June 23, 2013

Ancient steppe populations: hints of things to come

A reader alerts me to this research summary from a German government site (pdf). The research covered seems to be that of Joachim Burger's group.

The relevant chapter is:

Schritte im weiten Raum: Neue Blicke auf Zivilisationen der Eurasischen Steppe
[steps in the vast space: New Views on civilizations of the Eurasian steppe]

I invite my German readers to translate the most interesting parts of the chapter in the comments (or at least to summarize them). A few observations on what I've been able to make sense of:

  • Heterogeneity of North Pontic steppe groups with differences between Catacomb culture and earlier Yamnaya individuals
  • "European" light pigmentation but with darker eyes 
  • Iron Age nomadic horsemen of Central Asia/South Siberia were mixed West/East Eurasian


Here are some (little) processed Google Translate portions to whet your appetite:

[The first part of the project looked for copper and Bronze Age cultures of the steppe west and north of the Black Sea (Fig. 1). In the Late Bronze Age (around 3000 BC) came here the very mobile Yamnaya culture in appearance, their population and influence radius - as the investigations showed - apparently at the same time expanding consolidated. With the Yamnaya culture is a single burial rites used in so-called pit graves under kurgans (grave mound). Also this wont Halbno addition maggots strong trading relationships across the steppe. Around 2500 BC, they were replaced by the less mobile Katakombengrab-culture whose dissemination conduction region was significantly smaller. Population genetic analyzes of DNA occupied by the late copper to the Middle Bronze Age, a steadily increasing genetic distance between those cul tures. Between copper and time Katakombengrab culture is the genetic distance is greatest. Here the differences are much more pronounced than between early Chalcolithic cultures and Yamnaya population. This population genetic change could be an indication of discontinuity and population changes due to migration. An archaeological site of suspected immigration from eastern steppe areas but at least on the female side hardly taken place: For Central Asia typical DNA lines do not occur in the studied populations. Despite the genetic differences within the un the investigated groups are with them to popu lations, which are without doubt be described as European.Here the differences are much more pronounced than between early Chalcolithic cultures and Yamnaya population. This population genetic change could be an indication of discontinuity and population changes due to migration. An archaeological site of suspected immigration from eastern steppe areas but at least on the female side hardly taken place: For Central Asia typical DNA lines do not occur in the studied populations. Despite the genetic differences within the un the investigated groups are with them to popu lations, which are without doubt be described as European. DNA markers with known phenotype suggest a continuity between the North Pontic area of ​​4 / 3 Millennium BC and today's Europeans out. For instance, have all examined individuals tierungstyp on a bright pigments, as is prevalent in Europe today. Only the eye color has been dark in comparison to today.]

[The second part of the project was devoted to the population dynamics of early Iron Age peoples of nomadic horsemen in the Eurasian steppe belt. Here were 900-300 BC disseminated numerous highly mobile populations that are associated with the so-called Scythian or Sakian culture (Fig. 2). The groups studied are from the areas of eastern Kazakhstan, Altai mountains, Minusinsk Basin and Tuva. They all consist of a mixture of DNALinien, which today is a part of Central and East Asia and the other in Europe. Ity of the ground because this way the populations have a remarkably high level of genetic diversity that characterizes the Altai population today.

...

The Tagar Culture (Minusinsk Basin) this shows the greatest genetic - but also cultural - distance to all other groups. Although it chronologically corresponds to the Pazyryk culture of the Altai (5th-3rd century BC) seems to be present here genetic isolation. Between the Pazyryk culture and the significantly older findings from Tuva (7 / 6th century BC), however, the genetic distance in spite of the time interval is very small. Amazingly, has the Pazyryk culture also within its range a geographic substructure: Divided into Kazakh Altai, and Cuja Ukok plateau region, show the nomadic horsemen of Cuja region in relation to the other two groups increased genetic distance.]

63 comments:

Grognard said...

So as seems to always be the case, looks like pliny is confirmed.

I'll bookmark this for the next time I hear ancient persians or phoenecians labelled as arabic.

My guess is if you look back even further these guys were even further east back in time. Asia has thrown out wave after wave of migrants in historic times and there's no reason to think it didn't do so in prehistoric times as well. My guess is it's been the real center of most human evolution, which almost has to be the case as that's where evidence of highest populations are.

Onur said...

The Google translation is terrible. For instance:

"In the Late Bronze Age (around 3000 BC) came here the very mobile Yamnaya culture in appearance..."

Incorrect translation. Here, the correct translation has "Late Copper Age" rather than "Late Bronze Age". Huge difference. How does Google Translate still do such simple and stupid mistakes?

Davidski said...

Hey Dienekes, your Bronze Age Indo-European invasion of Europe is about to die a horrible death.

This study will show that there was a genetic shift on the North Pontic steppe from the late Neolithic to the early Bronze Age, which is exactly the same timeframe as the genetic shift in Central Europe (in which the descendants of the Near Eastern-like LBK farmers were replaced by the ancestors of modern Central Europeans).

The problem for you is that the groups moving into Central Europe and the North Pontic steppe weren't West Asians, they were closely related Corded Ware/Unetice/Catacomb groups from somewhere in Eastern Europe (north of the steppe).

What's the bet that the high incidence of the dark eyes in this ancient sample is due to the cultures that were replaced by these Catacomb people?

Dienekes said...

Hey Dienekes, your Bronze Age Indo-European invasion of Europe is about to die a horrible death.

It might, but certainly not on the evidence of this text; we do know, what a horrible death your "North European" TRB hypothesis has died.

What's the bet that the high incidence of the dark eyes in this ancient sample is due to the cultures that were replaced by these Catacomb people?

Inasmuch as I understand the relevant passage, I see no evidence that any of the groups involved were more or less dark-eyed. The only conclusion that can be extracted from the passage is that the investigated groups were darker-eyed than present-day Europeans.

As for the Catacomb culture itself, I also see nothing in the text as to why it was genetically different from Yamnaya groups (in what direction). Certainly the widespread practice of cranial deformation in Catacomb burials has no European parallels and points to a Near Eastern influence.

Davidski said...

Well, it looks like Central Europe and the Western Steppe experienced a genetic shift at more or less the same time with three related groups involved in the process - Corded Ware, Catacomb and then also Unetice. There was also a fourth unrelated one: Bell Beaker.

So my "North European" hypothesis is doing OK, especially if TRB elements were involved in the Corded Ware and Unetice Cultures, which I think they were based on the aDNA data I've seen.

In this context it really doesn't matter what fashions the Catacomb people borrowed after they got to the steppe, or who they mixed with there. All that matters is where they came from and what they looked liked (also genetically) when they first got there. Keep in mind that they ranged all the way from the Balkans to the Caucasus, so anything could have happened in due course.

But anyway, for your West Asian hypothesis to work out, you need the Catacomb Culture to arrive from West Asia, or at least from Southern Europe, and I suppose show more southern DNA (including darker eyes) than the east Balkan and Western Steppe Copper Age groups.

So when this paper comes out and we can compare how all the samples look in that context it'll be pretty obvious who's right and who's wrong.

eurologist said...

Looking forward to more details.

As it stands, the results are in agreement with what I have suggested previously: during climaticaly advantageous times, the West Pontic settlements were far richer and much more populous than the nomads to the northeast. When the climate got dryer, it was the (genetically then Central-to-SE European) West Pontic population that diversified into cattle-hearding on the steppes (while others moved NW to farm on wetter soils), and would have easily outnumbered and overpowered locals.

Finally, the catacomb culture has known associations with the west: corded pottery and polished axes. So, the genetic discontinuity appears to be an eastward expansion from perhaps NW Ukraine and beyond - not an influx from the East.

Rokus said...

It seems the post-war mission to impose upon us an eastern, preferably Europe-less solution to the Indo-European question, becomes increasingly difficult to continue. For the public important input on the issue still remains hidden or shrouded by English google-gibberish, but actually this report tells nothing new: the Kurgan entity is a hybrid construct that - if Indo-European at all - was certainly IE-ified from the outside: most probably from the west. It is well known that eg. the earlier Mikhaylovka group was strongly tied with Globular Amphora, Central Europe, that in turn is part of the Corded Ware horizon. Beaker and "Kurgan" cultures converge even more in later stages, though it wasn't exactly core Beaker territory adapting to the ways of the Steppes, culturally nor genetically. The Kurgan hypothesis is simply the cherished religion of an older generation that, albeit bound to disappear by natural courses, will never yield to any reasonable argument.

Dienekes said...

So my "North European" hypothesis is doing OK, especially if TRB elements were involved in the Corded Ware and Unetice Cultures, which I think they were based on the aDNA data I've seen.

The only available TRB sample is South European-like.

And, your theory doesn't explain the "West Asian" admixture in post-5kya Europe. You can't get that by mixing Gok4-like and Ajv-like populations.

Dienekes said...

But anyway, for your West Asian hypothesis to work out, you need the Catacomb Culture to arrive from West Asia, or at least from Southern Europe, and I suppose show more southern DNA (including darker eyes) than the east Balkan and Western Steppe Copper Age groups.

I'm not sure what you're saying here. We don't actually know much about the pigmentation of most ancient groups except the "light-eyed" Siberian reports and the "darker-eyed" hints of the current chapter.

Dienekes said...

Finally, the catacomb culture has known associations with the west: corded pottery and polished axes. So, the genetic discontinuity appears to be an eastward expansion from perhaps NW Ukraine and beyond - not an influx from the East.

That is of course possible; I have previously remarked that the abandonment of Tripolye sites due to climate change may have freed up a large population to "go pastoralist" to survive, since people don't just wait around to die (I've made the same argument for the settled populations of the BMAC with respect to the genesis of Indo-Iranian)

But, I don't quite see how shuffling the deck of Europe brings us any closer to the elephant in the room which is the arrival of R1a/R1b populations from the east -as they must have at some yet to be determined point.

Matt said...

On eye pigmentation, this might be an odd tangent, but, I have wondered whether the following is currently possible:

- Simulate a population of, for instance, "West Asian" individuals based on frequency data from ADMIXTURE clusters in contemporary populations.

- Take these individuals and use phenotype snp correlations in contemporary populations to estimate phenotype.

- Uses these estimated phenotypes to describe mean and variance in hair, eye, skin pigmentation in the ancient population the ADMIXTURE clusters theoretically maps to.

Of course, this would certainly not tell us that, for sure, ancient populations that map to clusters examined in this manner actually had these phenotypes, even if we accept that the clusters actually map to ancient populations. Because there could be patterns of selection that are independent of and just happen to correlate with cluster distribution.

But it still seems like it would be an interesting exercise, because for instance, taking the K=7 analysis with Southern, West Asian and Atlantic-Baltic components, dark eyes and hair seems to correlate strongly with the Southern component relative to West Asian and Atlantic-Baltic, but not so strongly the level of West Asian and Atlantic-Baltic relative to one another.

Kurti said...

Actually doesn't this study just shows, that most of Indo-European speakers in Europe do indeed stem groups in the Ponti-Caspian steppes? But this does not refute the idea that Proto-Indo Europeans did evolve somewhere in between West and Central Asia. In fact I always had the idea and said it earlier. that even though I think Proto-Indoeuropean evolved in the Middle East, most modern day Indo-European groups of Europe descend from Indo-European tribes in the Pontic-Caspian steppes.

My thesis, Proto Indo European tribes in Western Asia (somewhere south of the Caspian), these influenced groups in Central Asia and the Caucasus (Maikop culture). Later on the Maikop people influenced the Yamnaya culture from which in my opinion most North and West European Indo-Europeans derive from.

Simon_W said...

It's noteworthy how the text characterizes the Yamnaya culture: very mobile, but seminomadic. Important innovations they made use of: the cart with disc wheels, and draught animals. Needless to say that these things were also well known in the TRB and Baden culture, and theirs were not derived from the Yamnaya specimens. Then the text goes on to say that the 1st millennium BC saw the advent of nomadic horse-people. The 1st! Others have said that before, yet there are still people around who regard the Yamnaya culture as one of mounted nomads.

Interesting that the Catacomb culture differs genetically from the earlier cultures. We'll have to wait and see in what direction the difference goes. To me it's a surprise, because according to V. V. Bunak the Catacomb people were basically the same type as the Yamnaya people, i.e. a relatively Proto-Europid/Cromagnoid type with southern influences. According to Bunak the greater difference was with the later Srubna culture, a type with narrow, slightly flat face and somewhat broad nose which makes up the basis for the modern Great Russians.

Also interesting the dark eyes of these north Pontic cultures. I suppose this means that we shouldn't underestimate the southern influence in them, on the one hand from the Tripolye culture to the west, but probably also from West Asia south of the Caucasus. We don't know the pigmentation of the Dnepr Donets people, but as mesolithic folks they may have been lighter pigmented.

Finally, a few words on the third sub-project: The domestication of the Bactrian camel. According to the text, the modern wild Bactrian camels of Mongolia and China differ genetically a lot from the domesticated ones, both of ancient and modern age. The modern domesticated ones on the other hand are very similar to the prehistoric ones. The conclusion is, that the domestication didn't occur in the east, but in the southwestern steppe.

@ Grognard

The text doesn't deal with Persians, nor with Phoenicians. And while the iron age Scythian/Saka groups around the Altai were indeed mixed with central/East Eurasian and West Eurasian haplogroups, there are no central Asian mt-DNA haplogroups in the examinated chalcolithic/bronze age populations north of the Pontic.

@ eurologist

As far as I understood it, the Catacomb culture differed genetically both from the earlier north Pontic, but also from the investigated chalcolithic west Pontic groups, so you can't derive the former from the latter, at least judging from this report.

Simon_W said...

@ eurologist

Ah sorry, I misunderstood! You're suggesting that the western Pontic cultures went Yamnaya.

I think there's craniological evidence for even earlier influence of the Balkan neolithic on the northern Pontic area.

However, it has to be noted that the influence of y-haplogroup I-P37 ends quite abruptly at the Don. And neither does the „Southern“ autosomal component play a role on the Eurasian steppe and in central Asia. So, the IEs there were either associated with the „Atlantic_Baltic“ component or with the „West_Asian“ component which came from the south. Or a third option would be that they adopted it from their Balkan admixed western neighbours, but without genetical admixture. I consider this as the least likely possibility.

Simon_W said...

Kurti said:
Actually doesn't this study just shows, that most of Indo-European speakers in Europe do indeed stem groups in the Ponti-Caspian steppes?

No, this study shows no such thing. The main point of sub-project #1 is a relative homogenity of chalcolithic groups north and west of the Pontic, followed by some change with the arrival of the Catacomb culture; but the report doesn't mention in what way this differed. Furthermore it mentions a European type of pigmentation for all these groups, but relatively dark eyed. Moreover geoarcheological studies indicate that climatic changes leading to a drier climate probably have led to the formation of the mobile, seminomadic Yamnaya way of life. Sub-project #2 deals with iron age groups around the Altai (long after the time of PIE) and sub-project #3 with the domestication of the Bactrian camel.

Grognard said...

"Actually doesn't this study just shows, that most of Indo-European speakers in Europe do indeed stem groups in the Ponti-Caspian steppes? But this does not refute the idea that Proto-Indo Europeans did evolve somewhere in between West and Central Asia. In fact I always had the idea and said it earlier. that even though I think Proto-Indoeuropean evolved in the Middle East, most modern day Indo-European groups of Europe descend from Indo-European tribes in the Pontic-Caspian steppes.

My thesis, Proto Indo European tribes in Western Asia (somewhere south of the Caspian), these influenced groups in Central Asia and the Caucasus (Maikop culture). Later on the Maikop people influenced the Yamnaya culture from which in my opinion most North and West European Indo-Europeans derive from."

Historically there's lots of mention of scythians. To me this seemed to confirm their relative newcomer role from area of current day iran.

This is where central europeans stem from if you believe ancient historical sources.

The celts and protocelts are the other big group, for western europeans.

If you look at any group that worships bulls you are probably looking in the right place. I am thinking they are related to the canaanites as their culture fits in well.

Obviously a lot of mixing in there, too.

People shouldn't assume that anyone was "replaced" due to mtDNA evidence. Mathematically it will rush to weed out minority versions if they are neutral markers. If they are not (and probably they aren't, probably they are selected) it will rush towards fixation in the population.

Colin Welling said...

@davidski

you're right that there was a genetic shift in Central Europe and the western steppe around the same time, but the yamnaya were a source of some of this shift in Central and Western Europe. yamnaya > upper Danube and nearby > eastern bell beakers > Western Europe

http://distantconnections.wordpress.com/2013/06/25/steppic-discontinuity-during-the-metal-ages/

you can find another one of my lovely "paper" there

Va_Highlander said...

Simon_W:

"The conclusion is, that the domestication [of the Bactrian camel] didn't occur in the east, but in the southwestern steppe."

Could you be more specific? The Bactrian camel is assumed to have been domesticated in Bactria or Margiana. If you're saying that it took place elsewhere, that would be interesting.

Kurti said...

"No, this study shows no such thing. The main point of sub-project #1 is a relative homogenity of chalcolithic groups north and west of the Pontic, followed by some change with the arrival of the Catacomb culture; but the report doesn't mention in what way this differed. Furthermore it mentions a European type of pigmentation for all these groups, but relatively dark eyed."


But isn't the study actually saying that the genetic make up of these grups west and north of the Pontic do resemble modern day Europeans? So basically the end product of European Indo-Europeans. Isn't this exactly what I am saying? that European Indo Europeans have their roots ultimately in the Steppes of the Pontic sea. And we don't know anything specific about their genetic make up, just that they are typical European. if it says that they were almost identical to modern Europeans, than this means to me they must had quite reasonable West Asian and South European input too, since modern Populations have it.

And the populations in the Steppes didn't need to be entirely replaced by newcomers to be "influenced" by Proto-IndoEuropeans from somewhere else.


Grey said...

"the arrival of R1a/R1b populations from the east -as they must have at some yet to be determined point."

I think R1b at least partly came via the sea route to the Atlantic coast.

1)
If you imagine a "West Asian" population from Anatolia setting off to found a colony in say Greece with 80% married men and 20% single men (who find wives from among the locals) then the resulting population would be 90% Anatolian.

Some centuries later a group from the colony sails off to Italy also 80% married men and 20% single (who find wives among the locals) then the resulting population is (i think) 81% Anatolian.

Rinse and repeat from Italy to Sardinia then Sardinia to Southern Portugal etc in a sequence of colony hops and I think that is a plausible mechanism for gradually reducing the Anatolian signature along the route.


2)
If you assume for the sake of argument that the neolithic farming package worked adequately in the mediterranean ecozone but not as well e.g. crops less productive, in the Atlantic ecozone then expansion along the Atlantic coast may initially have been limited to the coast itself where seafood could compensate for less productive crops.

If at some later point the people along the coast developed a modified package that worked better e.g. by changing the proportions of the components e.g. increasing the use of cattle, then I think you have a possible mechanism for an expansion from the West.

DPNiemi said...

I wish they had identified the haplotypes known for each group.

Dr Rob said...

As ive mentioned before , it is worrying how some people attempt to equate language, culture and genetics into one neat picture. Aside from the fact that we don't even have anywhere near the whole picture , we must rememeb that all of the above variables are not co-eval. rather, the direction of population movement and language expansion might be orthogonal or even run counter-current(!) thus, even of we demonstrate an R1a shift during the bronze age from central Europe to Siberia, and we prOve this originated on Southern Russia, AND we prove that this was indeed a demographic phenomenon and not some accelerated drift or natural selection phenomenon, then it still would not PrOVE that IE originated in EE.
I know we are all offering friendly, unofficial speculations here, but we need to keep this mind, otherwise no genuine historian will (rightly) take heed of over-generalising , speculative conclusions from "genetics".

Simon_W said...

@ Colin Welling

The ultimate origin of R1b presumably lies in West Asia south of the Caucasus. The question is: Did it arrive in central Europe via the Yamnaya culture? Or rather directly from eastern Anatolia, for instance?

A-DNA seems to suggest that the eastern Bell Beaker folks were indeed rich in R1b. However, the typical physical type associated with them, i.e. a Dinaric skull, fits much better to a direct West Asian origin. This type of skull simply isn't the type associated with the Yamnaya people. On the other hand people looking similar like this are very often found in highland West Asia and areas nearby, like bronze age Cyprus.

Furthermore, judging by the modern haplogroup frequencies, the ancient Italics were not that overly dominated by R1b as the modern Celts, or even their central European descendants in general, are. Further, I note an incidence of about 4% R1a1a in central and southern Italy. To me it seems like a very plausible explanation that this, for the most part, is the Yamnaya input inherited from the Balkans.

On the other hand, the Ladin area in northeastern Italy and also the Tyrol region in Austria are much richer in R1b, and in eastern Tyrol for instance, R1a is only found in those areas, where the Slavs used to settle – in the others it isn't found. Now, this whole area used to be inhabited by the Raeti, a non-IE people.

In the comment section of this blog entry
http://dienekes.blogspot.ch/2013/05/genetic-structure-and-different.html
I explained why I regard the Bell Beaker East group, at least in the south, along the Danube, as presumably proto-Raeto-Etruscan.

Finally, to me there's also an open chronological question mark. At the time, when the eastern Bell Beaker culture started, around 2500 BC, were there not late Yamnaya groups in the Carpathian basin? In other words: Were these not two seperate, coexistent phenomena? I think this should be resolved first with up-to-date C14 datations.


@ Va_Highlander

The report isn't more specific, unfortunately. By the way, maybe it matters that the German terms for „Bactrian camel“, i.e. „zweihöckriges Kamel“ or „Trampeltier“ don't contain any reference to Bactria. The text seems to suggest that previously it was assumed that the domestication of this species occured somewhere around Mongolia or China. Literally it says, very unspecifically, „in the east“. But since the new genetic study showed that the wild Bactrian camels living there are very different from the domesticated ones, the text concludes, again very unspecifically, that the domestication must have happened „in the southwestern steppe areas“ (literally).


@ Kurti

A literal translation of the paragraph in question:

„In spite of the genetical differences among the investigated groups, these are populations which can doubtlessly be characterized as European. DNA-markers with known phenotype point to a continuity between the northern Pontic area of the 4th/3rd millennium BC and modern Europeans. Thus, for instance, all examined individuals have a fair type of pigmentation, like what predominates in Europe today. Merely the eye colour was darker in comparison to the present.“

Admittedly, it could be interpreted in the sense that modern Europeans are descended from these north Pontic groups. But personally I don't believe that this was what the author was trying to say. At least it would be a disputable conclusion if it was based on nothing more than markers with known phenotype.

We'll have to wait for more details to be able to judge for ourselves.

Grognard said...

"But isn't the study actually saying that the genetic make up of these grups west and north of the Pontic do resemble modern day Europeans? So basically the end product of European Indo-Europeans. Isn't this exactly what I am saying? "

Yes basically, assuming my ability to read through the babble fish soup is accurate.

"The text doesn't deal with Persians, nor with Phoenicians. And while the iron age Scythian/Saka groups around the Altai were indeed mixed with central/East Eurasian and West Eurasian haplogroups, there are no central Asian mt-DNA haplogroups in the examinated chalcolithic/bronze age populations north of the Pontic."

That's where scythians came from, that's not in debate really.

This study seems to be saying that their dna is what became modern central europeans.

As for "central asian" dna and pontic that's exactly what the study has said.

Modern "central asian" DNA is not the same as back then. We now know what the scythians were, which we didn't before (though this was the obvious culprit and fits in with all historic sources). They were already in persia previously so that's the end of speculation.

It would be interesting to see indus/gedrosian/vedoid DNA from pre bronze age, but we don't have it. I can make a guess, though, and expect that to come out in not too long.

Dr Rob said...

Wow ! So much speculation based on patchy and unfinished evidence. Some people need to learn that **haplogroups and language spread are two very different phenomena.**

SO: even if we establish that there was a generalized shift to R1a in central and eastern Europe, and central Eurasia, during the Copper Age -Bronze Age transition; and further prove that this shift was , in fact, due to demographic reasons (ie people moving, migrating, multiplying) rather than accelerated drift or natural selection, then this still tells us nothing about the spread of Indo-European languages. Language spread can be orthogonal to, or even counter-current to, demographic spreads.

If some of the commentators here were proper scholars publishing their 'theories', it would undoubtedly set back rather than improve our level of understanding. Thankfully, they're not.

eurologist said...

"...that European Indo Europeans have their roots ultimately in the Steppes of the Pontic sea...."

But before the climate turned dryer, much of that land to the W and NW of the Pontic was not steppes at all, but well-used farmland - and the inhabitants were clearly close to LBK-derived Europeans, with perhaps somewhat similar admixture of native HGs.

I believe that's why they were so similar - the real questions are when and where did the West Asian element enter (Bronze Age, mostly via the SE / Balkans / Mediterranean?), and when did R1a and R1b enter (perhaps much earlier? R1b with La Hoguette from S France; R1a when LBK and W Pontic agriculturalists met with Ukrainian HGs?)

Grognard said...

"As ive mentioned before , it is worrying how some people attempt to equate language, culture and genetics into one neat picture. Aside from the fact that we don't even have anywhere near the whole picture , we must rememeb that all of the above variables are not co-eval. rather, the direction of population movement and language expansion might be orthogonal or even run counter-current(!) thus, even of we demonstrate an R1a shift during the bronze age from central Europe to Siberia, and we prOve this originated on Southern Russia, AND we prove that this was indeed a demographic phenomenon and not some accelerated drift or natural selection phenomenon, then it still would not PrOVE that IE originated in EE. "

That's why papers like this are done now we know the genetics come from scythians and that's the important thing.

IE language itself doesn't matter and who knows where scythians got it, assuming that's what they really spoke which seems likely but is hard to be 100% on.

Grognard said...

"Wow ! So much speculation based on patchy and unfinished evidence. Some people need to learn that **haplogroups and language spread are two very different phenomena.**

SO: even if we establish that there was a generalized shift to R1a in central and eastern Europe, and central Eurasia, during the Copper Age -Bronze Age transition; and further prove that this shift was , in fact, due to demographic reasons (ie people moving, migrating, multiplying) rather than accelerated drift or natural selection, then this still tells us nothing about the spread of Indo-European languages. Language spread can be orthogonal to, or even counter-current to, demographic spreads.

If some of the commentators here were proper scholars publishing their 'theories', it would undoubtedly set back rather than improve our level of understanding. Thankfully, they're not. "

I have an uncanny knack for being proven right.

Which this article is nice proof for.

The IE stuff is a red herring, but that's why we have the genetic testing. Which goes along with what I always thought, which is the obvious, and is supported by every historical source that we have, which all seem to agree. It should not be a shocker to think that central europeans are descendants of the scythians. We can probably just put that down as fact now.

My guess is that the r1ba turned north of caspian sea and the west european branch turned south.

There's endless expansions out of asia all through history, and that's likely to have been true for longer than recorded historya s well. Caspian sea isn't the source of anything, just a stopping point.

As for people who comment on papers, since anthropologists don't seem to believe in math or evolution I am not sure that's an insult.

Simon_W said...

Dr Rob, I do have the impression that with your latest critique you're especially referring to my comments. But this misses the point. First of all, in my reply to Colin Welling my point was actually to call into question the theory that R1b (M269 and M73) was the PIE haplogroup and that it was spread by the Yamnaya people. Pnuadha elaborates on this theory on his blog, and Colin Welling keeps pointing it out here. (Pnuadha pretends that it's his own ingenious theory, but in fact this is exactly what has been promoted by eupedia for a very long time.) It's a theory that must be particularly appealing to R1b males with Kurganist opinion. Nonetheless it's of course wholly legitimate to support such a hypothesis. But when I see the confidence with which some people are supporting it, I can't help pointing out it's weak points. So rather than presenting an own theory about PIE and „their haplogroup“, I was trying to damage one made by others.

Now to some more general remarks to your sceptical concerns.

A lot has been written on ethnical interpretations of archeological cultures. My own view on the topic is that archeological cultures in the narrow sense, i.e. groups which really share most of their everyday pottery (not just a few high prestige objects), which have common funerary rites, common types of houses etc, are signs for tight communication networks. As such they make the presence of a common language likely. So, to me a homogenous archeological culture is rather a consequence of some linguistic homogenity, rather than the other way round.

Now it would be foolish to assume that each language must equal a specific y-haplogoup. Or that the speakers of any particular proto-language had to be genetically pure on the paternal side. Or that the same haplogroup cannot be found at high incidence in peoples with unrelated languages.

But certainly peoples (at a given time) can be characterized in terms of their haplogroups. Some are dominated by just one haplogoup, others have a particular blend of several haplopgroups which are common in them.

And languages are, especially in prehistoric times, bound to living people. And they were spread by people. I have a hard time imagining a language expanding without at least some people expanding together with it. Sure, a culture with a given language may attract and successfully assimilate a large number of foreign immigrants, but this wouldn't really be an expansion of the language, rather the maintenance of a steady state. And granted, when we reach the time of the Roman empire with it's highly developped state and administration, then this coeval expansion of people reaches a level that is probably no longer detectable by genetic means. But for the most part of prehistory we can expect at least some genetic trace of the original language bearers. However, this original impulse may have become progressively diluted in the course of the language expansion, during which it got adopted by people of different ancestries who became the main bearers in the next steps of the expansion. Actually, if you want to criticize someone for naive assumptions then your critique would apply better to the model proposed by Pnuadha who suggests that the original PIE people spread like the impulse of a wave, rushing through the whole continent, to end up the furthest away from the PIE Urheimat, at the western edge of Europe, in no way diluted, but even concentrated. (Still, I think it has to be unprohibited to discuss and support even models like this.)

Very often it's hard to decide by archeological means alone if certain changes in material culture took place because of a migration of people, or if only the ideas spread. Therefore it's really great that we now have the additional tool of genetics which brings new light into many debates that went on for decades without reaching a conclusion.

Simon_W said...

And to solve the problem of the origin and spread of the IEs we need all available evidence, liguistic, archeological, genetical, anthropological. The first priority must be assigned to linguistic opinions. But obviously linguists don't agree on the place of the PIE homeland; therefore other categories of evidence come into play and we have to balance the arguments. In no way I'm suggesting that we can find the origin by merely picking out one widespread y-haplogroup and detecting its origin. This would be silly.

As for the question how much the observed shifts in haplogoup frequencies are due to drift or selection: I think R1b and R1a may well have some selective advantage, but this advantage obviously didn't enable them to predominate in southeastern Europe, for instance, even though they're present there too, probably for millennia. The selective advantage cannot have been that decisive.

There's also the rather philosophical question how sceptical you choose to be and how firm a proof has to be until you accept a certain theory. Since the prehistorical cultures in question left no written evidence of their language, there will probably always be room for doubt. But generally I don't think science is about gaining certitude. Scientific theories should be based on the best of our current knowledge, but they are provisional and may change in the future. Keeping this in mind, I really have no problem with speculative theories, I never mistake them for firm knowledge, and when new evidence forces us to change our theories, I'm not disappointed, but excited. There are many rather speculative fields in science (e.g. cosmology, particle physics) and subjects where only very patchy, unfinished evidence is available, where one yet has to try to make the best out of it (e.g. paleoanthropology).

That said, you are right, we are just thinking and discussing speculatively, more or less trying to back our claims up with arguments. In the end it's the quality of the arguments that decides whether a proposal seems promising or not, I think every genuine historian knows that.

Dr Rob said...

Grognard, this a friendly forum which invites well-intentioned speculation. But yr comments like "endless expansions out of Asia" might well be correct, but are ultimately meaningless and derived from the historical hangover of the Mongol expansion. Even as late as 2000 years ago, things were far more nuanced than that. Expansion in Eurasia has more often been in north to south direction than sweeping east to west . So you commit the same sins as those who advocate a sweeping migration of blue eyed tall Europeans conquering large parts of Asia

Va_Highlander said...

Simon_W:

Many thanks. I find it somewhat bizarre that this article might assume that the Bactrian camel was domesticated in China or Mongolia. I've never encountered that suggestion before now and know of no archaeological evidence supporting such a claim.

"My own view on the topic is that archeological cultures in the narrow sense, i.e. groups which really share most of their everyday pottery (not just a few high prestige objects), which have common funerary rites, common types of houses etc, are signs for tight communication networks. As such they make the presence of a common language likely."

That archaeological cultures may be so narrowly and reliably defined is a somewhat dubious proposition in itself, given the inherent subjectivity of such definitions and the sometimes non-uniform adoption of cultural traits. But even if possible, a common language might represent a native tongue or it might merely represent a lingua franca. The mechanisms by which these two types of language expand or shift can be radically different.

Or consider the Old-Assyrian trading colonies of Anatolia. We know that they were colonists because they left behind textual evidence permitting us to identify their written language and place of origin. However, their settlements and material culture are essentially indistinguishable from those of native inhabitants.

While the model you're proposing might seem entirely plausible, as an intellectual exercise, empirical observation informs us that the real world has seldom been so simple.

Valikhan said...

What I think is that IE language spread from Tripolye culture. Tripolye people went as far east as Volga river basin, bringing Tripolye artifacts to there and probably were ancestors of those modern hg E folks.
R1a more likely were Eastern Europe aborigines, who adopted IE language and culture, including wheeled carts, horses, bronze work. What language they spoke who knows, it might be one or many languages of the so called Sino-Caucasus family.
I am not sure if the traditions of homosexual rituals were of Tripolye people or of native R1a folks.

http://csen.org/Articles_Reivews/Kangjiashimenji%20/Kanj-Text/Kanj.html#fn1

Simon_W said...

Va_Highlander, I think it varies; some archeological cultures are virtually homogenous along the entire range of categories, others are nothing more than pottery styles, again others meet some of the requirements, but not all of them. The Corded Ware for instance has a uniform burial rite, but the material culture varies by region; there are just some supra-regional objects which are spread across its entire range. I would suggest that the more homogenity is apparent, which is more often the case in small scale groups, the more likely is a common native language. Mixed cases like the Corded Ware are hard to interpret. There may have been a common lingua franca, or the local languages may even have been related. I know, there is the problem if archeological cultures can be objectively defined at all, and associated with it the problem that different types of finds do not always overlap. Sometimes, e.g. cultural provinces based on metallurgy differ a lot from the ones based on pottery and funerary customs, so the question arises, which aspect of the material culture is more closely tied to ethnicity.

Your Assyrian example reminds me of another one that is often cited in this context: The migration of the Scots from Ireland to Scotland; apparently this cannot be verified in the archeological record. Maybe the cultures involved were to similar? On the other hand the migrations of the Germanic tribes to Britain, or to Switzerland for instance are well established archeologically and pose no problem.

Dr Rob said...

No Simon W, I wasn't referring to you. You've always made sensible points !

A major problem with haplogroups is that they are a palimpsest, as we all know. And at the end of the day - all they can tell us is **possible** demographic movements. And unless we are to suppose that the "proto-Indo-Europeans" were moving into virgin territory or involved in large-scale population replacement (of which I'd take the PIE expansion was neither), I think we'd be very very ambitious indeed to expect to shine linguistic light from them.

This is not to make the whole exercise useless, but genetics should primarily be used to reconstruct local and regional demographic events before trying to make 'globalizing hypotheses'. Baby steps before marathon running; not vice -versa. :)

Dr Rob said...

And further reason for skepticism: apparently these scholars don't know anything about archaeology, for they claim (if the translation is in fact correct) that arcahaeology suggests that Yamnaya had 'eastern steppe' roots, which is refuted by their DNA evidence. I have not heard of a single source which claims an eastern (ie Oriental) origin for the Yamnaya type assembleges.
Geneticists really should just stick to genetic analyses, and leave the syntheses to more well-rounded people, becuase they wouldnt pass "History 101. "

Colin Welling said...

@simon

A-DNA seems to suggest that the eastern Bell Beaker folks were indeed rich in R1b. However, the typical physical type associated with them, i.e. a Dinaric skull, fits much better to a direct West Asian origin.

My theory is primarily about ydna, not autosomal DNA, which is what skull studies are best suited for. I don't even think skull type is that reliable for the given autosomal context.

judging by the modern haplogroup frequencies, the ancient Italics were not that overly dominated by R1b as the modern Celts, or even their central European descendants in general

Mute point

Further, I note an incidence of about 4% R1a1a in central and southern Italy

Most people agree that the indo European influence in Italy came from the north. So the clines you point out actually support my theory; that is, unless you priori assume I'm wrong and that r1a is an PIE marker but not r1b...

On the other hand, the Ladin area in northeastern Italy and also the Tyrol region in Austria are much richer in R1b, and in eastern Tyrol for instance, R1a is only found in those areas, where the Slavs used to settle – in the others it isn't found. Now, this whole area used to be inhabited by the Raeti, a non-IE people... I explained why I regard the Bell Beaker East group, at least in the south, along the Danube, as presumably proto-Raeto-Etruscan

So your saying it's more believable that Raetians left a greater concentration of r1b north of the Alps than Italo-Celts. OK...

when the eastern Bell Beaker culture started, around 2500 BC, were there not late Yamnaya groups in the Carpathian basin? In other words: Were these not two seperate, coexistent phenomena?

The yamnaya arrived at upper Danube area over a period of time.It wouldn't be a paradox mixed yamnaya and typical yamnaya coexisting.

Are there not mixed Mexicans and pure Mexicans coexisting in the USA?

Pnuadha who suggests that the original PIE people spread like the impulse of a wave, rushing through the whole continent, to end up the furthest away from the PIE Urheimat, at the western edge of Europe, in no way diluted, but even concentrated. (Still, I think it has to be unprohibited to discuss and support even models like this.)

Ya, you don't understand the model. That's why I didn't bother responding to your earlier post.

By the way, we both think that r1a traveled to Western Europe from the east, where the concentration of r1b is lower. That means you also believe that r1b concentrated in the West, far from its origin.

I could give a more nuanced answer, but you probably wouldn't get it.

Pnuadha elaborates on this theory on his blog, and Colin Welling keeps pointing it out here. (Pnuadha pretends that it's his own ingenious theory, but in fact this is exactly what has been promoted by eupedia for a very long time.)

Eupedia and I promote the idea that r1b spread to the west with yamnaya/PIE groups. But unlike eupedia, I argue that the yamnaya/ PIE were dominated by r1b and largely lacking m417. Furthermore, I think that m417 was largely introduced to the PC steppe from eastern central Europe post PIE.

Those are pretty big differences. If you can't notice that then you're really not one for nuance.

Va_Highlander said...

Simon_W, I think we are in agreement, for the most part. The frustrating thing is that it seems far easier to pick apart an argument for associating a given material culture with a particular language than it is to establish such a connection to the satisfaction of most observers. What we usually have are competing narratives, with plausibility depending too often upon the mindset of the observer more than the qualities of the observed.

Colin Welling said...

correction

By the way, we both think that r1b* traveled to Western Europe from the east, where the concentration of r1b is lower. That means you also believe that r1b concentrated in the West, far from its origin

Simon_W said...

Dr Rob, I'm not convinced that for geneticists it would be more reasonable to limit themselves to the reconstruction of local and regional demographic events. On the one hand, because the data is so scarce and scattered across big areas, on the other hand because the global genetic context, the phylogeographical background, is needed to understand the dispersal of haplogroups.

And while the haplogroup frequencies of modern populations may indeed be compared to a palimpsest, a-DNA gives us a direct glimpse of the past.

As for the **possible** demographic movements, sure, they're not certain, but still, I found e.g. the map with the dispersal of haplogroup N lately quite convincing, on the whole. There may be disagreement about details.

I don't know if the subject of haplogroups will be decisive in the end, when dealing with IE origins, or if it will be necessary at all. At least, the distribution and history of haplogroups strengthens the position of some theories and weakens the position of others. But R1a1a and J2a are both very widespread in the IE world, and both would be in line with different important theories about IE origins, so they can't be decisive on their own.

I think in every subject, be it archeology, linguistics or genetics, people should primarily base their results and assertions on their own type of evidence. Thus, it should become possible to describe for each archeological culture by which other cultures it was influenced, and which cultures it influenced in return. And linguists could tell us which other languages the speakers of PIE had contact with, or what environment and culture they were familiar with. And genetics could describe the actual movements of people, the geneflows.

The results of genetics (and prehistoric archeology) per se don't tell us anything about the language of the people involved. Nonetheless I think they're very interesting on their own! And in the end they may flow into a synopsis...

And further reason for skepticism: apparently these scholars don't know anything about archaeology, for they claim (if the translation is in fact correct) that arcahaeology suggests that Yamnaya had 'eastern steppe' roots, which is refuted by their DNA evidence. I have not heard of a single source which claims an eastern (ie Oriental) origin for the Yamnaya type assembleges. 

I think it was Marija Gimbutas who claimed something like this, that Yamnaya had its roots in the Khvalynsk culture, on the lower Volga. This however isn't the mainstream view, afaik.

Finally I would like to say that I cannot subscribe to the criticism of this genetic study in question, of which the present article is just a preview, after all. The article doesn't even touch upon the topic of PIEs, and IMO it doesn't intend to say that modern Europeans are descended from north Pontic chalcolithic groups, I think this is just formulated unhandily. The first sub-project is just about the chalcolithic and early bronze age groups north and west of the Pontic, so quite on a regional level, just as you demanded. We shouldn't mistake the assertions made by the article with the farther reaching conclusions made by some comentators here. Personally I'm glad this study was done and can't wait to see the genetic findings in more detail. In particular, it's going to be interesting in what way the Catacomb culture differed from the other ones...

Simon_W said...

@ Colin Welling

R1b-M269 and -M73 are descended from R1b-P297. If you look where the two sister clades of P297 occur, that is M335 in Anatolia and V88 in the Levant and northern Africa - hence it seems most likely that P297 originated somewhere in this Near Eastern area as well. From there it may possibly have crossed the Caucasus to flourish in the Yamnaya culture, but we don't know for sure. I know well that y-DNA has no influence on skull type. But as I said, the question seems to be if M269 reached central Europe via the Yamnaya people or directly from West Asia, and the skull type of the eastern Bell Beaker people, who evidentially seem to have been rich in R1b, favors a direct Near Eastern origin. I think this is an important point. Or do you think the accordance is merely a coincidence?

You didn't understand that genetic study on Italy. It found two principal components for the y-DNA. One seperates Sardinia from the rest. The other principal component has a northwestern pole and a southeastern pole, but not with much clines inbetween, rather with a narrow border belt. No matter what „most people“ think about the origin of the Italics, the truth is: The Italics strongly correlate with the southeastern pole. Correlating with the northwestern pole are the Celts and Ligurians, but also the Etruscans. Upon closer examination it turns out that the northwestern pole is strongly associated with R1b-S116 and R1b-U152. So rather than calling U152 an Italo-Celtic marker it would be more appropriate to call it Etrusco-Celtic. In fact, in this study it even peaked in Tuscany, not in the Gaulish-Ligurian northwest. And archeology tells us that the Protovillanova culture, with its indubitably northern origin, had its most lasting impact in Etruria. It was probably Etruscan. The Italics must have been already present in the middle bronze age Apennine culture.

And by the way, R1a1a is actually more frequent in the centre and south than in the north.

So your saying it's more believable that Raetians left a greater concentration of r1b north of the Alps than Italo-Celts. OK...

Exactly, or more precisely the common source population of Raeti, Etruscans and Lemnians. As long as you don't assume central Europe to be the PIE homeland, there must have been a pre-IE substrate population there. And just as the PIE impact in, e.g., Sri Lanka was probably much less biological than linguistic, that may be the case in central and western Europe as well.

Ya, you don't understand the model. That's why I didn't bother responding to your earlier post.
By the way, we both think that r1b traveled to Western Europe from the east, where the concentration of r1b is lower. That means you also believe that r1b concentrated in the West, far from its origin. 


But you seem to suggest that already the Yamnaya people were a large mass of R1b folks (just as Western Europeans today are) who merely migrated through the continent. I on the other hand think there may have been drift and founder effects involved, and the origin of R1b-M269 and -M73 may have been in rather small, unspectacular groups.

Furthermore you have to admit, it's counter-intuitive. In case of the Slavic expansion for instance, no one would suggest that the purest descendants of the proto-Slavs are to be found in northern Russia and the southern Balkans.

Simon_W said...

Eupedia and I promote the idea that r1b spread to the west with yamnaya/PIE groups. But unlike eupedia, I argue that the yamnaya/ PIE were dominated by r1b and largely lacking m417. Furthermore, I think that m417 was largely introduced to the PC steppe from eastern central Europe post PIE.

Well, the last time I read their theories, they suggested that R1b dominated in the southern steppe zone, and R1a more in the forest steppe to the north. Also one of their maps clearly showed that, in their mind, it was the Catacomb culture which re-introduced R1a southwards, and that it was also predominant in the Corded Ware. Even if it's not exactly the same theory as yours, it's in any case very similar.

Colin Welling said...

the question seems to be if M269 reached central Europe via the Yamnaya people or directly from West Asia, and the skull type of the eastern Bell Beaker people, who evidentially seem to have been rich in R1b, favors a direct Near Eastern origin. I think this is an important point. Or do you think the accordance is merely a coincidence?

Once again, my theory is primarily about ydna, not autosomal DNA! I make little concern about the autosomal composition of the Yamnaya to Beaker to West Europe movement, much less skull type... ydna does not remotely equate to skull type. The "concentration" of r1b in the West makes this painfully obvious.

As a side note, genetic studies have inferred a North European like autosomal shift in Iberia during the late Beaker phase.

The Italics strongly correlate with the southeastern pole. Correlating with the northwestern pole are the Celts and Ligurians, but also the Etruscans.

You suppose that northern Italy is non italic then "discover" that italics correlate with the southern half of Italy. That's absurd reasoning.

But you seem to suggest that already the Yamnaya people were a large mass of R1b folks (just as Western Europeans today are) who merely migrated through the continent. I on the other hand think there may have been drift and founder effects involved, and the origin of R1b-M269 and -M73 may have been in rather small, unspectacular groups.

You are not making a point.

Well, the last time I read their theories, they suggested that R1b dominated in the southern steppe zone, and R1a more in the forest steppe to the north. Also one of their maps clearly showed that, in their mind, it was the Catacomb culture which re-introduced R1a southwards, and that it was also predominant in the Corded Ware. Even if it's not exactly the same theory as yours, it's in any case very similar.

He still shows Yamnaya as, geographically speaking, being significantly composed of r1b and r1a (m417 presumably). He also argues that r1a peoples were a significant component in the formation of the yamnaya culture. [Actually, I think he used to show the southern yamnaya as being mixed with r1b and r1a, while the northern yamnaya were shown as dominant in r1a.]

So the guy from eupedia thinks that the yamnaya were a mix of r1a and r1b from the beginning. I think the yamnaya were dominated by r1b and largely absent of r1a, especially m417.

He thinks m417 was largely introduced to eastern-Central Europe by early, steppic, IE peoples. I think m417 was largely introduced to the post PIE steppe by eastern-Central Europeans. Thus we have m417 as being seeded in different cultures and initially having crossed the steppes in opposite directions, at different times.

Those are pretty significant differences/predictions for an IE theory.

Dr Rob said...

@ Simon W

Of course this study is useful, I was just questioning their archeological correlations - in which they claim 'eastern roots' for the yamnaya culture. All readings argue a local development from preceding hunter-gather groups, or at beast secondary colonization from the Tripolye, etc

Simon_W said...

@ Colin Welling

Y-DNA may be quickly decoupled from autosomal DNA, if the incoming males prefer local women for a couple of generations. But in the very beginning, the guys who had actually brought the haplogroup to central Europe will have had the autosomal make-up of their homeland. So if they had come directly from the Near East to central Europe, the first ones will have had a Near Eastern autosomal make-up. (I know, I know, you're saying the Yamnaya autosomal make-up may have been strongly diluted already, when they reached central Europe, but that's not the point.) The typical skull type of the eastern Bell Beaker people matches their apparently Near Eastern haplogroup. Do you think this is a coincidence? If their R1b did in fact come from Yamnaya steppe people, the eastern Bell Bearker people's skull type would have to have originated in a population with different haplogroups, but none would fit as well as R1b, except J2 possibly, but the ones tested so far didn't have J2.

As a side note, genetic studies have inferred a North European like autosomal shift in Iberia during the late Beaker phase.

From what I've read this wasn't the late Beaker phase, but the middle bronze age.

You suppose that northern Italy is non italic then "discover" that italics correlate with the southern half of Italy. That's absurd reasoning.

LOL, no this isn't absurd reasoning, you're just blatantly ignorant: Northern Italy was inhabited by:
- Ligurians (have been stated to be non-IE, but are now usually seen as „para-Celtic“)
- Celtic tribes (Gaulish and Lepontic)
- Raeti and Etruscans (related, non-IE)
- Veneti (close to Italic, but with other affinities, possibly with Illyrian, certainly with Germanic)

The Italics on the other hand were most of all the Oscan-Umbrian group of tribes and the Latino-Faliscan group. Maybe South Picene was a third group, but I think I read it was close to Oscan-Umbrian. In any case, these Italic tribes were in the centre (excluding Etruria) and south of Italy. The Umbri were the northermost of them.
These are the facts we know from ancient ethnography.

Finally some general words about your theory: It's actually a composite of two independent sub-theories:
1. That PIE and/or the centum languages are associated with R1b (M269 and M73)
2. That the Yamnaya people were dominated by R1b and actually the main source of European R1b.

In theory it's possible that only one of the two assertions will be proven to be true. Yet, at the moment I find neither very convincing.

I also have the feeling that there is some inconsistency in your theory. You suggest that the PIE were R1b and the satem languages were R1a. But the satem languages are descended from PIE too! So you're suggesting that in the case of the Centum branch, the expansion was mostly effected by the original PIE people, without much admixture from the substrate populations (as far as y-DNA is concerned). In the case of the satem languages on the other hand, you're advocating the opposite: R1a people adopted IE languages virtually without admixture from the original PIE.

Slumbery said...

Simon_W

Bell Beaker is later in Central Europe than in the West. It spread west to east in this region. This is a problem for the Pit Grave > Bell Beaker theory.

Middle Eastern is not a homogeneous type of people, especially not if time is a factor too. Do the Beaker people actually have detectable similarity to specifically to Pit Grave people?

The distribution pattern of the relevant R1b variants is not very friendly to this idea either.

I can't see any data that would clearly support such a connection.

Lancesf said...

Looks like the paper is out.
http://link.springer.com/article/10.1007%2Fs00439-009-0683-0

Lancesf said...

Looks like the "Siberian Ice Princess" was included in the study. It appears she did not have any East Asian markers.
http://siberiantimes.com/culture/others/features/siberian-princess-reveals-her-2500-year-old-tattoos/

Dienekes said...

>> Looks like the paper is out.
http://link.springer.com/article/10.1007%2Fs00439-009-0683-0

That is an unrelated 2009 paper.

Simon_W said...

Slumbery,

Yes indeed, the Bell Beaker package originated in the Iberian peninsula, afaik, at least most of its main characters. And that mt-DNA study lately has proven that there was also some geneflow from Iberia associated with it.

However, it was quite a superficial spread of just a few typical items, like e.g. the maritime Bell Beaker. The domestic, utilitarian pottery (or „Gebrauchskeramik“ in German) of the East group for instance, cannot be derived from Iberia. Accordingly, the study by J. Desideri on tooth morphology showed that at least in the East group, there was also some biological population continuity.

According to A. Gallay the domestic ware of the Bell Beaker East group can be derived from the Vucedol culture. (As a side note I guess this also explains why M. Gimbutas characterized the Bell Beaker culture complex as an amalgam of Vucedol and Yamna traditions. Those alleged Yamnaya influences however might be rather Corded Ware influence.)

Vucedol developped from the Baden culture. I think it was basically a late phase of Baden. And Baden had connections to eastern Anatolia. By the way, there is also a recent study by Joachim Köninger, Martin Kolb and Helmut Schlichtherle showing that there were elements from Baden and Boleraz in the humid soil settlements of southwestern Germany and that these possibly had some bearing on the transformation process of the local aeneolithic.

It's gonna be interesting what this anthropological study on the large Bell Beaker cemetery  at Hoštice za Hanou in Moravia is going to find:
http://www.muni.cz/press/research/projects/1863

The earliest findings of the Baden culture are from Moravia.

Middle Eastern is not a homogeneous type of people, especially not if time is a factor too. 

True, actually the short-/high-vaulted „Taurid“ types (like the modern Armenoids and Pamirids etc) are quite elusive in prehistory, in the Middle East, too. Most of the ancient Middle East was dominated by longheaded peoples. In the western part there was also an Alpinoid tendency, at times stronger, then weaker again... There is one early and clear centre of a Taurid type known, that is in bronze age Cyprus. I think in western Anatolia, there was a diluted influence. Compared to many modern samples these bronze age Cypriots had a somewhat less elongated upper face. The earliest known cranial sample of a Middle Eastern Taurid type with elongated upper face was from iron age Tepe Sialk in Iran.

Do the Beaker people actually have detectable similarity to specifically to Pit Grave people? 

If you mean the Bell Beaker people from western and central Germany and the Czech republic, the answer is a clear no. Rather they had some similarity with the TRB in Denmark. Maybe a consequence of intermixture, as Coon guessed.

Colin Welling said...

Y-DNA may be quickly decoupled from autosomal DNA, if the incoming males prefer local women for a couple of generations. But in the very beginning, the guys who had actually brought the haplogroup to central Europe will have had the autosomal make-up of their homeland. So if they had come directly from the Near East to central Europe, the first ones will have had a Near Eastern autosomal make-up. (I know, I know, you're saying the Yamnaya autosomal make-up may have been strongly diluted already, when they reached central Europe, but that's not the point.) The typical skull type of the eastern Bell Beaker people matches their apparently Near Eastern haplogroup. Do you think this is a coincidence? If their R1b did in fact come from Yamnaya steppe people, the eastern Bell Bearker people's skull type would have to have originated in a population with different haplogroups, but none would fit as well as R1b, except J2 possibly, but the ones tested so far didn't have J2.

You're still not understanding that my theory is primarily about ydna, not autosomal DNA. The question of whether or not there was direct autosomal influence from West Asia during the genesis of the eastern Beakers has no effect on my theory which asserts what happened with ydna.

That said, it is most certainly the case that r1b has been decoupled from autosomal DNA given the high concentration of r1b on the Atlantic fringe, as is repeatedly the case for other haplgroups. In fact the East-West divide of European ydna doesn't match the North-South autosomal divide. Now, I don't know if the Eastern Beaker skull type reflects a large amount of West Asian autosomal input, a merger of local neolithic elements and yamnaya, a population X which migrated north and south (to West Asia), the yamnaya themselves with genetically or culturally driven skull shape changes, or something in between the above suggestions.

But inferring the autosomal composition of ancient populations, to the resolution you are talking about, is generally very difficult and inaccurate. Then, to try and coordinate these inferences with the story of ydna amounts to drastically complicating the story and adding inaccuracies.

I opted to not make autosomal DNA an important part of my theory, both in terms of explaining the autosomal situation during that time period and in using autosomal DNA as supporting evidence. As such, autosomal DNA is largely outside the scope of my theory, though not entirely... So if you are wondering whether or not the yamnaya could have formed an important component of the Eastern Beakers when considering the available autosomal evidence, the answer is most certainly yes. It may even be suggested. In the late Neolithic much of Europe, including a good deal of Western Europe, seemed to be "otzi-like". There is a very good chance that Europeans became less otzi-like and more North European-like during the migrations of the metal ages. One of the best candidates for this transition is Western Europe is the Eastern Beakers (accompanied by ydna, mtdna, and language shifts). The inferred North European-like shift in Spain around 2k bce fits with the timing of Eastern Beaker influence in what's now Spain. The yamnaya were almost surely North European-like and they did settle in the same region where the Eastern Beakers set out from. Archeological and isotopic studies both point to this event. Heyde also claimed yamnays input. While I believe the possibility of yamnaya input into Eastern Beakers on the is certainly plausible on the grounds of autosomal, even more so than the evidence for direct autosomal input from West Asia, I don't pay too much attention to either. My theory is primarily about the travels of ydna.

Colin Welling said...

Continued...

The Italics on the other hand were most of all the Oscan-Umbrian group of tribes and the Latino-Faliscan group. Maybe South Picene was a third group, but I think I read it was close to Oscan-Umbrian. In any case, these Italic tribes were in the centre (excluding Etruria) and south of Italy. The Umbri were the northermost of them.
These are the facts we know from ancient ethnography.


The problem is that you assume Italic wasn't ever in Northern Italy. An even bigger problem is that you decouple the main IE influences in Italy, Celtic and Italic, then somehow suggest that it supports the r1a PIE theory over the r1b PIE.

Most linguists place Italo-Celtic, whether it be a language or a continuum, around the upper Danube region, whereby Italic shifts/ moves southward via North Italy and Celtic stays in the vicinity to exchange influence with Germanic.

Finally some general words about your theory: It's actually a composite of two independent sub-theories:
1. That PIE and/or the centum languages are associated with R1b (M269 and M73)
2. That the Yamnaya people were dominated by R1b and actually the main source of European R1b.

In theory it's possible that only one of the two assertions will be proven to be true. Yet, at the moment I find neither very convincing.


You do realize that I wasn't really trying to convince you? My theory would not be broken or solidified by these criticisms. No theory of the same nature looks great with every individual price of evidence; it would be amazingly lucky for all events to turn out just the way we expect. It all comes down to choosing the best sufficient theory IMO.



I also have the feeling that there is some inconsistency in your theory. You suggest that the PIE were R1b and the satem languages were R1a. But the satem languages are descended from PIE too! So you're suggesting that in the case of the Centum branch, the expansion was mostly effected by the original PIE people, without much admixture from the substrate populations (as far as y-DNA is concerned). In the case of the satem languages on the other hand, you're advocating the opposite: R1a people adopted IE languages virtually without admixture from the original PIE.

Essentially, yes. The earliest and thus "purest" IE settlements outside the steppe (Anatolia/Caucasus and Hittite, Altai Mountains and Tocharian, upper Danude and Italo-Celtic) are best connected by r1b ESPECIALLY when you consider how r1b peaks in the first two. Corded-Ware (?) and Satem is only one of these four "early" branchings and it is one step further removed from PIE compared to the rest.

Additionally, I see the ratio of r1a to r1b in Western Europe as far more troublesome for an r1b conversion than the ratio of r1b to r1a in Eastern Europe for an r1a conversion.

How could I believe that r1a Corded Ware, likely a region long dominated by Satem, taught r1b people Celtic. That whole r1a dominated area has nothing but Satem.

Simon_W said...

Again, I understand that your theory isn't about autosomal DNA, I always did! Please stop repeating that I don't. I was just trying to make an argument in favour of a Near Eastern origin of the Bell Beaker people's R1b, i.e. that it seems more likely than a Steppe origin. But thinking about it, I'm not very convinced myself, given the elusiveness of Dinaroid skulls in West Asia at that time, and given the similarity of the German and Czech Bell Beaker skulls with the Danish TRB skulls.

That said, I very much doubt that western European R1b has been completely decoupled from the autosomal DNA of its original bearers. In do see a correlation with the K12b Gedrosia component there. Also, just slightly less so, with the Dagestan-component of earlier analyses.

As for your speculations about the North_European shift, I'm not convinced either. What evidence do we have so far? There is one woman from the Swedish TRB. And there is Ötzi. That's all so far! The latter, even though for sure more Sardinian-like and more „Southern“ than modern north Italians and Alpine people, is still from Italy after all, from southern Europe, where the Cardium pottery once flourished. As to the Swedish TRB, even if it sounds pretentious, this may as well have been expected to have been rather „Southern“, autosomally, because craniometrically it was closely related with the LBK-derived Rössen culture. I strongly suspect that there were at the same time groups which were more North_European like. The Danish TRB for instance, differed craniometrically a lot from the Swedish TRB. It would be rash to suppose that they were autosomally the same. And morphologically we know that it included the massive „Borreby“ type. We may suppose that there were other hunter-gatherer derived pockets around in other places. As user Rokus keeps mentioning in the comments, the late neolithic cultures (beware, sometimes referred to as „middle neolithic“, also by Rokus) differed quite a bit from the eary LBK and LBK-daughter groups. I guess they were the result of a symbiosis of northern and southern influences. For instance, in that recent study on Bell Beaker mt-DNA, it was striking how the central German „middle neolithic“ was already similar to the modern Germans, and how much it differed from the LBK. This middle neolithic is probably late neolithic in different terminology, but one thing's for sure: It was earlier than the Corded Ware and Yamnaya time frame.

Not sure what you're talking about when referring to the influence of Eastern Beakers in Western Europe. The Eastern Bell Beakers were the eastern variant of the Bell Beaker phenomenon. Western Europe had its own variants, not derived from Eastern Beakers. What's true is that the Bell Beaker people from Britain appear to have come from the lower Rhine area, the Zoned Beaker complex, a Corded Ware-Bell Beaker fusion area.

The admixture date for the North European influence in Spain is 3600 BP +/- 400. This translates into roughly 1600 BC. This is the middle bronze age. The Bell Beaker culture in central Europe ended at 2200 BC, with the beginning of the early bronze age. So, it's very anachronistic to explain that North_European shift in Spain with an expansion of Eastern Beakers, at a time, when in central Europe the Tumulus culture was thriving.

Simon_W said...

The yamnaya did settle in the same region where the Eastern Beakers set out from. Archeological and isotopic studies both point to this event.

I don't think so.
First of all: The Bell Beaker package reached the eastern province from the West! The Bell Beaker phenomenon originated in Iberia.
Sure, as I have stated in a previous post, the domestic ware of the BB East group has been compared to Vucedol pottery. Vucedol influence has even been found in southern France. But afaik there was no real Vucedol substrate in the area where the BB East group developped. A large part of it was covered by the Corded Ware instead. Furthremore, from what I've read, the typical domestic ware developped later, after the installation of the Bell Beaker package; it wasn't part of a substrate. So what we've got are possible stylistic influences, nothing more – hardly enough to assume a strong genetic or linguistic impact. There's also the fact that the area that was really occupied by the Vucedol culture barely succumbed to the Bell beaker influence. There was a tiny Hungarian group, clearly intrusive, and then only centuries after the beginning of the BB East group, the syncretistic Maros-Perjamos culture and the Cetina culture.
I would add that furthermore, Vucedol and Yamnaya are quite different cultures which shouldn't be equated.

Heyde also claimed yamnays input.

In the BB East group? Quotation please, I don't believe this.

Simon_W said...

There is no evidence that Italic ever was in northern Italy, prior to the spread of Latin. Anyway, that's irrelevant, because what I noted was an undeniable, striking connection between the areas occupied by the Italics with the southeastern y-DNA pole on the one hand, and a no less striking connection of the formerly Etruscan areas with the northwestern pole on the other. According to the study in question the former shows connections with the Balkans and Anatolia, the latter with central and western Europe. My point was actually the derivation of the Etruscans from central Europe and, presumably, from the BB East group, at least along the Danube. I elaborated on this in the comment section of that Italian study. Actually I do think there was an Italic substrate in Etruria. Ancient sources report of an Umbrian substrate, and in the middle bronze age, this area belonged to the Apennine culture, just like the rest of the peninsula. But that's irrelevant, because later the Etruscans arrived and with them haplogroups which shifted Etruria away from the southeast.

No, it's not a problem to decouple Celtic and Italic, because after all they did split at some point.

And no, I don't subscribe to the R1a PIE theory, I just found it plausible that the R1a1a in central/southern Italy may be Yamnaya derived in the first place.

At the moment I tend to think that the Italo-Celtic nucleus was in the Vucedol culture. So, yes, around the Danube, but not in Southern Germany/Austria. What good arguments do „most linguists“ have for placing Italo-Celtic further west?

As for the correlation between haplogroups and languages, I see it like this:
Anatolian: J2a is strongest there.
Tocharian: possibly or even likely connected with R1b, though that's just an educated guess at the moment.
Italo-Celtic: yes, R1b is strong there, though with different variants, and in Italic J2a is strong too
Basque: strong in R1b
Iberian: ditto
Raetic and Etruscan: ditto

Also I would question the assumption that Italo-Celtic split particularly early from the rest. Obviously according to Ringe et al, that holds true, but the lexical data suggests otherwise.

And I don't think the Corded Ware can be reduced to Satem. Although early Balto-Slavic may well have been a part of it, there's no reason why the same shouldn't apply to early Germanic as well.

Furthermore, while most of the area dominated by R1a is indeed Satem, there is strong admixture of non-Slavic R1a in parts of the Germanic area. (I myself have got one of these.) Also note that Armenian and Albanian, in spite of being Satem-languages, have very little R1a. So the association isn't all that perfect.

But, on a second thought, I have to admit, maybe you're not completely wrong with your theories:
As I wrote, in my view there is a partial correlation between R1b-M269 / -M73 and the Gedrosia component. Also, just slightly less so, with the Dagestan-component of earlier analyses. Accordingly, the Lezgins from Dagestan are relatively rich in R1b.
Furthermore, the rare R1b-M269* haplotypes appear to have the highest STR-variance in Iran.
So maybe, M269 spread from Iran to the Maikop culture (as attested by cultural influence!) and got diffused from there? But this would be earlier than Yamnaya. And I don't believe this was an IE influence. I would deduce European IEs rather from the copper cultures of SE Europe.

Va_Highlander said...

Simon_W:

"Tocharian: possibly or even likely connected with R1b, though that's just an educated guess at the moment."

What is the basis of your educated guess? The Xiaohe mummies, from the Tarim Basin, were R1a1a but of course it is impossible to know what language they spoke and they predate the appearance of Tocharian by around a thousand years or more.

"So maybe, M269 spread from Iran to the Maikop culture (as attested by cultural influence!) and got diffused from there?"

That is an interesting and, at least at first glance, plausible possibility.

Simon_W said...


Having thought about the whole issue again, I came to the following conclusions.

I think Colin is right with his observation that R1b is linked with the earliest splits of IE.

I was wrong with questioning the early split-off of Italo-Celtic. (I was misled by the Bouckaert et al. Paper. The critique by Martin Lewis and Asya Pereltsvaig however has convinced me of dismissing its results.) Afaik Ringe et al. made two different trees of IE, one based on the lexical data, the other on morphological and phonological data. If we take the latter tree, the succession of splits is as follows:

1. Anatolian (dominated by R1b + J2)
2. Tocharian (R1b + J2?)
3. Italo-Celtic (R1b / R1b + J2)
4. Germanic (R1b + R1a (+ I1)) and Albanian (R1b + J2 (+ E-V13))
5. Greek (R1b + J2 (+ E-V13)) and Armenian (R1b + J2)
6. Indo-Iranian (R1a + J2) and Balto-Slavic (R1a / R1a (+ N))

I guess if we put more weight on morphology and less on phonology, then Balto-Slavic would move away from Indo-Iranian and end up between Germanic and Albanian. But in any case, the pattern is striking how the early splits are connected with R1b and J2.

However, this fact only suggests an R1b-rich Yamnaya if one is already convinced of the Kurgan theory! It could, quite to the contrary, also be used as evidence against the Kurgan theory.

As I said, the phylogeography of R1b rather suggests a West Asian origin. And the STR-variance of R1b-M269* points to an Iranian origin. (Maybe it was associated with pre-PIE, i.e. with IE before reaching the PIE homeland, which presumably was more to the west, close to J2.)

There is one important argument against a Yamnaya origin of R1b that I didn't mention yet: It's very likely that the Yamnaya people were rich in the Globe4 Amerindian component. And Armenians have a lot of R1b and a strong Gedrosia component, but their Amerindian component is 0%. Therefore it's extremely unlikely that their R1b has a Yamnaya origin.

Furthermore, the contemporary lack of R1b in the north Pontic area would entail a near complete population replacement, if we assumed that Yamnaya was dominated by R1b. However, according to all morphometrical analyses (by Bunak, by Menk, by Schwidetzky...), there was quite a lot of population continuity in the north Pontic area, starting from Yamnaya down through the ages.

So, I would suggest that Yamnaya was dominated by R1a. The Maikop influence from the south was probably rather associated with J2, as is suggested by the strong diffusion of J2 in southern Russia north of the Caucasus.

It's by the way striking how the West Asian admixture differs between the Celts and the Balto-Slavs. In the former this shows predominantly, in Ireland exclusively, as the K12b Gedrosia component. In the latter, Gedrosia is virtually nonexistent, and the West_Asian admixture shows mostly as K12b Caucasus. As I have suggested before, I think European Gedrosia is correlated with R1b. The Caucasus component on the other hand seems correlated with J2 admixture.

Simon_W said...

Hence, I would suggest that the West_Asian admixture in the Balto-Slavs, who are dominated by R1a, came from J2, probably via the Caucasus region. And I still think that R1b/Gedrosia came from the Baden culture, which had roots in eastern Anatolia.

This article by Roland Menk
http://slavanthro.mybb3.ru/viewtopic.php?t=929
reassured me in my opinion that the brachycephalic type of the eastern Bell Beaker folks had its origin in West Asia. At least Menk states that the same type is ascertained for the Baden-Vucedol culture. Apparently, this was really an important source for the eastern Bell Beaker people.

There is quite some disagreement between archeologists on the cultural and social importance of the Yamnaya impact in the west. Whereas M. Gimbutas for instance, also Heyd, it seems, ascribe a decisive role to the incursion of Yamnaya people to the Carpathian basin, A. Häusler for instance noted that there was only some weak Yamnaya immigration in the late phase of the Baden culture, and it wasn't culturally determining at all. Soon after their arrival, the immigrants gave up their funerary customs, which means that they were assimilated, not the other way round. He asserts that the Baden culture was superior, both in the construction of wagons and in copper metallurgy. Conclusion: If one isn't in the position to judge for oneself, at least it isn't advisable to blindly follow the Kurganist interpretation.

I'll stand by my opinion that the Raeto-Etruscans had their roots in the eastern Bell Beaker group. But possibly the latter also included some proto-Celts. At least, if R1b was originally associated with the IE spread and with Italo-Celtic, this assumption is likely, given the R1b in eastern Bell Beaker people. In fact, my theory requires the Urnfield culture to be multiethnic, partly Celtic, but also Raeto-Etruscan. And if that's possible for the Urnfield culture, it was no less possible for the Bell Beaker East group. Etruscans and Gaulish Celts seem to share R1b-U152, which is a subvariant of R1b-S116. And the latter is also shared between Celts and Basques. So in any case such a sharing between unrelated languages isn't unrealistic.

Simon_W said...

@ Va_Highlander

Dienekes wrote a blog entry on that topic:

http://dienekes.blogspot.ch/2011/05/on-tocharian-origins.html

Va_Highlander said...

Simon_W, many thanks. I read that some time ago and had forgotten the genetic component of his argument.

"Expect" might be too strong a word but I should certainly not be surprised to find DNA from south Central Asia in the Tarim Basin at a very early date. A material culture certainly seems to have expanded east as far as the Zarafshan Valley by 3000 BCE. Whether we can assign a language to any such expansion is another question entirely.

For the record, I find the conventional wisdom that pushes some ancestor of Tocharian across the steppe in the fourth millennium BCE most improbable. I am as certain as one can be about such things that the language arrived via the proto-Silk Route. My only question is when.

Blasonario Cremonese said...

To Simon W,

it is true that R1b in Italy peaks in Tuscany, but it would be a good idea to see "which Tuscany" is loaded by R1b people.

If you see, you could notice that the R1b loaded area of Tuscany is the northern Appennine one. Also, if you do a little research, you will see that the Appennine area of northern Tuscany is one of the blondest area in Italy (where blondism reach the same percentages as in Tyrol and Aosta Valley). Some Italian scholars, from the beginning of 20th century, sayd that that area was one of the Gaulish refuge from roman expansion.

If you, indeed, are in search of Etruscans in Tuscany, you have to look at the Casentino region, that has, according to Barbujani works, the most etruscan population. That part of Tuscany, however, is in the middle of Central Italy, next to Umbria, where a lot of R1b folks settled.

For what I am reading, I'm more convinced in a PIE population loaded with R1b, than in a PIE population loaded with R1a.

amrik singh said...

If we look at phenotype of ancien populations with Y haplogroups R1a1 dominant they show features that seem intermediary between proto-europides (robust bones strongly develop eyebrows ridge , medium to large size)and Mediterraneans (narrower face and straighter forehead)
The R1b ... Haplogroups came probably from Anatolia and apparently expended with stock breeding largely for dairy product as demonstrated convincingly by Jean Manco .
But the crucial point is that it is only R1b1a ant it's subclades that is a marker of the proto-Indo-Europeans (R-l21 , R-S28, R-S21 ...).The other R1a haplogroups came probably from NON_Indo-Europeans from Anatolia.
G2a3b1 seems to form (always?)a small companion haplogoup of Ria .
So it seems to me that the proto-Indo-Europeans east of the Don and in the forest steppe Indo-europeanised those living in the steppes west of the Don an that it is those who settled in western Europe.

amrik singh said...

I always think proofs must drive the theory and as new elements of knowledge accumulate I become more and more convinced that the first Indo-europeans are to be found in Khwalinsk and subsequent Yamnaya cultures . As the question to know if the Indo-european expansion is the result of "elite dominance" or partial "population replacement" both cases must have occurred : Sometime a war-band established itself as a dominant upper class ,other time they must have largely massacred the vanquished male population and keep the female population as wives or concubines as witness by the fact that Mt DNA is very divers but differ only in proportion between different European populations wile in some area Y DNA show a population replacement.
I am also impress by the apparent fact that Khvalinsk people is more gracille than east of the Dnieper steppe people and most of Andronovo .There seems to be a continuity between Kvalinsk,Repin,east of the Don Yamnaya ,Poltavka and Skrubnaya people and maybe Scytians.
I am very curious to see if Genetic confirm or contradict these hypotheses and to Know more about the Khvalinsk Paleo-anthropology . I think big discoveries will come soon and maybe forested steppes and broad lives forest peoples are more important for the Indo european question than previously thought.