Steppe Ancestry in western Eurasia and the spread of the Germanic Languages

[Anglesqueville]

^^ The problem, or rather the source of our difficulties of understanding (I say "our" out of prudence and modesty, I doubt in fact that it is only ours), is not only, perhaps not even essentially, the use of IBDs in itself, but the quantification which is made by McColl in the immediate line of Allentoft. This quantification, which uses ADMIXTURE in a very specific manner in supervised mode, is described in detail by Allentoft in the notes of his major 2022 study. My trouble comes from the fact that I see nothing in this method which can protect it from the phenomena of concentration of IBDs so natural to populations having suffered bottlenecks or isolates phenomena. Basically, what would happen to a group with a profile resembling my own case? I've said this a bunch of times before, but I'll repeat it in haste. I have a Finnish great-great-grandmother, everything else is a North-West European cocktail. I look at MyHeritage, which has the advantage of using a classic method (as far as I know) for their ethnic compositions, and a count of shared IBDs on imputed genomes for the count of genetic matches. It's probably not as refined as the Allentoft method, but it's pretty similar. As for my ethnic breakdown, look at my signature: 4% Finnish and around ten% Baltic, I have nothing to complain about, it's not that bad. Now look at the list of my genetic matches:
Etats-Unis d'Amérique
640
Finlande
545
Suède
177
France
172
...
If we refined this count it would be worse. For example, out of my 10 Spanish matches, 2 are half Finnish, and all the matches I have in common with them are from Finland. It's basically the same thing for all my Scandinavian and Russian matches. In short, my “Finnish” IBDs drown out everything. I am well aware that Finland is a special case, noticed for a long time. But what guarantees us that the problem, paroxysmal in my case, does not pollute in a more subtle way the quantifications made by the authors? I don't know, and really wish the authors thought about protecting themselves from this in some way.

[Queequeg]

^^ The problem...is not only, perhaps not even essentially, the use of IBDs in itself, but the quantification which is made by McColl in the immediate line of Allentoft. This quantification, which uses ADMIXTURE in a very specific manner in supervised mode, is described in detail by Allentoft in the notes of his major 2022 study. My trouble comes from the fact that I see nothing in this method which can protect it from the phenomena of concentration of IBDs so natural to populations having suffered bottlenecks or isolates phenomena.

Very good point indeed. While i'm not sure to what extent the use of ADMIXTURE in this case possibly leads into overinterpretations or even wrong, such as anachronistic interpretations, I somehow suspect that it might be the case.

[Anglesqueville]

^^ You are right to use the verb “suspect”. In an environment, that of amateur genomics, where people are inclined to say “for” or “against” anything, we must remain cautious and nuanced. I suspect the possibility of a problem, and declare that I will remain attentive to the details of the methods (as much as I am able). But if tomorrow a specialist proves to me that my concern was unfounded, I will be sincerely happy. Because the use of IBDs undoubtedly constitutes a significant step forward.

[Psynome]

Here's a recent paper that provides some great climate context for the earlier time periods of the genomic study: 

https://www.sciencedirect.com/science/ar...9123004390

Understanding climate resilience in Scandinavia during the Neolithic and Early Bronze Age (Bunbury 2023)

Highlights

Neolithic communities permanently settled in S Scandinavia ∼3800–3700 BCE.

The spread of farming in S Norway coincided with cooling periods after 2250 BCE.

Severe cooling periods (1850–1450 BCE) led to short-term demographic decline.

Economic intensification reflects human resilience in Southern Scandinavia.

Economic diversification reflects risk reduction in Arctic Norway

.

I'm especially interested in the Late Neolithic period, since that's the time of the theoretical formation of this mysterious Eastern Scandinavian cluster. While this paper unfortunately doesn't have data from east and central Sweden,  the paper does mention what I believe may be a crucial factor: the 4.2 Kya climate event, which led to sustained cooling in S Norway, S Sweden, and Denmark. We've seen before that climate shifts can have dramatic impacts on human migration and subsistence patterns. This could be another case.

[Russki]

Rodoorn,

>R1b U106 Z18 has an early presence on Zealand 2194-2026 calBCE.
>The NW Block has Unetice derived Z304/ Z381. Presence since BA.

R1b-U106 is quite rare among modern Irish, in particular Western Irish, although they have some English influence, it is also somewhat rare among Highland Scots and Welsh, from this I deduce that the Dutch Bell Beakers who replaced Neolithic British didn't carry R1b-U106 or it was quite rare among them. The influx of R1b-U106 into the area of Dutch Bell Beakers seemingly started after this event (replacement of Neolithic British) was finished, from Unetice or from later Jastorf people, probably from both.

>It's clear that Germanic is primarly Scandic, with a great touch of East-Scandic. Jastorf is imo the southern zone of NBA. With the Elbe-Havel group (aka Suebi) as the most southern Germanic tribe.
>The definite Germanization of the NW Block (especially the Western part) is after 400 AD, with the influx of the Germanics with at that moment the core on Zealand, central place Gudme.

The area of modern Denmark and Skåne was a periphery between the Nordic culture which was rich in R1a-Z284 and the Jastorf culture where it was more or less absent. Roman and pre-Roman Zealand samples have a significant frequency of R1a-Z284. I kind of doubt that it was the starting point of Germanic expansion because in this case they would have to go down South to the mainland Europe and spread R1a-Z284 in a genetically very Germanic places like Lower Saxony, but R1a-Z284 seems more or less absent there. Y-DNA evidence points to the one-road nature of migration with R1b-U106 expanding from South to North, but R1a-Z284 not spilling over from North to South. I1 was seemingly present in both Jastorf and Nordic cultures and doesn't provide much information about the direction of migration.

How is it clear that Germanic is primarily Scandic if one of the most frequent Y-DNA types of Scandinavians did not spillover to highly Germanic (autosomally) people like NW Germans?
It does demonstrate that the Germanic people are autochtonous to the North German plain, and from there they expanded Northwards.

Can you document this?

It's from the study itself which was reposted by Rodoorn. 24 samples of R1a-Z284 among 280 Zealand samples, which constitutes a frequency of 8.6%, supporting my previous claim that the area of modern Denmark was a periphery of the Nordic culture (rich in R1a-Z284) and the Jastorf culture (no R1a-Z284).

In a place like Lower Saxony you get no R1a-Z284 at all.

This is the extent of the Jastorf culture showing Zealand as a periphery/mixed zone:

[Capitalis]

Bear with me here...

Before creating a West/North/Central European Global25-based PCA plot, I always run an 'extended' West Eurasian PCA plot (WHG, EHG, CHG, Iran N, Israel N, Anatolia N, Morocco N) to check for signs of non-W/N/C European ancestry in the ancient samples, as significant non-W/N/C European ancestry will affect the 'spread' of the W/N/C European samples I am interested in. I also check for East Asian ancestry, as many ancient Hungarian samples have this.

I remove any ancient samples which plot significantly away from the main cluster of W/N/C Europeans. I leave any ancient and modern samples with only minor signs of non-W/N/C European ancestry. As an example, modern sample Netherlands27 appears to have some kind of minor (probably South-) East Asian ancestry, but it's not enough to impact the W/N/C European PCA plot.

I tell you this, to show that I do put some care into plots such as the one I posted at this forum last Sunday. Sample VK145, buried in Oxfordshire, England during the Viking Age, is highlighted with a red circle. This sample plots alongside modern Sweden, and not towards the eastern edge of modern Poland (i.e. the Baltics). This is clearly a Northwest European sample. Or so you would think.



Reading through Steppe Ancestry in western Eurasia and the spread of the Germanic Languages for the second time, on pp. 87-88 of Supplementary Notes 2-7, I noticed that sample VK145 has been identified by McColl et al. as belonging to their Baltic IBD cluster.



So I returned to p. 24 of the main paper to check my understanding of their Baltic IBD cluster. The Baltic IBD cluster samples in the image below are pink, and unsurprisingly are mainly found in the Baltic from the Bronze Age onwards.



From p. 25 of the main paper,

By using Iron Age sources for Western, Southern and Eastern Scandinavians (set 6, Extended
Data Figure 6), we are able to ascertain more specific source populations and regions for
migrations previously described more broadly to Northern Europe (Gretzinger, Langobards,
Stolarek). South of the Nordic region, the Jutlandic Iron Age source to be the primary
Scandinavian ancestry to the west (present day Germany, the Netherlands and England).
Further east, populations of present-day Poland, Lithuania, Latvia, Estonia, Öland and
Finland are primarily mixtures of Eastern Scandinavian and Baltic Bronze Age ancestries.

So this all just confirms that they are defining 'Baltic' in the way that I would expect it to be defined, and that they are using it appropriately in modelling.

Being a good citizen scientist (!), I opened Supplementary Table S2 to check McColl et al.'s cluster assignments for the two Viking England samples that plot near VK145, namely VK143 and VK263.



Having flashbacks to my past attempts to understand,

a) the complex supplementary section of The formation of human populations in South and Central Asia (2019), and
b) the modelling decisions of The Anglo-Saxon migration and the formation of the early English gene pool (2022),

I laughed silently to myself and wondered why I chose this sick hobby.

And then I closed all of my PDFs, all of my Excel spreadsheets, proofread this post, and clicked 'Post Reply'.

Good night. :-)

[Rodoorn]

^^ The problem, or rather the source of our difficulties of understanding (I say "our" out of prudence and modesty, I doubt in fact that it is only ours), is not only, perhaps not even essentially, the use of IBDs in itself, but the quantification which is made by McColl in the immediate line of Allentoft. This quantification, which uses ADMIXTURE in a very specific manner in supervised mode, is described in detail by Allentoft in the notes of his major 2022 study. My trouble comes from the fact that I see nothing in this method which can protect it from the phenomena of concentration of IBDs so natural to populations having suffered bottlenecks or isolates phenomena. Basically, what would happen to a group with a profile resembling my own case? I've said this a bunch of times before, but I'll repeat it in haste. I have a Finnish great-great-grandmother, everything else is a North-West European cocktail. I look at MyHeritage, which has the advantage of using a classic method (as far as I know) for their ethnic compositions, and a count of shared IBDs on imputed genomes for the count of genetic matches. It's probably not as refined as the Allentoft method, but it's pretty similar. As for my ethnic breakdown, look at my signature: 4% Finnish and around ten% Baltic, I have nothing to complain about, it's not that bad. Now look at the list of my genetic matches:
Etats-Unis d'Amérique
640
Finlande
545
Suède
177
France
172
...
If we refined this count it would be worse. For example, out of my 10 Spanish matches, 2 are half Finnish, and all the matches I have in common with them are from Finland. It's basically the same thing for all my Scandinavian and Russian matches. In short, my “Finnish” IBDs drown out everything. I am well aware that Finland is a special case, noticed for a long time. But what guarantees us that the problem, paroxysmal in my case, does not pollute in a more subtle way the quantifications made by the authors? I don't know, and really wish the authors thought about protecting themselves from this in some way.

Insightful you talked about it earlier, this makes your point ones and again clear Angels. Hopefully some authors of the paper are watching too!

But it also makes clear that some disbelieve that in crowded house NW Europe there are still some folks in "fringes"  that still come close to the original source population. All within perfect responsible statistical marges (@Jaska). To quote you in the case of the Qpadm about the Danish Isles IA samples and my parents: 'the p-values are astronomical.'

[Rodoorn]

Rodoorn,

>R1b U106 Z18 has an early presence on Zealand 2194-2026 calBCE.
>The NW Block has Unetice derived Z304/ Z381. Presence since BA.

R1b-U106 is quite rare among modern Irish, in particular Western Irish, although they have some English influence, it is also somewhat rare among Highland Scots and Welsh, from this I deduce that the Dutch Bell Beakers who replaced Neolithic British didn't carry R1b-U106 or it was quite rare among them. The influx of R1b-U106 into the area of Dutch Bell Beakers seemingly started after this event (replacement of Neolithic British) was finished, from Unetice or from later Jastorf people, probably from both.

>It's clear that Germanic is primarly Scandic, with a great touch of East-Scandic. Jastorf is imo the southern zone of NBA. With the Elbe-Havel group (aka Suebi) as the most southern Germanic tribe.
>The definite Germanization of the NW Block (especially the Western part) is after 400 AD, with the influx of the Germanics with at that moment the core on Zealand, central place Gudme.

The area of modern Denmark and Skåne was a periphery between the Nordic culture which was rich in R1a-Z284 and the Jastorf culture where it was more or less absent. Roman and pre-Roman Zealand samples have a significant frequency of R1a-Z284. I kind of doubt that it was the starting point of Germanic expansion because in this case they would have to go down South to the mainland Europe and spread R1a-Z284 in a genetically very Germanic places like Lower Saxony, but R1a-Z284 seems more or less absent there. Y-DNA evidence points to the one-road nature of migration with R1b-U106 expanding from South to North, but R1a-Z284 not spilling over from North to South. I1 was seemingly present in both Jastorf and Nordic cultures and doesn't provide much information about the direction of migration.

How is it clear that Germanic is primarily Scandic if one of the most frequent Y-DNA types of Scandinavians did not spillover to highly Germanic (autosomally) people like NW Germans?
It does demonstrate that the Germanic people are autochtonous to the North German plain, and from there they expanded Northwards.

Can you document this?

It's from the study itself which was reposted by Rodoorn. 24 samples of R1a-Z284 among 280 Zealand samples, which constitutes a frequency of 8.6%, supporting my previous claim that the area of modern Denmark was a periphery of the Nordic culture (rich in R1a-Z284) and the Jastorf culture (no R1a-Z284).

In a place like Lower Saxony you get no R1a-Z284 at all.

This is the extent of the Jastorf culture showing Zealand as a periphery/mixed zone:

The paper and the added info by people like Angles have imho made clear Russki that the core of the Germanic IA (Roman times/ Migration time) development was in nowadays Denmark, with Gudme as center.

Does this mean Jastorf is irrelevant with regard to the Germanic lineage. No I still think Nordic Bronze Age, at the southern outskirts of it there was the ELbe-Havel group that is founding for the Suebi/Elbe Germanics (= Jastorf).

Jastorf bordered in the west towards the NW Block, that's in the paper called East North Sea Cluster, basically Bell Beaker derived. From Kuhn et al (1962) to Kuzmenko (2011) and Schrijver (2017) seen as a folk speaking a language somewhere in the "Italo-Celtic" range....some think that these people spoke Germanic. But before migration ages there is no single evidence for it (pure speculative). We know for sure that these spoke a from of Germanic after the spread of the Saxons.

[Naudigastir]

Rodoorn,

>R1b U106 Z18 has an early presence on Zealand 2194-2026 calBCE.
>The NW Block has Unetice derived Z304/ Z381. Presence since BA.

R1b-U106 is quite rare among modern Irish, in particular Western Irish, although they have some English influence, it is also somewhat rare among Highland Scots and Welsh, from this I deduce that the Dutch Bell Beakers who replaced Neolithic British didn't carry R1b-U106 or it was quite rare among them. The influx of R1b-U106 into the area of Dutch Bell Beakers seemingly started after this event (replacement of Neolithic British) was finished, from Unetice or from later Jastorf people, probably from both.

>It's clear that Germanic is primarly Scandic, with a great touch of East-Scandic. Jastorf is imo the southern zone of NBA. With the Elbe-Havel group (aka Suebi) as the most southern Germanic tribe.
>The definite Germanization of the NW Block (especially the Western part) is after 400 AD, with the influx of the Germanics with at that moment the core on Zealand, central place Gudme.

The area of modern Denmark and Skåne was a periphery between the Nordic culture which was rich in R1a-Z284 and the Jastorf culture where it was more or less absent. Roman and pre-Roman Zealand samples have a significant frequency of R1a-Z284. I kind of doubt that it was the starting point of Germanic expansion because in this case they would have to go down South to the mainland Europe and spread R1a-Z284 in a genetically very Germanic places like Lower Saxony, but R1a-Z284 seems more or less absent there. Y-DNA evidence points to the one-road nature of migration with R1b-U106 expanding from South to North, but R1a-Z284 not spilling over from North to South. I1 was seemingly present in both Jastorf and Nordic cultures and doesn't provide much information about the direction of migration.

How is it clear that Germanic is primarily Scandic if one of the most frequent Y-DNA types of Scandinavians did not spillover to highly Germanic (autosomally) people like NW Germans?
It does demonstrate that the Germanic people are autochtonous to the North German plain, and from there they expanded Northwards.

Can you document this?

It's from the study itself which was reposted by Rodoorn. 24 samples of R1a-Z284 among 280 Zealand samples, which constitutes a frequency of 8.6%, supporting my previous claim that the area of modern Denmark was a periphery of the Nordic culture (rich in R1a-Z284) and the Jastorf culture (no R1a-Z284).

In a place like Lower Saxony you get no R1a-Z284 at all.

This is the extent of the Jastorf culture showing Zealand as a periphery/mixed zone:

I wouldn't assume all the R1a is going to be Z284. There are also more NW European continental variants like L664 and those of Balto-Slavic origin. I found eight probable Z284-samples in Denmark from the study in total (ISOGG R1a1a1b1a3 or downstream of it - from S2. Clustered ancient samples), two of which are related to Norwegian IA-influx in Northern Jutland. Regardless, Z284 among early continental Germanics is rare but not unheard of, I seem to remember there was one among the Lombards.

[Rodoorn]

I'm curios what the position of the Salian Franks is in this respect!

Latest theory about them is that they were a kind "resistance movement" or group against the incoming Saxons (see Van der Tuuk 2021 and Seebold 2013)

I guess they represent in genetic sense another cluster than the Saxons.

This leads me to Lao (2013) about the population structure in the Netherlands. There is a chart with an admixture, and the "clusters" weren't explained in the paper.

But I think  clear that yellow is connected to the incoming Anglo-Saxons and their Dutch Frisian and Saxon heirs, especially in the North-East.

The pink color are imo the Salian Franks, what is called Central North is about the Salland area in the Netherlands, and they also went to the South, towards Belgium, Northern France to found- what's in a name- the state of France.

Still guessing what the orange component is, seems like the East North Sea cluster (ESN) of the paper?

[Rodoorn]

Seebold (2013) after Hines (with some adds by me




Attested by the paper? I guess so....

[Russki]

Can you document this?

It's from the study itself which was reposted by Rodoorn. 24 samples of R1a-Z284 among 280 Zealand samples, which constitutes a frequency of 8.6%, supporting my previous claim that the area of modern Denmark was a periphery of the Nordic culture (rich in R1a-Z284) and the Jastorf culture (no R1a-Z284).

In a place like Lower Saxony you get no R1a-Z284 at all.

This is the extent of the Jastorf culture showing Zealand as a periphery/mixed zone:

I wouldn't assume all the R1a is going to be Z284. There are also more NW European continental variants like L664 and those of Balto-Slavic origin. I found eight probable Z284-samples in Denmark from the study in total (ISOGG R1a1a1b1a3 or downstream of it - from S2. Clustered ancient samples), two of which are related to Norwegian IA-influx in Northern Jutland. Regardless, Z284 among early continental Germanics is rare but not unheard of, I seem to remember there was one among the Lombards.

According to eupedia the frequency of R1a among modern Danes is 15%. According to Sanchez et al. 2004 the frequency of R1a-Z284 among Danes is 16.5%. So almost all R1a in Denmark is a Battle Axe-descended Z284 type, with little to none L664 or Balto-Slavic.






Rodoorn,

I agree that Gudme was the center of Germanic culture in the Roman times. But the transmission of Germanic language and culture together with migration of R1b-U106 Northwards, into the area of the Battle Axe-descended Nordic culture in Scandinavia, has started way earlier than that. Around 6th century BC, about 2600 years ago. The dominant culture in Europe at that time was the Greek culture.

[Rodoorn]

Rodoorn,

I agree that Gudme was the center of Germanic culture in the Roman times. But the transmission of Germanic language and culture together with migration of R1b-U106 Northwards, into the area of the Battle Axe-descended Nordic culture in Scandinavia, has started way earlier than that. Around 6th century BC, about 2600 years ago. The dominant culture in Europe at that time was the Greek culture.

Imho populations and population movements are in iron age not strictly one -Y-Dna- track minded....

With regard to R1b U106- taken to the core- lines like Z304 have a more NW Block kind of spread. Z18 is more Scandic.  Z18 came to Friesland in EMA with the migration ages. The (definite) Germanization of the NW Block was in EMA.

https://docs.google.com/spreadsheets/d/1...1221068102

[Naudigastir]

It's from the study itself which was reposted by Rodoorn. 24 samples of R1a-Z284 among 280 Zealand samples, which constitutes a frequency of 8.6%, supporting my previous claim that the area of modern Denmark was a periphery of the Nordic culture (rich in R1a-Z284) and the Jastorf culture (no R1a-Z284).

In a place like Lower Saxony you get no R1a-Z284 at all.

This is the extent of the Jastorf culture showing Zealand as a periphery/mixed zone:

I wouldn't assume all the R1a is going to be Z284. There are also more NW European continental variants like L664 and those of Balto-Slavic origin. I found eight probable Z284-samples in Denmark from the study in total (ISOGG R1a1a1b1a3 or downstream of it - from S2. Clustered ancient samples), two of which are related to Norwegian IA-influx in Northern Jutland. Regardless, Z284 among early continental Germanics is rare but not unheard of, I seem to remember there was one among the Lombards.

According to eupedia the frequency of R1a among modern Danes is 15%. According to Sanchez et al. 2004 the frequency of R1a-Z284 among Danes is 16.5%. So almost all R1a in Denmark is a Battle Axe-descended Z284 type, with little to none L664 or Balto-Slavic.

That's not what it says. It just details the frequency of any and all R1a-lineages in Denmark. Not very telling. Most likely the majority of R1a in Denmark is not below Z284. As indicated here, the R1a in Denmark are mostly of different clades than in Sweden and Norway (even given the small sample-size it's not likely to be random):
https://www.nature.com/articles/s41431-0.../figures/3

[Uintah106]

I wouldn't assume all the R1a is going to be Z284. There are also more NW European continental variants like L664 and those of Balto-Slavic origin. I found eight probable Z284-samples in Denmark from the study in total (ISOGG R1a1a1b1a3 or downstream of it - from S2. Clustered ancient samples), two of which are related to Norwegian IA-influx in Northern Jutland. Regardless, Z284 among early continental Germanics is rare but not unheard of, I seem to remember there was one among the Lombards.

According to eupedia the frequency of R1a among modern Danes is 15%. According to Sanchez et al. 2004 the frequency of R1a-Z284 among Danes is 16.5%. So almost all R1a in Denmark is a Battle Axe-descended Z284 type, with little to none L664 or Balto-Slavic.

That's not what it says. It just details the frequency of any and all R1a-lineages in Denmark. Not very telling. Most likely the majority of R1a in Denmark is not below Z284. As indicated here, the R1a in Denmark are mostly of different clades than in Sweden and Norway (even given the small sample-size it's not likely to be random):
https://www.nature.com/articles/s41431-0.../figures/3

Its obvious R1a got shoved northward in Scandinavia, where it thrived. I am amazed that I1 is ubiquitous From Finland west, in Northern Europe.