search this blog

Thursday, July 27, 2017

Yamnaya-related migrations into Iberia: infiltration rather than invasion (Martiniano et al. 2017)

The Martiniano et al. preprint that appeared at bioRxiv more than two months ago was published at PLoS Genetics today (see here). The paper packs a lot of supplementary information that wasn't included with the preprint. Below is the press release about the paper from the Public Library of Science (PLoS). Emphasis is mine.

The genomes of individuals who lived on the Iberian Peninsula in the Bronze Age had minor genetic input from Steppe invaders, suggesting that these migrations played a smaller role in the genetic makeup and culture of Iberian people, compared to other parts of Europe. Daniel Bradley and Rui Martiniano of Trinity College Dublin, in Ireland, and Ana Maria Silva of University of Coimbra, Portugal, report these findings July 27, 2017 in PLOS Genetics.

Between the Middle Neolithic (4200-3500 BC) and the Middle Bronze Age (1740-1430 BC), Central and Northern Europe received a massive influx of people from the Steppe regions of Eastern Europe and Asia. Archaeological digs in Iberia have uncovered changes in culture and funeral rituals during this time, but no one had looked at the genetic impact of these migrations in this part of Europe. Researchers sequenced the genomes of 14 individuals who lived in Portugal during the Neolithic and Bronze Ages and compared them to other ancient and modern genomes.

In contrast with other parts of Europe, they detected only subtle genetic changes between the Portuguese Neolithic and Bronze Age samples resulting from small-scale migration. However, these changes are more pronounced on the paternal lineage. "It was surprising to observe such a striking Y chromosome discontinuity between the Neolithic and the Bronze Age, such as would be consistent with a predominantly male-mediated genetic influx" says first author Rui Martiniano. Researchers also estimated height from the samples, based on relevant DNA sequences, and found that genetic input from Neolithic migrants decreased the height of Europeans, which subsequently increased steadily through later generations.

The study finds that migration into the Iberian Peninsula occurred on a much smaller scale compared to the Steppe invasions in Northern, Central and Northwestern Europe, which likely has implications for the spread of language, culture and technology. These findings may provide an explanation for why Iberia harbors a pre-Indo-European language, called Euskera, spoken in the Basque region along the border of Spain and France. It has been suggested that Indo-European spread with migrations through Europe from the Steppe heartland; a model that fits these results.

Daniel Bradley says "Unlike further north, a mix of earlier tongues and Indo-European languages persist until the dawn of Iberian history, a pattern that resonates with the real but limited influx of migrants around the Bronze Age."

Martiniano R, Cassidy LM, Ó'Maoldúin R, McLaughlin R, Silva NM, Manco L, et al. (2017) The population genomics of archaeological transition in west Iberia: Investigation of ancient substructure using imputation and haplotype-based methods. PLoS Genet 13(7): e1006852.

See also...

New resource: 67 diploid ancient genomes

Late PIE ground zero now obvious; location of PIE homeland still uncertain, but...


Samuel Andrews said...

Um...R1b P312 first appears in Iberia, Italy, France, Germany, England, and Ireland alongside Steppe ancestry. Uh ah, that must mean R1b P312 is from the moon.

Gioiello said...

@ Samuel Andrews

If you (and Others) are glad: prosit! R1b in the Martiniano samples is 100% and the Yamnaya introgression a little (and supposed due to a migration of Bell Beakers from Iberia to Central Europe and to a back migration from there!) Basques has R1b at 86% and steppe introgression close to zero... Prosit!

Samuel Andrews said...


The same tests that give Lithuanians 50% Steppe ancestry give Basque 25% Steppe ancestry.

Gioiello said...

Basque R1 is 100% R1b, Lithuanians (if you take off Jews, come from elsewhere for their R1b) very likely 100% R1a. Something doesn't fit.
R1b (above all only some subclades of R-L23) present in Yamnaya was pretty close to 0% when steppe people migrated to Lithuania. Someone thinks that they migrated before to Central Europe, but where the traces are? Something doesn't fit.

Samuel Andrews said...

It's pretty obvious R1b P312 derives from a single ethno-lingustic-cultural group that swept across all of Western Europe over the course of a few hundred years. The same is true for R1a Z283 in Eastern Europe (Corded Ware) and R1a Z93 in Asia (Sintashta).

I think R1b P312 folk knew what they were doing. It isn't normal for a single ethnic group to encompass 1/3s of a continent within 100-300 years. They definitely intentionally settled Britain and Ireland. The migration wasn't a gradual, accidental thing. R1b L21 folk probably learned there were two giant islands(Britain, Ireland) and then decided to leave mainland Europe and cover the entirety of both islands.

Samuel Andrews said...


Different Steppe groups had different Y DNA. Some had R1a, that's why Eastern European have lots of R1a. Some had R1b, that's where Western Europeans get their R1b.

bellbeakerblogger said...

Maybe it's just late, but Steppe but no teal?

Samuel Andrews said...


You should read this...

European R1b and R1a derive from genetically closely related populations. That's why both Lithuanians and Basque have Steppe ancestry.

Davidski said...


Maybe it's just late, but Steppe but no teal?

Yeah, but that's based on ADMIXTURE, so it doesn't necessarily mean much. Recent genetic drift hides a lot of stuff in ADMIXTURE output.

Gioiello said...

@ Samuel Andrews


You should read this..."

I read it and wrote 24 posts out of 439. I am R1b1a2-L23-Z2110-FGC24408, K1a1b1e... and you? My address is

Matt said...

Done a bit more analysis of their Table S9 from this paper, over what I posted up to the comments for "The Crisis" yesterday.

As there are a few links, just going to split this into a couple short posts to avoid the overzealous spam filter (sorry).

Comment 1: Means

1) .csv format for the populations means of ancient donation from CHROMOPAINTER Coancestry matrix:

2) the same data from 1), relabelled with the labels from their able S8 (which I think read more transparently):

This was the first set of data I computed from the table, then I had a look at the individual values, which I'll mention in the my next comment, with .csv for the data as I've formatted and some graphics.

Matt said...

Comment 2: Individuals

3) .csv format for a PAST3 datasheet for individuals, with population labeling and colour and format:
What I found with this when putting all the individuals through PCA was that some populations and individuals were *generally* low on haplotype donation with ancients, and that this produced a dominant PC1 relating to being generally low, or generally high on donation:

These seemed to be populations who had a history of small population size and high ROH / high inbreeding: Orcadians, Basque, Druze.

However, running in correspondence analysis, which removes these kind of "generally high / generally low" effects, produced all the populations in the *right* expected place for West Eurasian PCA:

Matt said...

Comment 3:

So, I produced:

4) as 3) but removing individuals with outlying PC1 score (in theory, inbred individuals):

This last one produces PCA which much less hypothetical "inbreeding" effect and are more conventional (as well as the same correspondence analysis).

Anyway, feel free to download the datasheets and plot for yourselves in PAST3.

Matt said...

Emphasizing as well, in the donation, lots of overlap between populations, for example comparing La Brana and Motala donation on the individual level,

It's only through the systematic population differences to lots of ancients that the "clean" separation in the PCAs, etc. I've posted above emerges.

Romulus said...

The height figure is interesting. CHG is even shorter than EEF and Paleo/Mesolithic Europe is the tallest, although BA Europe is close.

AWood said...

Can someone point me to the evidence of "steppe" ancestry in the Portugese_BA sample? Is it just because they are shifted right of the Atlantic_Neo in the PCA? There needs to be more evidence than that I think.

Romulus said...


It had no problems picking up the teal/chg in the rest of the samples.

Kristiina said...

"Notably, outgroup F3 statistics with modern populations (S5 Fig) reveal Portuguese BA samples to display highest shared drift with Basque populations, followed by Sardinians, as previously observed for a Spanish Bronze Age sample [17]. Portuguese LNCA and MN also share inflated levels of drift with Basques, though their highest affinities are seen for Sardinians, a recurring phenomenon in European Neolithic groups"

"Strikingly, the array of Y-chromosome haplotypes in ancient Iberia shifts from those typical of Neolithic populations to haplogroup R1b-M269 in each of the three BA males, of which two carry the derived allele at marker R1b-P312. Interestingly, modern Basque populations have the M269 variant at high frequency (87.1%)"

Shouldn't we therefore presume that Iberian Bell Beaker was Proto-Basque-speaking?

Romulus said...


I agree. The lack of CHG in the samples makes that hybrid model look very interesting.

Gioiello said...

@ Kristiina
"Shouldn't we therefore presume that Iberian Bell Beaker was Proto-Basque-speaking?"

And all that should make us think to all what I am saying from more than ten years:
1) some linguists think that Basque is linked with Indo-European as the same Etruscan (intermediate between IE and Caucasian as to Trombetti, even though it might have been the language of the agriculturalists from Aegean Sea to Central Europe)
2) a possible link of Basque with Sardinian
Thus the origin of R1b1 and subclades from the "Italian Refugium" is open more than ever...
Ask these Harvardians why there is a country practically not tested for aDNA: Italy.

Romulus said...

The Bronze Ages samples are from 1750-1430 B.C. so they actually post-date the Beaker Culture. Makes that lack of CHG even more interesting.

Palacista said...

Gioiello, nobody with brain links Basque and Indo-European, no connection at all.

Unknown said...

That makes perfect sense, Proto-Basques were an ally population of Protoindoeuropeans in Anatolia, both were close enough to share R1b-P312, but Protoindoeuropeans being from east Anatolia had much more CHG related ancestry. Obviously both invaded Europe through Italy. Once aDNA from Italy is sequenced this issue will be settled.

Gioiello said...

@ Palacista

Of course I am from the counrty of Alfredo Trombetti who wrote on 1925 Le origini della lingua basca, demonstrating the links with Caucasian languages, but I read a paper on another Italian (you may say he is an amateur) who linked the Basque roots with the Indo-European ones through a double Lautverschiebung. Perhaps it merits to exam it. Also Etruscan is thought not having anything to do with Indo-European, but many think the contrary, beginning from Morandi and many others.

Gioiello said...

@ Illya P. Constant
Very likely things both from the linguistic and genetic side are more complex than people usually think. About the origin from Anatolia unfortunately in the aDNA no old R1b has been found there so far, and very likely it there wasn't, being R1 the haplogroup of the European hunter-gatherers, and it seems that it wasn't either in the WHG migrated to Anatolia.
About Etruscan I propose that it were the language of the agriculturalists from Aegean Sea and Northern Anatolia who migrated to central Europe rather than Hattic as Schrijver thought. About Basque the link with Caucasian languages is clear in its grammar, whereas Etruscan demonstrates another development.
But much work is to do yet. Strangely my R1b1a2-L23-Z2110-FGC24408 finds matches in a French Basque and in an Armenian, beyond an Arab and an Englishman. The link may be 6000 years old as to YFull (I think also 7000 or 8000 years) with the Arab, more recent with the others. Unfortunately no one is tested deeply like me through a Full Genome.

Unknown said...

"unfortunately in the aDNA no old R1b has been found there so far, and very likely it there wasn't"
What? R-M269 has the highest variance by far in Anatolia, there is no doubt it was there.

Anonymous said...

Iberia really preserved pre-Indo European cultures for a long time, it was only after the Celtic arrival and the Roman conquest that it slowly started to change and to gain more Steppe.

Gioiello said...

@ Illya P. Constant s
"unfortunately in the aDNA no old R1b has been found there so far, and very likely it there wasn't"
What? R-M269 has the highest variance by far in Anatolia, there is no doubt it was there".

Illya, I could say that Anatolia has always been what many said to me about Italy (and it wasn't pleasant to me): a sink. We find there all what many peoples brought during millennia. The only proof is aDNA, and even though Anatolia has been deeply tested, much more than Italy, no old R1b has been found (Italy amongst only 6 samples tested has R1b1 14000 years ahgo and R1b1a2 4000 years ago). You say that it has to be. We'll see next. Science is made by proofs and not desires or presuppositions.

AWood said...

The YDNA of the Basque are very young R1b lineages. Don't be fooled by people who find an outlier like L23(xL51) and make false claims to suggest those are the founders of an entire group of west Europeans.

M153 is downstream of Z295+ which is a group that is linked heavily to the region of Germany to Iberia. M153 being found almost exclusively in southern France and Iberia. The brother branch to Z295+ is found almost exclusively in northern and western Europe.

Recall that the Neolithic people inhabiting modern territory we label as "Basque region" were I2-M223 and G2a2 in the Neolithic period, despite being a very close fit to the modern Basque population.

I think what the author states holds true, it's male mediated gene flow in the bronze period. I'm interpreting a closer affinity to the central European hunter gatherers + LBK rather than those of the Atlanto-Neolithic, but I see no CHG in their admixture, which would indicate they moved south west before the arrival of corded ware. Hopefully someone has more insight in the data, or can run their own analysis when it's made available.

Gioiello said...

@ AWood

"Don't be fooled by people who find an outlier like L23(xL51) and make false claims to suggest those are the founders of an entire group of west Europeans".


Gioiello said...

@ Palacista

Gianfranco Forni, Basque as an Indo-European Language – A Step-by-Step Introduction

Ryan said...

@ Kristiina - "Shouldn't we therefore presume that Iberian Bell Beaker was Proto-Basque-speaking?"

Pretty much IMHO. There's also the cultural continuity. It would seem strange to assume the Bell Beakers suddenly switched languages but had their culture left entirely intact.

@ Sam - "Um...R1b P312 first appears in Iberia, Italy, France, Germany, England, and Ireland alongside Steppe ancestry. Uh ah, that must mean R1b P312 is from the moon."

Either way we don't really have great sample sizes for either a Yamnaya-derived or non-Yamnaya derived source. There's 0 R1b from CWC to refer to and while 2/7 of Iberian (pre-Steppe) BB samples are R1b, P312 is only excluded for one, and the other we have no idea one way or another what subclade it belonged to.

@David - "Yeah, but that's based on ADMIXTURE, so it doesn't necessarily mean much. Recent genetic drift hides a lot of stuff in ADMIXTURE output."
I could just as easily say that R1b P312 first appears in Italy, France, Germany and Ireland alongside Bell Beakers."

Are you able to check it in a more robust way? The PCA in Olade et al isn't very helpful because it conflates an excess of WHG ancestry with a lack of CHG ancestry. Late Bell Beakers clearly had some Yamnaya-derived ancestry, but if there are also some Bell Beakers with EHG from a source not directly descended from Yamnaya, that would be extremely interesting to know.

I'm not sure the change in Y-DNA is entirely as abrupt as the paper suggests. Olade has Iberia's hunter-gatherer affinities becoming more eastern between the Early Neolithic and the Chalcolithic. I'd think gradually becoming more "eastern" hunter-gatherer (not true EHG, but eastern WHG) would push the Y-chromosome pool towards R1b as well.

AWood said...


I think we have enough data to suggest that R1b was not part of the Atlantic_Neolithic groups. Quite a bit now from Iberia and Britain to suggest it was a G2a2 and I2 thing.

Which (Iberian) BB remains were confirmed and reported as R1b?

Samuel Andrews said...

"2) a possible link of Basque with Sardinian
Thus the origin of R1b1 and subclades from the "Italian Refugium" is open more than ever..."

How does that prove anything?

"Ask these Harvardians why there is a country practically not tested for aDNA: Italy."

These Harvardians probably don't care where R1b P312 is from. They don't have an agenda. David Reich, their fearless leader, is Jewish. Why should he care? They are just trying to find the truth. And they found the truth; R1b P312 is from the Steppe.

Grey said...

"small-scale migration...striking Y chromosome discontinuity...male-mediated genetic influx"

implies they had a significant advantage of some kind

Unknown said...

Yamnaya-related? Those new R1b people in Iberia are 0% CHG which is proof they are unrelated to Yamnaya. Those researchers are like sheep, no balls to stand up for the facts against the "standard" view.

Simon_W said...

Keep in mind that even according to qpAdm Mako culture Hungary_EBA sample BR1 has 14.5% EHG, but only 0.2% CHG, that's close to none. So it's quite possible and even likely that Portuguese_BA was closer to BR1 than to the mainstream Europe_LNBA like northern Bell Beaker or Unetice.

Samuel Andrews said...


Hungary BA didn't have R1b P312. And because modern Iberians have normal Steppe ancestry it's safe to assume these BA Portuguese did to.

Gioiello said...

@ Samuel Andrews
"David Reich, their fearless leader, is Jewish. Why should he care?"

The answer is in my thousands of letters, now deleted or in blogs which are out. I don't add more, also for respecting Davidski, who is giving hospitality to us.

"They are just trying to find the truth. And they found the truth; R1b P312 is from the Steppe".

Not found so far there, neither its ancestor L51 and all the subclades, also of many L23 haplotypes. Don't forget that also the oldest R-M73 are in Western Europe, having Eastern Europe and Asia only the downstream R-M73-M478, and much other...

Grey said...

Mr Snow said...
"Yamnaya-related? Those new R1b people in Iberia are 0% CHG which is proof they are unrelated to Yamnaya."

Unless they moved before the CHG mixture.

Gioiello said...

@ anonymous said.
"Indo European my ass....Basque people spoke an agglutinative language like Turkic people etc. "

Also Etruscan had agglutinative elements with other inflected ones. Basque and Caucasian languages are very different, and the state of agglutinative, inflected and isolating languages changed many times, thus perhaps it isn't useless the attention to the roots as Gianfranco Forni did.

Samuel Andrews said...


I'd be thrilled if you would send me all your rich R1b data;

"Not found so far there, neither its ancestor L51 and all the subclades, also of many L23 haplotypes."

That's all modern DNA. Modern Eastern Europeans aren't representative of the Eastern Europeans who migrated into western Europe 4,600 years ago.

Romulus said...

Modern Iberians probably got their CHG from Gothic and or Islamic invasions.

Gioiello said...

@ Samuel Andrews


I'd be thrilled if you would send me all your rich R1b data;"

Which data? I changed many PC in my life after that they were out. I have no data, only thousands of printed sheets about all what I wrote and haplotypes from SMGF, FTDNA, YHRD etc, but I use above all my memory. Ask me, and I'll answer you.
Glad to know your address. So you are someone who laughs.

Matt said...

@Davidski, re: CHROMOPAINTER coancestry again, now I've had a chance to work out which RISE samples are which.

Turns out there are some Steppe_EMBA in the set, so we can compare the pattern of donation from Steppe_EMBA relative Kotias and Satsurblia

Yamnaya and Afanasievo vs Kotias:
Yamnaya and Afanasievo vs Satsurblia:

Overall patterns in these graphs:

Low Steppe_EMBA, low CHG: Basques and Sardinians and Southern Levant
High Steppe_EMBA, low CHG: Northern and Central Europeans and Spanish and Northern Italians - there is some marginal tendency for NW Europeans to show higher donation from Steppe_EMBA compared to NE and Central Europe, but this is very marginal.

Intermediate Steppe_EMBA, intermediate CHG: South Italians, Greeks, Cypriots, Iranians

Intermediate Steppe_EMBA, High CHG: South Caucasians - in particular Georgians

High Steppe_EMBA, High CHG: Lezgins - some Lezgin samples appear to have higher donation with Steppe_EMBA than do Northern Europeans.

Putting through PCA shows the patterns that the Northern Europeans tend to have *relatively* the highest donation from Steppe_EMBA compared to CHG, even though some Lezgin samples have the most:

Note, Chuvash don't show any special patterns with Steppe_EMBA samples...
Kind of cool to see that distinction there.

Also kind of nice, haplotype donation for RISE569 Early Slavic vs early Anglo Saxon and Iron Age Brits: The pattern's much less distinct (because the haplotypes are less distinct I guess) however you can see a nice bit of substructure there.

Unknown said...

Sorry to derail you guys from your obsessions, but the Haber paper on the Bronze Age Canaanites was just published and the BAM files are released. Open Genomes Foundation finds that the J1-P58 sample circa 1650 BCE is in fact derived for J1-FGC11, a SNP that is rapidly expanding with aTMRCA of 2000 BCE. It includes the major Arab J1 lineage---L222, as well as broadly distributed other J1 lineages. It is further derived than the J1-Z2324 sample from Bronze Age Jordan. This will be a critical finding for the Near Eastern archaeogenetics and in some ways mirrors the expansions of R1a and R1b in Europe.

Ryan said...

@AWood -"Which (Iberian) BB remains were confirmed and reported as R1b?"

Olade et al:

I0261 - R1b1a(xR1b1a1a2a)
I0257 - R1b1 (no downstream mutations were able to be tested)

That's 2/7 of this group of Iberian Beakers. They were found at Cerdanyola del Vallès in Barcelona and date to 2850–2250 BCE (for I0261) and 2571–2350 BCE (for I0257).

There's also sample I0410 from El Trocs in Spain. Early Neolithic R1b1c from Haak et al 2015.

It definitely wasn't the dominant lineage but it was there. Personally I think WHG from somewhere around the Danube is a more likely source (from WHG to MN to BB to today) but we don't have the samples really to tell. I don't think Yamnaya is a likely though - I don't think we'd have the extreme cultural continuity with a steppe elite wiping out the whole existing Y-DNA pool. If anything, I think a pool of somewhat EHG admixed WHG around the Danube is the source of R1b within both Bell Beakers and Yamnaya. I think that's more likely as I could see greater cultural continuity if the new substrate is from a closely related (WHG) culture, rather than a more distantly related ANE one.

Samuel Andrews said...

"Modern Iberians probably got their CHG from Gothic and or Islamic invasions."

Do you get a sick enjoyment from saying stupid things?

Chad said...

Oh boy. That no CHG thing is in Admixture. This is not to be trusted. There isn't even a separation of WHG and EHG. These need qpAdm and qpGraph to be sorted out. You can't trust Admixture output. This has been said over and over again.

Also, those two Iberian Beaker R1b's are not even M269. So, they're separated from the Eastern Beakers 10,000 years before this, or more.

Chad said...

If you don't know the difference between Admixture and Admixtools reliability, or even that they're two different things, don't even comment.

Rob said...

Great paper
Their analysis of WHG admixture was very interesting

Gioiello said...

@ Roy King
I thank you for having posted that here, because I don't want having anything to do with that person. I prefer speak with you. I read the paper and, some observations, shortly, because here is late and I hope that that is discussed in a dedicated thread.
1) The paper says what I said many times: that J (1 and 2) doesn't come from Middle East, but it is recent there: I said not more than 5000 years and also a subclade not older than 3300 years may have come even from Europe.
2) The expansion happened with the 4200 years ago from east Mesopotamia, and I said that very likely semite languages arose in the peryphery of the Natufian world, and amongst not Natufians, but Caucasians or "Iranians".
3) Perhaps you know that many time I denounced who funded these papers for demonstrating that Phoenicians survived the Roman wars and are present in the Mediterranean world. Now someone said that Lebaneses of to-day have not more than 10% of that ancestry. No old mt seems having descendants to-day amongst Lebaneses.
4) Look at the upstream haplotypes of FGC11 and you'll see that they are so old also in Italy that they cannot having come from Phoenicians or Canaanites.
5) More next.
6) Certainly R1a and R1b have nothing to do with Middle East as to their origin.

Chad said...

It would, just in my opinion, be better to exclude KO1 from analysis like this, since he is 15% Neolithic and about 20% EHG. He inflates WHG scores and shares more drift with about any farmer or LN/EBA sample. It kind of throws off the whole thing.

Rob said...

Yeah was going to say
And it's odd that it's groupie with LaBrana as "WHG 2".
I'm also weary about lumping Brana with loschbour in their analyses. I think it would skew results due to its higher El Miron ancestry

Ryan said...

@Chad - reread the paper penny post. We only known one of those Iberian Beakers is M269 negative. They weren't able to test any additional markers for the other.

If you look at Simon's comment you'll see qAdm also support the EHG without CHG scenario.

KO1 should definitely be treated carefully but I wouldn't exclude it as it is a good candidate for contributing directly to Bell Beakers.

@Grey - if they left before CHG then they aren't Yamnaya. Even in the Eneolithic CHG was present (though at lower levels).

Chad said...


Not a single M269 in Iberia, France, Germany, or Balkans yet before BB and Yamnaya. Thats about what now, 300-400 male ancients in that category? Yamnaya and BB is it. Put 2 and 2 together. It's not that hard.

Samuel Andrews said...

"if they left before CHG then they aren't Yamnaya. Even in the Eneolithic CHG was present (though at lower levels)."

Portugese BA got its R1b P312 from the same source as Northern Bell Beaker. Northern Bell Beaker had both EHG and CHg so it's safe to assume so does Portugese BA.

jeanlohizun said...

Just casually going to drop this D-Stats right here; they can be found in Table-S5

Mbuti Steppe Yamnaya Portuguese MN Spanish CA -0.0081 -3.045 39286 39924 802198

Mbuti Steppe Karasuk Portuguese MN Spanish CA -0.0064 -2.792 43054 43610 882743

Mbuti Portuguese BA Portuguese MN Spanish CA -0.0073 -2.745 40814 41417 814071

Mbuti Unetice Portuguese MN Spanish CA -0.0078 -2.487 30240 30717 622182

Now; let's have a look at the Portuguese Bronze Age Genomes:

Mbuti CHG Portuguese LNCA Portuguese BA 0.0021 0.729 40400 40229 816485
Mbuti CHG Portuguese MN Portuguese BA 0.0013 0.43 39881 39774 812649

Interestingly enough the Z score never goes above 3 when Portuguese MN is used instead of Portuguese LNCA:

Mbuti Steppe Karasuk Portuguese MN Portuguese BA 0.002 0.852
Mbuti Armenian BA Portuguese MN Portuguese BA 0.006 1.305
Mbuti Steppe Sintashta Portuguese MN Portuguese BA 0.0043 1.431
Mbuti Steppe Afanasievo Portuguese MN Portuguese BA 0.0057 1.884
Mbuti Steppe Andronovo Portuguese MN Portuguese BA 0.0063 2.253
Mbuti Steppe Yamnaya Portuguese MN Portuguese BA 0.0072 2.661

Now with Portuguese LNCA:

Mbuti Czech BB Portuguese LNCA Portuguese BA 0.0046 1.226
Mbuti Unetice Portuguese LNCA Portuguese BA 0.0047 1.587
Mbuti Estonian CW Portuguese LNCA Portuguese BA 0.0062 1.623
Mbuti Mezhovskaya Portuguese LNCA Portuguese BA 0.0073 2.203
Mbuti Armenian BA Portuguese LNCA Portuguese BA 0.0111 2.665
Mbuti Steppe Andronovo Portuguese LNCA Portuguese BA 0.0081 3.079
Mbuti Steppe Yamnaya Portuguese LNCA Portuguese BA 0.0083 3.233
Mbuti Steppe Karasuk Portuguese LNCA Portuguese BA 0.0069 3.279
Mbuti Steppe Afanasievo Portuguese LNCA Portuguese BA 0.0099 3.384

Mbuti Steppe Sintashta Portuguese LNCA Portuguese BA 0.0094 3.549

So does Portuguese MN have Steppe Admixture? Or what the hell is going on here?

Also Portuguese BA does not have CHG; at least not any excess of it when compared to Portuguese MN or LNCA. When compared to Spanish CA the Z score is 2.152 which isn't above 3; so suggestive but not conclusive.

jeanlohizun said...

Also @Chad; why are you ignoring the German site with the R1b1a2 related lineages which had ancestral and derived calls for M269.

Unknown said...

Thanks! I do think you tend to conflate all the geneticists at Harvard and Stanford into a unified Lux ex Oriente cabal. This is definitely not the case. My interest and knowledge, as I've emphatically stated, is in Near Eastern archaeology and archaeogenetics which I know quite a bit about. Italian and European prehistory are not my concern or part of my expertise aside from my knowledge and publications about Aegean archaeogenetics. If you could temper your criticisms of everyone associated with these two universities, scientists might take your observations more seriously. I do know that Peter Underhill, paradoxically, found the highest YSTR diversity of R1b-M269 lineages in North Eastern Italy--both L51 and L23(xL51).

Rob said...

Probably nothing but the impression I got with PCA data was that ATP 9 got its admixture from Balkans or Italy instead of Central Europe , but it wasn't super quality
Is the raw data for this available yet ?

Samuel Andrews said...


I don't think the story is going to change. Northern Bell Beaker represents the population that Western Europe got R1b P312 from.

IMO, it's pretty obvious the R1b P312 rich Northern Bell Beaker folk belonged to the same ethno-lingustic group just like the R1a M417 Corded Ware folk belonged to a single ethno-lingustic group. I don't know if they were proto-Celtic or IE, I don't really care.

Understanding that Northern Bell Beaker wasn't a bunch of random ethnic groups who happened to carry the same Y DNA is really important. It means Iberia and Italy almost definitely received R1b P312 from that ethno-lingustic group and therefore from people who had a similar autosomal makeup as Northern Bell Beaker.

Rob said...

Yes I understand that Sam.
But the middle Bronze Age is a different, more complex set of dynamics

Chad said...


Simple Dstats can mislead as there can be confounding factors, such as HG mixture differences.

Also, I do not ignore the German samples. There was no positive call for M269, but for only some at that level. This person is not ancestral to Europeans as L51 was already formed, and possibly P312 at that time. M269 is only found in Yamnaya and BB in Europe. M269 and L51 didn't form in Western Europe. To deny this flies in the face of hundreds of male samples across Europe.

Gioiello said...

@Roy King
For answering what you say I should exhibit all what I am being writing in these last ten years and more. Unfortunately the most part of that has been deleted, the blogs and sites are out, but strangely this post is always on the web. It is weird to read that now, with all what flew meanwhile. But there was all me, also my challenge with that close in old Latin. About Underhill I spoke a lot about, on his biases about the R1a found in Iran and no word about the samples found in Italy:

From: Gioiello Tognoni
Subject: [DNA] Umbilicus mundi
Date: Tue, 21 Aug 2007 19:04:09 +0200

Cristian Capelli wrote, after a deep analysis, being one of the young geniuses of Genetics, in “Y chromosome genetic variation in the Italian peninsula is clinal and supports an admixture model for the Mesolithic-Neolithic encounter”:
1) “The current set of data also provides a first frame for
testing the hypothesis of genetic continuity from Palaeolithic
to Mesolithic in Italy through the last Ice age. This
would point to the presence of an Italian Pleistocene refugium,
postulated for Iberian, Italian and Balkan peninsulae
for a number of species (Hewitt, 2001; Brito, 2005), but not
proposed for humans (Semino et al., 2000; Rosser et al.,
2000). This inconsistency could be possibly due to the fact
that so far no specific haplogroups have been identified at
Y chromosome level in Italy (Semino et al., 2000). However
this could be due to lack of resolution in the current set
of markers and other Y chromosome sublineages not yet
characterised might represent a specific marker for the Italian
area. Taking into consideration current European Y
chromosome hg distribution and data presented here, a
possible candidate could be within hg R1*(xR1a1). A comparison
of the genetic variation estimated as the variance of
the repeat scores averaged across loci of this group in both
Iberia and Italy did not show significant difference
(P > 0.05, data not shown, Brion et al., 2004; Bosch
et al., 2000, 2001). Looking at the pattern of haplotypes
sharing within hg P [that contains hg R1*(xR1a1)], only
28% of those are shared among the two populations. This
value is well below the one estimated when comparing Iberia
with areas re-peopled after last glacial maximum, as the
British Isles (47%) (Capelli et al., 2003). The opposite pattern
is instead observed when comparing haplotypes within
haplogroups whose dispersion was probably associated
with different and more recent events, as hg J (data not
shown). This suggests that the R1*(xR1a1) variation present
in Italy appear not to be a subset of the Iberian one.
More extensive analysis would give the opportunity to test
this hypothesis further”.

Gioiello said...

Aude maiora, Capelli, tu quoque Romanus es! The Basque language can’t have been introduced in Spain from Hg. R1b, being a language close to Caucasian (A. Trombetti, Le origini della lingua basca, 1923-25), but rather from Hg. G, who had a genetic “drift and bottlenecks” (p. 235) with the introgression to Spain of Hg. R1b, who there were not then, but came from Italy (the same for North Europe). The last findings demonstrate that also the “father” of homo sapiens sapiens were probably the Italian “homo of Saccopastore”, of ancient African Origin and come back to Africa.
If J2 and E3b came to Italy from Anatolia, R1b1c could be the haplogroup of Paleolithic Italy: West Europe 80-90%, Italy 40%, Turkey 15%.
1) 2) “Assuming proper identification of the source populations, these results suggest that in terms of demographic influence on the paternal Italian gene pool, the role of Neolithic farmers was greater than Greek historical colonizers of South Italy” (p. 237).
2) 3) “Similarly, given the sporadic and rare distribution of the E3b2 chromosomes, it is possible to conclude that North African gene flow, if any, left no significant evidence in the current Italian Y chromosome pool” (p. 237). The Normans deported to Lucera, and after a rebellion annihilated, the Arab population of Sicily. Someone escaped!
3) 4) The distribution of E3b* (xE3b1, E3b2) is clearly Etruscan (Central Tuscany 0.02, Elba Island (Tuscany) 0,01, Apennine Marche 0’07 (error for 0,03), Tuscany-Latium border 0,04, North-East Latium 0,02, West Campania 0,03): recent gene flow from Anatolia.

Annod uenturod Romai. Next year in Rome.

Gioiello Tognoni

Matt said...

@Rob: And it's odd that (KO1) groupie with LaBrana as "WHG 2".

That's really just a consequence of Bichon and Loschbour having a really tight relationship in shared haploptyes (also shows up in f3, but their clusters here are defined by CHROMOPAINTER) and splitting off into "WHG 1".

The "WHG 2" is really just WHG left over when you take out the tight "WHG 2" cluster. It doesn't indicate that KO1 and La_Brana_1 are more related to each other in haplotypes than either KO1 or La_Brana_1 is to Loschbour. In fact, all the evidence suggests they are less related.

Rob said...

Agree. So just a slightly peculiar choice of labels

Matt said...

I guess they could have called then "WHG General" for 2 and "WHG NW" for 1.

Still, that was my take on it. It's true that their comment is in the relevant supplement is

"Despite belonging to geographically distinct hunter-gatherer populations in Iberia and Hungary, LaBrana and KO1 were included in the same population. A recently published study [4] used f3 statistics to group individuals, and Hungarian KO1 and LaBrana were positioned in the same clade, however this has been interpreted as a possible artifact deriving from both samples being non-UDG-treated. In the present study, we excluded C->T and G->A changes prior imputation and therefore differences in UDG treatment between samples should not significantly affect our analysis. The positioning of KO1 and LaBrana in the same population may therefore represent real shared recent ancestry (S13 Fig).".

So I don't know.

Haplotypes can often identify more recent shared ancestry, and perhaps that's the case between KO1 and LaBrana1, despite differences in deep ancestry (e.g. UkraineHG like ancestry in KO1, ElMiron ancestry in LaBrana1). For instance the Bronze Age Hungarians with that clear signal of relatedness to each other and recent East-Central Europe, despite some deep ancestry differences, etc.


Separately for all some more correspondence analysis of their Table S9 data -

Grey said...

Ryan said...

"@Grey - if they left before CHG then they aren't Yamnaya. Even in the Eneolithic CHG was present (though at lower levels)."

right, it's a question of labels

if "steppe" is defined as "yamnaya" is defined as EHG+CHG then there must have been a time when EHG and CHG were separate so *if* there are populations with EHG but not CHG (and i'm only going by what i've read) then that implies the population got that component either before the mixture or from some region that didn't get that mixture.

which is an intriguing clue imo

Grey said...

"which is an intriguing clue imo"

for examples

1) for simplicity divide things into crops and herds:
- crop farmers expand to the northern limit of optimal crop farming (a)
- beyond that limit a hybrid crops/herds culture develops up to the limit of marginal crop farming (b)
- beyond that a fully pastoral culture develops (c)
then maybe region (a) is CHG, (c) is EHG and (b) is EHG+CHG


2) the EHG component dates from before the creation of the hybrid culture


3) other options i haven't thought of

it depends on whether this EHG without CHG thing is real or not - which dunno.

Rob said...

(C) can only be (B) or (Bii).

Matt said...

Few comments from their S6 Text which might have some link to the "CHG in Portuguese Bronze Age" question, re: Portuguese Bronze Age and a connection to Gok2 which shows up in some of their analyses:

From - S6 Text - CHROMOPAINTER and fineSTRUCTURE analyses

CHROMOPAINTER analysis I - aDNA samples only
"Portuguese Bronze Age (Portugal_BA)

In the present analysis, fineSTRUCTURE has identified the 3 Portuguese Bronze Age individuals as a genetically distinct population (S23 Fig). When compared to Central or Northern European populations such as Ireland [11], the degree of discontinuity between the Neolithic and Bronze Age in Portugal is not pronounced. However, despite the small sample size we have evidence suggesting complete discontinuity at the level of Y-chromosome lineages with all 3 male Bronze Age samples presenting derived alleles at marker M269.

Although in ADMIXTURE analysis we were not able to observe the presence of the CHG-related cluster in the ancestry proportions of the Portuguese Bronze Age samples, with D(Mbuti, X; Portuguese MN/LNCA, Portuguese BA) we find support for CHG/Yamnaya related introgression and also an increase in EHG ancestry.

In PCAs these samples are positioned between Central/North Western European Bronze Age and Iberian MN/LNCA Taken together, our results suggest moderate, yet significant admixture of an external source from populations with steppe-related ancestry, which have resulted in the differentiation of the Portuguese Bronze Age samples."

This is the analysis which is the basis for Fig 1 and Fig 2, and where in Fig 2, Gok2 and Portuguese_BA are on different subtrees of the "Atlantic" part of the cluster tree.


CHROMOPAINTER analysis II - aDNA samples and present-day Eurasians + Yoruba

The Atlantic cluster is a sister group to modern Sardinians. It now contains CB13, the Cardial Neolithic sample from Spain. Gok2 was now placed with Portuguese Bronze Age samples, perhaps because of having some level of EHG introgression. Neolithic samples from Turkey, Greece and Hungary, including LBK, are now in the same population, instead of separated in 2 distinct clusters.

CHROMOPAINTER analysis IV - Analysis with unfiltered genotype probabilities


As in Analysis II, the Portuguese Bronze Age cluster now contains the Scandinavian farmer Gok2 perhaps caused by EHG-related ancestry.

So I kind of wonder if we might have some masking of CHG in ADMIXTURE due to the true influx being a slightly larger more like a Gok2 (North Atlantic Middle Neolithic?) + Steppe thing, say 30%, rather than what looks like it could be a 18% Steppe thing alone.

Note Ireland also seems to be a hotspot for increased haplotype donation from Portuguese BA relative to Portuguese Neolithic, in their Figure 3.

Anonymous said...

I hope nobody minds me making a few points in favour of uncertainty...

- to the suggestion made above that "And because modern Iberians have normal Steppe ancestry it's safe to assume these BA Portuguese did too"... I can't say about what the BA Portuguese WERE like, but that assumption doesn't hold, IMO. We know that there have been at least three large-scale IE-speaking invasions of Iberia SINCE the early bronze age. There's also a known invasion by a non-IE steppe group in historical times. As a result, we really can't draw any huge conclusions about the past from the present.

- Basque as ancestral to the region: it's not clear. Indeed, the earliest attestations are not from Spain but France, and Basque isn't known in Iberia until about 1000 years ago. This may or may not be coincidence. Even if Basque per se comes from Iberia, there's no way to tell when, or whence, its ancestor ultimately came. And don't forget: even the earliest Aquitanian attestations are still thousands of years later than the IE invasions!

- on the Basques as being genetically ancestral to the region: the Basques have been relatively isolated for millennia, much less subject to Celtic, Roman and German invasions. It may be that that isolation has in itself skewed their genes away from their neighbours, even if they were earlier on more similar.

- regarding Basque as non-steppe: probable, but not certain. The invasions brought IE - we don't know if they might have brought anything else. Multilinguistic alliances and waves of invasion by different groups are both known from the later steppe. A 'steppe Basque' population would have to be closely genetically related to the PIEs, but not necessarily linguistically.

- on Basque as the 'BB' language: why? Why not Tartessian, Iberian, proto-Etruscan or anything else that hasn't survived? We know virtually nothing about pre-IE Europe linguistically. Basque has survived, but that doesn't mean it was once dominant. Could have been an isolate back then too.

- if I had to bet, of the languages we know about, the single safest candidate for early (non-steppe) iberian BB is Tartessian - it's furthest from any steppe invasions (although it is of course closer to potential invasions from africa or the mediterranean). That's not saying much, of course.

- if I had to bet, of the languages we know about, the best candidate for late (steppe, R1b) BB might be Lusitanian. If we're looking for a pre-Celtic IE language in Western Europe, Lusitanian is the only obvious candidate. Of course, the odds still aren't great.

- having said that, the exact natures of Lusitanian, Tartessian and Iberian will never be known, as attestations are too sparse for substantial analysis.

- Basque is not IE, and Forni's research is transparently balderdash - unforgivably sloppy both conceptually and in execution, not to mention prima facie immensely implausible. If there is any IE-Basque family connection, it seems to be at a very great time depth.

- similarly, both Basque and IE have been linked with Caucasian languages on grounds of appearance and typology, but this may be coincidence, or may reflect areal influences, or may reflect family relations at a very great time depth.

Anyway, just a few thoughts...

Davidski said...

I've got the haplotype dataset with the 67 ancients. It includes imputed data, but it should be a very useful resource in any case, because the imputation shouldn't skew the results too much, especially for the samples that already had around a million markers. The genomes include Afanasievo, Andronovo and Yamnaya (one of which is Ulan IV).

But to run f-stats and f-stat based models, I really need to wait for the pseudo-haploid dataset.

Gioiello said...

@ vacuouswastrel

"Basque is not IE, and Forni's research is transparently balderdash - unforgivably sloppy both conceptually and in execution, not to mention prima facie immensely implausible. If there is any IE-Basque family connection, it seems to be at a very great time depth.

- similarly, both Basque and IE have been linked with Caucasian languages on grounds of appearance and typology, but this may be coincidence, or may reflect areal influences, or may reflect family relations at a very great time depth".

At a very great time depth... of course, as R1b-L389, R1a-M420 etc., and all the intermediate haplotypes. The peoples were very likely the European hunter-gatherers.

Guy said...

Hi Folks,

Looking at supplemental figure 4 "Principal component analysis of 604 modern West Eurasians onto which variation from 224 ancient genomes has been projected." I notice that the modern Sardinians are farther out on PC2 than the LBK samples. Is this projection bias or is something pulling the modern Sardinians out that direction?


Anonymous said...


Basque shares some features with specifically Kartvelian family, which came to Caucasus from Anatolia.
Other Caucasian languages, supposedly indigenous to Caucasus, are starkly different.

Kartvelian and Basque have inflected verbs with pluripersonal agreement, agglutinative nouns, ergative-absolutive case system.
No congates are known however.

If the EEF language, potentially parental to Basque, is realted to the Kartvelian parent, they split at least 9000 years ago.

Davidski said...

I notice that the modern Sardinians are farther out on PC2 than the LBK samples. Is this projection bias or is something pulling the modern Sardinians out that direction?

Projection bias.

Davidski said...


Bullshit claim by batman that ElMiron belonged to Y-HG G2 has been deleted.

From now on, fact free posts such as these will be deleted as soon as I see them.

olga said...

Eduardo Blasco Ferrer (1956 – 12 January 2017) was a Spanish-Italian linguist and a professor at the University of Cagliari, Sardinia.[1] He is best known as the author of several studies about the Paleo-Sardinian and Sardinian language.
He wrote a very interesting paper connecting Paleo Sardinian and Paleo Basque.

"Substrata Residue , Linguistic, Reconstruction and Linking:
Methodological Premises,
and the Case History of Palaeo-Sardinian"

He postulates that detailed confrontation of Palaeo-Sardinian with
reconstructed morphological and phonological systems of Palaeo-Basque evince a vast
array of striking correspondances which are due, most probably, to the prehistoric
split of Pre-Proto-Basque into Proto-Basque and Palaeo-Sardinian branches in the late
Mesolithic / early Neolithic age. The paper provides a new Stammbaum model
to account for this split.

olga said...

There are some ideas circulating about basques and euskera that should be revised and actualizad.

All the attempts to link euskera and caucasian since the 19th century have failed. There is a lot of modern literature about it.And all the attemps to attribute basques a recent African origin too.
Clearly Euskera is not an indoeuropean language in its grammar, though it may share lots of words because of the natural cultural evolution.
The basque culture has matriarcal roots, that are somehow reflected in the traditional law. Women inheritated the land, they had special considerations in the family and they never lost her family name demonstrating that her family was as important as her husband's
There is a link between Euskera and Iberian, but as there are not many iberian written texts, it is very difficult to assert anything conclusive. They apparently were neighbours, they shared a numerical system and words referred to family links and of course they traded.
The core of this people lived in the Pyrenees and they are mentioned , not by the name of basques, or wascos or vascones, but the name of their tribes by greeks and romans. These tribes occupied both sides of the mountains from the river Garonne to the Ebro, and to the limits with the Cantabrians. And sometimes they reunited to fight the romans, suggesting they shared a certain common culture.
There are traces of the language in the geography, mountains, rivers, valleys, from the coasts of the Black Sea to Spain, and also as a substrate in French, Gascon, Catalan, Castellano and Aragonés, and perhaps in Sardo, according to the last researches.
Euskera was the language of the Kingdom of Navarra since its foundation in the 8th century, and became the jam of the sandwich between the Franks and the Visigoths who imposed their romance languages and try to grab their territories.
Their lands were rich in minerals, they were mostly shepherds, and they didn´t have a native word for war.
The actual word is Guda, very similar to War.
Apparently they worshipped a Goddess of the Earth, that created de Sun and the Moon, that were femenine figures.
All of these shows no relations with IE people, but more relation with Neolithic Agriculture.
The mistery of the Rb1, means that a feminine community imposed their languages and culture to a lot of men coming from the East, if Rb1 really comes from the East, or that an early migration from Anatolia or from the Danube or Balcans, of pastoralists speakers of an agglutinative language moved to western Europe. That could explain the geographical vasconic names through Europe.

batman said...


"Bullshit claim by batman that ElMiron belonged to Y-HG G2 has been deleted."

Twisting words, creating deceptions - to excercise blunt censorship. Joseph G. would be proud of you, David W.

Facts supporting the PCT-theory seem to be dangerous to your personal world-view, somehow. Unfortunately, the cognitive dissonance you experience can't be saved by editorial power used to censro and supress important facts, derail the academic discussion and dishonor everyone ready to participate in your public discussion about the history of past and present genomes.

We can all have a bad day. I still think You are better than that.

Samuel Andrews said...


Some of your ideas are interesting. Indo European is not a Paleolithic language, therefore there's essentially no Paleolithic linguistic continuation.

Also Y DNA I and G2 aren't representative of dynasties or languages. Only relatively young Y DNA haplogroups like R1b P312 could possibly be representative of languages and dynasties. Haplogroup I and G2 are way too old.The Y DNA I2 in Italy, Spain, and Britain in the Neolithic may have been apart of quite different branches.

There's no way to make any form of Paleolithic continuation believable when considering the current genetic data.

Samuel Andrews said...


The fact trading occurred across long distances in Neolithic Europe and the fact Neolithic Europeans spoke languages isn't evidence IE languages have been in Europe since the Neolithic. Language can change while the same trade routes continue to exist.

batman said...

The deleted texts refered to the LNE/BA gravefield of El Mirador - making an understatement of the fact that it is no more than a stone-throw from the famous 'Lady of El Miron'.

Discussing the paper I brought up the consquence of the Paleolithic/Mesolithic transition - as preliminary to the Mesolithic/Neolithic transition. To understand the LNE/BA transition within the arctic climate-zone.

The mentioned gravefields in the Basque province shows there's a close geographic and genetic continuity between paleolithic and mesolithic Europe. As one may know the late-paleolithic U5b lady from El Miron had U5b-descendants in LNE/BA Burgos.

The LNE/BA gravefield from El Mirador in Burgos is dominated by the "Atlantean" y-dna I2, at that time pre-dominant along the Atlantiuc Facade - from Biscany to Lofoten. While y-dna G seem to be completely predominant in the Meds - from the mesolithic to the violent transitions of BA/IA.

The occurance of the "Mediterranean" G2a2 in the "Atlantean" I2a-gravefield of El Mirador [MIRA 24] is an example of a close communion between these dynasties/etnicities. Sharing origin and graveyard proves their common origin was still known, apreciated and honored. Besides proving that a common, paleolithic herritage of an "eurasian" language that grew to cover both Northern and Southern Europe, well before the cattle-farmers found their way to Spain and Scotland.

Refering to Cunliffes "Facing The Ocean" (2001) I mentioned that the trade between the Atlantic and the Mediterranean worlds worked continously - from 9.000 BP to 2.000 BP. Combined with todays genetic maps of aDNA it's clear that these networks were initiated, developed and maintaind by the dynasties of y-dna G and I.

To make these networks work and grow - through millennia - a common language is required. Which is perfectly explained by their common, palearctic roots - from which their aconomical cultures, fenotypical adaptions and genotypic traits is explained. Already.

Looking at the species and cultures that spread during the Eurasian mesolithic, they all had a palearctic ancestry. So why not the main languages of the mesolithic?

It's pretty clear that A palearctic language once divided into an "atlantean" and a "mediterranean" branch, starting no later than the mesolithic. Thus we have all factors needed to explain the PIE language that spread across the continent – to diversitfy and specify during the mesolithic-neolithic-BA transitions.

Forming the hellenic languages in the south and the 'gottonic' branches in the north. As two new dynasties/etnicities of farmers spread into these areas - as R1a and R1b - they obviously adapted to these existing tongues. Thus the northern farnmers could adapt a gotic toungue, while the ox-drivers of the Meds adapted to the hellenic languages.

The Asian outliers of R1a/b - een when creating their own, large dynasties - had to adapt to the languages already developed - from a palearctic substate - under the dynastical branches of y-dna G2/G1 (Sumeria), J2J1 (Indo-Iran) and H1/H3 (S-India).

Outside of the semi-arctic climate-zone the immigration would result in non-IE languages. As cattles were bred into adaption to the tropics, their breeders reaching tropical Africa and Asia had to adapt to a langauge built on a tropical traditions, sounds and structuers.

After the recent repports on Eurasian aDNA the Paleolithic Continuation Theory have proven to stand on solid grounds - while the old, nostratic eplanations keeps dissipating. Anyone familiar with the major issues of this blog can check the PCT-theory and find the growing body of co-relating and confirming facts.

batman said...



"Some of your ideas are interesting. Indo European is not a Paleolithic language, therefore there's essentially no Paleolithic linguistic continuation."

Thank You. How do you know that IE - as in PIE - is NOT a paleolithic herritage?

Is such a claim even testable?

# "Also Y DNA I and G2 aren't representative of dynasties or languages. Only relatively young Y DNA haplogroups like R1b P312 could possibly be representative of languages and dynasties."

Beg to differ. The 'homozygosity' of G2a2 along mesolithic/neolithic Med says otherwise. Confirmed by a 'unisone' coverage of I2 along the Atlantic facade. Add the KNOWN, post-glacial distribution of hgs H, J and NO respectively - and it's quite clear that we have a clearly REGIONAL distribution - from 6-7 different LCA.

Who, in turn, had a LCA from Paleolithic Eurasia.

So we have "brother-lines" with a very brotherly distribution of lands and opportunities, challenges and responsibilities.Neatly explaining how the word "aet" ('large-family') is explained - as a predecessor to expressions and terms like "aetnic" and "etnicity".

Thne we have a very clear-cut continuatiuon within these specific regions - as no other than the son of a J2a2 could inherrit the positions of the former J2a2-men. Which is alerady explained by common anthropology as a "tribal system", built on an agnatic, patrilinear principle. Known from both ancient and modern Eurasia, as well as the tribal societies of the tropical hemisphere.

batman said...

@ Sammy

"Haplogroup I and G2 are way too old.The Y DNA I2 in Italy, Spain, and Britain in the Neolithic may have been apart of quite different branches."

May - and may not. What I am refering to is the descendants of F known to exist - or be created - at the very end of the Younger Dryas, when all Eurasia (north of the 40th parallel) were depopulated.

Which means that all paleoloithic samples becomes reference-points to the evolution of the Eurasian genome, but secondary to the realities on the ground after the tightest of known bottlenecks, as of 12 kyr BP.

What we do SEE after YD, as this enormous areas gets re-populated, is a clear-cut distribution of specific y-dna-lines along all of Eurasia. AFAIK most y-lines from the paleolithic have been deemed "extinct". I don'æt doubt that. Some survived, though, to recreate the same types and groups of alleles, but with a re-shuffled organisation.

As the re-distribution of y-dna repopulated Eurasia we got a new cast of y-dna F, into NEW varieties of F->GHIJK/LT. Each of them allowed to become funders of new dynasties - and as its king "become ancestral to a large and rich aetnicity" as old myths phrase it.

"There's no way to make any form of Paleolithic continuation believable when considering the current genetic data."

In that case there isn't any Paleolithic continuation believable. So from where did we get the post-glacial plants, animals and peoples in Eurasia - if not from the Eurasian paleolithic?

Please explain how that could happen without any form of 'con-tinu-ation' involved.

# "Language can change while the same trade routes continue to exist."

Sure - as an exception. But as an ordinary rule...?

On which principle should we base our assesments and assumptions, wether mathematical or contextual?

Adding the stable factors - the less that changes the less the communication changes.

Moreover; If we find the same routes thorugh an evolution of periods - and theres no shift of transport-vehicles + no shifts of beats of burdon + no shift of types of tradegoods + no shift in genetic make-up we should, i think, have reason to suspect that they had the same language too.

I see no reason why a large abd successful, rich and very well established family, with a highly developed and effective vocabulary, spoken across one entire land - should have ANY reason to change their language.

Major language-changes NEVER came easy, quick and traceless. Before people changes their mother-tounge they are always changed, radically, in most other ways, first.

Samuel Andrews said...

"In that case there isn't any Paleolithic continuation believable. So from where did we get the post-glacial plants, animals and peoples in Eurasia - if not from the Eurasian paleolithic?"

Yes of course 100% of every person's ancestry was represented by humans who lived during the LGM. But a very small percentage of modern European's ancestors lived on the piece of land they do today.

Maybe 10% of Iberian's ancestry goes back to the Iberian Paleolithic. Probably 10-20% of Balts ancestry goes back to the Baltic Mesolithic.

Region by region there's hardly any Paleolithic continuation. The important phases of settlement occurred long after the Paleolithic.

" The 'homozygosity' of G2a2 along mesolithic/neolithic Med"

"Mediterranean" isn't a genetically defined category. Sure there's some shared ancestry in some parts of the Mediterranean today but we don't know what the situation was in the Mesolithic.

It's Neolithic Anatolians, who became Neolithic Europeans, that had a lot of G2a2. At times G2a2 was frequent deep in Germany and France, far away from the Mediterranean. We don't know who the Mesolithic ancestors of Neolithic Anatolians are. Most may have lived somewhere deep in Asia, far away from the Mediterranean sea.

And we can be fairly confident Mesolithic Mediterranean Europe was basically WHG and dominated by Y DNA R1b1, I2, and C1a2 not G2a2.

"Thank You. How do you know that IE - as in PIE - is NOT a paleolithic herritage?"

First because there's no significant Paleolithic continuation anywhere in Europe. Second because the only significant genetic link between Scythians and IE-speaking Europeans is 'Steppe.' Third because mainstream lingustics don't consider it a Paleolithic language family.

Samuel Andrews said...

"So we have "brother-lines" with a very brotherly distribution of lands and opportunities, challenges and responsibilities."

Clearly often Y DNA does correlate with ethnicity; N1c in Uralic speakers, R1a/b in IE-speakers, and so on. I don't deny that. But I can't imagine that Y DNA haplogroups that are some 10,000-20,000 years old represent ethnicity.

"Confirmed by a 'unisone' coverage of I2 along the Atlantic facade."

I2 was popular all over Europe not just in by the Atlantic before R1 took over. Neolithic/Mesolithic Eastern Europe, far away from the Atalantic, had loads of I2.

Samuel Andrews said...

batman, you make cool connections to archaeology and stuff but you're finding correlations between different data sets where there really aren't any correlations.

The Britons Julius Ceasar interacted with were basically British Bell Beaker folk 2,000 years later. Their ancestors didn't adopt the language of I2 Neolithic British, they replaced I2 Neolithic British Y DNA and therefore definitely the native languages aswell.

Iberia is an exception. R1b P312 became dominate there while the Neolithic native's genes survived really well. You can argue some of their langauges that were spoken by I2 folk, BASQUE!, were adopted by R1b P312 folk.

Did R1b P312 from the Steppe adopt the languages of the I2 folk they mixed with in Central/East Europe? I really doubt that considering Celtic is apart of the same language family as Scythian.

mike said...

Basque Guda,for war sounds like a borrowing from proto Celtic katu

Rob said...

@ Batman

Where did you get the funny idea of mere one (Baltic) refuge which persisted through the Y.D. for all Eurasia ?

Ric Hern said...

@ batman

I think the Australian Aborigines and the Native American Languages can be considered relatively as good proof that Languages changed very drastically from the Palaeolithic to the present.

They diverged so much that many different Language Families formed with no linguistic connection between many....

batman said...


"The Paleolithic" is also a relative term, depending on what time-span you focus on.

The aboriginees of Australia seems to have a continous history that started before the Toba eruption, some 74.000 years ago. As have the indigenous Africans. Lately there's repports that even the Meso-American population have roots older than 40.000 years - in America.

Considering a minimum of 74.000 years on the vast Australian continent you may get a few variations in the various tounges developing in various villages and districts. Then add intra-regional marriages and migrations and the 'diversity' of the various toungues may develop into very different soundbites (phonemes), rhymes, rythms and tunes. Then you may the semantic connotations similarily and - voila - in 70.000 years you may find a pletora of dialects - from which some have even turned foreign and incomprehensible to some other...

Still, there's no-one denying that the major language-groups amongst the Aboriginees have 70.000 years long roots. There's actually NO proof of any major changes among these languages since the start of the Late Paleolithic.

Just as there are NO evidence supporting the idea that the Eurasian languages - both Uralian and I-E - can't have their historic roots in palearctic Eurasia. Since the surprising discovery of aDNA from palearctic Eurasia - such as Ust Isthin, Kostenki, Malta and Goyet - it' abundantly clear that both HG and Farmers have a common, palearctic ancestry. Moreover it became clear that their common ancestor have been the historical ancestor to all present Europeans - and most existing Eurasians.

The mesolithic Europeans descended in a straight line from the Palearctic Eurasia. There's NO reason to doubt that they already had a very well developed language - containing ALL the words, terms and comprehensions that may reflect the entire complexity of an ARCTIC kind of life.

You may find the ancestry of the survivors of the Younger Dryas - when Eurasia actually experienced some extreme, climate-conditions - refered to as the 'Fimbulwinter', the 'Deucalion' and the 'Great Flod'. The severe bottle-neck of this event may be reflected in the makrogroups y-dna F and mt-dna R.

Which is why y-dna F could be nicknamed 'Noah' after the Hebrew mythos - or Njord, Nerthus, Nep-tun and Po-Sidon as he was called in the old, I-E mythologies. As genetics now proves a direct and 'generic' co-relation between the Paleolithic and Mesolithic Europeans, there's NO reason left to AS-SUME that the language of their ancestral heroes - "the survivors" -
was foreign to ANY the Mesolithic Eurasians. Not before they started to produce the babylonic confusions...

batman said...


Did I ever state that the Baltic refugia was the o-n-l-y one?

AFAIK I always stressed that it is the only KNOWN one - to have survived BOTH the LGM and the YD Megafaunal Extinction Event.

From what I understand You seem to have some funny problem with accepting the facts of the matter. Why so?


The distribution of the post-glacial sperad of y-dna F->GHIJK/ isn't the only indicator of regional patterns covering vast territories. Such as A dynasty of hg G2 re-populating the northern shores of the Med and the Black Sea, while their ship-building brother-line I2 repopulates the Atlantic facade and the Baltic coasts - as well as the waterways across the continent.

"... the mtDNA lineages from Europe indicate a shared ancestry throughout the Atlantic zone, from northern Iberia to western Scandinavia, that dates back to the end of the last Ice Age."

PS: There's loads of evidence for various kinds of Palelithic continuation in Europe. All the arctic species roaming todays Eurasia have a paleolithic past in arctic Eurasia. Especially so at the mildest part of the northern continent - along the Atlantic shores.

What "mainstream linguists" represent is anyway highly disputed. As are the three major theories about the origin of the IE languages. Here's a small but important question that may give you some broader understanding of the nature if this debate - as it developed after WWII.

Please note how a lingusist looking to become 'mainstream' during the 1950-ties would use deceptive arguments (salmon = trout) to encounter what in those days were labeled "politically incorrect". That's what we get if we accept 'authorized opinions' rather than clear facts and stringent logic.

If you want to adress the lingusitcal part of history you may check on the various theories yourself - and check against your own, broader understanding of the genetics - before you figure which is the more 'mainstream' as of today.

batman said...


"Clearly often Y DNA does correlate with ethnicity; N1c in Uralic speakers, R1a/b in IE-speakers, and so on. I don't deny that. But I can't imagine that Y DNA haplogroups that are some 10,000-20,000 years old represent ethnicity."

Traded memories - as in 'oral traditions' - have long been furthered through traditional rites, celebrations and festvals - to encompas large and extensive populations throughout generations. Speaking of ancient ancestors and their achievments these traditions have been a part of most etnicities for thousands of years. Expalining both the ancestral origin and sucessions of their cheiftains/kings.

Since the industrial revolution these traditions have vained greatly, all across Eurasia. Some populations have been sheltered from out modern media and educational systems, and thus been keeping their old, indiginous traditions of professional story-tellers - travelling from village to village.

Some of these stories contain specific information about exceptional circumstances or phenomena -such as volcanos, earth-quakes, tsunamis. Thus there's been a renewed interest for the oral traditions of both America and Eurasia - as they start proving their historic reliability - and value.

Which is another evidence that words like 'family', 'tribe' and 'etnicity' were fully conscious and well known concepts during the LP, already. To keep, maintain and further concrete stories and descriptions as parts of a tradition - made to create 'common historical knowledge' within large populations - would normally require that you keep an etnic as well as a linguistic continuity - first.

When such traditions are proven to be 'true' - 13.000+ years after the fact - we better believe the story to be genuine. Indigenous descriptions of a glacial past or a YD refugia are more than one.

Here's one from N America now proving its reliability. As per consequence we have to presume that the language used to tell the story is of the same age - defined as it developed along with the etnicity speaking it.

The term, word and the concept of "Aetnos" is just as old as the terms "Family, Mother, Family". Aet-nicity simply means "grand-family" - from which the old, legal kingdoms could evolve into the ancient civilizations.

HistoryBuff39 said...

But its far fetched to say all R1b comes from Steppes to Iberia. R1b was in Villabruna in north east italy 14000bc. Niw a new paper suggests villabruna and Magdalenians HGs were mixing in Iberia very early R1b must have reached Iberia earlier yes its not found but doesnt mean ot wasnt there. How could R1b reach and got rid of all male lines from Steppes so quickly? We know R1 was in Siberia early Iberia had a vsriety of haplogroups then R1b dominated. It doesnt seem plausible that these regions had no contact after a recent paper suggests HGs in iberia were a variety of HG groups. The bronze Age seems no different to earlier connections. I think people are underestimating contact of different groups. Just because different dna doesnt mean groups arent related look at Aurignacians and Gravettians both unique dna but originally from same source population.

Davidski said...


But its far fetched to say all R1b comes from Steppes to Iberia.

Not all, but almost all.

The R1b in Iberia from the Copper Age and later is almost all R1b-M269, and that's definitely from the steppe.