I'm seeing increasing numbers of Bronze and Iron Age samples from Central Europe and surrounds with this peculiar set of traits:
- shared genetic drift with present-day Balto-Slavic speakers to the exclusion of most other Europeans
- and yet, an unusually low level of Yamnaya-related steppe ancestry
- so much so, in fact, that they're often outside the range of modern European genetic variation.
As far as I can tell, currently the best examples of this unusual population are HUN_Mako_EBA_o:I1502 (Mathieson et al.
Nature 2015) and HUN_EIA_Prescythian_Mezocsat_o1:I18241 (Patterson et al.
Nature 2021). Both are from the Carpathian Basin in what is now Hungary.
I ran a series of qpAdm mixture models to try and learn more about their origins. The most robust outcomes, out of about 50 different attempts, are these:
right pops:
CMR_Shum_Laka_8000BP
MAR_Taforalt
IRN_Ganj_Dareh_N
Levant_PPNB
TUR_Barcin_N
Iberia_Southeast_Meso
UKR_Meso
England_Meso
RUS_Karelia_HG
RUS_West_Siberia_HG
MNG_North_N
TWN_Hanben
BRA_LapaDoSanto_9600BP
HUN_Mako_EBA_o
Baltic_LTU_Narva 0.149 ∓0.028
POL_Globular_Amphora 0.613 ∓0.028
Yamnaya_RUS_Samara 0.238 ∓0.029
chisq 10.836
tail prob 0.370463
Full output
HUN_EIA_Prescythian_Mezocsat_o1
Baltic_LTU_Narva 0.186 ∓0.028
POL_Globular_Amphora 0.592 ∓0.027
Yamnaya_RUS_Samara 0.222 ∓0.029
chisq 12.492
tail prob 0.253499
Full output
Combining the two genomes produces a very similar result:
HUN_EBA-EIA_o
Baltic_LTU_Narva 0.160 ∓0.023
POL_Globular_Amphora 0.612 ∓0.023
Yamnaya_RUS_Samara 0.227 ∓0.023
chisq 14.653
tail prob 0.14524
Full output
Importantly, when I move RUS_Karelia_HG from the right pops to the left pops, to test whether HUN_EBA-EIA_o really has steppe ancestry, as opposed to closely related hunter-gatherer ancestry, I still get a very similar outcome:
HUN_EBA-EIA_o
Baltic_LTU_Narva 0.158 ∓0.027
POL_Globular_Amphora 0.605 ∓0.033
RUS_Karelia_HG 0.014 ∓0.038
Yamnaya_RUS_Samara 0.223 ∓0.053
chisq 10.461
tail prob 0.234171
Full output
So these largely Globular Amphora-related individuals do harbor as much as a quarter of steppe ancestry, which is to be expected considering the massive genetic turn-over that most of Europe experienced just before their time as a result of population expansions from the Pontic-Caspian steppe.
Nevertheless, this is ~20% less steppe ancestry than in the present-day populations of the region, and it clearly shows in any decent Principal Component Analysis (PCA) of West Eurasia. For instance:
At the same time, the relatively close genetic relationship between these ancients and present-day Balto-Slavic speaking populations shows up in fine-scale intra-European PCA.
The origins and implications of this population are still a mystery to me. I don't think it's native to the Carpathian Basin. Indeed, my qpAdm models suggest that it may have moved into this region from somewhere to the northeast, because its ancestry is best modeled with ancient groups from present-day Lithuania, Poland and Russia.
I'm adamant that these people weren't Balto-Slavic speakers, and certainly not proto-Slavs. Rather, I suspect that much like the Welzin warriors of Bronze Age North-Central Europe, they were closely related to a contemporaneous group that eventually gave rise to proto-Slavs. At best, they may have somehow contributed to the ethnogenesis of Balto-Slavs.
By the way, using the Global25 to model their ancestry is highly problematic, because of the strong Balto-Slavic genetic drift that affects some of the dimensions. So be careful when you try it, or better yet, don't try it at all, and stick to formal stats in this particular instance.
See also...
Tollense Valley Bronze Age warriors were very close relatives of modern-day Slavs
262 comments:
1 – 200 of 262 Newer› Newest»Nice work David, you’ve been on a roll lately, with only a few days in between posts.
Does this connect?
https://www.biorxiv.org/content/10.1101/2022.01.19.476915v1
I don't usually blog about papers or preprints now unless I can get the data, because it's hard to make any useful comments without running the samples.
IMO Balto-Slavs emerge from the “forts and striated pottery” group in the east Baltic and Belarus LBA
The Tollense warriors in Germany are the only cemetery size set of samples under this category.
When I run them through G25, their farmer ancestry always comes out as entirely of Hungary EF origin. They don't want Globular Amphora or anything western farmer admixed.
This makes me think maybe they formed in Hungary as a mixture of incoming northern hunter gatherers, incoming Yamnaya, and native farmers.
But as you say, maybe Balto-Slavic drift does skew who they chose as their farmer ancestry. Maybe, they formed in Corded Ware Poland.
But the Avar data is interesting. Not much in the way of typically “Slavic” individuals , a lot of EV13 . Makes you want to wonder who those “Sklavenoi” on the Danube were
@Genos Historia
The Tollense warriors might be from somewhere near Hungary.
Based on isotopic data, well before ancient DNA, the theory was that they came from southeast of Tollense, like from Silesia.
So they might have some sort of southern farmer input, because they're not nearly as unusual as HUN_EIA_Prescythian_Mezocsat_o1, with more steppe and less hunter-gatherer ancestry.
But anyway, their genomes are crap. So who really knows.
@Davidski,
All these Steppe-poor, Balto Slavic drifted individuals show the same type of farmer ancestry as the Tollense warriors when using the G25 PCA.
They always chose Hungary EF. If the Tollense warriors are from Hungary, then all of them should be.
In essence what I'm saying is the Tollense warriors aren't weird, they are apart of the same genetic cline/family as the others.
The oldest sample in this family, as we know, is N49 from Corded Ware Poland. He also choses Hungary EF which doesn't make sense because he lived in Poland.
He's good evidence this WHG-rich, Steppe-poor family formed in Corded Ware Poland.
Right, the G25 is probably compensating for something.
Things should look a lot better when we get more proximate samples.
The G25 has problems when it's forced into models that don't reflect reality in some important way.
Top 5 Global 25 matches for these two (excluding each other):
JAG82, VLI050, JAG78, MOK23, JAG58, MOK22, JAG06
Plotting of those on PCA: https://imgur.com/a/zW5y1Kf
I'm still quite unsure about their status re;Balto-Slavic drift...
Had a look at these again recently, and despite having comparable HG ancestry leve to Latvia Baltic BA, they don't have the same level of high distance to the model when modelled with Iron Gates, Afanasievo and Barcin N.
Now, if the excess distance of Latvia Baltic BA was purely caused by a shared HG element which these samples have in about the same excess, we would expect that the distance would be raised by the same amount...
For example: https://imgur.com/a/xpELepb
So still uncertain on that point.
Distance from present day people, also: https://imgur.com/a/UCAop9B
Generally given a choice in Global25, these samples seem to reject Globular Amphora and select a Hungarian farmer group like Baden - https://imgur.com/a/AvbLhOc
The y-dna for the above set of related samples so far, for males:
- JAG06, JAG82, JAG78; all come from a closely related set of males with G2a2a1a2a2a1.
- MOK22 (Grave 225); R1b1a2a2c1 (https://www.nature.com/articles/s41598-021-89090-x/tables/1)
All others are female (the two in the post, and all other above).
The preference for G2a and Baden LCA farmer seems reasonably as expected for the JAG samples for a Carpathian group.
Another way of quickly looking at the samples in terms of Balto-Slavic drift separately from deep ancestry proportions, is to the compare the samples to a set of simulations of them, using those propotions I computed up in Vahaduo (using Baden_LCA, Serbia_IronGates and Afanasievo).
Doing that and then projecting both on Vahaduo PCA: https://imgur.com/a/MBj8UTp
Projected on to the "North Europe PCA", both sets of real samples, the Carpathian group closest to I18241 and I1502, and the Baltic LBA group, are both displaced east towards the Balto-Slavic cline, compared to the simulations. That would suggest both are in some way more connected to the "marriage network" that was ancestral to present day Balto-Slavic cline.
But this is much stronger in effect for Baltic LBA. So it does seem like this would suggest that whatever the Balto-Slavic drift factor is, the LBA has a lot more of it.
In the West Eurasia PCA, the real and simulated samples basically sit on top of each other.
Genos, for formalities ... there are two such samples: N47 and N49.
David, if such individuals were closely related to the Balto-Slavs and participated in the ethnogenesis of the Balto-Slavic peoples, why couldn't they use Balto-Slavic dialects?
Based on the genetic structure of modern Balts and Slavs, the first people who spoke Balto-Slavic dialects had more steppe ancestry.
To assume that HUN_EBA-EIA_o were Balto-Slavic speakers, we would need to also assume that they represent another, extinct branch of Balto-Slavic, and then also that the main, surviving branch became more steppe shifted after they split.
However, this doesn't make much sense considering the steppe heavy character of other North and Central European Indo-European speakers.
@Matt
Yeah, the JAG and Mokrin samples have Balkan ancestry on top of this Balto-Slavic-related ancestry.
It showed up in a few basic qpAdm runs that I did.
Ok guys, unless there is no G25 I created oracle for Vahaduo using Admixture values for all samples from this table:) [url]https://www.biorxiv.org/content/biorxiv/early/2022/01/20/2022.01.19.476915/DC5/embed/media-5.xlsx?=true[/url]
Pannonian Oracle - [url]https://drive.google.com/file/d/1A-73p-2aGrr2_1YHdtTWVgmVC6J-tsj0/view?usp=sharing[/url]
Use ADC x0.25 or more, otherwise results are too fragmented.
So put samples with "OWN_" to Target to analyse properly Hun, Avar and Conqueror admixtures.
David, I don't think the percentages of the steppe matter here. The homogenization of the local gene pool could have been uneven and the decline of the pedigree collapse could perpetuated the farmers' genes.
as well as I know today's Germans, Poles and Slovaks from the Carpathians have a genetic heritage from Welzin_BA but the samples are not well covered?
N47 and N49 plot similarly to modern Lithuanians on the West Eurasia plot. I think they are closely related to proper Balto-Slavs, and not specifically connected to these processes in Pannonia.
This Pannonian ancestry is concentrated in cultures closely related to the Vucedol culture (Jagodnjak, Mokrin, Mako), other samples are outliers in their respective cultures (Vatya_o, Mezocsat_o, one of the Hungarian Bell Beakers) and can be explained as migrants from Vucedol.
Vucedol has archaeological similarities to the Kostolac culture which entered the Carpathian Basin from the east trough the Iron Gates a few centuries earlier.
The leaked PCA from the upcoming study on Pannonia shows samples with even more WHG than any of these, in EBA Serbia and Croatia, one of them exactly from the Iron Gates site.
The study on Mokrin suggested their increased WHG is coming from Central Europe ( I.e. Poland or Germany) but given all of this, it looks more as something from the wider steppe zone, distantly related to Sredny Stog (samples like Dereivka and UKR_EBA)
I don't think these were IE speakers at all, they just had some female mediated ancestry from Yamnaya. In it's pure form this population seems to be dominated by G2 and I2 (e.g. Jagodnjak, or Mokrin, when you remove the Yamnaya subclades)
Sheet with only modern samples from this table
https://drive.google.com/file/d/1rZhTCL6vdoCZYjQBglflzgQKYmlVawtz/view?usp=sharing
Tollense look like a Mako_EBA-related population which migrated north and, was mixing with BBC and Baltic_BA for some time.
Target: DEU_Tollense_BA
Distance: 1.2788% / 0.01278821 | R3P
40.8 DEU_Lech_BBC
40.4 HUN_Mako_EBA
18.8 Baltic_LTU_BA
However they were stil very pre-IE at this point, they still had a lot of I2, and genes for lactose intolerance.
Tollense has a rare I2 subclade (I2-L1229>Z2069) which is closely related to the one found in BA Pannonia (I2-L1229>S18331), which further confirms their connection.
I think Tollense could simply be originating from Western Hungary, which is very poorly sampled compared to eastern Hungary.
for example Slovenia_EIA:I5689 close to Western Hungary is also I2-L1229>Z2069, and autosomally Slovenia EIA has some Pannonian-WHG influence. Not sure about this exact sample.
@bce
Corded Ware N47 and N49 do look like Balto-Slavs genetically, though maybe with a little less steppe ancestry than expected.
But I don't think it's a given that they were Balto-Slavs considering their age.
They might represent the early formation of the Trzciniec culture, which then contributed to the formation of Balts and Slavs.
there will be a study on the west Hungarian branch of this population, of which we have no samples yet. they might as well turn out to be I2-L1229>Z2069.
"Our results indicate a population turnover from first to second horizon around 2200 BCE with a group of high (42%) Hunter-Gatherer component previously unknown from the period, which during the transition from the second to third horizon subsequently blended into the prevailing genetic pool of surrounding populations through female biased admixture.
The origin of this particular Hunter-Gatherer ancestry likely comes from unsampled regions of Eastern Europe, and contributed to various populations to some extent, while becoming most prominent in the Baltic region from the middle of the second millennium BCE."
"Genetic History and Life of Bronze Age Communities in Western Hungary" (Gerber et al)
@ bce
''Vucedol has archaeological similarities to the Kostolac culture which entered the Carpathian Basin from the east trough the Iron Gates a few centuries earlier.''
I 've heard of Cotofeni being from Transylvania, but Kostolac from the East ? What samples are they ?
If we look at Baden and the non-R1b Z2103 Vucedol, their Y-DNA (G2a; I2a1a) looks Alpine
@Rob
There are no Kostolac samples yet, I'm just talking about archaeology. The upcoming Iron_Gates_EBA sample might be a Kostolac descendant.
Kostolac started from the Iron Gates and then spread west towards Croatia. It's possible taht they are just an ofshoot of Cotofeni, athlough it's not 100% clear.
@Davidski, well, I think generally there's still a lot of room to determine where the source of farmer related ancestry is coming from in all these ancient and present samples who have shift towards present-day Balto-Slavic groups. It could be that Carpathian groups made some contributions to all of them. (The old argument is that if the population in which the full Balto-Slavic drift had developed was as HG rich as LVA_LBA, and it seems to be from everything the Global25 indicates, there must be some counterbalance to get to present day people, so a group lower in HG than GAC and richer in ANF must be making some contribution).
@bce: The study on Mokrin suggested their increased WHG is coming from Central Europe ( I.e. Poland or Germany) but given all of this, it looks more as something from the wider steppe zone, distantly related to Sredny Stog (samples like Dereivka and UKR_EBA)
Yeah, personally I think that's a really open question too. I see a relatively Central European refugia or survival location for the HG rich ancestry as most likely (rather than steppe zone) still because it just seems like the further we go towards Ukraine-Belarus, then the probability of a IronGatesHG or even Narva like group just gets lower and lower and it seems like groups from there should have other admixture. (Similarly why a late expansion of a core Slavic group from too far east seems less likely than somewhere at the intersection of Poland, Western Ukraine, Belarus, which is pretty much the mainstream idea I think. Just seems to me that people from too much further east at that time should be having some greater excess of steppe ancestry than fits for the predicted ancestor, and also some entry of the Siberian/Iranian related components, by this time. Like looking at the extent Ukraine samples by geography - https://imgur.com/a/Od90dVk. The MJ14 sample from Central Ukraine at 600 BCE is best overall match, but still it seems like you would probably want to go very slightly northwest from there to be safe?).
But it's possible that many disruptive migrations happened, and also many barriers to geneflow may have been higher, and people we don't expect to be in certain places were. Only the direct adna itself will tell. (If we can achieve it, despite practice of cremation and problems like this).
Well, it's rather unlikely that an Iron Age individual from the Carpathian Basin would just have Globular Amphora farmer ancestry.
This might not even be Globular Amphora ancestry at all, but something distantly related, or even just a composite that most resembles Globular Amphora.
But the models failed unless I had Globular Amphora in there, so that's all I really know for sure.
@ bce
o right. i thought you might have been suggesting well east, like Moldova
Dudic summarises - 'The origins of the Kostolac culture have been discussed by many authors, and most agree that it developed under the influences of the Baden culture on autochthonous Neolithic populations in eastern Slavonia and Syrmia (Dimitrijević 1979: 230; Durman 1988: 13), and that it existed independently until the formation of the Vučedol culture that used it as the base for the development of its ornamental style (Balen 2010: 87-89). '
But yeah, aDNA would be helpful to test hypotheses.
@ All
eastern Romania could be one region with yet unsampled HGs, although im not sure theyre the source for Mokrin or other west Balkan MBA groups
Hmm, yeah maybe Narva + BarcinN would fail, and even HG in LCA Carpathian is not enough, but the GAC HG+Narva composite covers it? OK, thanks, probably that's the limit of investigating that.
@Davidski
That's interesting, a likely WHG-rich population which had contact with GAC could be Neman, in east Poland/Belarus, or maybe Usatovo and Cotofeni in Romania/Moldova.
@Matt
well, Kostolac, Cotofeni and Usatovo further east in Moldova are connected in some way, but it's hard to say who originated from whom. I agree, more aDNA is needed.
Uasatavo is just synchronic with Cotofeni, but different (burials, habitats, etc). I would bet that Usatavo is a steppe group, whilst Cotofeni, Kostolac are essentially Balkan EEF
There are a few Usatovans on the leaked / presentation Anthony PCA - https://m.imgur.com/a/MWX38hE
As we know there is some close family relation between one of these and one of the Steppe Maykop.
@Davidski
„I'm adamant that these people weren't Balto-Slavic speakers, and certainly not proto-Slavs.”
Yes, I also think they were not proto-Slavs. Proto-Slavs came from CWC.
I agree with Arza that Balto-Slavic drift comes from HG-EEF mixed populations around Carpathian mountains because there is no such drift in Fatyanovo and other Eastern CWC groups. And that drift entered Balto-Slavic population after the separation of Indo-Slavs around 3000 BC.
https://postimg.cc/nsFyGj7Z
Some time ago I proposed following model:
https://postimg.cc/64y9tHdK
Interesting puzzle is why Balto-Slavic drift affected R1a CWC groups. The general idea of what puzzles me is presented here :
https://postimg.cc/w3vXJ6GN
One possible answer is that there were no linguistic, cultural, or religious barriers between R1a CWC groups and this may explain Slavic origin. According to linguists Proto-Slavic started to form around 3000 BC and never separated from PIE, i.e. all those red dots were speaking languages you would partially understand.
This looks like an Eastern Funnelbeaker group (hence Narva affinity) that migrated south prior to the Steppe migration and settled along the Danube. We see them in Slovenia, Hungary, and the Eastern Balkans too. The suggestion that they came from the Northeast into the Carpathian Basin or Bohemia in the BA will end up being false, there is no population like this in the Baltic pre or post CWC. They came from the South.
Then again the Tripyllians had this level of Steppe ancestry. Could be them too.
Some input from these populations into the ethnogenesis of Balto-Slavs would explain what we see on PCAs and, to a lesser extent, the results of tools based on formal statistics.
At the very least, it makes more sense the idea that Slavs are derived from Narva HGs and other types of nonsense that various ill-informed individuals have spouted throughout the years.
Ideally, you'd want more BA/IA samples from the Poland/Belarus/Ukraine border regions, but I don't see anything like that on the horizon.
I wonder if a similar scenario happened in other parts of Europe, some relict outlier groups, rich in WHG, contributing to a modern population.
The Basques are a pretty interesting case.
At first glance, most of their genetic distinctness from neighboring populations is down to isolation since the Iron Age. But they harbor an unusually high WHG
ancestry considering their ratios of EEF and Steppe.
@Altvred, I thought the consensus we'd come to on Basques was that they had a pretty normal MN farmer source for West Europe, in terms of its WHG:ANF ratio, but were enriched with WHG in formal stats because they just have lot more MN farmer share than typical and less steppe? The Sardinans have more EEF, but their EEF is much poorer in WHG share (more like Italian-South Balkan EEFs).
Re; "Some input from these populations into the ethnogenesis of Balto-Slavs would explain what we see on PCAs and, to a lesser extent, the results of tools based on formal statistics." I think that's the argument that arza and others have been making for last few years (or at least, that they share a ghost ancestor with these populations). I'm not sure about anyone proposing that Balto-Slavic languages spring from some pure Narva like population. That sounds very marginal.
There are a few crackpots online claiming that Slavs, and especially Poles, are mostly of Karelia_HG origin. It's happened here as well.
Some of them still occasionally try to post here, but I just delete their comments now.
The upcoming Hungary study will be the first to get a true population size set of samples from this mysterious genos, other than the Tollense battle.
So far we only have random outliers. These outliers comes from populations. We just have to find the populations and the culture associated with them.
I suspect the Maros culture was created by them.
@David - do you agree with @Matt's association of these samples with
JAG06, JAG82, JAG78 (G2a2a1a2a2a1) and MOK22 (R1b-P312)?
If so those are very weird haplogroups for pre-Balto-Slavs. Very First Farmers for G2 and very pre/early-Bell Beaker for R1b-P312. If Balto-Slavs get some of their ancestry from a related group, I'd suggest that they had VERY male-biased admixture from a later steppe group.
Sorry I was looking at an old tree. Mok22 is R1b-U106, not P312. Which does make more sense.
@Matt
Maybe Basques weren't the perfect analogy here, but I always found them odd. They score as much WHG, if not more than Northern Euros, whenever I try and model modern Europeans.
The consensus seems to be that they preserved a typical IA profile rich in WHG. At the same time, the rest of Iberia had it diluted with Roman and North African admixture.
I can't say I'm too familiar with Iberia, which is a blind spot for me. I only recently read the Olade 2019 paper.
https://docs.google.com/spreadsheets/d/1Tc_5kE3TEq3GD1LL3ArPz5Pi2VW1Y-Rdo9UWsoxOivM/edit?usp=sharing
"I think that's the argument that arza and others have been making for last few years (or at least, that they share a ghost ancestor with these populations)."
From what I understand, his thing was placing these exotic populations and, therefore, the formation of Balto-Slavs in the Carpathian, which given the combined linguistic and archaeological consensus, seems unlikely IMHO.
There was also that odd Chalcolithic Romanian sample that I think he initially brought up as having BS drift. It just looks like a Balkan-HG with high EEF; the sample also lacks Steppe ancestry, unlike those mentioned by Davidski in the main post.
@Davidski
As I'm sure you're aware, this field attracts droves of the mentally unstable. The "Slavs come from EHGs/don't have real Steppe ancestry" group is only one of many.
It all boils down to "Eastern Europeans couldn't possibly have more Steppe ancestry than Northwestern Euros(Me)"(excluding Scandis here).
I encountered this sort of behavior on AG several times, pathetic stuff.
@Ryan
If so those are very weird haplogroups for pre-Balto-Slavs.
I never claimed that the JAG MBA group were pre-Balto-Slavs. But they do have significant ancestry from HUN_EBA-EIA_o, as well as from Balkan farmers.
So they share Balto-Slavic drift with Balto-Slavs via a third party, and they might be ancestral in part to some Balkan Slavs.
I think the Slavic R1a subclades are obviously from Corded Ware, but some of the other typically Slavic Y-HGs might be from the group closely related to or perhaps even ancestral to HUN_EBA-EIA_o.
@Ryan
All the BA Pannonian samples we have are already mixed, autosomally and y-dna wise, between Baden (local EEFs) and 3 invading groups, Yamnaya, BBC and this WHG-rich population.
When the Baden, Yamnaya, and BBC y-dna is eliminated, these seem to be the original y-dna of the "HUN_EBA-EIA_o"
I-S18331
I-BY33138
I-Y13338
G-FGC2315
G-PF3239
I-S18331 is especially important, because it's sister clade I-Z2069 is found in Tollense.
@Altvred
In G25, The ROU_C_o sample strongly prefers Balto-Slavic or HUN_EBA-EIA_o sources.
Target: ROU_C_o
Distance: 2.8974% / 0.02897387 | R2P
59.4 HUN_Mako_EBA
40.6 HUN_Koros_N_HG
Target: ROU_C_o
Distance: 2.1287% / 0.02128669 | R3P
47.4 HUN_Koros_N_HG
31.2 DEU_Esperstedt_MN
21.4 Baltic_EST_BA
To me it looks like the ancestors of HUN_EBA-EIA_o were already living somewhere close to Romania, and ROU_C_o had some admixture from them, besides local WHG and EEF.
The nature of the Balto-Slavic drift is unimportant. Another point is important. The Balto-Slavic drift was supposed to be a marker of Slavic migration from north-eastern Europe, where it is currently most concentrated. However, the Balto-Slavic drift does not work as such a marker, as it appears en masse in Central Europe before the Middle Ages.
That's true, but it was largely erased by Celtic, Germanic and other expansions in Central Europe during the Iron Age.
Then it expanded again with Slavs, so that today it's associated with Baltic and Slavic speakers, as well as Germanic groups that plausibly have Balto-Slavic admixture.
Distantly recollecting, so we should double check, but the I2as in Tollensee & Vatya, Hungary are found earlier in western Europe (Wartberg, British Neolithic).
David, but there are also a lot of Iron Age samples from Central Europe with a high proportion of Balto-Slavic drift.
Most of these samples are outliers or heavily mixed.
That's probably why Balto-Slavic drift is so closely associated with Baltic and Slavic speakers today.
@Davidski
I asked you about the puzzle that interests me i.e. why Balto-Slavic drift affected R1a CWC groups. The general idea of what puzzles me is presented here :
https://postimg.cc/w3vXJ6GN
You seem to propose this initial scenario:
https://postimg.cc/5jJhMDJn
That is you believe that originally R1a CWC population with Balto-Slavic drift (i.e. Balto-Slavs) was constrained to some region in the east. And mixed with Celtic and Germanic populations during the migration period to produce modern distribution of R1a and Balto-Slavic drift. Am I correct?
So if we find R1a CWC populations with Balto-Slavic drift whose location on PCA corresponds to modern distribution but is dated before migration period, for example in Nitra, you are proved wrong. OK?
Balto-Slavic drift appears to have been initially derived from the Eastern European populations that Corded Ware mixed with, but of course it became more intense and more characteristic of Balto-Slavic speakers over time.
I'm not sure if Nitra will show any Balto-Slavic drift, based on what I've seen.
Also, no idea why you think that my theory includes Celts and Germanics?
The only reason that I mention Germanics is because some of them are mixed with Slavs and Balts, so they show Balto-Slavic drift, and also Y-haplogroups like M458 and Z280, so the drift is real.
Ok, Rob I am glad to finally get to the heart of your theory hunter gatherers survived just fine in Neolithic Europe.
I want you to know I looked at the linked map you shared from a Polish archeaologist claiming hunter gatherers suvrived alongside LBk farmers.
But I'll talk about that later.
Right now, I want to list some DNA facts from cases in Neolithic Europe and have you try to describe them in a way where the hunter gatherers were not replaced.
In Southeast Europe, your argument for the low levels of hunter gatherer ancestry in the farmers is hunter gatherers didn’t live there when farmers arrived.
But you can’t make this argument for Serbia and everywhere west and north of there, because we have ancient DNA from HGs who lived there. HGs did in fact live there.
Yet, despite this, in the huge number of DNA samples from farmers in Hungary ranging from the beginning to the end of the Neolithic era, HG ancestry never reaches above 15%.
What happened to the hunter gatherers in Hungary? Can you tell me how this can be the case if hunter gatherers in Hungary were not replaced?
Next Italy.
We know for a fact hunter gatherers lived throughout Italy. We have ancient DNA from them. Yet in the Neolithic farmer DNA we have from Italy, Hg ancestry never reaches above 15%, except in one early Neolithic sample.
Why is it that Neolithic Sicily had only 10% HG, why is it that Remedello only had 15%? What happened to the hunter gatherers in Italy?
Next Britain.
Look at neolithic DNA in Britain. British farmers don’t need any new WHG input from British HGs. Only in Orkney and one site in Ireland do we see intermarriage with British HGs. And a few surviving hunter gatherer tribes in the most remote parts of the Isles isn't a good argument for successful survive.
For the vast majority we no British HG ancestry in British farmers.
What happened to the hunter gatherers who lived throughout Britain?
@Rob,
When looking at these cases from Neolithic Europe, there's really only a handful of stories which can explain them. All of them tell a story of replacement.
One story, you can say is the hunter gatherers weren't violently replaced. They willingly joined the farmer society and became farmers. But because they had lower population size they made a small genetic impact.
This is a story of replacement. It still tells a story of hunter gatherer society being replaced by farmer society. It is by consent but it is still replacement.
This is especially relevant for the cases in Early Neolithic France and Serbia where we see true hybrid hunter gatherer/farmer populations. These hyprid societies were still farmer societies. The farmer culture replaced the HG culture. The HGs assimilated into the farmer society.
And after the initial admixture event, Hg ancestry was diluted. In the end both France & Serbia became lands of farmers with mostly Anatolian ancestry. In the end Anatolian ancestry mostly replaced Hg ancestry and replaced Hg culture. This is a story of replacement.
Also, low HG population size is not an argument against replacement. A story of a smaller HG population, assimilated into a bigger incoming farmer population, is still a story of replacement.
The second story you can tell is, HGs simply lived segregated from farmers. They were not absorbed by the farmers, they were not replaced. They lived on their own as if nothing had changed. This is a story of survival.
This is what you are arguing happened with LBK. It is undeniable this happened in Neolithic Europe. Yet, this can't be said to be the overarching story for Neolithic Europe. Eventually it is impossible to give this explanation.
Because, eventually there were no hunter gatherers in Europe. At some point, they stopped living on their own. Eventually all hunter gatherers joined a farmer or Kurgan pastoralist society. In the end, the farmer/pastorlists ancestry became predominate everywhere.
And very much so in Neolithic Europe. I just gave good examples. Hunter gatherer ancestry never reaches above 15% in Hungary farmers all the way to 3000 BC. The same so far is the case for Italy. If hunter gatherers survived segregated in Italy till 3000 BC, they must have been replaced by the incoming IEs during the Bronze age, because we see no traces of those last Italian HGs in Etruscans. What happened to them?
The same can be said for HGs in Britain. If no British Hg ancestry in British farmers is because British HGs lived segregated from British farmers, where is the resurgence of HG ancestry in bronze age Britain? What happened to the British HGs? Eventually, they had to have been truly replaced.
Rob, I have to say I enjoy these conversations. Because we are cutting at what really happened on the ground.
You have changed my views on Neolithic Europe.
In reality, I believe in a mixed story for Neolithic Europe. But, I think replacement, one way or another, of HGs is mainly what happened.
Ancient DNA proves pockets of hunter gatherers survived in some areas. But I want to know where is the archaeological documentation of this so we can understand it.
The twofold fact that farmer lifestyle in the end replaced hunter gatherer lifestyle in most regions, the fact farmers are more predominate in archeology, the fact Anatolian ancestry was predominate in all farmers, in the end I think makes a story of replacement undeniable.
Yet, what type of replacement? The fact, as you point out, HG paternal lines predominate in western Europe suggests HG males weren't slaves, poor chaps that farmers graciously let survive. But, the fact those HG males became farmers, the fact the paternal clans they founded had mostly Anatolian ancestry, to me in the end tells a story of HG replacement.
@Rob,
I'm going to make a video at some point about hunter gatherer survival/replacement, and would like you to contribute to the video.
Here is my email....
thepopulationgeneticschannel@gmail.com
If you could send a link to the work of that Polish archaeologists saying he has evidence of hunter gatherers surviving alongside LBK, that'd be great.
Like I said, I understand hunter gatherers survived segregated from LBK to an extent. I said so in my Bohemia video. I just don't know to what extent they did. I doubt they did very much.
If I acquire enough relevant studies like this from you, by the time I make my video it will be very informed.
@David - sorry, should have said para-pre-Balto-Slavic or something like that I guess.
Re: Germanic, there are shared linguistic features between Balto-Slavic and Germanic, at least according to wiki.
"Languages in which the instrumental, dative, and ablative plurals, as well as certain singulars and duals, exhibit endings beginning in -m-, rather than the usual *-bh-."
I'm a bit puzzled by where this contact occurred though? Shouldn't proto-Germanic and proto-Balto-Slavic have been separated by a pretty big stretch of the Baltic Sea?
@Davidski, it does seem like it's kind of still there with the "Hungarian Scythian" samples, although 500 BCE. Though these are actually dated earlier than the Hungary IA La Tene samples. So I guess the period in Pannonia between those and the Avar samples is a unknown territory. Maybe they were supplanted by the La Tene population or maybe a genome like that persisted.
As one way of sort of looking at the question, I looked for any ancient samples whose population mean was closer than 0.05 distance on scaled G25 from either Croatian, French, Lithuanian, Finnish, Irish. (That kind of actually does exclude any core population I guess but would concentrate around them). Then PCAd all of them together.
Quick plot of that: https://imgur.com/a/WWpVx1f
I don't think there are a lot of insights from that - most of the populations that plot towards the obviously Balto-Slavic area are from the Viking Age or medieval. But I did find it interesting that the Fuzesabony Middle Bronze Age samples from 1900 BCE also had a fair amount of shift in this direction. They do show up as having some shift toward Balto-Slavic cline on Vahaduo's PCA.
As another side note but related to the question, they way I'd currently label all the HUN_La_Tene based on Global25 affinities is as follows:
HUN_IA_La_Tene_Main: I20752, I20774, I18220, I18110, I18488, I18526, I18527, I18528, I18529, I18530, I18531, I18834, I18838, I18839, I18840, I25508, I25510, I25512, I25516, I25518, I25519, I25522, I4996, I18489, I18491, I18492, I18493. Big cluster with a centre relatively close to France, among moderns, but with some more shift to Eastern Europe. Including ancients, closest to Slovenia and Slovakia IA and prescythian HUN_IA. It looks like this cluster may represent SE European / Hungarian continuity judging by affinity to other ancients?
HUN_IA_La_Tene_oNorth: I18182, I18183, I18226, I25509, I25524. This is a set of samples that overlaps in Vahaduo's North Europe PCA with the Balto-Slavic cline from present day Hungary to Lithuania. Their average is pretty close to Ukrainians or Poles, with a distance in the realm of 0.018-0.02, and the distance to Krakauer_Berg_MA is the same as well.
That's a bit higher than modern Polish and Ukrainian get with each other, but not really high and still pretty close. Dates from 210-150 BCE. If I do modern averages and Top 5 ancient averages, this grouping tends to come in 3rd or 4th for Slavic populations, after Avar Szolad and Krakauer Berg MA.
This cluster does show that it's possible to "make" a cluster that's pretty close to present day Slavic population means from samples dated pre 0BCE from Hungary. But it is based on selecting a set of outliers, so doesn't prove that a similar population was living near Hungary.
HUN_IA_La_Tene_oSouth: I4998, I18832. Seems to overlap best among moderns with present day Tuscany / Corsica. Dates from 250-210BCE. This is the sample already labelled with oEast in the normal set, plus I4998, but to me it seems more Southern than Eastern. Including ancients, they're closest to HRV_CA_Steppe sample POP39 from 2600 BCE. So again it's plausible this represents continuity but with more links to South?
HUN_IA_La_Tene_oWest: I18837. Looks to be close to Germans, Dutch. Among ancients close to the Knoviz_LBA and La_Tene from Czech Republic. Might represent some link between La_Tene horizon back to the North and West?
The main one that seems worth separating out at the moment is the group I've labelled "North", as the other groups are closer to being a big sprawling cline, and these are just a bit outlying. One other sample who I'm not sure is LaTene or not is I25525 and he seems similar to the preceding Fuzesabony_MBA samples.
Visually on PCA and list of closest 15 to the averages: https://imgur.com/a/kxgNHym)
Pastebin: Relabelled HUN_IA_La_Tene samples and their averages: https://pastebin.com/gcuNx0Z9
I've been intrigued by these HG-rich Carpathian basin individuals for a while.
Perhaps they originated from a mix of 'typical' (pre-proto-/para-)Balto-Slavs and some WHG-rich 'farmer' group? Seems like the Balto-Slavic ancestry was female-mediated in that case, considering the Y-DNA. I3528 is dated to 2559-2301 calBCE, by which time there were still people without steppe ancestry around in that area, as demonstrated by e.g. I2741.
In any case, the Transdanubian Encrusted Pottery culture samples (HRV_Jag_MBA) descend mostly from this population. Hajdu et al. 2016 says: "According to the latest archaeological observations, the origins of the Encrusted Pottery culture can be directly related to the Corded Ware culture that had eastern origins²⁸." Implying that these people were CWC-derived in some capacity.
As another side note but related to the question, they way I'd currently label all the HUN_La_Tene based on Global25 affinities is as follows:
HUN_IA_La_Tene_Main: I20752, I20774, I18220, I18110, I18488, I18526, I18527, I18528, I18529, I18530, I18531, I18834, I18838, I18839, I18840, I25508, I25510, I25512, I25516, I25518, I25519, I25522, I4996, I18489, I18491, I18492, I18493. Big cluster with a centre relatively close to France, among moderns, but with some more shift to Eastern Europe. Including ancients, closest to Slovenia and Slovakia IA and prescythian HUN_IA. It looks like this cluster may represent SE European / Hungarian continuity judging by affinity to other ancients?
HUN_IA_La_Tene_oNorth: I18182, I18183, I18226, I25509, I25524. This is a set of samples that overlaps in Vahaduo's North Europe PCA with the Balto-Slavic cline from present day Hungary to Lithuania. Their average is pretty close to Ukrainians or Poles, with a distance in the realm of 0.018-0.02, and the distance to Krakauer_Berg_MA is the same as well.
That's a bit higher than modern Polish and Ukrainian get with each other, but not really high and still pretty close. Dates from 210-150 BCE. If I do modern averages and Top 5 ancient averages, this grouping tends to come in 3rd or 4th for Slavic populations, after Avar Szolad and Krakauer Berg MA.
This cluster does show that it's possible to "make" a cluster that's pretty close to present day Slavic population means from samples dated pre 0BCE from Hungary. But it is based on selecting a set of outliers, so doesn't prove that a similar population was living near Hungary.
HUN_IA_La_Tene_oSouth: I4998, I18832. Seems to overlap best among moderns with present day Tuscany / Corsica. Dates from 250-210BCE. This is the sample already labelled with oEast in the normal set, plus I4998, but to me it seems more Southern than Eastern. Including ancients, they're closest to HRV_CA_Steppe sample POP39 from 2600 BCE. So again it's plausible this represents continuity but with more links to South?
HUN_IA_La_Tene_oWest: I18837. Looks to be close to Germans, Dutch. Among ancients close to the Knoviz_LBA and La_Tene from Czech Republic. Might represent some link between La_Tene horizon back to the North and West?
The main one that seems worth separating out at the moment is the group I've labelled "North", as the other groups are closer to being a big sprawling cline, and these are just a bit outlying. One other sample who I'm not sure is LaTene or not is I25525 and he seems similar to the preceding Fuzesabony_MBA samples.
Visually on PCA and list of closest 15 to the averages: https://imgur.com/a/kxgNHym)
Pastebin: Relabelled HUN_IA_La_Tene samples and their averages: https://pastebin.com/gcuNx0Z9
@ Sam
Yes, not all HGs became Farmers. Some were indeed replaced, for example in Iberia, at least their males. Because Iberian HGs had Y-DNA C1a, I1; but post-Cardial Iberian Neolithics had a lot of I2a instead, perhaps from France. So agree, that only some HG groups were able to ride the farming wave, especially those in France; but also the southern Baltic/ south Scandinavia/ northern Europe (future TRB); some around the lower Danube (Iron Gates, Dudesti/ Malaki Presvlajevts), these hypothetical pre-balto-Slavic HGs, so forth.
The first key point is this - the extent of their contribution is underestimated using % admixture estimates with G25, qpADm, or whatever other admixture method.
But the second key point is I don't think LBK or Impressa did that replacing. The replacing was done by HG-Farmer creoles, during the 'second wave', like those which moved from France to England. The third aspect is some late HGs (e.g. British, Italian, south Balkan) were apparently already dwindling away before farmers even got there. So the story is not simply of ''Farmers replacing HGs'', then HG 'bouncing back'.
@Genos Historia “ Like I said, I understand hunter gatherers survived segregated from LBK to an extent. I said so in my Bohemia video. I just don't know to what extent”
What happen to Tollensee Valley HG rich pops?
@Davidski “Balto-Slavic drift appears to have been initially derived from the Eastern European populations that Corded Ware mixed with, but of course it became more intense and more characteristic of Balto-Slavic speakers over time.”
Do you mean that the substrate in Balto-Slavic owes to GAC and TRB, mostly? I can’t see any other group - Narva HG, Combed Ceramic, Tripolye or Baden contributing much.
@Davidski @Andrzejewski
"Balto-Slavic drift appears to have been initially derived from the Eastern European populations that Corded Ware mixed with, but of course it became more intense and more characteristic of Balto-Slavic speakers over time."
I'm assuming this implies that the 'Balto-Slavic drift' in G25 is actually a combination of several things that set northeastern Europeans apart from northwestern Europeans, including slight differences in the sources of 'farmer' ancestry as well as drift accrued by a late CWC population?
Until recently, I believed that this signal was overwhelmingly the result of intense genetic drift, probably due to a rapid expansion of a late Corded Ware group with a low effective pop size.
But now it's clear that it existed in some other population in Eastern Europe, and it became more specific to Balto-Slavs in more recent times.
The Gerber et al. quote that bce posted above is very interesting in this context:
Our results indicate a population turnover from first to second horizon around 2200 BCE with a group of high (42%) Hunter-Gatherer component previously unknown from the period, which during the transition from the second to third horizon subsequently blended into the prevailing genetic pool of surrounding populations through female biased admixture.
The origin of this particular Hunter-Gatherer ancestry likely comes from unsampled regions of Eastern Europe, and contributed to various populations to some extent, while becoming most prominent in the Baltic region from the middle of the second millennium BCE.
Rob said...@ Sam
"Yes, not all HGs became Farmers. Some were indeed replaced..."
It might be useful to note that sometimes the HG communities adopted a different kinds of food production strategies than the first farmers. LBK and its daughter cultures were to some extent locked into the loess farming strategy. This meant that they didn't have the "technology" to farm in other environments. Here's a quote from Karsten Wentink's Neolithic Depositions in the Northern Netherlands:
"The LBK agricultural techniques were adapted to a specific
ecological zone, namely the loess soils. This technique could not just be extrapolated to the sandy soils of the north, for their physical attributes necessitated a different technique of agricultural food-production. The LBK farmers possessed the “formula”, to quote Bakker (1982), for subsistence on the loamy loess soils, but not for the sandy soils of the north.
The TRB culture restricted themselves to the sandy “islands”
of the north not crossing the loess boundary (Bakker 1982,
88) (There was a minor exception with TRB in Poland,)... this could explain why the rapid spread of agriculture slowed when the borders of the loess expansion of central Europe had been reached."
The TRB communities -- featuring a fair amount of HG ancestry -- did not "replace" the LBK successor cultures as much as it offered a different subsistence strategy in different areas that may also have been more adaptable to environmental change.
Of course, everybody is descended from hunter-gatherers, including EEFs. But the LBK successors did not revert back much to hunting or seaside gathering. TRB, CWC and Bell Beaker did. Fot example, the mass of the meat identified in the German Corded Ware settlement is from wild animals not domestic cattle.
Nobody stayed pure blooded through these processes. But the northern WHGs did manage to build their own approach to adopting farming while retaining some of the advantages of a modified hunter gather strategy.
Hope this helps
@Davidski, funnily enough, I've been shifting a bit to the view that the signature we find today (and well represented by Baltic LBA?) is a composite of drift in that HG group (the view as arza has championed and I also thought likely and Gerber's research also looks to offer some support for), which may have happened at virtually anytime during the Mesolithic or Neolithic, along *with* the drift effect within some post-early CWC Copper Age-Bronze Age population like you suggested.
Basically because 1) these early HG rich samples do seem to show some of the shift, but not all of it, and 2) if all the drift was in the one part of the ancestry that is still relatively small, it would have to be enormous to have the effect it does, and that seems probably too enormous a drift effect in a single population to be likely. So I think it's from a HG-ancestry group but supplemented by another effect. I think those patterns have become more apparent as more samples came out. (And probably more samples will continue to affect our view).
@Davidski
"But now it's clear that it existed in some other population in Eastern Europe, and it became more specific to Balto-Slavs in more recent times."
Okay. I'm still a little confused though; when you say it became more specific to and intense in Balto-Slavs, doesn't that imply that they acquired more drift after the admixture?
I suppose this means that these Carpathian basin individuals needn't have any specific relation to Balto-Slavs either, but rather just have admixture from the same pre-IE group.
Yes, basically what Matt said as well.
Would it make sense to look for more samples got from sites around the Vistula river system?
That would seem to be the obvious connection between at least the Baltic coast and Czech areas. Or is that naive?
I had a look at the varying shares of steppe ancestry and WHG around central Europe, from eastern France to Slovakia, from the EBA to the LIA.
I obtained these figures on Vahaduo using Yamnaya Samara, WHG and Barcin_N as the sources:
Grand Est EBA = 42.2 Steppe, 17.6 WHG
Switzerland EBA = 41.6 Steppe, 15 WHG
Singen EBA = 40.4 Steppe, 16 WHG
Lech EBA = 46.8 Steppe, 13.6 WHG
Hungary Mako EBA = 27 Steppe, 28.4 WHG
Slovakia EBA = 59.8 Steppe, 14.8 WHG
The above results for the EBA show similar figures in Grand Est, Switzerland and Singen, slightly more Steppe on the Lech, the very pecculiar low steppe/high WHG ratio in the Mako culture of Hungary, and finally a rather high steppe share in Slovakia.
Czech Tumulus MBA = 38.8 Steppe, 16.2 WHG
Hungary Fuzesabony MBA = 50.8 - 51 Steppe, 17.8 WHG
Hungary Vatya MBA = 29 Steppe, 13.6 WHG
The above results for the MBA show pretty normal figures in the Czech Tumulus culture and strongly diverging results for the cultures in Hungary.
Grand Est LBA = 29.6 Steppe, 17.6 WHG
Halberstadt LBA = 52 Steppe, 14 WHG
Czech LBA Knoviz = 41.6 Steppe, 16.2 WHG
Hungary LBA = 34.6 Steppe, 18 WHG
Tollense BA = 38.2 Steppe, 21.8 WHG
Hungary LBA Halva = 39.4 Steppe, 20.4 WHG
The above LBA results show a reduction of Steppe in Grand Est, probably due to southern admixture; Halberstadt on the other hand has rather high steppe; the Czech Knoviz culture has similar figures as Czech Tumulus; and then there are Tollense BA and Hungary Halva resembling each other in their unusual composition with moderate Steppe and high WHG - I guess these must in part be derived from the earlier Mako culture: If you take Mako and mix it with something that has higher steppe and lower WHG, that is, with a more "normal" population, then you get the Halva and Tollense figures.
Now the EIA in Hungary obviously brought new genetic input from the Balkan-lower Danubian area, from populations with only modest Steppe and very low WHG:
Hungary EIA = 27.2 Steppe, 6.4 WHG
Hungary IA Syrmian Srem = 34.4 Steppe, 4.4 WHG
Whereas the Vekerzug people of Slovakia look pretty average; however towards the LIA there is input of a population with more steppe and less WHG:
Slovakia IA Vekerzug = 38 Steppe, 11.6 WHG
Slovakia LIA = 45.4 Steppe, 5.6 WHG
I'm now coming to the Hallstatt and La Tene samples. I noted the closest modern population behind each ancient sample; the belt of broadly French-like people from France to Slovakia looks impressive:
Alsace Hallstatt: French_Nord (39.6 Steppe, 14.2 WHG)
Czechia Hallstatt = French_Nord (39.2 Steppe, 16 WHG)
Hungary Hallstatt = French_Auvergne (33.6 Steppe, 14.2 WHG)
Belgae (Hauts-de-France) = French_Nord
Champagne La Tene = French_Brittany
Alsace La Tene = French_Nord (42.8 Steppe, 14.8 WHG)
Occitanie La Tene = French_Occitanie
Southeastern France La Tene = French_South
Czech La Tene = French_Brittany (42.6 Steppe, 15 WHG)
CHE_IA = Spanish_Castello (27.2 Steppe, 13.4 WHG)
Austria La Tene = French_Auvergne (33.6 Steppe, 12 WHG)
Hungary La Tene = French_Auvergne (35 Steppe, 13 WHG)
Slovakia La Tene = French_Occitanie (37.4 Steppe, 14 WHG)
Croatia La Tene = Italian_Veneto
However, it's strinking how Austria, Hungary and Slovakia resemble more southerly French people than the samples from Czechia and eastern/northern France. This is reflected in the lower shares of Steppe ancestry in the former. I could imagine that they are a mix of the Hungary LBA samples and the Hungary EIA/IA samples, but haven't tried it yet.
Well, I made a ghost, just for fun:
Target: HUN_EIA_Prescythian_Mezocsat_o1:I18241
Distance: 0.0036% / 0.00003616
48.6 POL_Globular_Amphora
29.0 HUN_EIA_I18241_HG_Ghost
22.4 Corded_Ware_early
For what it is, this works surprisingly well for modelling Baltic LBA:
Target: Baltic_LVA_BA
Distance: 3.6021% / 0.03602097
67.2 Corded_Ware_CZE
32.8 HUN_EIA_I18241_HG_Ghost
Not sure how close this is to the truth, but there it is.
Davidski, a tangential question: at what point, if any, are there distinct "Slavic" and "Baltic" autosomal signals?
In terms of genetics, Slavs are basically Balts with foreign admixtures; Germanic in the west, Balkan in the south, Uralic in the east.
David, when we take this point of view, the East Slavs will also be Balts with a Germanic admixture. Even modern Balts will be Bronze Age Balts with a Germanic admixture.
This issue was once discussed by Arza on his blog:
https://slavicorigins.blogspot.com/2021/05/surplus-eef-ancestry-in-modern-day-slavs.html
It seems that you're confusing Balts with the peoples of the Bronze Age East Baltic.
They're not the same populations.
I don't think there has been no genetic continuity in Pribaltica since the Bronze Age. There is no indication of that.
Mittnik:
"Modern Eastern Baltic populations cluster with Baltic BA on the PCA plot and exhibit among all modern populations the highest shared genetic drift with ancient Baltic populations (Supplementary Fig. 2), but show substantial differences to samples from the Bronze Age. The statistic D(Lithuanian, Baltic BA; X, Mbuti) reveals significantly positive results for many modern Near Eastern and Southern European populations (Supplementary Fig. 6a). Limited gene-flow from more south-western neighbouring regions after the Bronze Age is sufficient to explain this pattern, as nearly all modern populations besides Estonians, especially for Central and Western Europe, have a higher amount of farmer ancestry than Lithuanians."
There's usually some continuity when there are population replacements, but it's obvious that Balts are a different population.
Mittnik's explanation is just based on autosomal affinity and simple parsimony, and totally excludes uniparental markers, archeology and linguistics.
Honestly, anyone who claims that East Baltic speakers have been in the East Baltic since the Bronze Age needs to get another hobby.
"Honestly, anyone who claims that East Baltic speakers have been in the East Baltic since the Bronze Age needs to get another hobby."
An example would be the Gauja river in the East Baltic, which has a centum IE etymology.
@Davidski
„In terms of genetics, Slavs are basically Balts with foreign admixtures; Germanic in the west, Balkan in the south, Uralic in the east.”
And where is the Iranic admixture?
I think Balts are Slavs with some HG/Uralic admixture. Germanics are a mix of Celts, Scandinavians and Slavs, Balkan are a mix of Slavs and Greek/Italic/Celtic etc.
Ethnicity is a much more complicated problem than you think.
Try to define Slavs, Balts, Germanics, when and where they originated, show their origin on PCA and on the geographical map, tell us which cultures they originated from starting from the beginning i.e. HG, EEF, steppe. Without it we have only pointless speculation. I don’t know what you mean by Slavs, Balts, Germanics.
To understand ethnic origin you have to look not only at genetics, but also at language, culture, religion etc.
Yes, culture, religion are important. Where did Sintashta get the cremation from? In the beginning they didn’t cremate at all. In the end it was very popular and the cremation was the mark of the arrival of Aryans in India.
I think they got cremation from the people from which they got the words related to the fire and solar cult i.e. Slavs. Slavs were the last IE people in Europe who cremated and I think, and there is a lot of evidence for this, they were also the first IE people in Europe who cremated, worshiped fire and the sun, believed in soul and body duality, linking the soul with the sun. It is well preserved in Slavic folklore.
Before Slavs cremation was practiced by some HG and HG-EEF mixed drifted groups around Carpathian mountains which are difficult to describe genetically because they cremated. Mixing of R1a CWC Proto-Slavic people with this HG-EEF drifted people around Carpathian mountains and the birth of new religion took place after 2000 BC and then that solar religion and cremation started to spread, reached Sintashta, Greece etc. Traces of that religion are best preserved in Slavic folklore and language. It all makes sense, you will see. Other theories, including yours, do not make sense, you will see it the moment you will try to formulate them properly.
@ Davidski
''Mittnik's explanation is just based on autosomal affinity and simple parsimony, and totally excludes uniparental markers, archeology and linguistics.
Honestly, anyone who claims that East Baltic speakers have been in the East Baltic since the Bronze Age needs to get another hobby.''
Agree that all these lines of evidence should be used.
But do you mean speakers of Eastern Baltic, specifically, are new to the region (vs West Baltic), or all all Baltic speakers in the East Baltic ?
@Rob
I don't know yet, because I need to see some ancient DNA from West Balts to make up my mind.
But as things stand, I'm very skeptical that those Baltic BA genomes represent any sort of Baltic speakers.
The Mittnik paper was a fairly typical Max Planck effort in that it presented the ancient DNA evidence in a total vacuum, and to be honest, it didn't make any sense linguistically.
@EastPole
If you can give me a plausible explanation for all of that U106 and I1 in Slavs, especially West Slavs, then let me know.
But like I say, it has to be plausible, not something like "U106 was in Corded Ware, so it's a Slavic marker since the Late Neolithic".
In regards to Iranian speakers, it's interesting that they lack that specific Baltic affinity and Sintashta lacked it as well. Hmmm...
@Davidski
"Honestly, anyone who claims that East Baltic speakers have been in the East Baltic since the Bronze Age needs to get another hobby."
What about those who legit still think Fatyanovo was Baltic lol?
"But as things stand, I'm very skeptical that those Baltic BA genomes represent any sort of Baltic speakers."
Even with the most superficial analysis, it's evident that modern Balts and Baltic_BA genetically represent entirely different populations.
Yeah, Balts assimilated some Baltic_BA substrate, but there was a rather wholesale population replacement by and large.
https://docs.google.com/spreadsheets/d/17Aa9j9_sDQVaqsdAuH9DPf_EI4n5oqmrF2zAKSaQa4E/edit?usp=sharing
The earliest "Baltic_BA-like" sample is Spiginas2, dated 2132-1749 BCE.
Way too early for any Balto-Slavic split.
Looking at both the autosomal and Y-DNA (CTS1211), these guys are related to Balto-Slavs, but it's a pretty distant relationship.
At the latest, whatever Baltic_BA spoke would have probably split from "Para-Proto-Balto-Slavic" around 1800-2000 BC.
@Davidski
“If you can give me a plausible explanation for all of that U106 and I1 in Slavs, especially West Slavs, then let me know.
But like I say, it has to be plausible, not something like "U106 was in Corded Ware, so it's a Slavic marker since the Late Neolithic".
Germanic homeland and the source of U106 was around lower Rhine i.e. Holland. After 1100 AD there was massive migration from that area to conquered Slavic lands. Here is the map of eastward migration of German peasants after 1100 AD:
https://i.imgur.com/9ohrFJT.png
Then with Prussians it came to Poland, but Prussians were 50% Germans and 50% Slavs. What is the percentage of U106 in Poland? If it is 5% then what is the Germanic autosomal contribution? 2.5%? Are you sure it has changed the genetic map of West Slavs?
I1 is a different story, it is probably much older and probably from the North. Nothing to do with Germans.
“In regards to Iranian speakers, it's interesting that they lack that specific Baltic affinity and Sintashta lacked it as well. Hmmm...”
I don’t know what you are talking about. I was joking about Iranic admixture in Slavs. We know there is none.
The link of Slavs with Sintashta is cultural, not genetic, although the ancestors of Sintashta and Slavs came from the same Indo-Slavic R1a CWC tribe.
But after 2000 BC there were still strong cultural links between Urals and Slavic lands:
https://postimg.cc/DWPt1kWZ
And they got cremation, and the cult and vocabulary that goes with it from us. Because it was first around the Carpathians.
"But as things stand, I'm very skeptical that those Baltic BA genomes represent any sort of Baltic speakers."
I think I've seen this idea thrown around before, anyway some of them could be the source of the 'North Baltic' substrate in Finnic, as it is called by Petri Kallio (Kallio 2008, On the 'Early Baltic' Loanwords in Common Finnic).
Excerpts:
The "Early Baltic" loanwords must therefore have been borrowed at the time when the Finnic and Saamic branches were still linguistically uniform but already areally diverged, which according to the traditional Fenn(o-Ugr)icist chronology would mean the second and first millennia BC (cf. Kallio 1998).
we may arrive at the conclusion that the "Early Baltic" loanwords in Finnic consist of the two chronologically different layers: while the earlier stratum corresponds to Kortlandt's Proto-Balto-Slavic stage (1989: 43-46), the later one represents an otherwise unattested Balto-Slavic dialect
"Where did Sintashta get the cremation from? In the beginning they didn’t cremate at all. In the end it was very popular and the cremation was the mark of the arrival of Aryans in India."
Sintashta did not cremate. Fedorovo/late andronovo is when you start seeing signs of it. And fedorovo is 100% R1a Z2121, no relation to south asia.
Figure out where fedorovo got the practice of cremation from, that will be more worthwhile.
@Simon_W
As impressive as Matt previous posts on French-like IA Central European populations ...
Don't want to bother anyone, but what does it mean to you and what's your take on Modern French ancestry?
@EastPole, Arza
How do we know that these and other samples were reliably tested?
I asked Davidski about it, but he did not answer me, see:
Do you have any idea what proportion of the samples have already been retested and the results have been confirmed to be correct and reliable?
January 23, 2022 at 4:07 AM
https://eurogenes.blogspot.com/2022/01/mistaken-identity.html?showComment=1642939658386#c1358422007383767397
Since nothing is absolutely certain about these or other similar genetic data / samples, logically, this whole discussion and others like this one are rather pointless… :-(
@SKRiBHa
Do you have any idea what proportion of the samples have already been retested and the results have been confirmed to be correct and reliable?
Is there any reason why you can't check this yourself. Are you in some way mentally retarded?
@EastPole
U106 is just one obvious marker (and it doesn't reach 100% in any population, so 5% is actually quite a lot).
There are a number Y-chromosome and mtDNA markers that we can look at to study the overall male and female admixture.
@Erik Andersson
North Baltic is a very hypothetical concept, and so are those inferences of an ancient linguistic relationship between Balto-Slavs and Uralics.
Baltic BA is an extinct population. They were almost totally replaced by modern Balts.
@ Dave
But some of the Baltic LBA samples, e.g. Turlojiske date to 1000 BCE. Where are we getting the idea of complete replacement ? Modern Balts are that, + extra EEF.
So Valter Lang's theory that the Brushed pottery groups which appear there could be early Baltic speakers. Another, more radical idea Ive read is Balts are from a Milograd culture migration c, 500 BC
@Davidski
O! Thank you for your comprehensive and polite reply. You could have just written that you did not know… Do not need to be so tense. Cheers!
@SKRiBHa
I do know, but considering that you've been interested in this for years, I'm totally baffled why you need my help.
@Rob
I didn't claim that there was a complete replacement.
But modern Balts look like a different population, even compared to the Turlojiske samples.
For instance, modern Balts have a lot of N1c, while it's totally missing in Baltic BA.
David, are there any y-dna haplogroups associated with Baltic BA?
@ Dave
The VL29 in Southeast Baltic, seems like a founder effect from Fenoscandia, right ?
@Davidski,
"Honestly, anyone who claims that East Baltic speakers have been in the East Baltic since the Bronze Age needs to get another hobby."
I disagree with this. You can't expect everyone to know the details of Balto Slavic languages to say there's no way Baltic BA was Baltic speaking.
Because, genetically, Baltic BA are really similar to modern Balts. They are like uber-Balts, they are a more extreme version of what modern Balts are.
Which isn't that surprising. You expect there to be intermarriage with neighbors after 3,000 years to cause southern shift in modern Balts.
I'm not saying Baltic BA was Baltic speaking. But it isn't crazy for someone to think so.
Treating Baltic BA as alien to modern Balts, is like treating British Bell beaker as alien to modern Scottish.
You guys are going overboard.
Btw, I tried to use DATES to estimate when the Narva_HG/GAC and later Steppe admixture events happened.
I got overlapping dates for both of the samples David mentioned in the main post.
Estimating the date of admixture for Globular Amphora and the Narva_HG-like ancestries:
Hungary_EBA_Mako
dates_expfit version: 200
step (Morgans) :: 0.001000
fitting 1 exponentials + affine
after initialization: 0.000002 0.947
gslsetup called
gslans: 20 0.000002
0.947
error sd: 0.001563
halflife: 12.738
mean (generations): 54.418
0.004100854 0.000331889
##end of run
jmean: 0.000 std. err: 46.657 jmean (years): 0.000 std. err(years): 1353.053
Hungary_EIA_Prescythian_Mezocsat_o1
dates_expfit version: 200
step (Morgans) :: 0.001000
fitting 1 exponentials + affine
after initialization: 0.000004 0.900
gslsetup called
gslans: 19 0.000004
0.900
error sd: 0.001898
halflife: 6.589
mean (generations): 105.195
0.007682831 -0.000013079
##end of run
jmean: 0.000 std. err: 94.930 jmean (years): 0.000 std. err(years): 2752.970
Date of admixture of GAC/Narva_HG for HUN_EIA_Prescythian_Mezocsat_o1_I18241:
775 BCE (mean of 900-650 BCE) + 2752 = 3525 BCE
Date of admixture of GAC/Narva_HG for HUN_Mako_EBA_o_I1502:
2084 BCE (mean of cal carbon date) + 1353 = 3437 BCE
Now testing for the date of Steppe admixture with Steppe_EMBA (Yamnaya_Samara, Yamnaya_Ukraine, Yamnaya_Caucasus) and Poland_GlobularAmphora as admixing pops.
Hungary_EBA_Mako
dates_expfit version: 200
step (Morgans) :: 0.001000
fitting 1 exponentials + affine
after initialization: 0.000001 0.968
gslsetup called
gslans: 21 0.000001
0.991
error sd: 0.001209
halflife: 75.439
mean (generations): 9.188
0.000844166 -0.000024700
##end of run
jmean: 0.000 std. err: 4.489 jmean (years): 0.000 std. err(years): 130.181
https://pastebin.com/GvDAMJhH
Hungary_EIA_Prescythian_Mezocsat_o1
dates_expfit version: 200
step (Morgans) :: 0.001000
fitting 1 exponentials + affine
after initialization: 0.000002 0.952
gslsetup called
gslans: 20 0.000002
0.952
error sd: 0.001509
halflife: 14.115
mean (generations): 49.109
0.001985311 -0.000060488
##end of run
jmean: 0.000 std. err: 41.728 jmean (years): 0.000 std. err(years): 1210.112
https://pastebin.com/69mzD8XY
Date of admixture of GAC/Steppe for HUN_EIA_Prescythian_Mezocsat_o1_I18241:
775 BCE (mean of 900-650 BCE) + 1210 = 1985 BCE
Date of admixture of GAC/Steppe for HUN_Mako_EBA_o_I1502:
2084 BCE (mean of cal carbon date) + 130 = 2214 BCE
@Rob
I don't know where the Baltic N lineages are from, but they became common in the East Baltic at about the same time as the overall genetic structure shift there.
@Genos Historia
Three important points:
- genetic replacements were often by geographically proximate and thus genetically similar populations, so they're hard to detect in ancient DNA
- modern European languages represent a small portion of the linguistic diversity that existed in Europe during the Bronze Age
- everything (autosomal DNA, Y-DNA, linguistics and archeology) points to a significant population replacement in the East Baltic during the Iron Age
@Davidski,
Most definitely. I'm just saying it takes someone who really understands this topic to know that. They aren't necessarally stupid if they think otherwise.
I suppose I took your guys shit talking personal, because I couldn't help but at one time consider Baltic BA to be the best proxy for proto-Balto Slavs. There's a lot of reasons to think so.
When you deal, with closely related neighboring pops moving around it difficult to know what is going on.
I know very well from my work on my Bohemian video, that most cases of population replacement are done by closely related neighboring populations.
@David/all - the ancient Uruguay results recently published - has anyone taken a here taken a look at them yet?
The associated paper seems to be suggesting a major (26%) pre-Columbian input from a distantly West Eurasian source.
http://secher.bernard.free.fr/blog/index.php?post/2021/12/23/Le-g%C3%A9nome-d-un-ancien-individu-d-Uruguay
https://www.biorxiv.org/content/10.1101/2021.11.11.468260v1
Having a hard time believe this one.
It’s hard to appreciate a population replacement on the archaeological record of the Baltic lands since LBA .
I’ve not seen any genetic evidence for it either . I wouldn’t call shifts from Sambia to the Daugova, and cross frontier admixture a “replacement” event
Ongoing contact with Central Europe and Ukraine can explain the shifts in EEF vs Narva ratios, appearance of iron , cremations etc
Hunter Provan’s phylofinder suggests a FebiScandian origin of Baltic N
It seems to me there was a migration from the east during the Iron Age, bringing at the very least East Baltic languages to the East Baltic.
@Ryan
Genetiker was right ! Solutreans crossed the Atlantic
Not sure what Febi Scandian is, I think I meant FenoScandian
these Uruguayans have no West Eurasian ancestry
fake news
@Romulus - What do you base that on? The article contradicts but I'm not going to take that article as gospel either mind you.
I'd bet on extra ANE before Solutreans too.
@Rob
Baltic N1c clades (around 50% of their Y-dna) all have TMRCAs of 2100-2200 ybp. I think this can't be the result of a gradual ongoing contact. Either a tribe came from somewhere else (for example modern Belarus) and had a massive founder effect on the Baltic coast, or the locals went trough a mass extinciton and repopulation event which changed their y-dna.
btw, do you know of any archeological signs of mass replacement in the east Baltic around 1900 BC? Genetics show CWC being replaced by a Balto-Slavic-related population at this time, but in archaeology all I'm finding is something about possible contact with Trzciniec and Lusatian. Am I misssing something?
Saag:
"The individuals of EstBA, EstIA, IngIA, EstMA, and modern Estonia are quite similar to each other on average, indicating that the relatively high proportion of WHG ancestry in modern Eastern Baltic populations compared to other present-day Europeans traces back to the BA.
Outgroup f3 and D statistics do not reveal apparent differences when comparing EstBA to EstIA, EstIA to IngIA, and EstIA to EstMA (Data S1). These results highlight how uniparental and autosomal data can lead to different demographic inferences—the genetic change between CWC and BA not seen in uniparental lineages is clear in autosomal data and the appearance of chrY hg N in the IA is not matched by a clear shift in autosomal profiles."
@ambron
There's a clear shift in autosomal DNA from the Baltic BA to the MA.
I have no interest in what Saag was able to pick up with her f3 stats.
Here's how this "no clear shift in autosomal profiles" really looks like.
https://1.bp.blogspot.com/-buEHbc2Y1mY/XyuMHqzJHvI/AAAAAAAAJGE/HQnRBtuOHF04iqMCYzNmS4TtAVEp9sRCwCLcBGAsYHQ/s721/G25_East_Baltic_BA-IA_transition_scaled_0LS10.png
This is not a clear change when the distance of a modern Latvian to a Bronze Age Latvian is the same as that of a modern Latvian to another modern Latvian.
The shift in Latvia is the least noticeable, but it's still there.
In Estonia it's enormous. Saag really should've done better there.
@ambron, the rest of her paper says there are autosomal shifts though and talks about the Nganassan related fraction and how "EstBA individuals (are) clustering together with Latvian and Lithuanian BA individuals (while) EstIA, IngIA and EstMA individuals project between BA individuals and modern Estonians (on PCA), partially overlapping with both"? The outgroup f3 stat is only comparing f3(Yoruba,Est_BA,Est_CWC) and is not very sensitive to Nganassan related change.
One indicator of turnover (of some unknown level) over time is the reduction of female EEF bias on the X chromosome, which suggests some form of admixture from other populations that don't have this bias ("We also tested for sex biases by comparing outgroup f3 statistics calculated on autosomal (A) and X chromosomal (X) data. The high X to A ratio of European-early-farmer-related ancestry observed in Estonian CWC decreases over time and disappears by the MA (Figure S2C–F, Data S1).").
@ bce
After CWC, there seems to be a population decline in East Baltic. So the local CWC either moved elsewhere or almost became extinct, esp. in NE Baltic. Apart from some scattered finds here & there, there isn't a huge amount happening until late M2. Then things accelerate after 1,000 BC, with a succession of inflluences incl. eastern/ forest, Finnic, Milograd, Carpathian, Pomoranian, Lausatian influences. So there might have been a few movements here although overall it might manifest as 2 or 3 major ones
Off Topic:
Yersinia pestis genomes reveal plague in Britain 4,000 years ago
http://secher.bernard.free.fr/blog/index.php?post/2022/01/27/La-peste-%C3%A9tait-pr%C3%A9sente-en-Grande-Bretagne-il-y-a-4000-ans
https://www.biorxiv.org/content/10.1101/2022.01.26.477195v1.full.pdf
@Rob
the earliest Baltic_BA-type sample in the east Baltic (Spiginas2) is dated around 1900 bc, they seem to completely replace CWC at this time, and in Latvia they remain genetically unchanged at least until 600 BC, in Lithuania and Estonia they do get slight central European and Finnic influence later.
It looks like during this bronze age gap, the east Baltic was in fact settled by the Baltic_BA population. Maybe it was a very sparse settlement for the first few centuries.
@Davidski
I do know, but considering that you've been interested in this for years, I'm totally baffled why you need my help.
You're kidding me, aren't you?
What is weird about that? I am asking for an opinion of a world-famous expert in genetics, who is (according to yourself) the only infallible source of knowledge in that field. When it comes to ancient genetics, you are an authority to me something like the Delphic oracle, the Pope and the TSUE joined together.
Many times has it been found that important samples were ‘mislabeled’, so I want to find out what is credible and what is not.
I am professionally interested in linguistics, especially in the Polish language.
As far as I know, genes do not transmit languages. People, traditions and cultures do.
This is because I do not deal with genetics, nor analyse any samples myself, etc. All I do is to logically organise the available language, genetic and other data. I look for reliable data, ask the wiser (like you) and compare like meerkat. Simpuls!
…..
I am totally baffled why you knowing the answer, do not want to give it quickly, but you waste your time on such nagging. Could you just stop it, provide this data and some source for it, or will you continue to tease me like a cute girl? What is wrong with you? Does this question about reliable data affect you somehow?
How many percent or how many ancient DNA samples have already been retested so far, so that their results are 'pretty solid stuff' - 1, 5, 10, 15..?
@SKRiBHa
Your question is pointless, so I refuse to spend any more time on it.
Anyone who has any sort of serious interest in ancient DNA knows that some samples are more reliable than others, and naturally this is taken into account by most of us in discussions about these samples.
So, for instance, things like the type of sequencing, depth of coverage and marker overlap with reference datasets are noted and their implications weighed up.
It doesn't take much effort to look up the specifications of ancient DNA samples in papers and at the David Reich Lab website, and check whether they've been resequenced recently, but I won't be doing that for you.
@Davidski, completely off topic for Slavic groups, but I noticed that the Reich lab have added another modern dataset onto their list of datasets for 2022, for a paper currently "in press" - https://reich.hms.harvard.edu/datasets
It looks like a set of present day individuals from Southeast Asia and some are Tibetan-Burman language speakers.
It's possible these are the modern day individuals from this upcoming paper focus on Tibet and Tibeto-Burmans which includes ancients and modern people (https://www.ebi.ac.uk/ena/browser/view/PRJEB41752?show=reads). The paper doesn't look like its out anyway.
If it is not too difficult, would you be able to transfer these to G25?
Particularly some of the samples are Maniq people, who are a "negrito" minority in Thailand, so those could be interesting to see where they sit on Global25.
(On a side note, they provided their .anno file in the genotypes file and I noticed that they included some extra fields around levels of Runs of Homozygosity in samples; total shortRoH and total LongRoh. If that's a new thing for their resources as a whole, where they use Harald Ringbauer's methods to estimate it for sufficiently covered ancient samples, it will be helpful in identifying inbred and outbred ancient individuals.)
@George, re; your OT comment, yeah, it would be interesting to see if they can get any human adna from the massacred individuals, and see if they are atypical in any way or just normal BBC-EBA people from the time (2000 BCE).
(further to my OT comment, actually the paper by Reich Lab I mention upthread where the genotypes are uploaded, is probably this preprint - https://www.biorxiv.org/content/10.1101/2021.01.21.427591v1.full - "Indian genetic heritage in Southeast Asian populations" - Changmai et al)
@All
Since our all-knowing and infallible but capricious host does not want to provide any answer to my simple and very important question, about a reliability of ancient genetic data sampling, so I ask all of you the same:
Do any of you have any knowledge how many percent or how many ancient DNA samples have already been retested so far, so that their results are 'pretty solid stuff' - 1, 5, 10, 15..?
@ Dave
do we think that BOO does not lie on the extant F-U cline ?
BOO is sitting on the Uralic cline or very close to it in all of the main dimensions, so I think it's close enough.
@SKRiBHa - they don't need to be rested to be "pretty solid stuff"
@ bce
“ the earliest Baltic_BA-type sample in the east Baltic (Spiginas2) is dated around 1900 bc, they seem to completely replace CWC at this time, and in Latvia they remain genetically unchanged at least until 600 BC, in Lithuania and Estonia they do get slight central Europea”
Yep but there is still a large gap in sampling, see the table in Mittnick
This means that no one was being buried in the east Baltic maritime between 1750 & 1200
It might be useful to look a subtle differences between the early Baltic BA type group and the later ones.
But it might be their homeland was mostly in SE Lithuania / North Belarus
There's something which doesn't sit right about the idea that FU is associated with S-T. It seems off geographically and chronologically.
There's definitely no close correlation between Seima-Turbino and the expansion of Uralic languages.
But I think that Seima-Turbino was very important to the expansion of Uralic languages, one because Uralic languages were probably spoken by at least some of the people who took part in it, and two because it seems to have encouraged mobility and cultural exchange across North Eurasia.
@Rob
An issue is that a lot of what you see in the STP sites is grounded in traditions from West-Siberia and the Altai-Sayan, and western Uralics tend to have next to no WSHG related ancestry (and Finnics dont need steppe_mlba). It also doesnt match well archaeologically with a significant population movement and migration. I've been real skeptical of associating the STP sites in western Siberia but particularly eastern Europe with a rapid, military expansion of Uralic speaking warriors.
Also some other issues such as demographics and subsistence economies etc.
Western Uralic people also seem to have roughly 1000 bc admix dates for their european and asian ancestries which doesn't match up well either.
I think its still a good context to explain early Indo-Iranian contacts as well as potentially a driver to start migration from east to west but we will see how that develops over time.
I think ST was, as Chernykh said, a “transcultural phenomenon”
If anything , I see it as a diffusion of western metallurgical techniques, stimulating the Altai-Sayan nucleus
That & other issues the Axeman said . Many sites are epithets and ceremonial deposits rather than actual burials
Is there any reason why we can’t entertain a link with Ymyykhakhs, if we for a moment put a side the claims of loanword evidence which allegedly mandates a western FU homeland ?
You can't explain the spread of CTS10760 to Europe in any way other than with ST. Their last connection in Siberia is the Avar group Y16323 deep in Mongolia. CTS10760 decays in 2100 BC already in Europe, leaving no trace anywhere in Siberia. In Russia, a genetic study of the population of the north-western regions (Novgorod and Pskov) was conducted and it turned out that this population has common roots with the Komi and Permians. Most likely, this group penetrated through the Middle Urals as part of ST and subsequently expanded in the European part through the founder effect. I doubt that it was this group that brought the FU language, but the style of textile ceramics apparently belongs to this group. This group apparently originated from some ST site in the European part, for example, Turbino is located near Perm. This group quickly assimilated with the local population of the European part, and in general at some point apparently became a Baltic couple. The real FU passed later is the Z1936. This group left its mark in Siberia in the Sargat culture. Its subclades Y13850 belong to both Magyars and Khanty. Subclades of this group exist in Tatarstan and Bashkiria and throughout the middle belt of the European part of Russia. In the European part, they appeared as part of the Maklasheevo culture. It's hard to say in Siberia yet, I think it's something like Cherkaskul culture, Suzgun culture, Pakhomovo culture. In general, this is the LBA period.
@Davidski:
"I didn't claim that there was a complete replacement. But modern Balts look like a different population, even compared to the Turlojiske samples."
I'm pretty sure that modern Balts (especially Lithuanians, but Latvians too) have some Slavic admixture - also from relatively recent times (from the last several centuries). It is evident for instance when you examine Lithuanian samples from Urnikyte et al. 2019. And even when looking at census data from the 19th century and 20th century makes it obvious - for instance even in Kaunas Uyezd - in the heart of Lithuania - around 35% of the population were various Slavs, as of year 1897.
The genetic shift I'm talking about starts during the Iron Age. Compare...
Baltic_EST_BA,0.1321485,0.1213556,0.0992205,0.111662,0.051671,0.040746,0.0158398,0.0204684,-0.0029042,-0.050607,0.0009094,-0.0182389,0.0359462,0.0400344,-0.0180239,0.0018297,0.0103654,-0.000684,-0.0013952,0.0046272,-0.0024957,-0.0062692,0.0106487,-0.017388,0.002359
Baltic_EST_IA,0.12881,0.1064613,0.093903,0.0930778,0.0434952,0.0322583,0.0097137,0.0131533,0.0019773,-0.036356,0.0054128,-0.0121143,0.0222,0.0148862,-0.0057682,0.0081983,0.010474,-0.0006547,0.003289,0.011193,0.0060727,-0.00237,0.0028553,-0.0118087,-0.0021555
@rob
Judging by the fact that Y6058 decayed 2700 BC, it means that this happened before the formation of ST. The Y16323 subclades, which later became Avar and localized in Mongolia, obviously remained in Siberia. Whether he was related to ST is not known and it does not matter. It is important that its sister subclades CTS10760 are absent in Siberia and are widely represented in Europe, and its subclades VL29 and Y28526 were formed in the European part in 2100 BC. Finding two subclades Y16323 and CTS10760 of the same group at such a distance from each other and without intermediate populations needs to be explained somehow. The most logical explanation would be to move along with the ST groups.
It's a completely different matter with the Z1936 subclades, which split with the CTS10760 also in 2700 BC. But it is found everywhere in Siberia, both in ancient and modern populations. This means that this population walked slowly through Siberia, leaving its subclades along the way. This is completely different from the campaign of an armed group. That is why they reach about the LBA/EIA in the European part.
@ Vladimir
“ You can't explain the spread of CTS10760 to Europe in any way other than with ST. Their last connection in Siberia is the Avar group Y16323 deep in Mongolia. CTS10760 decays in 2100 BC already in Europe, leaving no trace anywhere in Siberia. In Russia, a genetic study of the population of the north-western regions (Novgorod and Pskov) was conducted and it turned out that this population has common roots with the Komi and Permians. Most likely, this group penetrated through the Middle Urals as part of ST and subsequently expanded in the European part through the founder effect. I doubt that it was this group that brought the FU language, but the style of textile ceramics apparently belongs to this group”
- if you mean the N3a3 in BOO, I doubt ST has much to do with it. Rather a possible link with a post-Ymyykhakh expansion to Kola
- Avars do not come from Mongolia. There is a paucity of hg N in Mongolia . The link with Ruruan is invented by modern scholars just to fit a jigsaw puzzle
@rob The BOO group was L1026* extinct without leaving a trace in modern populations. I also doubt that the Avars are from Mongolia, but a new article about the Huns, Avars and Magyars claims that the Avars are from Mongolia. Personally, I think that the avars of the Y16323 group were localized in the area where kra001 was detected.
@ Vladimir
Yeah I know the article well, I helped out in it. I disagree with that particular aspect
Do we know that L1026* in BOO is extinct or is it rather non-resolved / ancestral
@rob Yeah I know the article well, I helped out in it. I disagree with that particular aspect
Do we know that L1026* in BOO is extinct or is it rather non-resolved / ancestral
Even if it is not extinct, it marks cultures of the circumpolar zone with wafer ceramics (asbestos ceramics). As depicted on these maps in the work of the Kuzmins, these cultures are depicted in yellow
https://www.evrazstep.ru/index.php/aes/article/view/177/209
Margaryan:
Some individuals have strong affinity with Eastern Europeans, particularly those from the island of Gotland in eastern Sweden. The latter likely reflects individuals with Baltic ancestry, as clustering with Baltic BA individuals is evident in the IBS-UMAP analysis and through f4-statistics.
Genetic clustering using IBS-UMAP suggested genetic affinities of some Viking Age individuals with Bronze Age individuals from the Baltic. To further test these, we quantified excess allele sharing of Viking Age individuals with Baltic BA compared to early Viking Age individuals from Salme using f4 statistics. We find that many individuals from the island of Gotland share a significant excess of alleles with Baltic BA, consistent with other evidence of this site being a trading post with contacts across the Baltic Sea.
The population of Baltic BA could not die out, since it was still alive in the Middle Ages.
"If anything , I see it as a diffusion of western metallurgical techniques, stimulating the Altai-Sayan nucleus"
Also the metallurgical techniques ultimately are western (asian), the diffusion of metallurgy actually went southern central asia > dzungaria > Siberia > Eastern Europe.
Eastern sites and goods tend to be older than the western ones and so far the oldest casting mould typical of STP goods was in one of those poorly-studied Chemurchek burials.
@ambron
The Medieval Baltic-like Gotland samples mostly cluster with modern Balts and even Slavs.
Only about 3 or 4 are closer to Baltic BA, so they may have come from some unusual population that managed to survive into the Middle Ages.
Margaryan et al. didn't check this, probably because they only had the samples from Mittnik et al to work with.
David, in general, my point is that the genetic changes in the Eastern Baltic are most simply explained by the Slavic and Finnish admixture to the Baltic BA, and not by population exchange.
You're wrong, because the southern shift happens during the Iron Age in places where there's no reason to assume any sort of Slavic input, like Estonia.
And, of course, you can't seriously suggest that this southern shift is due to Uralic admixture.
The only way it can be explained by the Uralic expansion, is if the Uralic speakers were mixed with someone else, like Balts.
And in case you don't know yet, southern Finland was home to Saami people back then.
Modern and Iron Age Estonians look as if the Bronze Age Estonians were given a Slavic and Finnish admixture at the same time.
I just told you there's no way that these Iron Age Estonians could have any Slavic or Finnish ancestry, because their closest neighbors that we know of were Saami people.
Aside from genetics, an expansion of Baltic speakers into IA Finnic territories on the East baltic is quite commonly accepted on linguistic grounds is it not? Hydronym layers certainly seem to suggest something like that.
Well, the Finns came from the east, and some southern neighbors of Estonians looked genetically in the Late Bronze Age like modern Poles.
@ambron
Modern Finns are a mix between Balts, early Finns, Saami and Scandinavians, so the Finnish cluster that you're looking at didn't exist back then.
There also weren't any Polish-like people anywhere close to Estonia in the LBA, as we know from ancient DNA.
You're really not very good at this.
These people Baltic_LVA_BA and Baltic_EST_BA were basically R1a-Z280\CTS1211\FT92022 and R1a-Z280\CTS1211\YP4932. These are not the most widespread subclades right now, even in the Baltic States. Now typically Balto-Slavic are R1a-Z280\CTS1211\Y35, R1a-Z280\Z92 and R1a-Z280\CTS1211\YP340. So Baltic_LVA_BA and Baltic_EST_BA are either some unknown Indo-European lines or a para- of Balto-Slavs.
@ Vladimir
Right, understand what you're saying.
I had a thought that the route of Kra-001 like ancestry point to a circum-polar route, with differential Y-hg founder effects rather than 2 -waves of kra-like migration.
I have a feeling the late Srubnaja like groups , Sintashta_o, etc would have occupied the circum-ural southern forest/ forest-steppe zone until ~1300 BC. At this point, the settlement system collapses, emptying the Ural zone, just when steppe ancestry appears in Swat, funnily enough.
@Ryan
@SKRiBHa - they don't need to be rested to be "pretty solid stuff"
Unfortunately, your answer does not make any sense, see for example Ancient Bohemian I20509, Alexandria I6561, Locomotiv etc. It is obvious that you did not understand my simple question which was not about whether any of them need to be retested in order to be 'pretty solid stuff'.
I asked about:
Do any of you have any knowledge how many percent or how many ancient DNA samples have already been retested so far, so that their results are 'pretty solid stuff' - 1, 5, 10, 15..?
@SKRiBHa
I20509 and I6561 don't need to be retested. The first one needs a C14 date and the second one a correct C14 date.
Some of the Locomotiv samples that were done with outdated PCR technology have been retested with shotgun sequencing. No R1a in them.
You should know all of this. Anyone who visits this blog regularly should know it.
Unfortunately, you appear to be mentally unstable, which is OK with me, but you're cluttering up my comments section.
@Rob So you think it was one migration. Subclades Y16323 remained at the starting point. Subclades Z1636 separated in western Siberia. The CTS10760 subclades separated somewhere in the European part, conditionally in the Komi Republic or Perm region, and BOO reached the Kola Peninsula?
Reasonably working models I get for present day Finnish in G25 Vahaduo is:
15.2% FIN_Levanluhta_IA (Saami like, 600 CE), 29.4% LVA_Baltic_BA (550 BCE), 53.8% NOR_Mid_VA (950 CE), 1.6% Nganassan (present day).
It looks pretty close - https://imgur.com/a/5Wf5yPy.
Not sure how to model FIN_Levanluhta IA.
Was thinking about the model presented in the new paper on Hungarians (https://www.biorxiv.org/content/10.1101/2022.01.19.476915v1.full.pdf) on page 8, where the proto-Hungarian conquerors are from a group that formed at the interface between Mezhovskaya and Nganasan, then Nganasan migrated away. If that is the case, then is it possible that Nganasan acquired their language during this stage, without much gene-flow, then branched away early? I guess the objection to this would be "Linguists 'feel' Samoyedic is too differentiated to date to such a time".
@Davidski
You wrote that you would not answer me, but you changed your mind, so thanks for the partial info.
@ D and re :"I just told you there's no way that these Iron Age Estonians could have any Slavic or Finnish ancestry, because their closest neighbors that we know of were Saami people."
According to some yet unpublished results even in Finland there were at least three distinct IA populations, sort of creating a triangle in a PCA. One overlapped with Levänluhta group and modern Saami, but the features of the other two (Luistari near the West Coast, Hiitola in the Ladoga area) are a bit of a mystery to me. Odd enough, the Ladogan group was based on mtdna more related to farming type of populations, the western Luistari group on more HG type of mtdna.
Some Germanic groups may have been there already, possibly similar to the Levanluhta outlier.
Speaking of Finnic groups and Sami, has anyone figured out the source for approximately 1/3 of Sami vocabulary which is from a PaleoEuropean language(s)? Could it be from SHG/Pittware (who largely replaced TRB in Southern Scandinavia)?
Did the non-IE non-Uralic hydronyms in the East Baltic derive from a Narva HG language (WHG) or from a Combed ware one (EHG, likely related to Volosovo)?
Also, the south western Luistari group was based on males with paternal N: https://terheninenmaa.blogspot.com/2019/03/iron-age-finns-in-southwestern-finland.html. They may of course have spoken Germanic too. Then, there are Meryanic looking toponymes in the Hiitola surroundings of the Ladoga area. We don't yet unluckily have the results from Pre Slavic Suzdal, one of the Meryan centres, but we already know that the people living there in the IA were "similar to Uralic speaking NE Europeans" what ever that means. I'd assume that the Luistari people were basically similar to IA Estonians, something like V10_2. Hiitola, remains to be seen, but maybe something slightly biased towards Fatyanovo even?
David, I think each of us will stick to our task. For me it is obvious that today's Balts are Baltic BA with a Slavic admixture. For example, some Lithuanians and Latvians locate centrally between the Baltic BA and Poles, and some Late Bronze Age Lithuanians looked genetically like modern Lithuanians and Poles.
Maybe a possible point of departure for those turbo guys:
https://en.m.wikipedia.org/wiki/Rzucewo_culture
@ambron
For me it is obvious that today's Balts are Baltic BA with a Slavic admixture.
This view can't be reconciled with ancient DNA, uniparental markers, archeology or linguistics.
You should try to be more objective, because in the end you're only fooling yourself.
@ Vladimir
yep something like that, + added local shifts which might explain the patterning, although I know each subclade has its own inferred TMRCA
Seems like there was 'domino effect' type migrations from Baikal to regions north, then further afield which all began this
@ Anyone
what are the chances that BOO represents a Saami-type branch, whilst later movements brought Finnic
some of the late Baltic BA individuals from Kivutkalns plot within Iron Age range
yet the (low res) late antiquity individual plots back with the LBA cluster
Looking at multiple PCA, seems like Baltic BA shifted towards Baltic IA and Uralics, which I think makes sense considering the Y-DNA.
@Huck Finn
Shouldn't we expect the Luistari group to resemble Levanluhta_IA?
@Rob & Vladimir “ yep something like that, + added local shifts which might explain the patterning, although I know each subclade has its own inferred TMRCA”
I’m starting to believe that the supposedly discredited “Ural Altaic” macro family is indicated, with Ulchi-like East Asian in Native Americans, Altaic and Nganasan Uralic maybe originating in the same pop ~25K-20K ya.
@ambron
I think Slavs are CWC/Unietice mixed with some HUN_Mako_o/Vatya_o like populations. Balts are Slavs (i.e. CWC/Unietice plus HUN_Mako_o/Vatya_o) plus some HG like UKR_Meso. Most eastern Poles like myself have some Baltic admixture.
Target: EastPole_scaled
Distance: 2.8437% / 0.02843726 | R5P
17.8 Baltic_EST_BA:s19_0LS11_1
17.2 POL_Unetice_EBA:RISE154
15.6 HUN_MBA_Fuzesabony:I20750
15.4 HUN_Mako_EBA_o:I1502
13.8 Baltic_EST_BA:s19_V9_2
12.2 Baltic_EST_BA:s19_X17_2
5.4 Corded_Ware_POL:pcw350
2.6 Corded_Ware_POL:pcw211
Without Baltic_BA:
Target: EastPole_scaled
Distance: 4.3456% / 0.04345639
41.0 HUN_Mako_EBA_o:I1502
17.6 Corded_Ware_Baltic:Spiginas2
14.0 HUN_MBA_Fuzesabony:I20750
13.6 Corded_Ware_Baltic_early:Gyvakarai1_10bp
9.2 Corded_Ware_DEU:I1538
2.6 Corded_Ware_POL:pcw211
2.0 Corded_Ware_POL:pcw350
Target: Corded_Ware_Baltic:Spiginas2
Distance: 3.5155% / 0.03515473
28.8 Corded_Ware_POL:N49
24.0 HUN_MBA_Vatya_o:RISE479
@ Copper Axe
'Also the metallurgical techniques ultimately are western (asian), the diffusion of metallurgy actually went southern central asia > dzungaria > Siberia > Eastern Europe.
Eastern sites and goods tend to be older than the western ones and so far the oldest casting mould typical of STP goods was in one of those poorly-studied Chemurchek burials.'
Could the significant levels of BMAC-ancestry in some Chermuchek individuals be a sign of this purported Asian affinity. I can;t recall Chernykh making that link. Who did ?
The Gotland VA samples only barely overlap with Baltic LBA.
In North Euro PCA: https://i.imgur.com/KNoJC5O.png
DA171 looks like a better match, at least for VK434 and VK439. He apparently also belongs to N-L1025.
DA171, VK432, VK434 and VK439 compared against the Latvia BA average:
Distance D3: ( AC - BC ) / ( AC + BC ) ↑
A: Baltic_LTU_Late_Antiquity_low_res:DA171
B: Baltic_LVA_BA
C: ↴
-0.05235072 SWE_LN_low_res
-0.04750159 VK2020_DNK_Sealand_LNBA
-0.04468099 CZE_IA_La_Tene_oFennoscandian
-0.03907451 VK2020_NOR_North_IA
-0.03765669 NLD_LNB_EBA_Bell_Beaker_low_res
Distance D3: ( AC - BC ) / ( AC + BC ) ↑
A: VK2020_SWE_Gotland_VA:VK432
B: Baltic_LVA_BA
C: ↴
-0.06569482 VK2020_RUS_Kurevanikha_VA
-0.06124681 DEU_Tollense_BA_o2
-0.05668340 HUN_Avar_Szolad
-0.05278618 VK2020_UKR_Lutsk_MA
-0.04981984 Lithuanian_PA
Distance D3: ( AC - BC ) / ( AC + BC ) ↑
A: VK2020_SWE_Gotland_VA:VK434
B: Baltic_LVA_BA
C: ↴
-0.08281393 Finnish_North
-0.08278077 Finnish_Southwest
-0.07401234 Finnish_Central
-0.06451720 Polish_Kashubian
-0.06241301 Vepsian
Distance D3: ( AC - BC ) / ( AC + BC ) ↑
A: VK2020_SWE_Gotland_VA:VK439
B: Baltic_LVA_BA
C: ↴
-0.06569053 CZE_IA_La_Tene_oFennoscandian
-0.05341236 CZE_IA_Hallstatt_low_res
-0.05214952 FIN_Levanluhta_IA_o
-0.05051438 NLD_LBA
-0.05035341 VK2020_NOR_North_IA
VK432 is 'normal Balt' shifted, whereas the others appear Scandinavian/Finnish shifted (or shifted towards a combination of Finnic and 'generic northern European' ancestry?).
There's also VIII5 (RUS_Ingria_IA), supposedly dated to AD 75-300, who clusters with Baltic LBA, but that one isn't carbon dated, so make of that what you will.
@ Rob and re:"what are the chances that BOO represents a Saami-type branch, whilst later movements brought Finnic".
No way in linguistic sense at least. Saami is based on the same West Uralic as Finnic however including a sizeable substratum of some unknown language. Saamic expanded into Lapland i.e. into northern parts of Finland etc. during the first centuries CE. Meryanic in Central (!) Russia apparently represents a sister clade of Saamic i.e. they descend from the same West Uralic sub branch.
@ D and re: "Shouldn't we expect the Luistari group to resemble Levanluhta_IA?" Luistari is basically located inside the Finnish Tarand area and even nowadays there are people on the southern coastline resembling a lot the people in Estonian Tarands. But, there were possibly a very different group from Levänluhta Core living even next to them. For instance one of the main Finnish ship burials is located not far from Levänluhta. Luistari, on the other hand and as a burial, somewhat resemble those of continental Germans. So, some (female, because of paternal N in the burial?) samples of Luistari might look like Levänluhta_o?
David:
"This view can't be reconciled with ancient DNA, uniparental markers, archeology or linguistics."
I really don't understand what you are talking about. The continuity of Baltic BA in the eastern Baltic was demonstrated in three studies. Bronze Age uniparental markers remain in the eastern Baltic to this day. There has been a cultural continuity in this region since the Lusatian culture. The Eastern Baltic is one of the areas of the highest concentration of the old Baltic toponymy.
It seems to me that we are arguing about what should be called admixture and what should be called population exchange.
EastPole, I am inclined to the concept of Arza.
CWC assimilated the local demographic substrate in the Carpathian region. The Slavs stayed where they were and the Balts moved to the eastern Baltic, but there were still population flows in both directions between the Carpathians and the Baltic.
Erik, the VK432 is simply proof of the survival of the Baltic BA population into the Middle Ages.
@ambron
There's clearly a significant genetic shift in the East Baltic from the BA to the IA.
You don't see it, because you don't want to see it.
@ Huck Finn
''. Saami is based on the same West Uralic as Finnic however including a sizeable substratum of some unknown language.. Saamic expanded into Lapland i.e. into northern parts of Finland etc. during the first centuries CE.''
yes the paleo-lakelandic substratum.
Linguists and scholars have proposed theories ranging from the Mesolithic to Medieval. What do we base the 1-300 AD model on ? Is there inscriptions, Roman accounts or archaeological evidence of a Saami migration at that time ?
'Meryanic in Central (!) Russia apparently represents a sister clade of Saamic i.e. they descend from the same West Uralic sub branch.''
yes i can see that on the F-U trees. for ex. this one
I see no problem. To me, it looks like the Perms of Central Russia (!) moved down into their new homes from the sub-arctic.
I think everyone has been imagining that FU moved west then north through some epi-corded groups, Volosovo or whichever. But it might very well be that they were in fact moving west and then south
@ RE Seima Turbino
I think this is an overrated thing. Dates from western Siberia arent younger than in Mongolia or the Altai. And then you have a region going from nothing to producing fine quality metals. Doesn;t make sense.
The know-how was brought by western herders, and then there developed its own metallurgical tradition with some reflux of ideas and people.
@ Davidski
Are you able to delineate these Baltic shifts in terms of Y-hg ?
@Rob
Modern Balts have different R1a subclades and a lot of N.
The Baltic BA R1a subclades are still there, but clearly a new population moved in considering the accompanying autosomal shift.
I think there is population continuity since at least 500 BC
We can't make claims of population discontinuity from GW data
hg N is expected FUY-related introgression, making sizable founder effects due to the fact that Baltic groups were often fighting and shifting around, e.g. from west to east Lithuania, and back.
David, of course I can see this shift. I can see a little Uralic admixture. It can hardly be called a population exchange.
In today's Balts, either this Ural admixture is minimal or it does not occur at all. Only the Baltic N1c remains her distant echo.
Right, Balts have very little Siberian ancestry and their own N lineages.
Ergo, the new population that moved into the present-day Baltic speaking areas during the Iron Age wasn't Uralic.
Ok Dave
You suggest that “Balts” came in pre-formed, with N1 and R1a-CTS1211 in tow after 500 bc from Djakovo or Milograd ?
As opposed to a more gradual in situ formation
I don't know where they came from yet, but modern Balts are not mostly of Baltic BA origin.
Of course, it was not the Uralic population. It was the population which genetically pulled the Balts from Baltic IA direction to the south-west direction.
Proto-Balts had more GAC ancestry than Baltic BA, that's why it looks like Baltic DNA was pulled southwest genetically, but they could have come from the south or even east.
Each model shows Baltic BA as the base component of modern Balts.
Post a Comment