered by Asus. Hopefully it works. 1T was 13 bucks per year, sounds cheap.

https://www.asuswebstorage.com/navigate/a/#/s/DFCA3622FFDC4F82BBC4FB01E12337CB4

Davidski said...

Thanks, I'll check them out tomorrow.

Samuel Andrews said...

@Anthony Hanken,
"N being from Asia is completely irrelevant. I am talking about N-L1026, a specific subclade relevant to Uralic speakers. Are all R1b sublades from Asia because R* is? We currently don't have enough data to know where N-L1026 was born 6300ybp."

There's no Y DNA N in Europe predating Bolysho Oleni. Bolshyo Oleni is the first Y DNA N in Europe plus they have about 50% East Asian ancestry. So, it looks like Y DNA N in Europe "recently" arrived from Asia.

Anthony Hanken said...

@Samuel Andrews

"There's no Y DNA N in Europe predating Bolysho Oleni. Bolshyo Oleni is the first Y DNA N in Europe"

The oldest N in Europe sampled so far. There are still huge swaths of Russia north of the steppe that remain completely unsampled. Unfortunately this is the area most key to understanding N-L1026's expansion in to Europe.

"plus they have about 50% East Asian ancestry. So, it looks like Y DNA N in Europe "recently" arrived from Asia."

BOO also has 50% EHG ancestry. N obviously entered Europe from Asia but believing an isolated population on the Kola peninsula is the be all end all of all Eurpean N is flawed thinking. By 1500BC N-L1026 was probably already spread along the forest-steppe and could have entered a circum-arctic group anywhere along the way at that point.

Mauri said...

This is only a prompt reply to this Levanluhta outlier question. Using Dstat, wrong or right, it looks like to be more Latvian than Swedish. Differences in direction or another in comparison with Swedes and Balts are minimal. In my opinion, Dstat results are not reliable enough to make conslusions. I use 1 megasnp data.

Latvian Swedish Levaluhta_outlier Mbuti1M 0.0016 0.004316 0.361 7556 7533 125310
Lithuanian Swedish Levaluhta_outlier Mbuti1M -0.0020 0.004339 -0.463 7524 7554 125310
Estonian Swedish Levaluhta_outlier Mbuti1M -0.0011 0.003749 -0.284 7551 7567 125310

We will see something more when it is time...

Davidski said...

@M. Myllylä

Global25 data...

Distance to: FIN_Levanluhta_IA_o:JK2065
0.04504230 Icelandic
0.04544055 Swedish
0.04696795 Norwegian
0.04951976 Orcadian
0.05114096 Scottish
0.05125024 Irish
0.05367936 Welsh
0.05371792 English
0.05386352 Dutch
0.05500297 English_Cornwall
0.05507112 Shetlandic
0.05849415 French_Brittany
0.06212786 German
0.06560219 German_East
0.06694346 Czech
0.07111936 Belgian
0.07136844 Finnish
0.07258106 French_Nord
0.07321472 Austrian
0.07333296 Slovakian
0.07514122 Polish
0.07626756 French_Paris
0.07677411 Hungarian
0.07705314 French_Alsace
0.07784415 Ingrian

Mauri said...

Someone wanted to see Dstat. I compared Levanluhta outlier to modern populations in terms of Dstat drift. On PCA things look different, not at least because of every sample is compared to a global sample map. Therefor admixtures play bigger role on PCA than in pairwise comparison.

Davidski said...

@M. Myllylä

You really can't tell that JK2065 is Northwestern European?

Mauri said...

@Davidski,

if I recall right I wrote that she was likely Scandinavian, because archaeological evidences prove it. Vendel type artefacts. I also wrote that nothing proves about Finns or anything else but Scandinavians and Saami-like people, although Saamis left no artefacts, only Scandinavians.

EastPole said...

“The Precursors of Proto-Indo-European”

https://brill.com/view/title/55752

A lot of articles, some open access:

https://brill.com/view/book/edcoll/9789004409354/BP000013.xml

Indo-Uralic hypothesis, if true, would suggest that the precursor of PIE was EHG language IMO.

Davidski said...

@M. Myllylä

Have you tried a rare allele analysis or something? That might help. You should get the same results as I did with the Global25.

Anonymous said...

@EastPole "Indo-Uralic hypothesis, if true, would suggest that the precursor of PIE was EHG language"

There is no doubt about the truth of the Indo-Uralic hypothesis. However, after separation in Mesolithic (or before), the PIE and the proto-Uralic languages were not in contact before the proto-Indo-Iranian period (Napol'skikh).

Ryan said...

@Archi - "There is no doubt about the truth of the Indo-Uralic hypothesis."

What? They don't even have the same counting base.

Anonymous said...

@Anthony Hanken ". By 1500BC N-L1026 was probably already spread along the forest-steppe and could have entered a circum-arctic group anywhere along the way at that point."

The zone of distribution of Finno-Ugric was a dark coniferous taiga, and Finno-Ugric penetrated Europe along with the spread of dark coniferous taiga, which displaced Fatyanovians.

Rob said...

If Indo-Uralic is true then F-U can’t be associated with N or Altai
Uralic is EHG; whilst PIE is originally from eastern WHG

Anonymous said...

@Rob "If Indo-Uralic is true then F-U can’t be associated with N or Altai
Uralic is EHG; whilst PIE is originally from eastern WHG"

It's full absurd. Uralic was Siberian before Iron age. EHG was European. You always don't understand any topic you're trying to talk about. The Indo-Uralic hypothesis concerns the kinship of languages during the Palaeolithic period.

Davidski said...

Y-haplogroup N is not an East Asian lineage, but a Siberian one. It's found in West Siberian hunter-gatherers and even samples associated with the Poltavka culture (Mereke_MBA).

Keep in mind also that Eastern European and West Siberian hunter-gatherers are distantly related and were probably always in contact with each other in the Urals region.

And just because modern Uralic speakers don't have much West Siberian hunter-gatherer ancestry, it doesn't mean that Proto-Uralics didn't, nor that Uralic-specific subclades of N aren't derived from populations rich in West Siberian hunter-gatherer ancestry.

So Proto-Uralic can be from Siberia and at the same time related to Proto-Indo-European.

Samuel Andrews said...

@Davidski, West Siberian Hgs had tiny (15-20%) East Asian ancestry. Also, what about the ancient relationship between y-haplogroup N and y-haplogroup O (the main hg in China, Korea, Japan, Vietnam, etc)? Can't imagine Y-haplogroup N isn't ultimately from East Asia.

Davidski said...

N might be ultimately from East Asia, but even if that's true, it's native to Siberia, because it's found in indigenous Siberian foragers with, as you say, very little East Asian ancestry.

Do we have any indigenous East Asian foragers belonging to N? Not that I'm aware of.

So the low levels of West Siberian ancestry and relatively high levels of East Asian ancestry in Uralic speakers might not have anything to do with the origins of their subclades of N.

Anthony Hanken said...

@Archi

Your right, I should have said north of the forest-steppe.

@Samuel Andrews

Again if N is from East Asia (Northeast China?) we are talking about over 20,000ybp, completely irrelevant to Uralic or neolithic WSHGs.

Samuel Andrews said...

@Anthony Hanken, By East Asia I mean East Asian genetically not geographically. East Asian derived people lived in Siberia by at least 10,000yo. Bolysho Oleni descended from Siberian peoples of mostly East Asian origin. West Siberia HGs were part East Asian btw which could easily explain their Y haplogroup N.

There's no Y-haplogroup N recorded in ancient Europe yet predating BOO. EHG, Baltic Hgs are all R1b, I2, R1a, and a few Q. The chances N-L1026 derives from this low, just like it is a low chance R1b L151 derives from Neolithic Spain.

Ryan said...

@David - "N might be ultimately from East Asia, but even if that's true, it's native to Siberia, because it's found in indigenous Siberian foragers with, as you say, very little East Asian ancestry."

N is also found in Vietnamese populations but I've never seen you suggest they have even a sliver of Siberian ancestry.

I'd say rather that WSHG was the western limit of N's ancient range, but that range extended pretty far East, and reasonably far south. Wasn't the Longshan culture along the Yellow River rich in N?

I wouldn't be surprised if the ultimate origin of N and the ancestors of Uralics is somewhere between the Yellow and Amur rivers.

Anthony Hanken said...

@Samuel Andrews

I agree with the first part of what you said however we don't have any Mesolithic East Asian N auDNA so no one can say for sure.

"There's no Y-haplogroup N recorded in ancient Europe yet predating BOO."

I know that, but not all of eastern Europe has been sampled. I'm not expecting 100% EHGs with N in neolithic Europe but N-L1026 in the Kama region predating BOO (3500ybp) is probably likely.

Anthony Hanken said...

@Ryan

https://science.sciencemag.org/content/361/6397/88

N doesn't show up in Southeast Asia until the iron age.

As far as Longshan goes the subclades are unknown. There are plenty of old Chinese branches of N Longshan could be but if they aren't N-L708 then they can not be the ancestors of Uralics.

Ryan said...

@Anthony - Agreed on both points, but that doesn't make N any less East Asian in origin.

Rob said...

@ Davidski

“So Proto-Uralic can be from Siberia and at the same time related to Proto-Indo-European.”

That’s not really viable
Unless Uralic is an EHG language; or indo-Uralic isn’t real

Ric Hern said...

I think it all comes down to where K2a and K2b split from each other. Oase 1 and Ust Ishim is already K2a from which N and O came...looking at these ancient samples they already lived significantly to the North. So no reason why their distant cousins did not also live there. Basically just like P in Yana and downstream R and Q. If Uralic is closer to Indo-European why then are people with Haplogroup Qs Languages not closer to Indo-European when considering their Paleolithic till Bronze Age closeness and contact ? This can only mean closer contact in later periods I think.

Ric Hern said...

Or if Proto-Uralic and PIE indeed were closer than Haplogroup Q Languages it could be that Haplogroup N was bordering Haplogroup R1s in the North for significantly longer than Q bordered R1s to their East...so maybe the Sub-Arctic was always N territory and they only sporadically made excursions Southwards.

Ric Hern said...

Maybe this Northern distribution is why Uralic share both somethings with both Altaic and Indo-European (R) and Altaic (Q) and Indo-European are more distant from each other ?

K33 said...

Hey David!

Off topic but do you think you'll get a chance to process the Fulani genomes from the recent paper? (doi: 10.1186/s12864-019-6296-7)

They estimated ~25-30% Eurasian admix in Fulani but for some reason they used modern populations like CEU and Mozabite as references in their qpGraph, we need to get these in the G25!!!

Data is here: https://www.ebi.ac.uk/arrayexpress/experiments/E-MTAB-8434/

Davidski said...

@K33

I'll add them if and when the genotypes become available somewhere. Those raw sequences are too big.

Rob said...

@ Ric

It's hard to guess, but I think Siberia & EE went their own ways after 15000 BP. But some occasional hg Q, mtDNA R1b -associated admixture contnued to trickle in via boreal path.
I think there might be something to the ice-path theory for N1

Queequeg said...

Proto Uralic ceased to exist around 2000 BCE. Based on all we know, for the time being, it was spoken near Ural area, possibly next to Volga-Kama interfluve. Pre Proto Uralic then might have been West Siberian or not, we don't know.

Also, really old mutation levels of N predating even Ice Age don't have anything to do with Proto Uralic or even Pre Proto Uralic, in terms of scientific discussion. Some next-to-Ural-area specific younger sublineages of N such as still pretty Uralic N-Y9022 TMRCA 3900 YBP in my opinion however do support the idea that N has resided in the area already during the emergence of Proto Uralic language. This, even if there are younger sublineages such as very much (Siberian) Turkic N-M2019 TMRCA 3800 YBP.

That being said, this sentence of D makes a lot of sense, especially after Narasimhan et al have shown that WSHG type of genebase was apparently present also in that Volga-Kama interfluve:

"And just because modern Uralic speakers don't have much West Siberian hunter-gatherer ancestry, it doesn't mean that Proto-Uralics didn't, nor that Uralic-specific subclades of N aren't derived from populations rich in West Siberian hunter-gatherer ancestry."

Then there are posts, however for instance those of Anthony's excluded, which don't make a lot of sense.

Rob said...



@ Huck Finn

“Also, really old mutation levels of N predating even Ice Age don't have anything to do with Proto Uralic or even Pre Proto Uralic, in terms of scientific discussion”

For sure; my point was that the idea of paleolithic FU-PIE family is clutching at straws

Anyhow; let’s see if L1026 was sitting around west Siberia; as your “sensical” post suggests

Queequeg said...

@ Rob: Which ideas?

Ric Hern said...

The Subarctic stretch further South near the Urals than most areas of its Southern border....

Queequeg said...

@ Rob: as long as people say discuss EHG or WHG languages, somebody has to be. Genes don't speak languages, especially very old genes.

Mauri said...

@Davidski
"Have you tried a rare allele analysis or something? That might help. You should get the same results as I did with the Global25."

Off-topic, because it is a consensus among scholars that we have evidences only about Scandinavians there and a thing I hate pretty much is unnecessary work. Knowing well the history of Baltic Finns, as well the Swedish history, I have my goals and way to reach something based on this steady ground. Speaking about Levanluhta or any other spot finds is useless without taking into account the wider context, so for example Levaluhta admixtures can't be isolated from the big picture, even though some pre-Viking Scandinavian traders had there a stronghold.

Kristiina said...

@ Anthony
These are Chinese Neolithic samples from Cui et al, 2013:
Niuheliang Hongshan, 5000 ypb: 4 N (xN1a, N1c),1 C3e,1 O3a3
Halahaigou, Hongshan-Xiaoheyan culture, 4500 YBP 12 x N1(xN1a, N1c)
Miaozigou, Inner Mongolia, Yangshao Culture, 5500 YBP 3 x N1(xN1a, N1c)
You see that they are all xN1a and N1c.

Instead, N1a (N1a2-L666 according to ISOGG 2019) and N1c (N1a1-M46/TAT according to ISOGG 2019) have been detected in the Baikal Neolithic, and the oldest published Baikal sample is dated to 7123 ybp. Although Uralic N which is under N1a-F1206 is closer to Baikal N1a2-L666 than to Chinese Neolithic N1b-F2930, the typical Uralic N1a1a1a1a-L1026 has not yet been found in the east. N1a1a1a1a-L1026 was detected in Zhizhitskaya culture ca 2500 BC close to Smolensk. I hope that other samples from this site would be analyzed with modern techniques. Moreover, the oldest split in the N tree is between N2-Y6503 (Botai) and the rest, and N2 has only been found in the west.

If we have a look at the issue from a broader perspective, we see that Ust Ishim in Siberia ca 45 kya is K2a. Then Tianyuan ca 33 kya close to Beijing is K2b, Yana samples ca 30 kya are P1. Malta1 24 kya is R*. The tropical hunter-gatherers in Southeast Asia are D and C, respectively. Ydna O is Neolithic in Southeast Asia and N arrives with Iron Age. The ancient evidence has been consistently against Southeast Asian origin of yDNA K2.

Considering the structure of K2 haplogroups and the current ancient evidence, one must conclude that K2 was the first haplogroup to move to the northern latitudes. N and Q remained there, but O headed to the southeast and R to the southwest.

Rob said...

@ Kristiina

Very interesting; you could be onto something.

Of course, as Huck Finn has thankfully pointed out, Uralic L1026 isnt from the Dinosaurs, and people need mouths to speak. We must keep that in mind

Davidski said...

@M. Myllylä

You were basically suggesting that my Global25 analysis of JK2065 was wrong, and that this sample was closer to Latvians, at least judging by the D-stats you ran.

But D-stats or not, this individual is obviously of Germanic origin, and even more western than most modern Swedes. This can be verified in many ways.

Whether it's off topic here is another matter altogether.

Queequeg said...

@ Rob: I hope that you do. I'm somewhat afraid that you don't.

Anonymous said...


Linguists do not doubt the relationship between the PIE and the Uralic languages, there are unambiguous phonetic correspondences and common vocabulary, which cannot be explained by borrowing. This commonality is determined by the final Upper Paleolithic (maybe including the beginning of the Mesolithic). Uralic languages lived in Siberia, as their vocabulary makes clear; they were not present in Europe until the Iron Age. EHG+R1a came to Europe from Siberia in the final Upper Palaeolithic-Early Mesolithic. EHG is related to WSHG, so says genetics. That unequivocally proves that only EHG+R1a is a PIE, but not a WHG and CHG, in Siberia from Baikal to the Urals at that time were inhabited by their Uralic-speaking relatives with WSHG.

This is the way it should be according to the theory, the carriers of N should have absorbed the Indo-Uralic language from proto-EHG+R1a+(?).

Neolithic Kitoi Russia Lokomotiv, Irkutsk [LOK_1980.006 and LOK_1981.024.01] 5500-4885 BC M R1a1-M17 2 samples
Neolithic Kitoi Russia Shamanka II [DA245, SHA_2006.076, Grave 76] 6065-5916 calBCE (7123±37 BP, OxA-26456) M N1c2b2-L666
Neolithic Kitoi Russia Shamanka II [DA248, SHA_2005.063.01, Grave 63-1] 5755-5635 calBCE (6815±38 BP, OxA-25327) M N1c2b2-L666
Neolithic Kitoi Russia Shamanka II [DA247, SHA_2004.051, Grave 51] 6856 ± 40 BP, OxA-21526 M N-M231
Neolithic Kitoi Russia Shamanka II [DA250, SHA_1998.006, Grave 6] 6483 ± 37 BP, OxA-27054 M? NO1-M214
Neolithic Kitoi Russia Shamanka II [DA251, SHA_2008.108.03, Grave 108-3] 6373 ± 32 BP, OxA-21503 M N1-M2291
Neolithic Kitoi Russia Shamanka II [DA362, SHA_2004.049.01, Grave 49-1] 6319 ± 33 BP, OxA-24793 M? N1c2b2-L666

Neolithic Isakovo Russia Ust'-Ida [DA345, UID_1995.056.01, Grave 56-1] 4730 ± 70 BP M N1c1-M46-M2080


The Zhyzhytsky sample is dated very indirectly, it is somewhere there stashed at the bottom without radiocarbon dating and research of the reservoir effect, so its dating cannot be trusted.

Anonymous said...

@Kristina "N1a1a1a1a-L1026 was detected in Zhizhitskaya culture ca 2500 BC close to Smolensk."

There is not N1a1a1a1a-L1026 there. There is simple N1c there.

Neolithic Zhizhitskaya Russia Serteya II (Smolenskaya oblast') [A6] after 2500 BC M N1c Chekunova E. M. et al. (2014)

Mauri said...

@Davidski,

"
You were basically suggesting that my Global25 analysis of JK2065 was wrong, and that this sample was closer to Latvians, at least judging by the D-stats you ran.

But D-stats or not, this individual is obviously of Germanic origin, and even more western than most modern Swedes. This can be verified in many ways.

Whether it's off topic here is another matter altogether."

No, I was not suggesting anything like that. I only made a Dstat test, because someone wanted to see Dstat. It is not my fault that different methods give different results, or should I say different perspectives to even same questions. Give me a hint where I wrote that Levanluhta outlier is not western. What I wrote was that other Levaluhta admixtures are not Bolshoy+Levanluhta outlier. Levanluhta outlier is western and all local archaeological evidences support it. Actually there is a chain of evidences connecting Vendel Swedes, as well as Levanluhta area to British Saxon culture.

Reto said...

@Kristiina
"If we have a look at the issue from a broader perspective, we see that Ust Ishim in Siberia ca 45 kya is K2a. Then Tianyuan ca 33 kya close to Beijing is K2b, Yana samples ca 30 kya are P1. Malta1 24 kya is R*. The tropical hunter-gatherers in Southeast Asia are D and C, respectively. Ydna O is Neolithic in Southeast Asia and N arrives with Iron Age. The ancient evidence has been consistently against Southeast Asian origin of yDNA K2."

I wonder what you think on the possibility of more than one early "wave" from South to North, not necessarily just one. There's a margin for the age estimations involved, so it seems possible that K2 was originally in the South and that K2a reached North firstly, followed by P1 a bit later. M and S, <- K2b1, are pretty "Southern", and there is even P* among the highly isolated Andamanese people, not to mention the basal P1 branch in Phillipines. Also, K and K2 TMRCAs are very close to one another. A Siberian origin of K2 could perhaps imply "complications" regarding the origin of other close clades, including K (also above LT).

Queequeg said...

@ Archi and re "Uralic languages lived in Siberia, as their vocabulary makes clear; they were not present in Europe until the Iron Age."

Where do you people get all these strange ideas? I assume that you're able to provide a scientific source which supports your claim.

Kristiina said...

Yes, there may have been several waves.

We would need very old yDNA data from India in order see if P or K2b1 were there.

However as Tianyuan is K2b, it is possible that S and M are ultimately derived from a yDNA that reached Island Southeast Asia from the north.

However, I see that there is K2* in Australia, K2c in Bali, K2d in Java and K2e in South Asia. So, it is probable that there was K2 in India that spread early to Southeast Asia/ Island Southeast Asia.

EastPole said...

The GenomeAsia 100K Project enables genetic discoveries across Asia


https://www.nature.com/articles/s41586-019-1793-z#Sec6


https://www.nature.com/articles/s41586-019-1793-z/figures/2


Lots of European genes in Asia

Mauri said...

Here is a short study made by a reputable Finnish scholar. Little to tell about in Levaluhta Vendel area, a small piece of the history. Not a big deal, but Levaluhta itself was only a small incident in history.

https://www.jstor.org/stable/2847025?seq=1

Anonymous said...

@Huck Finn It's not a strange idea, it's a proven academic science fact. What I have written is mainstream academic science, but alternative opinion is a "strange idea" that has long been rejected by science.

Anthony Hanken said...

@Kristiina & Huck Finn

I think you should reread what I said. I was making the point that it doesn't matter what subclade the Chinese N is, if it does not belong to N-L708 it is not relevant to European N. Which I believe, is the exact same point you are making.

Looking at it objectively N probably does originate somewhere in the northern half of China. There is a lot of subclade diversity showing up there with modern testing and aDNA with possible N-Tat from the early Neolithic being released soon.

With that being said Davidski is likely right about N being in Siberia for a long time, the suggestion that PPU and N-L1026 went straight from China to Europe makes no sense. Its been firmly esatablished that PU was spoken near the Urals and that's were N-L1026 should be ~2500-2000BC. I wouldn't hold my breath expecting much (if any) N in Europe before that though unless its something weird like N2.

Ryan said...

@Archi - Just making up BS about what linguists believe doesn't make it true.

Queequeg said...

@ Archi: so you're not able to provide a source, as expected.

@ Anthony: yes, makes sense.

Anonymous said...

@Ryan It's just that your statements don't make them true. If you don't know anything, it doesn't mean your bullS is true.

Ryan said...

@Archi - Prove it. Show that "all linguists agree Indo-Uralic is true."

Kristiina said...


@ Anthony

The genetic history of China is still very poorly understood.

However, at the time of the formation of N1 and N2, 21700 ybp, Northeast China was inhospitable for humans and depopulated. It was repopulated after the LGM, and scheletal evidence points to a migration from the north.

https://postimg.cc/yJL12WTV

Source: Late Pleistocene climate change and Paleolithic cultural evolution in northern China: Implications from the Last Glacial Maximum
(https://www.researchgate.net/publication/222924019_Late_Pleistocene_climate_change_and_Paleolithic_cultural_evolution_in_northern_China_Implications_from_the_Last_Glacial_Maximum)

Source: Craniometrics Reveal “Two Layers” of Prehistoric Human Dispersal in Eastern Eurasia(https://www.nature.com/articles/s41598-018-35426-z)

TLT said...

Looks like there was some talk about yDNA N and yDNA K2 in general here. I'll put in my own 2 cents then. Yes Ust Ishim- a very old K2a carrier, and Oase, another very old K2a carrier were found quite northward and quite westward for early east Eurasian individuals. However, I recall that around this time, there was a west Eurasian metapopulation which was distributed over a large region, a branch of which would have spread all the way to northeastern Siberia in the form of the mixed ANS people (only a few data points to go off of so there aren't any known pure west Eurasian samples in this region from 30,000+ years ago yet). Quite a lot of long distance activity going on, which would make sense since this time between the first Eurasian spread and the LGM had milder weather.
So with that established, I think its quite reasonable to find that there was an early east Eurasian metapopulation which could have been sprawled out to Siberia in the north and parts of Europe in the west. There seems to be some east Eurasian affinity in Goyet Aurignacians to corroborate this along with the outright presence of mtDNA M in GoyetQ116-1. The K2 haplogroup and subclades might or might not have originated in southeast Asia but they would have definitely originated in an early east Eurasian population. Though it would be strange if it didn't spread out from southeast Asia originally given the diversity of K2b subclades over there. This can be explained as a combination of milder pre 30K BP weather along with pushing the out of Africa migration back to something like 70,000 years ago instead of 50,000 years ago; and to not think of it as one event, rather a back and forth movement in the general pre-upper paleolithic MENA region until the 3 Eurasians started to become distinct by ~50,000 years ago. The question now is how were the K2 carriers spread early on from 45,000 to 30,000 years ago.

Anthony Hanken said...

@Kristiina

I know nothing about the prehistoric climate in Northeastern China but the DNA data we have so far would suggest somewhere in that general area. There is no problem finding an origin for N farther north except when we are talking about its deep origins YDNA O has to be taken in to account as well.

According to a post Ebizur made a while ago, modern day O has its greatest diversity in the plains region of central China. I'm not sure it makes sense for O to have a Siberian origin as well but who knows I guess.

I will also add that on Yfull certain Chinese branches seem to be older than the LGM but will concede that modern day distribution does not nessacarily tell the full story.

Anonymous said...

@TLT "The question now is how were the K2 carriers spread early on from 45,000 to 30,000 years ago."

So funny that pre-NO was found in the west in Europe - Palaeolithic Romania Oase 1 41640-37580 cal BP M K2a* (pre-NO*), although the branch NO is undoubtedly eastern Eurasian, but the western pre-P was found in the east of Eurasia - Palaeolithic China Tianyuan Cave, Zhoukoudian cave system, Beijing [Tianyuan 1] 38170-36880 calBCE M K2b.

Ryan said...

@natsunoame - That's my point.

Anonymous said...

@Anthony Hanken "I'm not sure it makes sense for O to have a Siberian origin as well but who knows I guess. "

Why not?

Palaeolithic Russia Ust'-Ishim, western Siberia [Ust_Ishim] 46364-40844 calBCE (41400±1400 BP, OxA-30190)] M K2a* (pre-NO*)

https://commons.wikimedia.org/wiki/File:Y-Haplogroup_Paleolithic_Migrations.png

Anthony Hanken said...

@Archi

Well its not impossible but Ust-Ishim and Oase are dead lineages. O is rarely found outside of Asia in ancient or modern DNA.

Anonymous said...

Anthony Hanken "Well its not impossible but Ust-Ishim and Oase are dead lineages. O is rarely found outside of Asia in ancient or modern DNA."

What's impossible? It's possible and still as! Specific populations in Europe and China are extinct, but not the Siberian branches themselves!

Palaeolithic Russia Yana river, north Siberia [Yana1 and Yana2] 32047-31321 cal BP (27940 ± 115) M P1 2 samples

So it's impossible to claim now that they couldn't have happened in the north.


epoch said...

"On the other hand, the views that the Uralic language family is native to Northern Europe and/or closely associated with the CWC are fringe theories usually espoused by people not familiar with the topic or, unfortunately it has to be said, mentally unstable trolls."

Peter Schrijver, not a troll and a pretty well respected linguist, considers a Uralic substrate in Germanic, concluding that Germanic was a IE languages imposed on native Uralic speaker.

His line of reasoning is that a very typical Germanic sound shift, Verner's law, has a pretty exact replica in Finnish. There, it is considered a so called onsonant gradation, a phenomenon known in more Uralic languages, although with different characteristics.

https://en.wikipedia.org/wiki/Consonant_gradation
https://en.wikipedia.org/wiki/Verner%27s_law

I think also Parpola at least for some times considered the option.

epoch said...

I don't think there is any other evidence for Uralic languages in Scandinavia - not a trace in toponyms - so there must be a different explanation. One could be that the Finnish consonant degradation is simply a result from CWC or Nordic Bronze Age in Finland, i.e. it is independent from such phenomena in other Uralic languages.

Or some older shared substrate.

It *is* interesting though. No other IE language has such a thing where sound shifts depend on stress of syllable following it, if I understand it all correctly.

Anonymous said...

"There, it is considered a so called onsonant gradation, a phenomenon known in more Uralic languages, although with different characteristics."

The very origin of the consonant gradation in the Finnish is mysterious, as it does not exist in other Uralic languages. Most likely, it dates back to the ancient European(?) substratum.

"Peter Schrijver, not a troll and a pretty well respected linguist, considers a Uralic substrate in Germanic, concluding that Germanic was a IE languages imposed on native Uralic speaker."

Therefore, he is mistaken, he simply does not consider the fact that in Finnish it is a unique phenomenon even in relation to other Uralic, and therefore could not be passed on to anyone from the Finns, because the Finns borrowed it themselves.

Kristiina said...

"I will also add that on Yfull certain Chinese branches seem to be older than the LGM"

When I right now look at the yfull N tree, I see that TMRCA of N1-Z4787, formed 21700 ybp, is only 17800 ybp, which means that it arose when the climate started improving after the LGM.

TMRCA of the Chinese N1b branch is 15600 ybp. TMRCA of N1a branch which covers Baikal Neolithic, all Uralic N and East Asian TAT is 16 000 ybp. TMRCA of N1a2b-L666 which is the Baikal Neolithic line is 8600 ybp. TMRCA of Chinese TAT is 6700 ypb and TMRCA of Western TAT is 10800 ybp.

https://www.yfull.com/tree/N/

epoch said...

Samoyedic languages have a form of consonant gradation:

https://en.wikipedia.org/wiki/Consonant_gradation#Samoyedic_languages

Rob said...

@ Ryan

“Prove it. Show that "all linguists agree Indo-Uralic is true."”

Has Archie yet been able to prove anything ?

Anthony Hanken said...

@Kristiina

So you suppose that N* was in a Siberian refugium during the LGM spread to China when the climate became hospitable? If so would O* have been in a similer refugium?

It sounds plausible but wouldn't that still make European N lines decended from the post expansion Chinese subclades? They are still decended from N-Z4762 after all with the exception of N2.

Anonymous said...

@epoch "Samoyedic languages have a form of consonant gradation"

They just seem similar, but they function differently. Therefore, in Samoyedic languages it is called an alternation consonants, not a consonant gradation.

Kristiina said...

@ Anthony

"So you suppose that N* was in a Siberian refugium during the LGM spread to China when the climate became hospitable? If so would O* have been in a similer refugium?"

First of all, I am waiting for ancient DNA data from China. Modern diversity is not very relevant.

TMRCAs of O clades are clearly older than TMRCAs of the two existing N lines, N1 and N2. Main branches O-F265 and O-M122 are 29300 ybp and 28800 ybp, respectively. TMRCA of O-M268 which includes the Japanese and Korean specific old O2b and Austroasiatic O2a is 28000 ybp. TMRCA of O-F265 which covers Taiwanese and South Chinese old O1a is 17600.

N and O separated already 36800 years ago which is really a very long time ago.

I presume that O entered China before LGM, my guess is more than 30 000 years ago.

"wouldn't that still make European N lines decended from the post expansion Chinese subclades"

Why? TMRCA of the Chinese N1b branch is 15600 ybp while TMRCA of N1a branch which covers Baikal Neolithic, all Uralic N and East Asian TAT is 16 000 ybp, which means that N1a is slightly older than N1b. As the dfference is only 400 years, N1 was probably relatively close to Northeast China. Baikal fullfils that criterion.

Ric Hern said...

Denisovans seems to have been spread from the Altai as far as Southeast Asia...How did some groups avoid admixture when they seemingly originated in that area.

My guess is that K, K2a+b originated somewhere is Anatolia/Caucasus/Northwestern Iran then moved North. Some migrated East using a Northern Route basically sidestepping the Denisovans ?

Mauri said...

@Davidski,

I am analyzing our Levanluhta case and updating my blog tomorrow.

Anthony Hanken said...

@Kristiina

I agree we need more aDNA from China and that modern DNA is not especially useful.

If we are looking at modern DNA, on YFull N-L792 looks basel in China as well as N1b. There is even a basel N-L792* branch from Shaanxi. The oldest BHGs are only from the 7500ybp IRC and there is a branch of N-L666 with a TMRCA of 7800ybp in China. Of course this branch may have come from Baikal but we can't say for sure that this is the case.

The fact of the matter is we know there are diverse and relatively old clades of N in both the Baikal and Chinese neolithic and on Yfull subclades look most basel in China. This is as a pretty objective observation however maybe the fact that NE-China was inhospitable points to Baikal.

Anonymous said...

@Ric Hern "My guess is that K, K2a+b originated somewhere is Anatolia/Caucasus/Northwestern Iran then moved North."

Do you think that people there lived in the early Upper Paleolithic? But you will be surprised, practically did not live. At that time, there was an area there that was practically not adapted to human life. There are practically no ancient people's camps there. A man enters Turkey from the Balkans to the Bosporus region.
Human camps appear in the North Caucasus, but not in the South.

Homeland K where F, Near East cannot be homeland F.

Rob said...

@ epoch

“”
I don't think there is any other evidence for Uralic languages in Scandinavia - not a trace in toponyms - so there must be a different explanation. One could”

Toponyms are an almost useless line of evidence; long misused
What’s the common thread ? Battled Axe..

Anonymous said...

@Rob gehn "early cwc wasnt IE"

Go to Carlos Quelles, you'll be welcomed there, you've never been right about anything, and you haven't been able to prove anything to anyone.

@Anthony Hanken "The oldest BHGs are only from the 7500ybp IRC and there is a branch of N-L666 with a TMRCA of 7800ybp in China."

You can't see that, you consider marginal branch that was at the beginning it to be a basal, but that's a mistake. The fact is that China borders both with the Baikal region and Amur river. Therefore, everything that is located on the territory of eastern Baikal and Amur, all this inevitably gets to China, but this is North Chine. You can see how far migrations have gone, that there are Koreans and Japanese in the late branches, as well as Vietnamese. In general, there is no sense in separating eastern Baikal, Amur, and Northern China (Manchuria). The branch N-F1101 most likely belonged to Altaic languages such as Manchurian or most likely Japan-Korean.


Rob said...

@ Archie

Prove what language BAx spoke. If Germanic substrate is related to FU then BAx is a possible link
In any case; Germanic arrived late to Scandinavia
And don’t get too lippy; you’d drop in a second

Davidski said...

@Rob

Battle-Axe is considered Indo-European based on linguistic, archeological and genetic data.

It's the ancient culture that links Germanic speaking groups to Balts, Slavs and Indo-Iranians archeologically and genetically.

The idea that Battle-Axe was a Finnic culture doesn't stand a chance. It'll never be accepted by any serious scholars because it doesn't make any sense in terms of linguistics, archeology or genetics.

Davidski said...

By the way, Battle-Axe need not have been the sole precursor to Germanic to be the link between Germanic, Balto-Slavic and Indo-Iranian.

Its people may even have spoken a now extinct Indo-European language, but Indo-European nonetheless. A theory like this is outlined in this recent thesis:

The shared lexicon of Baltic, Slavic and Germanic

Rob said...

I’m aware of that; but that’s a rather primordialistic / rigid model
There are nevertheless later north European contacts
But somehow; it seems that BAx is important for FU.
I see no need for mutual exclusivity

«Oldest ‹Older   1 – 200 of 327   Newer› Newest»