search this blog

Wednesday, July 17, 2019

Viking invasion at bioRxiv


A new preprint featuring hundreds of Viking Age genomes has appeared at bioRxiv [LINK]. Titled Population genomics of the Viking world, it looks like a solid effort overall, although I'm skeptical about its conclusions. I might elaborate on that in the comments below, but I'll have a lot more to say on the topic if and when I get to check out the ancient genomes with my own tools. Details about the new samples, including their Y-chromosome haplogroup assignments, are available here. Below is the abstract, emphasis is mine:

The Viking maritime expansion from Scandinavia (Denmark, Norway, and Sweden) marks one of the swiftest and most far-flung cultural transformations in global history. During this time (c. 750 to 1050 CE), the Vikings reached most of western Eurasia, Greenland, and North America, and left a cultural legacy that persists till today. To understand the genetic structure and influence of the Viking expansion, we sequenced the genomes of 442 ancient humans from across Europe and Greenland ranging from the Bronze Age (c. 2400 BC) to the early Modern period (c. 1600 CE), with particular emphasis on the Viking Age. We find that the period preceding the Viking Age was accompanied by foreign gene flow into Scandinavia from the south and east: spreading from Denmark and eastern Sweden to the rest of Scandinavia. Despite the close linguistic similarities of modern Scandinavian languages, we observe genetic structure within Scandinavia, suggesting that regional population differences were already present 1,000 years ago. We find evidence for a majority of Danish Viking presence in England, Swedish Viking presence in the Baltic, and Norwegian Viking presence in Ireland, Iceland, and Greenland. Additionally, we see substantial foreign European ancestry entering Scandinavia during the Viking Age. We also find that several of the members of the only archaeologically well-attested Viking expedition were close family members. By comparing Viking Scandinavian genomes with present-day Scandinavian genomes, we find that pigmentation-associated loci have undergone strong population differentiation during the last millennia. Finally, we are able to trace the allele frequency dynamics of positively selected loci with unprecedented detail, including the lactase persistence allele and various alleles associated with the immune response. We conclude that the Viking diaspora was characterized by substantial foreign engagement: distinct Viking populations influenced the genomic makeup of different regions of Europe, while Scandinavia also experienced increased contact with the rest of the continent.

Margaryan et al., Population genomics of the Viking world, bioRxiv, posted July 17, 2019, doi: https://doi.org/10.1101/703405

See also...

They came, they saw, and they mixed

Who were the people of the Nordic Bronze Age?

Asiatic East Germanics

268 comments:

1 – 200 of 268   Newer›   Newest»
Synome said...

@Davidski
I
looks like a great effort overall to me. Which parts are you feeling skeptical about? There's certainly many details here to be looked over.

Davidski said...

I suspect that the authors are taking the Chromopainter results too literally, but it's hard to judge that for now.

Dave the Slothtopus said...

I'm excited because they finally found another DF19 (VK333) other than the one from Roman York, but they label him as "non-local" to Oland (and their graphics are nearly illegible for my screen), so I don't know how much info one can glean, there.

Razib Khan said...

david, tbh had same thought about chromopainter...

Dave the Slothtopus said...

Looks like VK333 had filed incisors, so whatever he was, it looks like he was partaking in the Viking lifestyle.

Davidski said...

What are filed incisors?

Dave the Slothtopus said...

https://www.history.com/news/did-viking-warriors-bare-groovy-teeth

Davidski said...

Polish-like Viking on Funen Island...

Denmark_Galgedil VK139 Y-hg: R1a1a1b1a1a1c1 (L1029)

More details here...

https://www.biorxiv.org/content/biorxiv/early/2019/07/17/703405/DC3/embed/media-3.xlsx?download=true

andrew said...

Obvious observation of the day: "We find that the period preceding the Viking Age was accompanied by foreign gene flow into Scandinavia from the south and east"

I would have been shocked indeed if there was foreign gene flow from the North, and would be very surprised foreign gene flow from the West at least prior to Lief Erkison, or if there was more than de minimus gene flow from the west at any time.

"we sequenced the genomes of 442 ancient humans from across Europe and Greenland ranging from the Bronze Age (c. 2400 BC) to the early Modern period (c. 1600 CE),"

I would be particularly interested to see if there was population continuity or population change between the early and late parts of the Nordic Bronze Age as the archaeological record suggests that there may have been discontinuity at that point and it is quite relevant to the linguistic history of the region. There is some indication that the early Nordic Bronze Age may have had a Bell Beaker source which was at least partially replaced with a Corded Ware derived source in the late Nordic Bronze Age.

andrew said...

"I would be particularly interested to see if there was population continuity or population change between the early and late parts of the Nordic Bronze Age"

So much for that vain hope. Only two individuals in the sample are prior to 70 CE, one woman and one man from the early Nordic Bronze Age (VK214 and VK531). One can get a little bit of autosomal comparison out of that, but it doesn't say much and is prone to outlier effects, even at the autosomal level with a group known for maritime mobility from a very early date.

Romulus said...

Anyone able to reconcile these Y HG assignments with SNPs? this yLeaf software they used must be using a new(er) tree.

could be some errors here, but a rough attempt at adding up the most common hgs
297 Y DNA Haplogroups total

94 I1 32%
84 R1b 28%
61 R1a 21%
17 N1 6%
14 I2 5%

Romulus said...

They use the 2018 or later version of the ISOGG tree FYI.

Davidski said...

@Romulus

For the Y-haplogroup round up, check out page 12 here...

Supplementary Note 2

Davidski said...

Can anyone point me to a single, clear figure or table in the preprint showing the genetic shifts in Scandinavia from the LN through the BA and IA to the VA?

Bob Floy said...

So these Vikings had almost even amounts of R1a, R1b, and I1?
Interesting.

dsjm1 said...

I was hoping they might have looked at R1b-S1194(CTS4528/DF100) as all my own investigations suggest our appearance in UK was dominantly due to Danish incursions & occupation in the Viking Era. The paper by Myres et al. (2010) had pointed out Sth Baltic as a hot-spot for S1194 (back then they called us L11*). They even pinpointed the Island of Bornholm as possibly 10% L11*. The only other hot-spot was Sth Alps and this adds some contention. One line of reasoning says Sth Alps was a destination of invading Germanic/Scandinavians (particularly the Lombards & possibly the Burgundians (Burgundians are linked to Bornholm)). A second line of thinking by others argues that Sth Alps was the point of origin & Sth Baltic & UK came later. Personally I favour the 1st possibility.

But the third line of argument by a few amateur researchers is that S1194 (L11*) originated in Sth Holland and went to the other 3 places. This line is hardest for me to follow as IIRC the area they think of, Sth Braban,t was a swamp until comparatively modern times and never likely to be the starting point 4,500 or so years ago for S1194 (L11*).

As with all ideas on origins, we can but look at current studies and make the best assumptions and worry later about what gets found as more data emerges.

Doug M

dsjm1 said...

@Bob Floy ...
"So these Vikings had almost even amounts of R1a, R1b, and I1?
Interesting.

July 17, 2019 at 9:12 PM"

From what I see in my data, that seems to be the same today but I1 density seems to increase across Denmark towards Sweden (from Jutland, over Fyn and into Sylland (Zealand). And, R1b-U106 (which is strong in Jutland, diminishes to a similar % moving across the same locations. I1 seems to be the strongest today, around Stockholm in Sweden. R1a seems pretty consistent throughout Sth Scandinavian countries. U106 is very strong all up Jutland and of course strongest in what was greater Frisia (Nth Netherlands, Nth West Germany, Jutland & Sth Norway.

AWood said...

@dsjm1

Stockholm is also the approximate source of the men from the Salme ships, which was rich in I1 (apparently) then as it is today.

Bastian Barx said...

I haven't had time to look at The paper yet. Is there any phenotypical data in it? Like hair and eye color.

Davidski said...

@Bastian Barx

As far as I can tell, pigmentation traits aren't listed for the individual samples, but a cohort of 262 of the Scandinavian Vikings is compared to various modern populations in terms of pigmentation and other traits starting on page 75 here...

Supplementary Note 2

Samuel Andrews said...

Macro-haplogroup mtDNA frequencies match modern northwest Europe. Natural selection by that time had shifted frequencies.

Natural selection changed mtDNA frequencies in Europe. It brought H up to 40% everywhere. Baltic Bronze age is the first group in Europe recorded with 40% H frequency. This is surprising because they were mostly of European hunter gatherer decent (who only carried mtDNA U). It shows the power of natural selection on mtDNA.

The Viking age Scandinavians also have 40% H.

Drago said...

@ Dave

See Extended Data Fig 6
Not very obvious , tbh

Wr.t. this question (having been dsicussed before), and Andrews above comment, from what I can see, compared to Battle Axe, later Scandinavians (e.g. NBA, Iron Age) are just more EEF & Baltic-WHG; but too few samples to really see a pattern

This main cultural sequence were

CWC/ BAx 2800 BC : R1a arrives
Dagger Horizon 2300 BC R1b-U106 arrives
Nordic Bronze Age 1600 BC. :

Somehow, within these periods, I1 became the preminent Nordic Y -marker. Perhaps social mobility enacted within new social codes and privileged alliances ?

Samuel Andrews said...

The abstract suggests change in pigmentation after the Viking age......
"By comparing Viking Scandinavian genomes with present-day Scandinavian genomes, we find that pigmentation-associated loci have undergone strong population differentiation during the last millennia."

Then the supplementary info contradicts this by showing Viking age samples have the same allele frequencies as modern Danes. The data also shows Finns are more blonde/blue eyed than Norse/Scandinavians.

Samuel Andrews said...

Finns descend from a inbred population so them being more blonde/blue eyed than any other pop in Europe would not be a surprise.

Interesting that natural selection drove up frequencies of blonde hair for all the post-Neolithic groups who live around the Baltic sea: Baltic Bronze age, Scandinavians, and Finns (who are mostly Scandinavian & Baltic).

Samuel Andrews said...

Viking age Scandinavians also had modern level frequencies of lactose persistence. I think Baltic Bronze age, Poland Iron age are the only other data sets with modern frequencies.
British Bronze age doesn't.

Matt said...

@Davidski, IA->VA shift in Southern European* should be visible in Fig4 as well.

BA->IA is qpadm only I believe, as the fine drift / haplotype methods would probably not work on well on such ancient individuals, and a relatively smaller sample set.

The focus is mostly on fine ancestry distinctions between relatively high steppe flows using fine IBS / IBD / ChromoPainter chunk structure, e.g. more British backflow to Scandinavia than outflow from Scandinavia, net Danish Viking flow within Scandinavia to colonise the north, etc.

*Foggia Viking Age, Southern Italian genetically and spatially

Gaska said...

@Sam Andrews-"This is surprising because they were mostly of European hunter gatherer decent (who only carried mtDNA U)"

That's NOT true, in Iberia Mit Hap-H is a typical WHG marker- For example, it has been documented in the famous cave of el Miron- Only some subclades have something to do with Neolithic farmers.

Samuel Andrews said...

They say Viking age samples fit in Norwegian, Danish, southwest Swedish, other Swedish clusters & this substructure still exists in Scandinavia today. Norwegian IA are related to Norwegian Viking age.

There's mention of Vikings in Orkney Scotland of local Pictish origin. David mentioned one Polish Viking.

Samuel Andrews said...

Salme, Estonia burial is interesting.

Viking ship burial. 700-750ad. 41 men. All them died violent deaths. All were buried with expensive weapons.

DNA was successful taken from all 41 individuals. DNA can specify they were from Sweden. DNA also shows they were from the same village-region-chiefdom. Although from the same tribe, the only family members were a group of four brothers buried side by side.

Davidski said...

Can't wait to get these genomes. I'll probably do a blog post for each one of the ancient pops.

Joey said...

@Samuel Andrews

>Finns descend from a inbred population so them being more blonde/blue eyed than any other pop in Europe would not be a surprise.

The proto-Finnic Tarand grave samples were already overwhelmingly light pigmented pre Finnish bottleneck. Sami are also from an even smaller founder population, yet they have fewer light eyes and hair. Too many times people try to explain everything about Finns with either the small Siberian contribution or their specific drift, as if thats all they can be.

Grey said...

"Sth Braban,t was a swamp until comparatively modern times and never likely to be the starting point 4,500 or so years ago for S1194"

swamps are under-rated imo - at various times in history i think coastal wetlands will have had the highest population densities (plus maybe more resistant to various plagues and famines).

Gaska said...

In terms of R1b lineages, the R1b1a1b1b-Z2105 and R1b1a1b1a-L11 clades have distinct distributions, with the first being typically associated with the Yamnaya and distributed today around the Caucasus and Volga-Uralic regions, and the latter, more commonly found in Late Neolithic, Bronze Age and later periods in Central and Western Europe. A derived status at R1b1a1b1b-Z2105 was found in a single sample from Italy (VK535). The vast majority of other R1b samples belonged instead to the R1b1a1b1a-L11 clade, and were mostly distributed within its two main subclades: P312/S116-R1b1a1b1a1a2 and U106/M405-R1b1a1b1a1a1191, of western and eastern distribution relative to the Rhine river basin"


It seems that slowly, everyone is assuming reality, that is, L51/L11/P312 is Western, they can start looking for other explanations to justify the presence of the steppe-Yamnaya ancestry or as they want to call it in Western Europe (I'm waiting for someone explain it carefully)

On the other hand, it is evident that there were population movements from southern Europe, probably linked to the bronze trade and later in the Iron Age. No people that we meet in Europe from the Iron Age will be genetically homogeneous.

The presence of R1b1a1 (pre R1b-P297) in Norway 2,400 BC complicates the situation to explain the massive steppe migrations, although surely someone is willing to swear that the CWC reached the Arctic Circle

Typologist said...

NJ tree based on autosomal admixtures from published table.

Colors are assigned to regions like Denmark, Estonia etc., I think it is self-explaining.

https://i.imgur.com/eMBrM7e.jpg

Matt said...

High level Y Chromosome Haplogroup across all their VK samples by ancestry proportions from Table S6: https://imgur.com/a/QJxdLrD

Gaska said...

Thanks Matt

Matt said...

Correlations, where available between individual level and group level Chromopainter Table S6 proportions and group level Table S7 qpAdm WHG+Anatolia+Yamnaya model proportions: https://imgur.com/a/0ov7GOE

Looks like essentially no correlation between differences in intra Scandinavian+NW European ChromoPainter components and differences in group level Yamnaya/WHG/Anatolia.

That suggests that it's unlikely that any of these clusters are being assigned by ChromoPainter as a pseudo-effect that's really driven by underlying variation in Yamnaya+WHG+Anatolian.

There's some modest correlation between Anatolian and Italian Foggia proportion, however, the plot positions of samples which are enriched in this and the extended qpAdm models seems to suggest to me this is generally *real* Southern European related ancestry (not from relict incomplete panmixa of Anatolian+Yamnaya+WHG ancestry).

Matt said...

Attempt at using the ChromoPainter Table S6 values for groups to predict qpAdm values from Table S7: https://imgur.com/a/BwI2C5Q

Some prediction based largely on the southern Italian chromopainter component is possible, but the remaining components don't predict very much.

Drago said...

So, the analysis theyve used is the classic 3-way (Yamnaya, Barcin, WHG) + using modern modals (.e.g Northwest Atlantic -like)
Wouldn't it be more optimal to do qpAdm models based on LBA samples; for example they could use Halbestadt, LBA Baltic, LBA Hungary, proto-Scythians, etc? Might be an issue with overlapping ancestry

Davidski said...

They seem to have decided to generally keep things simple and use the most familiar models of this type, which is understandable, because it'd be easy to overdo things and risk confusing most readers.

But I think there's certainly some complexity in their new data that isn't being explained very well, such as the precise character of the Southern European-like gene flow into Scandinavia after the LNBA, or even the type of LNBA ancestry that seems to have survived in southwest Sweden into the Iron Age.

This might be due to the fact that they don't have the right tools and time available to look at this sort of stuff in such fine detail yet, but the data should become available at some point so we'll try and work it out, and I reckon it should be possible.

Drago said...

Yep; & it isnt a critique on the paper. Its just an interesting issue, evident as papers move into more recent periods, where analysis is going to be more complex.

Samuel Andrews said...

National Geographic article did a big article on the genetic history of Europe. Another, example where Leftist politics try to ruin genetics.

The first Europeans weren't who you might think.
https://www.nationalgeographic.com/culture/2019/07/first-europeans-immigrants-genetic-testing-feature/

It shouldn't be controversial to say this national geographic article is a political interpretation of genetic data & that this kind of take on genetics isn't appropriate. I hope David Reich & others stand up the people who do this.

Bob Floy said...

"https://www.nationalgeographic.com/culture/2019/07/first-europeans-immigrants-genetic-testing-feature/"

Ugh.
I wouldn't count on Reich&Co. standing up to this kind of thing, since they're probably playing for the same team.

Davidski said...

These sorts of companies and labs are under some pressure because population genetics and ancestry tests have the potential to be used to encourage racism.

So I suppose they feel that they have to communicate regularly to the public, and especially to those people who are concerned, that this isn't about racial purity, and that such a concept doesn't have much sense in the context of human population genetics.

In fact, I think if they don't do this, then they run the very real risk of losing funding from advertising, scientific grants, you name it. Of course, it'd be nice if they didn't go too crazy about the whole thing, because that will hurt them too in the long run.

Bob Floy said...

"population genetics and ancestry tests have the potential to be used to encourage racism"

Only if they're presented in a really incorrect way to really stupid people. To a non-idiot all of this new information makes it clear that modern European ethnic groups originated from a blend of ancient ethnic groups, which came from disparate regions. They could just present the truth in a straightforward way, without politicizing it, and everything would be fine. But they don't.

Davidski said...

Well, there are a lot of "stupid people" around. I just deleted a racist comment from a banned troll who was raving on about a conspiracy theory.

But yeah, I totally agree, this sort of thing shouldn't be politicized either way, by the right or left.

Davidski said...

Here we go again.

Gioiello, piss off, you're banned from this blog.

Bob Floy said...

@David
"this sort of thing shouldn't be politicized either way, by the right or left."

Maybe one day, I still have hope.

"Here we go again"

He's like a bad little kid, that guy. 70 year old man acting like a middle school student in detention, ought to be f***ing ashamed of himself.

Ric Hern said...

I think they misinterpret the difference between Migrant and Immigrant. Since there were no real Borders thousands of years ago I will rather call Humans Migrants and not Immigrants. Neanderthals were not plentiful and there most probably were wide open spaces not used by Neanderthals and suited Modern Humans better....

Bob Floy said...

@Ric

Very good point, although to me they're willfully misinterpreting that.

Matt said...

@Drago, they have used other qpAdm models in the supplementary table:

Anatolia.SG_MLBA:Yamnaya.SG_EBA:Barcin.SG_EN:Loschbour.SG_M
Armenia.SG_MLBA:Yamnaya.SG_EBA:Barcin.SG_EN:Loschbour.SG_M
Botai.SG_EBA:Yamnaya.SG_EBA:Barcin.SG_EN:Loschbour.SG_M
CaucasusHG.SG_M:Yamnaya.SG_EBA:Barcin.SG_EN:Loschbour.SG_M
Saami.SG_IA:Yamnaya.SG_EBA:Barcin.SG_EN:Loschbour.SG_M
SarmatianRussia.SG_IA:Yamnaya.SG_EBA:Barcin.SG_EN:Loschbour.SG_M
Tagar.SG_IA:Yamnaya.SG_EBA:Barcin.SG_EN:Loschbour.SG_M
XiongNu.SG_IA:Yamnaya.SG_EBA:Barcin.SG_EN:Loschbour.SG_M

(Though most of these infeasible for most pops, and they have reported in main text on samples who get improved fit).

They have not bothered with qpadm testing the IA->Viking Age samples as themselves composed of other populations who can be well modelled with Yamnaya.SG_EBA:Barcin.SG_EN:Loschbour.SG_M, since I think they have doubts about the ability of qpAdm to detect geneflow between these populations, compared to the IBD and ChromoPainter chunk methods they use. (Which seems reasonable enough to me, as I don't think I have ever seeing a truly convincing qpAdm analysis of this sort?).

a said...

Davidski said...
"Here we go again.

Gioiello, piss off, you're banned from this blog."

Davidski how about we have some real diversity of ideas. Some of us Europeans are closely [related to various degrees] to each other by thousands of years[10 thousands] of isolated evolution, as evidence by our unique language, food, culture, physical customs, physical concerns genetic prone weakness etc.... I propose a new approach.How about setting up something like zoom[electronic communications-face to face] where you can interact directly with multiple European members at the same time face to face, where we can share our concerns on a genetic level- with you and other kindred spirits? Perhaps there is no trickery and treachery, we won't know unless we can communicate and exchange our diverse ideas and our genetic concerns, not on an open public forum with political stalkers who we are not related to genetically and or come from a different Gotra[Sanskrit: गोत्र) is commonly considered to be equivalent to clan].

Davidski said...

@a

Gioiello was banned for racist trolling after being given many warnings. Here are the rules if you're still confused...

New rules for comments

zardos said...

Most of the yDNA variation of modern Scandinavaians is present, but I wonder whether the carriers of J, G and E from Scandinavian places are all "regular Vikings" or show traces of recent immigration to the North? Anyone looked into that?

Which paternal lineages have those with increased Neolithic and Caucasian ancestry?

Joey said...

At least Laziridis already slammed the article on Twitter and was confuced about the African HG thing. The people in the fields cited in the article were the usual suspects, so I'm not that surprised or worried.

a said...

Blogger Davidski said...
@a

"Gioiello was banned for racist trolling after being given many warnings. Here are the rules if you're still confused..."
Yes of course. You gave warning, so as a mediator/mod- you have to lay down some basic rules.

However, I was thinking more along the lines of setting up a genetic conference exclusive for Polish surnames and genetic issues related Poles for example. A conference that we can exchange diversity of thought about issues related to Polish cultural heritage. You could mediate and answer questions like an electronic conference similar to zoom; or even invite guests and field questions --
https://www.youtube.com/watch?v=6i-NA563Ojk

Survive the Jive said...

I was quite shocked to see this quote in the supplements. "Modern brits have 9-18% ‘Italian’ ancestry, plausibly associated with the Normans and associated increase in population movement during that era. This is a two-way process, with high fractions of the ‘UK-like’ ancestry in a sub-population of Italians."

Can anyone explain to me why this was more plausibly associated with the Normans and not directly left by the Romans? Does this imply the Normans picked up Italian ancestry from Gauls/Franks who had it from the Romans? They do not specify. or Is the sub-population of Italians a region where Normans went like Sicily, indicating that there was Norman movement between England and Sicily?

I am also curious as to why previous studies have not picked up such a significant Italian influence on British DNA. I appreciate any informed input.

zardos said...

Italian and other Romance speakers beside Normans came to Britain under Norman rule, but those numbers you quote are way off for those relatively small contingents, so I wonder too what they want to imply with this.

Davidski said...

@Survive the Jive

Can anyone explain to me why this was more plausibly associated with the Normans and not directly left by the Romans?

There's actually no good reason to claim that this supposedly Italian ancestry in the UK was mediated via Norman population movements.

The authors just seem to be interpreting their Chromopainter output in a very narrow context basically limited to the Viking story and their sampling strategy.

It's more likely that this Italian or Italian-like ancestry arrived in the UK at various times and from various places, including Italy. I can think of several potential sources: Celts, Romans, Normans, and Industrial Revolution migrants from Southern Europe.

Survive the Jive said...

@Davidski

Thanks for your response. I find that very odd considering this is such a well researched and thorough study. Why would they taint it with ill -founded claims? I had thought perhaps that the absence of this signal in pre-Norman English genomes had led them to presume the admixture was Norman-related but even if they had compared the pre-Norman and post-Norman (your mention of Romans would imply they did not) then that wouldn't rule out a more recent source for the italian ancestry.

It is funny also because Reich too hinted tht his team had detected a revival of Neolithic farmer ancestry in Britain, and he attributed this not to Romans but to the outlandish idea of an Asterix village style refugium of Neolithic farmers in Britain! I was keen to learn what he was referring to but it was never elaborated on.

Davidski said...

@Survive the Jive

There are a few Roman British genomes available and they don't show any recent Italian-related ancestry, but they're all headless gladiators. So I don't think that's enough to claim that at least some of this admix didn't arrive in the UK during the Roman period.

Actually, one of these gladiators may have been from Egypt or surrounds. That's what his ancestry suggests. So it's likely that Italians were there too.

@Matt

I pulled one of those posts of yours from the spam folder. That formatting never had a chance with the spam filter. LOL

Andrzejewski said...

@Sam how would natural selection account for the increase in mtDNA H? What has one thing got to do with another?

Andrzejewski said...

@Drago Dave mentioned that it may be because of the GAC people (mostly farmer but with I1 Haplogroups from forager groups most likely either Baltic HG or Erteboelle-like) became a founding lineage within Scandinavia

Andrzejewski said...

@Joey Finns or even Tarand have a strong (60%) European admixture so no wonder about the blond eyes.

Andrzejewski said...

@Gaska No. H, W, T, J, K and N are all Neolithic farmer ones. The diversity of mtDNA may indicate that Anatolia_Paleolithic was itself a diverse population but they were pretty homogeneous once they reached Europe

Matt said...

Ahah.

Davidski: It's more likely that this Italian or Italian-like ancestry arrived in the UK at various times and from various places, including Italy. I can think of several potential sources: Celts, Romans, Normans, and Industrial Revolution migrants from Southern Europe.

About the Industrial Revolution bit, I'd tend to doubt anything too recent and lumpy, as it still would show up in chunks in big population samples (Biobank and probably PoBI), so older is better for the big panmictic spread in big British Isles samples.

And in the 18th-19th century we also have direct good record keeping which suggests no trend of increased rates and that compared to total population migration flows from Southern Europe. (Like, I think the estimate here about 245 Italians per year 1800-1945 (40k/145 years), and that probably mostly at the tail end - https://en.wikipedia.org/wiki/Historical_immigration_to_Great_Britain).

Earlier migration when migrant populations could be larger relative to pop size seems possibly more plausible too (when France was a demographic juggernaut and Britain had relatively small pop, it is easier to have an effect).

But yeah, continuous migration surfaces through the Middle Ages and prior, seem a pretty solid and reasonable bet to make against a single Anglo-Norman Era pulse, from Normandy alone.

FrankN said...

Dave: "I think there's certainly some complexity in their new data that isn't being explained very well, such as the precise character of the Southern European-like gene flow into Scandinavia after the LNBA.."

I am looking forward to your (and others) respective analyses, once the DNA becomes available.

In those early medieval S. Baltic emporiae, e.g. Gr. Strömkendorf (near Wismar) or Wolyn/Vineta/Jomsborg, among the classic "triade" of "Nordic" boat burials, Saxon "human & dog/horse", and Slavic graves, there occasionally were also Byzantine burials found. Varangian trade with Byzanz is well evidenced by coinage, we have reports about both Arab and Byzantine travellers/ emissaries visiting Haithabu, and there was of course the famous "Varangian guard" of the Byzantine court.
In fact, the relation seems to already have started during the MBA, when around 50% of Swedish copper was sourced from Greece (Laurion) and Cyprus, corresponding with a massive increase of amber finds in Mykenae and Egypt. Goths a/o traded fine E. Pontic glass ware to the Danish Isles. Golden jewellery of most likely E. Carpathian (Chernakhiv) provenience reached Denmark and even Ireland in late Roman times.

In short - I could well imagine that a good part of that South European gene flow was rather from Greece and the Pontic area than from Italy.

Otherwise, I find it noteworthy that they couldn't tell apart the "Danish" from the "Anglo-Saxon" signal. This implies to me that N. German / Old Saxon ancestry might also have been quite similar to their Danish cluster, albeit possibly somewhat southern-shifted, in the case of E. Holstein/ Mecklenburg also eastern/ Slavic shifted. As such, well attested Hanseatic League influence on Scandinavia, most notably Bergen, Gotland and Stockholm (with 50% German-speaking city council members until 1645) may well have contributed to spreading both the "southern" as the "Danish" signals.

Last but not least, a number of Swedish soldiers that fought in the Thirty Years War returned home with (south) German wifes. Afterwards, Sweden ruled over much of N. Germany until 1721 (Great Nordic War). Wismar and Stralsund remained Swedish until 1815. Trans-Baltic geneflow has continued till to date. I have several personal acquaintances with one Swedish grand-parent. Sweden of course was also a major German emigration target during Nazi rule (Willy Brandt, Kurt Tucholsky, etc.).
across

FrankN said...

Short correction: Skip "major target" for German emigration to Sweden during 1933-1945. Apparently, we are talking about some 5,000 emigrants (USA: 130,000), which shouldn't have been too relevant genetically.

FrankN said...

Andrzej: "how would natural selection account for the increase in mtDNA H?"
The mitochondria control metabolism, and play a major role in immune defence.

Studies have a/o shown that:
- Men belonging to haplogroup H have the lowest risk of asthenozoospermia (reduced sperm motility);
- Members of haplogroup H had increased chance of survival from severe sepsis;
- AIDS progression was slower for members of haplogroups H3, I, K, U, W and X (a conclusion that may probably be extended onto other, possibly cancerogenic retroviruses);
- Haplogroup H is associated with a higher maximal oxygen uptake (VO2 max) compared to other haplogroups, and was found to be strongly overrepresented among Polish top athletes.

Downsides - at subclade level - include increased risks for developing Alzheimer (H5a), or Diabetes type 2 (various H1 subclades).

Shaikorth said...

@FrankN

They made an attempt to tell the Anglo-Saxons and Danish Vikings apart: apparently Danish Viking Age samples have too much "Swedish" component to be responsible for more than about 6% of the Danish-like ancestry in England and the rest could be attributed to the Anglo-Saxons.

Vinitharya said...

Also with mtdna there are a lot of Haplogroup J, which has an adaptation making higher body heat that was advantageous in Scandinavia but not so much in southern New England during a heat wave. (Winter is Coming is our haplogroup motto as a statement of relief). And there were two Rurikid samples, neither having the family's supposed N1c subclade. One was I2a-Din and the other R1a-M458-L1029, both common clades among Slavic peoples and their central European neighbors.

Joey said...

Isnt the ydna with a clear connection to the area were Rurik was originally from, Central Sweden, a more likely candidate than the local haplogroups? Considering N-L550 was found both in Sigtuna and now in the Salme ship burials, I would say the case for Rurikids being N-L550 became even more likely. Why would a man from Scandinavia have typical Slavic ydna anyway?

Matt said...

In light of this, wonder if it may be worth re-examining the conclusions of https://science.sciencemag.org/content/360/6392/1028.long -
"Ancient genomes from Iceland reveal the making of a human population"

Ebenesersdóttir et al suggest that Finally, the D-statistic test D(YRI, X; Gaelic, Norse) also revealed a greater affinity between Norse and contemporary Icelanders (0.0004, 95% CI 0.00008–0.00072) than between Norse and ancient Icelanders (−0.0002, 95% CI −0.00056–0.00015).

But that may be confounded if this paper is accurate on geneflow back into Scandinavia, and ongoing Danish flow into the rest of Scandinavia.
Ebenesersdóttir et al note - "Some shift in ancestry must also be due to later immigration from Denmark, which maintained colonial control over Iceland from 1380 to 1944 (for example, in 1930 there were 745 Danes out of a total population of 108,629 in Iceland) (18)."

There also might be difference in that this paper suggests Orkney represents Pictish continuity and cultural switching to a "Viking" way of life, while Ebenesersdóttir exclude Orcadians from their analysis...

Proportions of ancestry in the Icelandic settlers should be more easily interpreted using other ancients rather than working back from contemporary Scandinavians and Irish.

....

From the paper some proportions on Iceland, Orkney and Ireland and archaeology: https://imgur.com/a/F7cHXNh

In Iceland, it looks like a sample from an isolated valley is least Orcadian-like, while the samples from the churchyard of related people run the gamut.

Davidski said...

@Joey

Why would a man from Scandinavia have typical Slavic ydna anyway?

This is actually possible. Here's an example from the Viking paper...

Denmark_Galgedil VK139 Y-hg: R1a1a1b1a1a1c1 (L1029)

But yes, it's rather unlikely that a Slavic lineage, or two, first made it to Scandinavia and then moved to Kiev with a Norse dynasty.

Bob Floy said...

@Joey
"Why would a man from Scandinavia have typical Slavic ydna anyway?"

Because he has a Slavic or Slavic-adjacent ancestor?
Why was Thomas Jefferson carrying Y-haplogroup T?
This kind of thing happens all the time.

Joey said...

You dont think its more plausible that the local subclades are just that, local, and not from people who first moved to Scandinavia and then migrated back? The one with a clear connection to the historically attested area of origin and one that is not common in the local populace is miles more likely to be the real deal in my opinion.

I dont know about Jeffersons lineage, but I would bet that its of a subclade of T thats plausibly Neolithic European. But I fail to see the relevance.

Drago said...

Why would a man from Scandinavia have typical Slavic ydna anyway?

Because he’s not from Scandinavia ?

Joey said...

I thought Rurik originating in Scandinavia was pretty well established. Did I miss something?

Drago said...

The earliest account of Rurik is ~ 300 years after his life
That Swedes comprised an important component of Rus is clear; however local
Balts Slavs & Finns obviously participated
So the Rurik genealogy might very well be Scandinavian; but his patriline could have been local

FrankN said...

Why would a man from Scandinavia have typical Slavic ydna anyway?

Because the Baltic Sea was always more of a connection than a separator, see Motala aDNA. There are some 50 Slavic toponyms on the southern Danish Isles, and the main ferry port from Holstein to Denmark bears a Slavic name: Puttgarden <-pod gorod "at/below the castle". The "Balti-Culti" area, a mix of Nordic, Saxon, Slavic and Baltic elements, is for the early medieval attested on the S. Baltic coast all the way from Stargard/Oldenburg (Holstein) in the west to Prussia in the East, but was in all likelyhood also mirrored by similar structures on the Swedish coast, and quite certainly on Gotland. The Lübeck Law - defining urban self-government in a similar way to the Magdeburg Law, but with a stronger focus on protecting the interests of foreign merchants and therefore preferred to the latter in trading centers as far as Talinn, Narva and Veliky Nowgorod - is believed to in parts having been derived from the legal system of Slavic Liubice "the lovely" (Old Lübeck).
https://en.wikipedia.org/wiki/Lübeck_law

The House of Mecklenburg, or Niklotids, of Slavic paternal origin, long held an (unsecessful) claim to the Swedish throne. The Niklotids followed a tradition already established by preceding West Slavic dynasties to acquire their consorts from Scandinavian nobility - Sweden in the case of the Niklotids/ Mecklenburgians, Denmark for earlier Obotrite leaders.
https://en.wikipedia.org/wiki/House_of_Mecklenburg

Drago said...

What are people’s thoughts on ‘uniform manifold approximation and projection
(UMAP) ?

Samuel Andrews said...

@Survive the Jive,

There is a slight change that occurs in Britain/Ireland from the Beaker period/Bronze age to the Iron age. There is a slight increase in farmer ancestry, slight decrease in Steppe ancestry, slight decrease in WHG ancestry. This could be because of the Celts. R1b L21 remained dominate so it is safe to assume British Beaker ancestry remained dominate.

All analysis I've seen indicates Spanish & French have Roman Italian (or southeast European) ancestry. If Britain has Roman Italian ancestry it's much less than what Spain, France have.

Plus, if there's southern European ancestry in Britain from Rome it would be expected for Welsh to be more 'southern' than English. But actually Welsh are slightly more 'northern' than English.

Samuel Andrews said...

@All,

Take a look at the German samples in G25 using 'pure' Germanic (Lombard, German_Medieval, Ducth), Slavic, and French references.

Very few German samples are mostly Germanic. Almost all have a lot of French-like ancestry. East Germans are pretty homogenous but about 50% Slavic. Austrians are not homogenous. They are on average about 50% east European (Hungarian, Slovak-like).

This is still a poor dataset from Germany. More samples are needed. But the picture emerging is Germany is one of the least homogeneous, pure countries in Europe. It brings up questions about German history, especially why admixture with neighboring Slavs was so common.

The Lombard samples are key to understanding Germanic tribes because they look like they come from completely untainted blood lines. They cluster with Scandinavians yet had lived all over central-southern Europe for 100s of years.

Samuel Andrews said...

*Very few Germans look to be overwhelmingly Germanic. But, many are about 50%.

There should be a good chance most of the foreign admixture is east of the Elbe river & south of the Danube.

Drago said...

“ Sam

“The Lombard samples are key to understanding Germanic tribes because they look like they come from completely untainted blood lines. “

Do you mean that according to myth; they’re the most noble of Odin’s people; or do you mean the current sample set they’re least admixed ?

Katharós said...

"But the picture emerging is Germany is one of the least homogeneous, pure countries in Europe. It brings up questions about German history, especially why admixture with neighboring Slavs was so common."

I think one major aspect of importance, is that there was an effort to merge all subjects under a common rule , that inevitably also includes a common religion. For instance in 1059 the Bishop of Bamberg (Franconia-Germany) complained that the Christianized Slavic inhabitants of his bishopric, were still devoted to their pagan ways. In an older 10th century text from Würzburg (also Franconia-Germany) these heathen Slavic practices are even directly named, trebo&hougir. So the customs of a land are directly linked to the culture of a land and regardless how we define culture. The term is still linked to religion, as proper cultus, or the proper way to honour God. So it can be said, that those who do not participate in the proper cultus of the land, are perceived as people “die nicht dem Volk angehören” quasi non-German. A modern day example of such a mechanism is America. You have evangelical groups trying to influence current affairs and legislation in the state. With the ulterior motive to unify America under the proper cultus.

Slumbery said...

@Samuel Andrews

I am just reading now, but I have some off-topic advice for you. Please choose better the words you use. Don't use the word "tainted" when talk about population genetics, because that word does not simply mean mixed, but specifically means polluted/contaminated/mixed with something inferior. "untainted blood lines" sounds seriously eugenicist and I am pretty sure this is unintended in your case.

FrankN said...

Sam: "There should be a good chance most of the foreign admixture is east of the Elbe river & south of the Danube.

West of the Rhine (Gallo-Romans) needs of course to be added, but it's actually much more complicated. Katharós adds a valid point, namely religion, albeit this was less related to ideology and more to finance.

Free German(ic)s didn't know taxation. Instead, they had to follow their leader's (kings, dukes, counts) call to arms - at their own expense (& profit). Consequently, the king/emperor had little financial resources aside from the revenue from a handful of royal domains ("Pfalzen"), and taxes levied on groups outside tribal Germanic law. These groups were

(i) Church institutions (monastries, bishoprics);

(ii) "foreign" merchants, especially Jews (hence Pope-instigated progroms, e.g. along the Rhine during the First Crusade, in order to weaken the Emperor's tax base);

(iii) Free/ Imperial cities (from the High Medieval onwards).

So, Frankish missionising had, aside from a certainly existing ideological motivation, primarily the political functions to enhance the tax base and create a literate (monasterial) administrative structure, for which the king/emperor had control over leadership assignments.

This created intriguing coalitions that often ran contrary to ethno-linguistic lines. Most importantly, Catholic Franks allied with partly-catholised West Slavs against Pagan Saxons. These Slavs were subsequently awarded with substantial parts of Saxon territory in order to diminish Saxon potential for upheaval. This concerns a/o East Holstein, but especially the Altmark and the Wendland, both west of the Elbe. Both areas today have predominantly Slavic toponymy, and in the Wendland, Polabian was spoken until 1756. Yet, early Medieval toponyms, e.g. the Höhbeck n. Gorleben (castellum huobeccum) or Werben/Elbe (Wiribeni iuxta Albim) were exclusively Germanic, and in spite of apparent strong Slavic settlement after 800 BC, both areas remained formally integrated into the HRE's bishopric structure.

Similarly, the "Wendish Crusade" that marks the onset of the German "Ostsiedlung" wasn't a German (Duchy of Saxony) <> W. Slavic war, but again ran along confessional lines. Catholic Slavic leaders, such as the a/m Niklotids/Mecklenburgians, were rewarded for their participation on the Catholic side by becoming recognised Imperial nobility. Subsequent Germanisation followed very different patterns, where strong Germanic (Flemish, Westphalian, Frisian, Saxon) colonisation of Pagan territories contrasted with a gradual assimilation/ acculturation process of already Catholicised areas. As such, you probably find much more Slavic ancestry in - say - E. Holstein than on Rügen.

FrankN said...

Another issue was monasterial colonisation. Free German(ic)s paid their dues to the king on the battle field, making them rather unsuitable for populating monastry-controlled villages where the focus was on extracting regular grain deliveries. Consequently, the monastries turned towards recruiting either Slavic or Gallo-Roman settlers. Slavic toponyms, and names such as "Wenddorf", "Böhmerdorf" etc., are quite common in former monasterial lands also far outside of what reasonably can be considered as early medieval Slavic territory, e.g. in Thuringia, Würtemberg or the Black Forest. IOW - Germanic colonisation in the east had a Slavic counterpart in the west - albeit assymetric in terms of social status/ individual rights.

Last but not least, there was substantial Polish westward migration within the Prussian state during the industrial revolution. This concerns in particular the Ruhr area, but didn't stop there. In the Hannover suburb I grew up, half of my primary school mates had Polish last names. A good tracer of these migrations are football clubs: "Borussia" (Prussia) [Dortmund, Mönchengladbach] usually means ethnically Polish immigrants, to be contrasted with German(ic) "Arminia" [Bielfeld]. "Rot-Weiss" [Essen, Oberhausen] is equally obvious - the red & white club colours of course emulate the Polish flag.

The migration continued after 1919, this time once more under confessional lines, when a lot of Protestant Kashubians and Kujawyans placed more trust into the multi-confessional Deutsche Reich than the newly-created, Catholic-dominated Polish state.

In essence: Germany is a recent construct. Possibly as recent as the Napoleonic Wars, albeit there are good arguments to connect the origins of a German identity and unified language to the Luther Bible. Genetically, Germans (pre-1960s "Gastarbeiter") may be approximated as roughly 30% each Germanic, Slavic, and Gallo-Roman (with Balts, Finno-Ugrics, Alans, Avars, Ashkenazi, etc. accounting for the remainder) in proportions that vary widely among regions. The linguistic impact of each of these strains on High German (IMO at least as differentiated from the West Germanic "mainstream" as is English) was somewhat different, with Gallo-Roman clearly being the major super-/adstrate, while Slavic only ranges at par with Baltic and Alanic (Ossetian).

Drago said...


@ Frank
“Genetically, Germans may be approximated as roughly 30% each Germanic, Slavic, and Gallo-Roman (with Balts, Finno-Ugrics, Alans)

I’m not sure what those figures are based on
The use of “Gallo-Roman” is a bit anachronistic (strictly speaking) when discussing the Ostsiedlung of medieval Germany

Modern Germans demographically overwhelmingly descend from Carlongianised west Germans
Modern Germans plot southwest of ancient Germanics

Davidski said...

Slavic ancestry is significant in about a third of Germany; in most of the former East Germany and in the Ruhr district in western Germany. Apart from that, it's obviously well represented in the descendants of Germans who moved west from the former, pre-WWII German lands, but they're a minority practically everywhere.

There's also a lot of Scandinavian ancestry among the Germans near the Danish border.

So Germans are still a fairly homogeneous population, and I wouldn't characterize them as roughly an equal mix between Germans, Slavs, Gallo-Romans and Medieval invaders. They're more like West Germanics with 0-100% Slavic ancestry, depending on the region in question.

By the way, to counter Frank's experience with a classroom filled with kids of eastern origin, as a kid I lived in western Germany, but not anywhere near the Ruhr, and my name actually seemed to confuse a lot of people. A school teacher actually thought I was the kid of an American soldier rather than a Pole. Crazy.

Davidski said...

@All

Does anyone know how to line up these ID codes from the Viking paper and Harvard's dataset? I'd like to cross check their ancestry proportions from Chromopainter with the Global25 North Euro PCA.

VK557 Sigtuna.SG_VA
VK558 Sigtuna.SG_VA
VK559 Sigtuna.SG_VA
VK560 Sigtuna.SG_VA
VK561 Sigtuna.SG_VA
VK562 Sigtuna.SG_VA
VK564 Sigtuna.SG_VA
VK565 Sigtuna.SG_VA
VK566 Sigtuna.SG_VA
VK567 Sigtuna.SG_VA
VK568 Sigtuna.SG_VA
VK569 Sigtuna.SG_VA
VK570 Sigtuna.SG_VA
VK571 Sigtuna.SG_VA
VK572 Sigtuna.SG_VA
VK573 Sigtuna.SG_VA
VK574 Sigtuna.SG_VA
VK575 Sigtuna.SG_VA
VK576 Sigtuna.SG_VA
VK577 Sigtuna.SG_VA
VK578 Sigtuna.SG_VA

vik_2072.SG Sweden_Viking.SG
vik_84001.SG Sweden_Viking.SG
vik_84005.SG Sweden_Viking.SG
vik_97002.SG Sweden_Viking.SG
vik_97026.SG Sweden_Viking.SG
vik_97029.SG Sweden_Viking.SG
vik_bns023.SG Sweden_Viking.SG
vik_grt035.SG Sweden_Viking.SG
vik_grt036.SG Sweden_Viking.SG
vik_gtm021.SG Sweden_Viking.SG
vik_gtm127.SG Sweden_Viking.SG
vik_KAL006.SG Sweden_Viking.SG
vik_kal009.SG Sweden_Viking.SG
vik_kls001.SG Sweden_Viking.SG
vik_nuf002.SG Sweden_Viking.SG
vik_stg020.SG Sweden_Viking.SG
vik_stg021.SG Sweden_Viking.SG
vik_stg026.SG Sweden_Viking.SG
vik_urm035.SG Sweden_Viking.SG
vik_urm045.SG Sweden_Viking.SG
vik_urm160.SG Sweden_Viking.SG
vik_urm161.SG Sweden_Viking.SG

FrankN said...

Dave: Your reference to East Germany as having significant Slavic population is somewhat imprecise. On one hand, relevant parts of East Germany, namely Thuringia, Western Anhalt and Eastphalia (the region between Brunswick and Magdeburg) weren't affected by the early medieval Slavic expansion. OTOH, early medieval Slavic settlement also reached W. Germany, especially E. Holstein, NE Lower Saxony (Uelzen, e.g., is believed to be a Slavic toponym), and the Bavaria Slavica (actually rather Franconia Slavica) that was mentionned in Katharos' comment above. On the latter, see
https://www.hdbg.eu/karten/karten/detail/id/140

Late 19th century intra-Prussian mobility wasn't restricted to the Ruhr, but brought Poles from Prussian-controlled areas to all emerging industrial centres, including Hannover, Mönchengladbach (in the Rhineland), certainly also Berlin. A good indicator are Weimar Republic election maps: Votes for the Catholic Centrum Party in historically Protestant areas in all likelyhood came from Polish immigrants (Hannover's CDU was until well into the 1980s known for a pre-dominance of Polish last names among Party officials). [Apparently, you spent your time in W. Germany in what used to be the post WWII American occupation zone, i.e. outside pre-1935 Prussia.]

My great-grandfather in law migrated from somewhere between Wroclaw and Posznan to Scscezin, my mother in law (with a Polish maiden name) falls into the category of post WWII displaced persons. She isn't exceptional in this respect - I have other friends whose parents were replaced after WWII and have partly Polish roots. Still, my impression is that the majority of post WWII displaced persons rather brought Baltic (Old Prussian) than Slavic ancestry with them, as the displacement was much more radical in East Pommerania and East Prussia than it was in Silesia.
(Descendants of) displaced Pommeranians/ East Prussians, btw, today make up for the majority of the population of Schleswig-Holstein, and are demographically significant in Hamburg, Mecklenburg and Lower Saxony. Bavaria took up some 2 million (post-)WWII refugees and displaced persons, mostly from Czechoslovakia, accounting for nearly a quarter of the 1951 Bavarian population.

Swedish said...

https://www.biorxiv.org/content/biorxiv/early/2019/07/17/703405/DC2/embed/media-2.pdf?download=true

It only mentioned R1a1a1b and Z284 subclade and that makes sense considering the modern distribution like you can see here https://i.imgur.com/vhRuKlx.jpg. But is there any chance to check the raw data for the quite rare Haplogroup R1a1a1a with subclade L664?

R1a 2016 Tree
https://i.imgur.com/vhRuKlx.jpg

JuanRivera said...

Did a lot of modeling. Yana_UP has better fits when using K14 than when using Sunghir. Also, fits get better when Hoabinhian is included in the model. Indigenous Australians model as mixtures of Onge and Papuan, with traces of English admixture. Given that Indigenous Australians are a sister population of Papuans, Papuans (and Indigenous Australians) themselves may be mixtures of Onge and an AMH population related to the first settlers of Sahul (such as Mungo Man).

Davidski said...

@Swedish

It only mentioned R1a1a1b and Z284 subclade and that makes sense considering the modern distribution like you can see here https://i.imgur.com/vhRuKlx.jpg. But is there any chance to check the raw data for the quite rare Haplogroup R1a1a1a with subclade L664?

The genomes haven't been publicly released yet, so no one but the authors can check the raw data.

But, as far as I can see, there's no L664 in the results. The only Z284- samples belong to typically Central and Eastern European subclades, and indeed the samples that carry them are significantly Polish-like in terms of autosomal ancestry.

Swedish said...

Sorry wrong link, distribution here: https://i.imgur.com/WXj5kh0.jpg

Swedish said...

@Davidski
Alright thanks

Swedish said...

@Davidski
Isn't Z284 uniquely Scandinavian and Northwestern? It doesn't show up at all in Eastern Europe according to https://i.imgur.com/WXj5kh0.jpg

Davidski said...

@Swedish

Z284 shows up occasionally outside of Northwestern Europe. I know of examples from Poland, Russia and Switzerland.

But there's also M458 in Scandinavia, and there are samples with this subclade in the Viking paper, including this one.

Denmark_Galgedil VK139 Y-hg: R1a1a1b1a1a1c1 (L1029)

Andrzejewski said...

AMH=?

JuanRivera said...

Anatomically modern human.

Ric Hern said...

@ JuanRivera

I wonder if Polynesians somehow fits into this picture ? The Polynesians seems to originate from the Philippines...Any connection between Mungo and Polynesians ?

Shaikorth said...

@Davidski

The Sigtuna samples included there are probably the ones with most overlap with modern datasets so:
84001
84005
grt035
grt036
gtm021
kal006
stg021
stg026
urm160

The chromopainter fits roughly match their positions on the PCA:
https://eurogenes.blogspot.com/2019/06/they-came-they-saw-and-they-mixed.html
The exact numbers compared to G25 models make more sense when Estonians are added in the G25 fit.

Davidski said...

Yes, but which ones are which. Any way to find out?

Shaikorth said...

Without asking the authors there is no way to be certain. Modern Swedes in this model are supposed to be something like 10% British 65% Scandinavian 5% Polish 5% Italian 15% Finnish. (S11.14)

Based on that...
VK564=grt035
VK565=grt036
VK568=KAL006

VK557, VK573, VK574 = stg021, stg026, 84001 (not sure about which one is which)

VK558, VK566, VK577 = gtm021, 84005, urm160 (ditto)

Arza said...

@ Davidski

IDs are not needed. Just calculate their G25 coordinates based on Chromopainter and if there will be a strong correlation with original G25, it will be obvious which sample is which. And if it won't be obvious, you'll have your answer too. The only question here is which samples should be used as a stand-in for the Danish-like component. Maybe some mix of Germans and Swedes/Norwegians?

Joey said...

>I would have been shocked indeed if there was foreign gene flow from the North
Time to be shocked then I guess.

"We identified two ancient individuals (VK518 and VK519) originating from northern regions of Norway (Nordland), which have affinities to present-day Saami. This signal is weaker for VK519, indicating that he might have also had Norwegian-like ancestors. Given the geographic provenance of these samples, it was not unexpected to find individuals with Saami-like ancestry among the VA samples. However, as VK519 is indeed an admixed individual with both Norwegian-like and Saami-like ancestries, it appears that genetic contacts between these groups were already underway in VA Norway."

Cy Tolliver said...

@Juan Rivera

What program are you using for these models? Are you able to come up with any models for Onge itself?

JuanRivera said...

Using nMonte. Haven't tried yet for Onge. Also, since there are no ~40kya australian samples, my hypothesis remains a hypothesis.

Davidski said...

@All

I've extended the Viking Age Sigtuna sample set in the Global25...

SWE_Viking_Age_Sigtuna:vik_84001
SWE_Viking_Age_Sigtuna:vik_84005
SWE_Viking_Age_Sigtuna:vik_97002
SWE_Viking_Age_Sigtuna:vik_97026
SWE_Viking_Age_Sigtuna:vik_97029
SWE_Viking_Age_Sigtuna:vik_grt035
SWE_Viking_Age_Sigtuna:vik_grt036
SWE_Viking_Age_Sigtuna:vik_gtm021
SWE_Viking_Age_Sigtuna:vik_gtm127
SWE_Viking_Age_Sigtuna:vik_KAL006
SWE_Viking_Age_Sigtuna:vik_kls001
SWE_Viking_Age_Sigtuna:vik_nuf002
SWE_Viking_Age_Sigtuna:vik_stg020
SWE_Viking_Age_Sigtuna:vik_stg021
SWE_Viking_Age_Sigtuna:vik_stg026
SWE_Viking_Age_Sigtuna:vik_urm035
SWE_Viking_Age_Sigtuna:vik_urm045
SWE_Viking_Age_Sigtuna:vik_urm160
SWE_Viking_Age_Sigtuna:vik_urm161

Samuel Andrews said...

Major language families in the world 'proven' to spread with genes.....

Indo European.
Turkic.
Bantu.
Slavic
Polynesian.
Afro-Asiatic-into East Africa.

So, the pots=/=people theory & language=/=genes theory, seems to be universally wrong in most cases around the whole world....not just sometimes wrong.

If you see a language or very specific cultural traditions or very specific genetic markers common across a large stretch of land, then it is safe to assume that at some point a very expansive/large/successful people group caused that to happen in a short/rapid period of migration.

Geneticists have been making the wrong assumption that the expansions/founder events in modern DNA are the markings of the first human migrations. Most of the expansions/founder events in modern DNA are actually the marks of 'recent' migrations.

Gabriel said...

@Samuel Andrews

So do you believe Afro-Asiatic originated in the Levant?

Bob Floy said...

@Sam
"Most of the expansions/founder events in modern DNA are actually the marks of 'recent' migrations."

Yes.

Samuel Andrews said...

@Gaberial,
"So do you believe Afro-Asiatic originated in the Levant?"

That would make sense. The only for sure case for Afro-Asiatic gene change is East Africa. Their Middle Eastern ancestry is much more Natufian than Levant_N was however.

JuanRivera said...

I'd add Chukotko-Kamchatkan to that list, since modern Chukotko-Kamchatkan speakers are overwhelmingly (85-95%, in the Chukchi) to near-completely (95%+, in the Koryak and Itelmen) Magadan_BA. Interestingly, Magadan_BA also shows up in the Nivkh, in line with the proposed relationship of the Nivkh language with Chukotko-Kamchatkan outlined by Fortescue 2011. Eskimo-Aleut also may be added to the list, though I didn't yet model modern Eskimo-Aleut speakers. I expect them to be Ekven_IA+Uelen_IA+Paleoeskimo+Athabascan-like Amerindian groups, with the marker component being Ekven_IA and probably Uelen_IA.

JuanRivera said...

Modeled the present-day Eskimo-Aleut populations. Eskimos model as Ekven_IA+Magadan_BA; Neo-Aleuts model as Ekven_IA+Paleo-Aleut+Athabaskan_1100BP; Greenlanders model as Magadan_BA+Athabaskan_1100BP. All have a large amount of Ekven_IA, except the Greenlanders, which have none. The model of the greenlanders hint at the presence of high amounts of Saqqaq ancestry. Since there's no file for Saqqaq in nMonte, there's at present no way to check it. Magadan_BA may be Saqqaq+Kolyma_Meso+Steppe, which could account for the Magadan_BA affinity in Greenlanders.

JuanRivera said...

Uelen_IA doesn't seem ancestral to anybody, and trying to model it the same way as Ekven_IA produces dismal fits. However, modeling as Ekven_IA greatly improves the fits, with the highest fits being produced by a Ekven_IA+Magadan_BA+Anzick model.

Gaska said...

@Andr No. H, W, T, J, K and N are all Neolithic farmer ones. The diversity of mtDNA may indicate that Anatolia_Paleolithic was itself a diverse population but they were pretty homogeneous once they reached Europe

And Again, That's NOT true. We have already discussed this other times and I have sent the Iberian deposits where Mit Hap-H appears from the Paleolithic, and of course in the Mesolithic. Then at least in Iberia certain clades of H including H1, H3, H4, H6 and H7 are markers prior to Neolithic migrations, and from Iberia recolonized the north during the Neolithic.

Davidski said...

@Gaska

Can you show me an example of anything apart from mtDNA haplogroup U in a pre-Neolithic Iberian sample sequenced with capture or shotgun methods and reliably radiocarbon dated?

Gaska said...

@Davidski-Can you show me an example of anything apart from mtDNA haplogroup U in a pre-Neolithic Iberian sample sequenced with capture or shotgun methods and reliably radiocarbon dated?

Yes of course-

1º- Many of the samples of ancient dna that are handled throughout Europe (both U and H), come from old studies with which neither are absolutely reliable

2º Ancient DNA in the Cantabrian fringe populations: A mtDNA study from Prehistory to Late Antiquity. Concepción de la Rúa, Neskuts Izagirre, Santos Alonso, Montserrat Hervella (2.015). Five samples associated with a hunter-gatherer way of life were collected in 5 archaeological sites from the Cantabrian fringe, La Chora, La Pasiega and El Mirón in Cantabria, Erralla in the Basque Country and Aizpea in Navarre.

"However, the genetic composition of Paleolithic and Mesolithic hunter gatherers in the Iberian Peninsula showed a more complex interpretation with some haplogroups existing both in hunter gatherers and extant populations suggesting continuity from Paleolithic Mesolithic times"

+ Cueva del Mirón (Ramales de la Victoria, Cantabria)-In the hunter-gatherer groups at the end of the Upper Paleolithic of Central Europe, no individuals have been found to present the ht2 haplotype described in El Mirón and La Pasiega, nor have the haplotypes obtained in La Chora been found- Mit-Hap- H (probably H1), Date-Middle Magdalenian (14,000-12,000 BC)


3º-Cueva del Pirulejo (Priego de Córdoba, Córdoba) - In the excavation campaign of 1991 in sub-level 4B (inventory number 256) corresponding to the Magdalenian, various human remains appeared, specifically two fragments of parietal, a milk canine, a molar lower milk and three premolars probably lower. Mit-Hap-Sample 2PI-HV- Sample 2PI-Mit Hap N*-Date Middle Magdalenian (13,500 BC)-

4º- +Cueva de la Pasiega (Puente Viesgo, Cantabria)-(16.200-15.740 AC). The most recent stratum was also relatively rich, with several burins, punches, bone objects and perforated teeth that could belong to the Archaic Magdalenian. Important cave paintings- M. Hervella et al (2012) -Mit Hap-H

5º-+ Sado Estuary (Central Portugal) - 8 mitochondrial haplogroups belonging to the Mesolithic were recovered (9,500-7,500 BC) - Helen Chandler, Joao Zilhao et al (2005). Results Mitochondrial Haplogroups- H (2), H1b (1), H7 (1), U4 (1), U5b1 / c2, N1b, N5.


There are many more cases but I do not want to bore you, it has even been shown the appearance of H, H1, H2, and H14 in the north of Morocco in Magdalenian sites with material cultural identity that the Iberian deposits (harpoons etc.) -

The curious thing is that except for the Lady of El Mirón (Mit Hap-U5) none of these samples has been analyzed carefully to establish clearly what is the genetic makeup of the WHG. When the geneticists do it (I hope they do not take long), then we will have many surprises and hundreds of pages and genetic articles should be reviewed.

Precisely for this reason we distrust the autosomal analyzes that geneticists do, they only use two samples in Iberia (El Mirón and La Braña), when they could (and should) use at least two dozen. The scarcity of samples analyzed is not enough to establish clearly what is the autosomal component of the WHG - Exactly the same happens with the SHG, EHG, and EEF, and of course with the famous steppe signal (we still do not know which culture it belongs), maybe you can clarify it.

It is boring to have to resort to these topics, but only the absolute ignorance of the Paleolithic of Western Europe can lead someone to think that H is a Neolithic lineage.

I hope you consider these examples as sufficient to clarify your doubts. If you are interested I can send you other studies both Paleolithic and Mesolithic. And if you have any doubts about the dates, then you would have to question the rest of the dates of the European Paleolithic sites where the U lineage has been found. For example, in the Miron cave, both U and H appear, the first, in somewhat older stratigraphic levels.

Katharós said...

Slightly off topic, but I bumped into this book that may be of interest.

Heidenfrage Und Slawenfrage Im Deutschen Mittelalter: Ausgewählte Studien 1953-2008
ISBN-10: 900416751X

Synopsis:
The christianization of Central and Northern Europe, fundamental for the formation of the unity of our civilization was considered by earlier scholarship only in terms of what took place but not in terms of the official norms of the medieval church. On the other hand, the spiritual starting point of so-called 'missionary objects' was left largely out of view. Consequently, anachronistic terms came in to use and actual facts became distorted. 26 Studies, published over more than 50 years in four different countries, discuss these problems especially against the background of Carolingian Saxony, and the Slavonic tribes between Germany and Poland, -most of whom may also be seen as the ancestors of modern-day Germans-, with special reference to the strange "Wendenkreuzzug" (Wendish crusade) of 1147.

Angantyr said...

A few days late, but...

@Gaska

"The presence of R1b1a1 (R1b-P297) in Norway 2,400 BC complicates the situation to explain the massive steppe migrations, although surely someone is willing to swear that the CWC reached the Arctic Circle"

VK531 complicates nothing, as there's nothing steppe about him:
* R1b-P297 is way too old to be considered a steppe marker, and R1b-P297 of course even less so.
* In 2400 BCE northernmost Fennoscandia was home to the late Combed Ware culture.
* We have earlier Combed Ware individuals from the Baltics, and they are autosomally mostly EHGs with some shift towards SHGs/WHGs.
* Unsurprisingly, VK531 also sits right there in Figure 2.
* (Unfortunately we don't have much Combed Ware Y-DNA yet, but R1b-P297 was present in the Narva culture that preceded Combed Ware in the Baltics.)

Drago said...

Is Carlos suggesting that M458 is Germanic (Based on the ~ zero ancient Germanic samples it’s turned up in )?

Davidski said...

I'm sure Quiles has some very interesting things to say about the Viking paper and especially its ancient Y-chromosome data. Interesting, that is, if you're into science fiction.

Gaska said...


@Davidski-In this I agree with you, science fiction, lack of rigor in the study of genetic data and constant errors in their attempt to square a theory that clearly collapses. I don't think he will be able to rectify in the future, the truth is very disappointing, because he could leave an open door to correct his immense mistakes. Others have done so, and can continue to maintain a certain "genetic" and "linguistic" prestige. I respect all those who continue to see a link between P312/Df27 and the Indo-European Language, but from the mouth of Carlos Quiles I have not heard today a reasonable explanation of the evident genetic continuity in Iberia from BB culture to the historical peoples of the Iron Age. He has gone through the matter giving ridiculous and unacceptable explanations, and I am worried because he is Spaniard and seems more concerned with belittling all those who think differently. Its main objective is the Basques and the R1a, apparently all of us dangerous racist or stubborn fools

pnuadha said...

"because population genetics and ancestry tests have the potential to be used to encourage racism"

That's not really why. They don't want normal whites to have an identity; its just an anti white expression. Where are the similar articles making salacious claims about Indians being migrants and not from India, Japanese not being from Japan, modern Jews largely being non ancient Levantine, or Native Americans not being homogeneous or the result of several invasions?

This kind of political manipulation of the science only happens to one people, and its precisely the people who are becoming minorities in their own countries. If that is the goal then it makes sense to not let native Europeans be aware of who they are. This is propaganda.

Andrzejewski said...

* Was Combed Ware an Uralic speaking ancient population or was it a non-IE non-Uralic one of the Baltics?

* Was Narva mostly autosomally EHG? Or was Combed Ware representative of a brand new replacing population?

Andrzejewski said...

@Angantyr * One more thing: would you account for the assimilated Combed Ware in the Baltics/Fenniscandia for the excess EHG in Corded Ware (75% Yamnaya-like)?

Angantyr said...

@Andrzejewski

I am quite convinced that Combed Ware spoke a non-IE and non-Uralic language, and that Uralic languages arrived with Y-DNA N and a bit of Siberian ancestry later on. (Davidski has explained the reasons why pretty well in a number of blog posts.) Combed Ware may have provided (parts of) the non-IE and non-Uralic substrate found in the Saami languages.

The Narva genomes reported by Saag 2017, Mittnik 2018 and Mathieson 2018 were autosomally generally more WHG than EHG, but with high variability. As the Combed Ware people reported by the same studies + Jones 2017 were more EHG heavy, the material culture change must have involved a migration, but likely not a complete replacement. (By the way, as Combed Ware or if you like Pit-Comb Ware was so widespread, I am not sure it represents a single "people" as such.)

I am not sure I understand your last question, but that Corded Ware people in the Baltics were more (E)HG ancestry heavy than their brethren further to the south and west may have been a combination of absorbed Narva and Combed Ware genes, and a relative lack of absorbed TRB/GAC.

Andrzejewski said...

@Angantyr *Maybe Narva Culture provided some non-IE non-Uralic substrate in the Saami languages as well. Also, as the Saami substrate is very different than the reconstructed PIE it may rule out the latter one being anything close to an EHG language, (if it’s the case that part of the Saami substrate is EHG).

Andrzejewski said...

* as for European Uralic (Finnic speaking in particular) people, I know that they have Baikal HG (East Asian affinities) and that they are associated strongly with y-dna N1c1. I also know that they have a majority European autosomal DNA but I’m yet to see their genetic PCA as in WHG, EHG, EEF, Yamnaya etc. I think someone said in a few blog entries back that they assimilated lots of EHG but I’m not sure about that. As far as I remember they are over 50% Yamnaya.

Shaikorth said...

@Angantyr
The non-Indo-European non-Uralic linguistic substrate in Saami is dated to the Iron Age. This means it can't be EHG or Comb-Ware (too late for that) but more likely something that arrived later, like the language of the Bolshoy Oleni Ostrov people. From "An Essay on Saami ethnolinguistic prehistory":

"The Palaeo-Laplandic substrate in the Saami lexicon and place-names also provides us with a straightforward method of dating the language shift. The fact that unetymological vowel combinations are found in numerous substrate words shows that they have been adopted after the Great Saami Vowel Shift. Even more importantly, substrate words also include initial consonant clusters in the West Saami area, quite like Proto-Scandinavian loans. Occasionally, one also fi nds the phoneme /f/ in likely substrate items (e.g. North Saami uffir ‘rocky seashore’, láfol ‘Eurasian Dotterel’, skuolfi ‘owl’). This shows that the adoption of substrate vocabulary must still have continued after the break-up of Proto-Saami and at the time of the emergence of Northwest and Southwest Saami proto-dialects in Scandinavia. The same is, of course, implied by substrate place-names as well as by the fact that numerous substrate words show very restricted distributions in the Saami languages, and cannot thus be reconstructed to Proto-Saami.

Thus, we arrive at the surprising conclusion that substrate infl uence of Palaeo-Laplandic languages was contemporaneous with the adoption of Proto-Scandinavian loanwords. As the Proto-Scandinavian contacts in Scandinavia must be equated with the Early Runic era ca. 200–700 AD, we thus have an absolute dating also for the spread of Saami languages to Lapland and for the disappearance of the unknown Palaeo-Laplandic languages. All evidence indicates that it has taken place only after the beginning of our era, likely about 1500 years ago."

Andrzejewski said...

@shaikorth @Angantyr Bolshoy Oleni Ostrov people spoke Proto-Uralic. It was theorized that the Volosovo people were responsible for the non-IE non-Uralic substrate but I guess they are identical with the Bolshoy complex?

Shaikorth said...

It's unlikely that the Bolshoy spoke proto-Uralic which was probaly still far to the southeast during that period, this was referenced in the Lamnidis study about them (ca. 3500 yBP predates most linguistic estimates of the spread of extant Uralic languages to the area) and no evidence of Uralic languages older than Saami have been found in Lapland, as toponyms or as a substrate in Saami. Instead the Saami have a late non-Uralic non-IE substrate. Volosovo too is far too old (and in the wrong place) to be related to Saami substrate.

Andrzejewski said...

Ok. So who spoke Proto-Uralic? (And please don’t say “Tarand”).

Angantyr said...

@Shaikorth

Asbestos ceramics were made in northern Sweden well into the Iron Age, so someone carried that tradition since the Combed Ware days. Maybe they were mixed with more recent immigrants, but the Bolshoy Oleni Ostrov individuals already looked heavily mixed with "locals" and not like recent arrivals from the east, so what language did they speak?

I know that we'll probably never know what was spoken in northernmost Fennoscandia before the Saamis and Norse arrived in the Iron Age, but I wouldn't rule out something Combed Ceramic derived, thus my "may".

Shaikorth said...

@Andrzejewski
During the Bronze Age, linguists generally say, Uralic would have been close to the Urals, possibly the northern neighbours of Sintashta which would explain the generally assumed Indo-Iranian substrates. Tarand is generally connected proto-Finnic, but never heard of anyone claiming it's proto-Uralic. But as people have shown with qpAdm the earliest Tarand samples have a component from something like the Sintashta outliers with excessive WSHG ancestry compared to normal Sintashta.

@Angantyr
Cultural tradition can be preserved over linguistic changes too, I doubt the very eastern end of where Comb-Ware pottery types abounded (Manchuria) spoke the same language as Baltic Comb Ware even though the pottery type is oldest in the east. I agree though, we may never know the various extinct substrate languages for sure. Maybe someone should take on the herculean task of digging through languages in the Baltic and the North Eurasian forest zone for unclassified substrates and making a comparison...

JuanRivera said...

Those ceramic makers must have suffered from cancer, knowing what asbestos do to lungs.

FrankN said...

A specific feature of North Germanic is a heavy tendency towards syncopation, i.e. creation of consonant clusters by loss of unstressed vowels. Compare, e,g. English "do not" (original) vs. "don't" (syncopated) [English syncope btw. betrays substantial NGerm linguistic influence]. This process took place sometime between 500 and 800 AD, when Old Norse emerged as distinct language out of the Proto-Germanic dialect continuum. A fair assumption is to see these shifts as substrate language influence.

Syncopation is completely alien to Uralic, which generally prohibits intra-syllabic consonant clusters. In fact, Uralic has effected the reverse tendency on East Slavic - injection of additional vowels to break up consonant clusters. Compare, e.g., OCS *grad vs. E. Slavic gorod "city, castle". As such, the linguistic substrate that worked on North Germanic can't have been Uralic.
It probably also wasn't the language that provided Germanic with its ca. 30% non-IE basic vocabulary (drink, sing, hand, etc.), because these substrate terms entered (pre-)proto Germanic already much earlier and are shared by N. and W. Germanic.

This leaves us with a specific Nordic non-IE language, either from North Fennoscandia, or - in view of the "N. Atlantic" signal - from Scotland and the Shetlands ("Pictish"). I don't know enough about Scotch-Gaelic to ultimately tell if syncopation was/is at work there as well, setting it apart from the remainder of Insular Celtic (but have at least found such feature nowhere mentionned). To any Scottish out there: Do locals pronounce Edinborough as "Ednbro", Glasgow as "Glsgo"? If so, "Pictish" needs to be considered. Otherwise, the view goes to NC Fennoscandia. As Shaikorth noted correctly, that wouldn't have been Pitted Ware - a LN phenomenon - anymore, but probably a successor [I haven't been able to figure out anything specific about the post-PWC cultural sequence in inland Fennoscandia.]

In any case, Scottish LN Grooved Ware isn't only namewise somewhat recalling Pitted Ware. Ultimately we might have had a cultural (also linguistic?) continuum of fishing-foraging cultures in the N. Atlantic already during the LN (communication between Norway and Scotland in fact dates back to the Epi-Paleolithic).

Andrzejewski said...

@FrankN don’t you think that if 30% of Germanic is non-IE (highly exaggerated in my opinion, I would say not more than 5% max is non-IE) that it would be either a Farmer language (LBK, TRB, GAC, Laryngeal, Rossen) or a WHG one (Erteboelle?) or even both?

Andrzejewski said...

Moreover, the hypothesis about 30% of Germanic being non-IE was suggested back in the 1930s and by now it’s all but obsolete and abandoned.

FrankN said...

Andrzej: I did a check on the Swadesh 100 list. There, a good third of Proto-Germanic terms don't have a PIE analogy.
The Swadesh lists, as all similar approaches, are to some extent questionable. One issue are doublettes - e.g., "this"/"that", and "who"/"what", each commonly represent the same root, but are still dually included in Swadesh 100. Also, it is well documented that across most language families, "I"<>"one" and "thou" ("you" sg.) <>"two" correspond, yet all four terms are included in the Swadesh lists. Another problem are "paleo-words", e.g. "water"-words on *wä, or *k(w)u (satemised *(a)su) "dog" as an Epi-Paleolithic wanderwort of global distribution. This results in some 15-20% "basic stock" common across numerous language families. Against this background, a good third of PGerm. terms w/o clear PIE connection in fact means rather close to 50% proprietary PIE vocabulary not present in Germanic.

Vennerman reports a study of Class III German strong verbs - these are verbs of the "sing-sang-sung" type. Only 25% of them were doubtlessly derived from IE, another 15-20% could with considerable linguistic (semantical/ phonological) effort someway be connected to IE. Strong (in English called irregular) verbs are generally believed to represent a relatively ancient stratum of (pre-)Germanic. Their formation seems to have stopped after the early Medieval ("write", "ride" as borrowings from Latin). As such, the a/m result is another indicator of substantial non-IE substrate in Germanic.

Did it come from a farmer language? Well, some pre-IE substrate has been identified across several European IE languages. Compare, e.g. Slavic
bielo "white" (colorless) vs. Lat. pallidus "pale", Engl. to bleach, bleak - the irregularities of the initial consonant point towards several borrowings from the same, non-IE language. This language, for its apparent widespread distribution, should have been related to EEFs.
But most of the non-PIE Germanic words lack analogies in other European IE languages. Therefore, I think we are dealing with Erteboelle-derived TRB terms.

Bob Floy said...

@Andre

Yeah, the proposed non-IE substrate in Germanic, if that's what it is, would definitely amount to more than 5%.

Andrzejewski said...

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

“More recent treatments of Proto-Germanic tend to reject or simply omit discussion of the Germanic substrate hypothesis. Joseph B. Voyles' Early Germanic Grammar makes no mention of the hypothesis, like many recent publications on the Germanic language family.“

Bob Floy said...

Ok but are you ready to hang your hat on that? There's a reason why for ages tons and tons of scholars have believed that there was a substrate. Also, Wikipedia.

Andrzejewski said...

@FrankN that’s insane! 50% non-IE???

What language do you think TRB spoke, a Farmer LBK one or a forager Erteboelle one? A creolization of both? How about GAC?

Bob Floy said...

@Andre

How could anyone possibly know?
It's unrecorded, missing history. It sucks, but we'll never know.
I personally would like to think that the Erteboelle language(assuming there was only one) became some kind of prestige language, like Latin, and survives today as the non-IE Germanic substrate, but of course that's just fan fiction.

Andrzejewski said...

Beside the irony of history that the so-called self-professed “Master Race” spoke a language that (according to @FrankN) was almost lexically half non-IE and if you will - have the LEAST amount of Steppe Yamnaya Kurganite autosomal DNA - I think both of you exaggerate the amount of non-IE influence on Proto-Germanic.

Did LBK speak a Pelasgian Lemnian (“Tyrsenian”) language close to either Basque or Etruscan, or was it completely unrelated?

What about Funnelbeaker? Was it from forager Erteboelle or from from Farmer LBK? A creole?

Did GAC speak TRB or Rossen? Were they EEF languages?

Do you both propose the the German language is a tri-partite of forager WHG Erteboelle, Farmer EEF TRB and Yamnaya/CWC Indo-European, all in one package?

Andrzejewski said...

Erteboelle would leave a substrate in Baltic languages too. And I think some substrate in Saami languages would derive from WHG as well, but Germanic stands out.

My theory is that if European languages share a cognate across the board then it’s likely that they came from PIE or Corded Ware.

Davidski said...

Erteboelle would leave a substrate in Baltic languages too.

Awesome theory.

But the problem is that Erteboelle had nothing to do with anyone in the East Baltic and regions further to the east were modern Baltic languages came from.

Bob Floy said...

@Andre

Normally I despise this topic, because it's been beaten to death by everyone, but what about the irony of the "master race"'s "untermenschen" actually being genetically closer to the historical Aryans? That's the thickest irony, I'd say.

"I think both of you exaggerate the amount of non-IE influence on Proto-Germanic."

As far as I'm aware, it's still a topic of debate, but I've never seen a figure as low as 5%. It would barely be worth discussing in that case.

"Did GAC speak TRB or Rossen? Were they EEF languages?

Do you both propose the the German language is a tri-partite of forager WHG Erteboelle, Farmer EEF TRB and Yamnaya/CWC Indo-European, all in one package?"

Again, anything we could say is 100% speculation, although I personally am inclined to think(now, after that Iberia paper) that Vasconic is probably descended from a farmer language, maybe GAC, maybe TRB, maybe even something more western. I'm not prepared to be so specific about what Germanic is made of, whether it's got three components or not.

Bob Floy said...

A farmer language which had come to be spoken by the westernmost BB groups, that is.

Drago said...

@ Bob

''Vasconic is probably descended from a farmer language, maybe GAC, maybe TRB'"

GAC is too eastern . For your theory, you'd need to hold something like TRB-West

Andrzejewski said...

@Bob Floy "Again, anything we could say is 100% speculation, although I personally am inclined to think(now, after that Iberia paper) that Vasconic is probably descended from a farmer language, maybe GAC, maybe TRB, maybe even something more western. I'm not prepared to be so specific about what Germanic is made of, whether it's got three components or not."

It's remarkable how Vasconic is completely unrelated to Etruscan and its Lemnian group of Eastern Med, purportedly a descendant of some farmer language back east...

Bastian Barx said...

Like Andrewrzjniz, I was under the impression that the Germanic substrate hypothesis was more or less deprecated, since etymological origins for most Germanic words of non-PIE origin had been found. I'm not going to argue either way, since I clearly don't know enough about that to have an opinion - but if there really is such a huge non-PIE substrate in Germanic languages, it would undoubtedly come from TRB.
Jutland was SGC territory for a century or more, while Funen and Zealand still was TRB. Also, a lot of continuity is seen between the two cultures, to a degree that Danish archeologists as late 2013 still thought it was the same people, just with new pots. I guess the TRB people of Funen and Zealand would have been "sandwiched" between the SGC of Jutland and the boat-axe culture of Scania for odd 200 years, which is optimal for some kind of merger of language and culture to have happened.
In contrast to the south, where the replacement of farmers with pastoralists would have been much more abrupt. So basically geography comes as a handy explanation of where and why there's such a larger non-PIE substrate in Germanic (if there really is one)
It's also interesting to note, that religious sacrifices to the bogs seems to have been performed since the early Mesolithic in southern Scandinavia and all the way up to the Viking age. Some direct continuity in culture is easy to imagine.
Personally I find it unlikely that TRB people would have spoken an Ertebølle language, although some words for local plants or animals clearly could have entered the TRB language as a substrate.
The Ertebølle language would have missed a lot of words for new concepts and technology brought by the farmers. And it would have been a case of Ertebølle people joining the TRB, not the other way around. But of cause, I'm just guessing here.

Bastian Barx said...

Forgot to post this.
https://rootsofeurope.ku.dk/roe_sommerskole/Iversen_RoESS_2018.pdf

Bob Floy said...

@Drago

I know that you're right, and that we're probably talking about something close to the Netherlands, but, GAC is just so damn sexy, lol.
Of course I could also be totally wrong about all of this, but that bronze age Iberian R1b injection must have brought some kind of linguistic change, and to me Vasconic/Aquitanian/Iberian does not look native to Iberia.

*waits for impassioned three page speech from Gaska*

Bob Floy said...

@Andre

"It's remarkable how Vasconic is completely unrelated to Etruscan and its Lemnian group of Eastern Med, purportedly a descendant of some farmer language back east..."

Etruscan bothers me :p

Andrzejewski said...

@Bastian Barx What's interesting in the Graph you shared is that Neolithic languages influenced (apparently) all branches of IE languages while Germanic was the only one to be also influenced by foragers.

Why would any forager language still be around to influence Porto-Germanic is beyond me.

Bob Floy said...

@Drago

In the past you've also expressed a belief that BB=Vasconic, if I remember right. What's your idea?

Drago said...

@ Bastian

''Jutland was SGC territory for a century or more, while Funen and Zealand still was TRB. Also, a lot of continuity is seen between the two cultures, to a degree that Danish archeologists as late 2013 still thought it was the same people, just with new pots. I guess the TRB people of Funen and Zealand would have been "sandwiched" between the SGC of Jutland and the boat-axe culture of Scania for odd 200 years, which is optimal for some kind of merger of language and culture to have happened.
In contrast to the south, where the replacement of farmers with pastoralists would have been much more abrupt. ''

Essentially, you're endorsing the Copenhagen model, which is rather problematic. and frankly, rather facile.
There is nothing particularly special about Danish TRB. ''Farmers'' (as you call them), did not simply disappear everywhere. 'In the south' (presuming south of the Baltic Sea), such cultures continued for 400 years - GAC, Vlaardingen, did not just end in 2800 BC. So even we take their model at face value ; how do we know that the non-IE substrate comes from TRB , and not GAC ?

Delving deeper, Kristiansen/ Iverson/ Kroonen attempt to take the genesis of proto-Germanic to the Late Neolithic (Danish TRB + CWC), which is probably 2, 000 year premature. Apparently, they seem to think nothing much happened in Europe after 2800 BC.

Bob Floy said...

@Andre

"Why would any forager language still be around to influence Proto-Germanic is beyond me."

When you consider that WHG ancestry survived much better in those parts, it shouldn't be too difficult to imagine. From what I've read, they and the farmer folk up that way were co-existing more or less peacefully at least some of the time, especially in the Erteboelle culture's area. Pretty sure there is archeological evidence of this. Maybe the religion of the foragers in that region was adopted to some extent by the incoming farmers? All speculative(and that graph is hypothetical), but, *something* happened which left northeastern Europe with a ton of WHG.

Drago said...

@ Bob

"" but, GAC is just so damn sexy, lol. ''

I agree. Those Amphoras & cattle burials really do it for me

''In the past you've also expressed a belief that BB=Vasconic, if I remember right. What's your idea''

I don’t think I made such an explicit connection. My current view continues to doubt the prevailing understanding of BB genesis; as it’s based on some rather flawed premises. Given that B.B. was such an ''in-group'' (& predominantly but not exclusively male-centred) phenomenon which spread so rapidlly, it would be not unreasonable to surmise that its bearers spoke a common tongue.
What that is, and where they ''picked it up'' is more speculative; but might be worth revisiting in 3 or 4 papers time, when the studies from Switzerland, south Germany, Italy & France become available to the general public

Bob Floy said...

@Drago

Lol, those cattle burials and chevron designs...really hot.

"I don’t think I made such an explicit connection."

Awhile back you said, "if Vasconic didn't come with BB, where did it come from? Phoenician colonists?". Seemed to mean that to me.

"What that is, and where they ''picked it up'' is more speculative; but might be worth revisiting in 3 or 4 papers time, when the studies from Switzerland, south Germany, Italy & France become available to the general public "

I can't disagree with that.
And, whether or not the current ideas about BB's origins are correct, it seems really, really likely to me that Mallorca_EBA and the Rathlin island guys were speaking the same language.

Andrzejewski said...

Ever wondered how come there are some many mtDNA markers associated with the Neolithic revolution? H, T, J, HV, N, K, W, Z, etc...

Bob Floy said...

@Andre

It's because the steppe ancestry in modern Europe comes mostly from males, the farmer ancestry mostly from females.

Bastian Barx said...

@Drago
I don’t find their hypothesis too far-fetched, but rather well-founded. The reason to think the said substratum would come from southern Scandinavian TRB, would obviously be because this is where we know proto-germanic was later spoken – Obviously -
It’s generally assumed that some kind of “pre”-proto-germanic was spoken in the Nordic bronze-age. Not too crazy to assume.
And while big migrations clearly did occur, many people obviously stayed for generations were their ancestors had lived. I don’t think it’s too far-fetched to claim, that a language could have roots in a language spoken 1000 years prior in the same area. (Nordic bronze starts 1700 BCE. ) The religion of the nordic-bronze age was clearly PIE derived, going by the imagery (sun wagon, holy twins etc.) So assuming a corded ware derived PIE language was spoken isn’t far out, since we know what language people spoke later in that area.
…..and here we are in the that same timeframe. What you call “since 2800BC”.
No, there is nothing special about “Danish” TRB except the geography, which obviously did mean a difference. That is a fact. Jutland was better suited for a pastoralist lifestyle, while Funen and Zealand was heavily forested. And SGC and TRB did share Denmark for some 200 years.
Could a Germanic substrate come from GAC? Of cause it could, I just don’t see any archeological or genetic reason to posit that hypothesis. “Because it’s possible” is just a really piss-poor argument.
“The Copenhagen model”, as you call it, has stronger arguments and thus has a higher probability. But it doesn’t mean that it is true. It’s just a hypothesis. One you clearly have an aversion too, for some reason.
By the way, when I said, “in the south” I obviously meant “south of Denmark”.

Ric Hern said...

What we should not forget is that some words could have been PIE derived. The difference could be that not all Indo-Europeans preserved the use of some words which other Branches did...we see this happening in recently developed Languages. Not all words in the Parent Languages are used and becomes redundant because of various reasons...

Andrzejewski said...

No, I mean BEFORE the Indo-Europeans: the farmer mtDNA must’ve been highly diversified before they migrated to Europe

Andrzejewski said...

Did Cucuteni Tripolye contribute ANYTHING to any modern (Balkan?) populations or did they all get wiped out like the Botai?

Ric Hern said...

@ Andrzejewski

A while back there was a paper on Anatolia Neolithic samples... Yes they had high MtDNA diversity...

Ric Hern said...

@ Andrzejewski

https://www.biorxiv.org/content/10.1101/032763v1

Ric Hern said...

@ Andrzejewski

https://www.biorxiv.org/content/10.1101/016477v2

zardos said...

Another question is how different TRB and GAC actually were. The shift to the HG male lineages and intensified cattle breeding was happening in earlier TRB already. So probably they spoke related dialects anyway?

Obviously Proto-Germanic must have been influenced by both groups, but definitely more by TRB culture because of the long and strong local interactio.

Back to the threat starter I wonder whether the Neolithic/Southern yDNA, G,J and E came from survivors or later migrants. If later recent VA or earlier and when?
Do they have signs of recent admixture?

Also I dont get some remarks, as Germanic people were, and still are, among the populations with the highest proportions of Corded/Yamnaya-related ancestry.
And Germanics were closer to the original culture and IE way of life than most other groups as well. Only Balts and old Slavs come as close.
Which has a lot to do with their relative isolation obviously, but also with ancestry proportions.

Drago said...

@ Bastian

“It’s generally assumed that some kind of “pre”-proto-germanic was spoken in the Nordic bronze-age. Not too crazy to assume.”

I think that’s a reasonable hypothesis, and actually quite a different one to the Copenhagen model


“The Copenhagen model”, as you call it, has stronger arguments and thus has a higher probability.”

I wasn’t not actually proposing that GAC should be preferred ; it was merely a point
Given that neither R1a (CWC) & I2a1 (TRB) are the dominant lineages in ancient Germanic speakers; your lads model doesn’t seem to pass a basic sanity check

Andrzejewski said...

@zardos “And Germanics were closer to the original culture and IE way of life than most other groups as well.“

How? Even their language is not as “pure” Indo-European as the Balto-Slavic branch lol

Andrzejewski said...

It is not “Germanics” who have one of the highest Yamnaya: it’s the Nordic people in particular; modern Germans are mostly Neolithic Farmers (its funny how Germans and Turks hate each other in Germany, when in reality both groups have a high of 30% -40% Anatolia_N, but that’s just a side note, a foot note).

As for Scandinavians, they are about 40% - 50% Yamnaya, 30% - 35% Anatolia_N and about 25% - 30% WHG, with very small traces of SHG.

Drago said...

“Another question is how different TRB and GAC actually were.”

The might plot similarly; but they’re quite distinctive

Andrzejewski said...

@Drago I’m not sure who’s behind the transition from TRB into GAC. They both seem like WHG-rich Farmer groups. Some posters here attributed it to either Rössen or Laryngeal introgression so we’ll never pin it down 100%.

mtDNA T is common in Poland, N1a which is common in the Middle East nowadays was a staple of the Southern European Cardial Pottery derived cultures (like Sourhern France), whereas Ashkenazi Jews share lots of mtDNA K ( wherever the heck it came from!). All the while, Cucuteni Tripolye had RO, HV (assumed to be due to Paleolithic continuity) but on the other hand ever since Bronze Age H became very common.

Does it have anything to do with pigmentation? I don’t know but maybe.

The reason I’m so piqued about the CT Culture is that they used to be the hallmark of Neolithic “Old Europe” for decades according to Gimbutas and her outdated theories. I used to think that Paleo-Balkan populations (Thracians, Dacians, Illyrians, Paninians, etc) had at least partially something to do with it.

Did CT vanish completely off the face of the earth or did could any modern population: Austrians, Romanians, etc. reliabily claim any descent from them?

zardos said...

@Andre: Balts and Slavs are obviously the only other groups as close. You can now discuss who was or is closer, but thats rather meaningless.

As for the cultural aspect: Even under the presumption of Germanic language having more non-IE influence than other IE dialects, the social organisation and way of life of Germanics was closer to that of the IE.

The Neolithic ancestry is best represented by Sardinians,not Turks, as you should know. Abatolia today has no particularly high proportion of this component. Its higher in Southern Europeans. Modern Turks are quite different from the ANF.

And not that it is particularly important, but Germans are diverse, but even on average they have still more Yamnaya related ancestry than most other people. Especially if looking for larger world regions in comparison. But I spoke of Germanics anyway...

@Drago: I know they were different, but probably just two branches of the same tree. Ethnolinguistically I mean, speakers of related dialects.
Or do you see an influence big enough for thinking otherwise? Any migration from the South, change of lineages? I dont, but you know better.

Andrzejewski said...

Germanic people did not have cattle or use horses in their cavalry until the Middle Ages, nor did they bury their dead in Kurgans so I don’t think their social organization was closer to original PIEs. Besides, their religion bears similarities to the PIE one but so do other IE peoples’. The most remarkable facet of their social organization mimicking PIE was the distinction between priests, warriors and free men v. slaves (like the yeomen of medieval England), and maybe the reason England is so class divided almost like India is due to that fact. But I don’t see Germanic people otherwise resembling PIE more than others.

zardos said...

@Andre:
"Germanic people did not have cattle or use horses in their cavalry until the Middle Ages, nor did they bury their dead in Kurgans"

Is this a joke?

Germanics further developed what Corded people started. Its most obvious with cattle breeding and raw cow milk consumption. Or do you think one of the highest rates of lactase persistence is a mere coincidence?

They made warrior burials in a clear tradition, lived in patriarchal clans and of course they used the horse.

You heard of Tollense? There was cavalry.
Read Julius Caesar? Germanic cavalry played a decisive part in his campaign in Gallia.

Inform yourself.

Matt said...

Andrzejewski: The most remarkable facet of their social organization mimicking PIE was the distinction between priests, warriors and free men v. slaves (like the yeomen of medieval England), and maybe the reason England is so class divided almost like India is due to that fact.

This is really strange comment on multiple levels.

If we go by the OECD's projections, Europeans who speak Germanic languages today span from the most to least socially mobile / "class divided" Western Europeans* and the full gamut of income and wealth inequality.

England tends to be in the middle, funnily enough, with the extremes of high mobility in Scandinavia today, and low mobility in Germany today.
But even where these societies are more or less egalitarian or mobile in wealth, the rank order was different 100 years ago (Norway and Sweden in 1890-1910 were less egalitarian than the United States). So it is a nonsense to suggest these go back time way further than this.

It is also a nonsense to compare England, probably one of the most outbred and genetic structure free large countries in Western Europe, to India's Caste system. (The kind of nonsense I expect from Anglo and Celtic left wingers who are ignorant of the world outside the British Isles, obsessed with social class and obsessively chippy about it, and insist that the caste system of India was produced by the British, but not something that I would expect to ever see on the Eurogenes comment section, which if dopey, is usually not dopey in that direction).

*Eastern Europeans are no OECD so not comparable, other than Hungary, which is less socially mobile than Germany.

JasonK said...

@Andrzejewski
"Germanic people did not have cattle or use horses in their cavalry until the Middle Ages, nor did they bury their dead in Kurgans"

How do you frequent this blog and post such an ignorant statement? Let's disregard the fact that cattle were an important livestock since before the indo-europeans reached north central europe and Scandinavia among the TRB culture and point out that the norse rune "Fehu" which literally means "cattle" was synonymous with the term wealth/money and there are kurgan burials from the ancient battle axe culture into the end of viking age in discussion here. Frankly a discussion about who is more culturally close to proto indo europeans is pointless, but either you are misinformed or trolling.






Andrzejewski said...

The caste system of India was produced by the Andronovo “Aryans” and a similar division between priests, warriors, etc existed in other historical IE societies including pre-revolution France. In England there IS some sort of class hierarchy absent here in the USA between “royalty” v. “Commoners” where it comes to patterns of speech or the difference between those who attended Eton or Uxbridge like Boris Johnson. I think that it dates back to the PIEs but I may be as well wrong

zardos said...

To compare Catholic France in pre-revolutionary times with Indian priest castes and use that for your IE social division argument is another step into the wrong direction imho. Sorry, but it seems at least today you make it worse with every comment you post.

zardos said...

The British have social stratification and its more extreme than in Germany I'd say, but this has nothing to do with the Germanics in general, but was a long process which started with the Norman (=foreign) conquest.

Ric Hern said...

@ Andrzejewski

Germanic people surely had horses and cattle as reported by the Romans. Frisian Tribes had White Cattle and a Tribe migrating from Hesse to Frisia had Black Cattle....even the Angeln and Harz Breeds seems also products of earlier Germanic cattle. And where did you think the Fjord Horse came from ? Etc.

EastPole said...

A new book on Indo-Europeans will be published soon:

https://www.amazon.com/Tracing-Indo-Europeans-archaeology-historical-linguistics/dp/1789252709

Here is the first chapter:

https://www.academia.edu/39840697/Tracing_the_Indo-EUropeans._New_Evidence_from_archaeology_and_historical_linguistics

They are convinced without any reasonable doubt that the archaeological Yamnaya complex is the cradle of (Core-)Indo-European languages.

a said...



Steppe horses evolved on the steppe.Since no settlements have been found, a strategy for Yamnaya males to protect the clan would have to be put into use while traveling vast distances both east and west on the steppe.

For example the nomadic/life of Yamnaya clan/herd had battle for alpha ranking order and was much more violent than male Horse battle/fights. :)

"Horses fighting for hierarchy of herd order, a natural behavior to be used in behavioral training"

https://www.youtube.com/watch?v=SvE6Apx0rfI

Arza said...

@ EastPole

Here is the first chapter:

The idea for the present volume was fostered at a workshop held at the University of Copenhagen in December 2012

Oh.

https://www.youtube.com/watch?v=_xQNVexhSJQ
Round table discussion concluding the seminar "Tracing the Indo-Europeans: Origin and migration", organized by Roots of Europe - Language, Culture, and Migrations, University of Copenhagen, 12-14 December 2012

I wonder if barbarians will be worth mentioning.

Gabriel said...

@Andrzejewski

Isn’t the caste system likely as old as IVC though?

George said...

Hi

Off Topic

The Genetic History of France
https://www.biorxiv.org/content/10.1101/712497v1

"In this study, we analyzed SNP-chip genetic data from 2 184 individuals born in France who were
enrolled in two independent population cohorts." ... "In conclusion, by performing the
first exhaustive study of the genetic structure of France, we fill a gap in the genetic studies in
Europe that would be useful to medical geneticists but also historians and archeologists. "

"... the population can be divided into subgroups, which match geographical regions ..."

It appears to be similar to the paper: The fine-scale genetic structure of the British population, Nature volume 519, pages 309–314 (19 March 2015)

George

JasonK said...

@Gabriel
"Isn’t the caste system likely as old as IVC though?"

No, I've never even heard of fringe theorists that believe this. Although the other side of the spectrum is some academics pushing the misconception that the IVC was extraordinarily egalitarian or pacifistic, somewhat equivalent to how some viewed neolithic europe.

Fanty said...

@Andrzejewski
"Germanic people did not have cattle or use horses in their cavalry until the Middle Ages, nor did they bury their dead in Kurgans"

Haha. Ridiculous claim.

Tacitus (Roman historian, wrote the book "Germania" in 99 AD, thats looooong before the middle age). talks about cattle of the Germanic people.

he also mentions horses. He says the priestess using sacred white horses for reading the future from the horses behavior.

And he mentions Germanic cavalry. He says, the majority of the Germanic warriors fights on foot, but there are some who ride horse. These horses are not beautifull or fast and the warriors use them with shield and lance. No armor. Also no stuff that is supposed to improve looks or make the rider apear badass. Just a man on a horse with shield and lance who fight mixed with infantry (no pure cavalry units)

Davidski said...

@Andrzejewski

Like I said before, you need to do some background reading.

This is not a place to learn the basics. You need to do that yourself before you get into discussions here.

And when you do it, make sure you read more than just some random Wikipedia and other online sources, because they can be very misleading.

self-consumer said...

@Andrzejewski

I thought asking "AMH=?" was rock bottom for you. Sadly, I was wrong.

FrankN said...

Seemingly, my earlier comment on non-IE substrate in Germanic has been somewhat misinterpreted, might also have been formulated misunderstandably. Let me clarify a bit on that Proto-Germanic Swadesh 100 list. It is structured as follows (my detailed assessment fell victim to a HD crash, numbers are approximate and from memory).

1. A handful of terms (6-7) are rather useless. Some are ill-defined across a very wide semantic field, resulting in a rather arbitrary recording of representation in a specific language. This concerns "stone" (gravel, pebble, rock, boulder, jewel, fruit kernel, non-organic construction/ tool making material, etc.), "path" (a/o trail, track, road, street, passage/ crossing, c.f. proposed PIE *pónth₁s, meaning "bridge" in Latin), and "belly" (suggested PIE *úderos rather means "womb" than with "belly").
Equally problematic are colour terms. Most reconstructed proto-languages only distinguished "coloured" and "colourless" (c.f. for the latter English "bleak", "black", "blue", "pale"). Individual colour terms often tend to be derived from very specific flowers/ substances, e.g. "violet", "cyan", "purple". "Red" is in many languages associated with "blood", see PIE *krewh₂- "blood", PIE *h₁rewdʰós "red" (Dravidian is another family with "blood"="red" association), and as such redundant. [Note otherwise "blood" as an example of a non-PIE proto-Germanic root.].

2. Some 20 terms qualify as what Pagel e.a. call "ultraconserved words", with a linguistic half-life of up to 70,000 years, i.e. OOA II or even before. These are mainly personal, demonstrative and interrogative pronouns, but a/o also include "water" and "fire" terms of global distribution. Other examples are Malay dua "two" (PIE *dwó), and Ojibwe men-ikk "many" (PIE *monogʰo-), both of which can be traced back to the respective proto-languages, i.e. the 4th mBC, and as such aren't recent borrowings.

3. Another some 15 terms are shared by PIE with more than one (North) Eurasian language family (Uralic, Altaic, Afro-Asiatic, Kartvelian, NW/NE Caucasian, Dravidian, Nivkh, Chukotko-Kamchatkan) - somwhat akin to "Nostratic", whereby Nostratic doesn't include Afro-Asiatic. Examples include "heart" (if not derived from the globally-distributed *kar "stone, core" paleo-root), "sit", and "name". These terms may actually have originated from (pre-)PIE, to have been borrowed by other candidates, especially Uralic and Kartvelian. But they may equally reflect external linguistic influence on (pre)PIE. We'll probably never know for sure. In a conservative assessment, these terms are to be qualified as Neolithic Wanderwörter/ "aerial terms".

4. Somewhat around thirty terms qualify as proprietary PIE roots. This, however, includes both

a.) PIE roots of universal representation, i.e. including Anatolian, Tocharian and Indo-Aryan, such as PIE *pṓds "foot", *gonu "knee", péth₂ "feather", and

b.) Roots restricted to Western IE, unattested for Anatolian, Tocharian and Indo-Aryan. They may constitute late Western IE innovations, but equally borrowings from EEF substrate. Western IE *séh₁mn̥ "seed", as a farming-related term, was in all likelyhood borrowed from EEF. I haven't yet checked on other Western IE terms, nor calculated the ratio of "universal" vs. "Western" IE terms.

FrankN said...

5. Finally, another good thirty Proto-Germanic terms lack a credible IE etymology. Among them is "hand", which Shrijver connects to similar terms in Saami and ultimately intepretes as Fennoscandian substrate. Also included are terms for which an IE etymology is IMO semantically or phonetically questionable. "Bone", e.g., is via PGerm *bainaz "straight" being connected to PIE *bʰeyh₂ "to hit, strike, beat" ["Hit me with your collar bone"...]. "Drink" has been connected to PIE *trenkʷ "to push, to press" (German "drängen", dialectal N. English "thring")- a root poorly represented outside Germanic except for Baltic, and thus in itself a prime candidate for non-IE substrate. "Hair" is being related to PIE *ḱer "to grow, nourish", c.f. "create", but that leaves the lengthened grade unaccounted for, and *ḱer is only attested for OGrk and Latin.

My above "50%" statement related to an approximately equal ratio of proprietary IE terms (#4) to those w/o credible IE etymology (#5) in the Swadesh 100 list, but shouldn't be taken as an absolute measurement of non-IE Germanic substrate.

This holds even more true for the fact that - for unknown reasons - the Swadesh 100 list apparently does an exceptional job in singling out this substrate. Atkinson/Gray (link below) have performed phylogenetic analyses on both the Swadesh 100 and Swadesh 200 lists. The Swadesh 200 list yielded the standard result of Germanic clustering with Italic and Celtic (a similar result as e.g. produced by Jaeger's 2016 analysis based on the Swadesh 40 list). In contrast, the Swadesh 100-derived phylogeny showed Germanic as having split very early from the IE mainstream, prior to Indo-Aryan and Balto-Slavic.

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.799.7328&rep=rep1&type=pdf

«Oldest ‹Older   1 – 200 of 268   Newer› Newest»