The mystery of the origin of the Indo-Europeans may be solved within the next 2 years

By Razib Khan | July 1, 2012 1:27 pm

Dienekes has a post up, The Bronze Age Indo-European invasion of Europe. The crux of his argument is as such:

But there is another component present in modern Europe, the West_Asian which is conspicuous in its absence in all the ancient samples so far. This component reaches its highest occurrence in the highlands of West Asia, from Anatolia and the Caucasus all the way to the Indian subcontinent. It is well represented in modern Europeans, reaching its minima in the Iberian peninsula….

Thanks to the public release of genetic data Dienekes has developed his theories in part out of his own analyses of said data. Though I’ve run fewer analyses, with smaller data sets, some of the same patterns jump out at me. In particular, there is a component which is modal in northern West Asia (e.g., the trans-Caucasian region) which seems to drop mysteriously between the French generally and French Basques, and the Basque vs. non-Basque Spanish samples. There are also similar, though not necessarily easy to map across the two regions, disjunctions in South Asia between geographically close Indian groups.


Ultimately model-based clustering algorithms and PCA is going to get us only so far. Remember that the clusters generated from these methods don’t give us reality as such, but particular patterns which map back to reality. You can’t read from cluster A to population X without a non-genetic interpretative frame. Nevertheless, I do think within the next few years we may solve the “mystery” of the demographic origin of Indo-European languages and culture through genetics.

First, we have to posit a hypothesis in the fashion which Dienekes proposes. That is, Indo-European languages began to spread rapidly ~5,000 years ago from a small core population. This rapidity leads to both cultural integrity, and some genetic signal, which spans Indo-Eurpean groups. There are two dimensions, time and space. As Dienekes notes ancient DNA data points will get thicker. If the West Asian component in European ancestry begins to show up >5,000 years B.P., that’s going to be highly suspicious. Second, our data set of extant populations is going to get thicker. You’ll be able to independently contrast relatively similar Indo-European vs. non-Indo-European populations. For example, Sinhala vs. Tamil and Swede vs. Finn. By and large the two tips of the two clades should exhibit genetic similarity, but if the thesis of Indo-European demographic expansion is correct then you’re going to see a subset of matches which correlate with language family, and not proximity. You can then construct a “synthetic” genome from these matches across independent pairs. Finally, you can compare that genome to present populations and ancient DNA samples.

Addendum: I assume that methods like looking at identity-by-descent tracts are going to be important, even though recombination will have broken apart the regions quite a bit.

CATEGORIZED UNDER: Anthroplogy
MORE ABOUT: Indo-Europeans
  • https://twitter.com/#!/christopherburd Christopher Burd

    It’s one thing to know where the I-E’s came from, but are we ever going to find out how they pronounced their laryngeal phonemes?

  • http://dispatchesfromturtleisland.blogspot.com ohwilleke

    Color me skeptical.

    We have lots of well dated data points on archaeological cultures. We have lots of linguistic data points. We have a fair number of historic era or near historic era data points. We have lots of uniparental genetic data points. But, we also have all sorts of IE origins theories that are alive and kicking that sometimes hinge on obscure ponts like whether pre-Hittite languages in a particular part of Anatolia were IE or non-IE.

    Being able to characterize the major Bronze Age to early Iron Age expansions of IE into Europe and South Asia as consistent with a particular genetic profile would be an accomplishment and it would be great if linguistics and prehistorians could focus on the areas of consensus rather than the areas of disagreement.

    But getting from the genetic profile of the expansion period to IE origins greatly reduces resolution and begs some of key questions in the field, like whether the late Bronze Age, early Iron Age IE expansion supplanted a non-IE world in Europe, or whether they were simply substituting one IE linguistic population for a previous IE linguistic populations (as was the case, for example, when the Romans imposed Latin in lieu of Celtic languages).

    Pre-Bronze Age expansion, all sorts of theories can be consistent with the evidence. And, of course, few Trans-Caucus populations today are IE speakers.

  • Wim Van Dijk

    “But there is another component present in modern Europe, the West_Asian which is conspicuous in its absence in all the ancient samples so far.”

    hmm…
    We don’t have any data from the numerous population of the Tripolye-cucuteni yet. What if they were high in west Asian component? Kurgan populations would have to mix with them in their spread west, Indo-europeanization was not a flash process.

    In the Yamna period (thought to be the time of dispersal of PIE in the Kurgan hypothesis) the west and south of this cultural group was different from the north and east (proto-europoid). The autochtonous (formerly) hunter-gatherer population was heavily admixed with cucuteni populations (more mediterranean type), and maybe some little Caucasian input (some south Russian (adyge)/caucasian hg are apparently found in ancient steppic remains).

    And are there any archeological matches with this bronze age west asian IE surge from Anatolia in archeology? AFAIK no. We have hints of influence from south-east europe in Anatolia during bronze age, but not the reverse AFAIK.

    The west Asian component in India could also be the sign of a non-IE new arrival. Dravidian languages have been associated with the Elamite language of south-west Iran by some and the Brahui language (related to Dravidian) of Pakistan might be a remnant of this time (and rigvedic sanskrit (likely confined to north-western India at first) had borrowed a lot to Dravidian for local terms).

    This theory ignores a lot of linguistic, archeological and cultural points. It’s a shoot in the dark just because of the apparition of this west asian component in Eurasia that can still have other explanations.

  • Paul Ó Duḃṫaiġ

    Hopefully we will start seeing alot more in the way of Ancient DNA particulary with regards to the likes of Y-Chromosome haplogroups (SNP testing) over the next three years. Already on what little results that are in from Mesolithic/Neolithic samples there appears to be a disconnect between modern day frequency and these periods (no R1b samples let found — Haplogroup G and I predominate). Of course it’s still early days but given technological advance I do hope it will gives us some interesting insights.

    Of course the earliest we can put an IE language in Europe would be “Mycenaean Greek” which is during the Bronze age — Linear B tablets (14th century BC)

  • http://dienekes.blogspot.com Dienekes

    So far, all the ancient mtDNA we’ve gotten from the steppe has shown a mixed Caucasoid-Mongoloid gene pool, and this extends all the way to Ukraine in the west:

    dienekes.blogspot.com/2012/06/ancient-mtdna-from-neolithic-ukraine.html

    This does not bode well for the identification of the steppe area as the PIE homeland, because East Eurasian mtDNA is lacking in most European populations or occurs at trace frequencies. Even if a supposed migration carried only males (explaining the non-existence of East Eurasian mtDNA) it would still carry East Eurasian autosomal component, which is similarly lacking.

    To top it all off, we know for a fact that the steppe has been affected from outside: the East Eurasian mtDNA has been reduced to irrelevancy in modern populations of the Western steppe, and the R1a that dominated in the Siberian steppe has been supplemented by a whole host of other haplogroups.

    Nor is Europe a better candidate for the PIE homeland, but for a different reason: the most European of Y-haplogroups is I and that is completely lacking among eastern Indo-Europeans. And, the legacy of the Neolithic people in Europe (the “Southern” component) is also lacking in the Indian subcontinent.

    Some have also proposed that the “North_European” component that seemed to be lacking in Oetzi may have been a signature of a post-5ka migration from the steppe. Such a migration would explain why the “North_European” component exists today in North Italy, whereas Oetzi did not possess it and was much more “Southern” and Sardinian-like.

    However, the discovery that Mesolithic Iberians were North_European-like indicates that there is no need to postulate a late appearance of this component, but rather this element is now known to have spanned Europe (from Iberia to Gotland) in hunter-gatherer times.

    The PIE homeland can be circumscribed by a process of exclusion: South Asia is out because its indigenous ASI component is lacking in Europe; Europe is out because of the lack of Y-haplogroup I and the “Southern” component that was established with the farmers. The steppe is out because of its East Eurasian elements. The most coastal parts of the Near East where presumably the early agriculturalists originated, bearing the “Southern” component into Europe are also out because they don’t seem to have affected the East.

    Through a process of elimination and by considering the totality of the evidence, it appears that the mountainous regions of West and Central Asia are the only possible candidate that would satisfy the requirements of a PIE homeland. The fact that its modal component is also ubiquitous in Europe and seems to have been added there (*) post-5kya is also consistent with that identification.

    (*) As for Tripolye and the Balkans, of course I think it’s highly likely that the West_Asian component may have been present there earlier than in the rest of Europe. This makes geographical sense. How early, remains to be seen. Actually, I have previously hypothesized that the abandonment of Tripolye mega-sites due to climate change may have spurred the Indo-Europeanization process in Europe, just as the abandonment of the BMAC mega-sites in the East may have played a similar role in the east. People don’t sit around to die, and the huge populations in both these cases certainly did not. Collapse is followed by increased mobility, and mobility is what is necessary for spreading languages.

  • Wim Van Dijk

    @ Dienekes:
    “The PIE homeland can be circumscribed by a process of exclusion: (…) The steppe is out because of its East Eurasian elements”

    “This does not bode well for the identification of the steppe area as the PIE homeland, because East Eurasian mtDNA is lacking in most European populations or occurs at trace frequencies. Even if a supposed migration carried only males (explaining the non-existence of East Eurasian mtDNA) it would still carry East Eurasian autosomal component, which is similarly lacking.”

    Err…. Let’s stay cool-headed.

    a/ It’s true that mtDNA C had a surprising important presence in the mentionned study’s samples, but we can’t really be completely sure of how representative it was on a few samples.
    We can notice, for instance, that the bronze and iron age south Siberian samples of Kayser et al, 2009 are likely mostly representative of the Afanasevo substrate despite the date of the samples and we know that this culture is derived from east Yamna (It seems pretty well established) and that the oldest samples were overwhelmingly (we could almost say “exclusively”) west eurasian with matches in modern samples in Europe.

    As you said, we can also assume a male-biased propagation. We can’t assume that the presence of east Asian mtDNA hgs implies a sizable east Asian component either, because we don’t know when the mixing took place and its exact importance in the populations that would have been implicated in the spreading of IE languages. AFAIK, nothing in the morphology of these populations betray an east Asian influence either. It could be from a very old mixing.

    The 60% of Y-DNA and ~5% of mtDNA (IIRC) of east Asian hgs among the Finnish population aren’t a good indication of the percentage of the east Asian component within it (it’s much less that we could expect).

    b/ As you mentionned, there is mtDNA C found in the east european populations (namely mtDNA C5c), from my memory ranging from Poland to Romania. Even such a low percentage can be congruent with movements from the steppes carrying a few east Eurasian mtDNA hgs IMHO, especially if the genetic impact was particularly male-centered.

  • http://dienekes.blogspot.com Dienekes

    Err…. Let’s stay cool-headed.
    a/ It’s true that mtDNA C had a surprising important presence in the mentionned study’s samples, but we can’t really be completely sure of how representative it was on a few samples.
    We can notice, for instance, that the bronze and iron age south Siberian samples of Kayser et al, 2009 are likely mostly representative of the Afanasevo substrate despite the date of the samples and we know that this culture is derived from east Yamna (It seems pretty well established) and that the oldest samples were overwhelmingly (we could almost say “exclusively”) west eurasian with matches in modern samples in Europe.

    My head is perfectly cool.

    Not sure what you are arguing here. Actually, the Keyser samples also have East Eurasian mtDNA, so if they’re supposed to be an example inconsistent with the Ukrainian results, they are not a very good one.

    Your contention that Afanasevo is derived from east Yamna is flawed, and certainly not as certain as you seem to think.

    http://dienekes.blogspot.gr/2012/02/frachetti-on-multiregional-emergence-of.html

    There is also apparently new evidence that Europe was a destination rather than a source vis a vis Central Asia.

    http://dienekes.blogspot.com/2012/06/smbe-2012-abstracts-part-ii.html

    “Using an approximate Bayesian framework, we find that present patterns of genetic diversity in Central Asia may be best explained by a demographic history which combines long-term presence of some ethnic groups (Indo-Iranians) with a more recent admixed origin of other groups (Turco-Mongols). Interestingly, the results also provide indications that this region might have genetically influenced Western European populations, rather than vice versa.”

    As you said, we can also assume a male-biased propagation. We can’t assume that the presence of east Asian mtDNA hgs implies a sizable east Asian component either, because we don’t know when the mixing took place and its exact importance in the populations that would have been implicated in the spreading of IE languages. AFAIK, nothing in the morphology of these populations betray an east Asian influence either. It could be from a very old mixing.

    You are wrong re: their morphology

    “On the Origin of Mongoloid Component in the Mitochondrial Gene Pool of Slavs, B. A. Malyarchuk, M. A. Perkova, and M. V. Derenko

    Concerning the population of Eastern Europe, it should be noted that the forest zone of Eastern Europe was the area of intense population admixture [35]. It seems likely, that formation of the complex of Mongoloid traits happened not later than in Upper Paleolithic. For this reason, it is suggested that East Siberian populations could have much time for migration to Eastern Europe [35]. The number of such migrations still remains unclear, since in the northwest of Eastern Europe Mongoloid component is detected 10000–8000 years ago; in Dnepr–Donetsk tribes, 7000–6000 years ago, and on the territory of Ivanovo oblast (Sakhtysh), 6000–5000 years ago [35, 36].

    We have mtDNA haplogroup C and Mongoloid paleoanthropological influence. I think there are pretty good chances that they carried a Mongoloid autosomal component.

    b/ As you mentionned, there is mtDNA C found in the east european populations (namely mtDNA C5c), from my memory ranging from Poland to Romania. Even such a low percentage can be congruent with movements from the steppes carrying a few east Eurasian mtDNA hgs IMHO, especially if the genetic impact was particularly male-centered.

    A few males into Eastern Europe do not Indo-Europeanize an entire continent.

    There is also the added difficulty that a few males into Eastern Europe do not turn a G/I gene pool of Western Europe into a predominantly R1b one either.

  • http://www.buildinghistory.org/distantpast/ Jean M

    “Ultimately model-based clustering algorithms and PCA is going to get us only so far.”

    So true. Whatever the West Asian component here is, it can’t be a major IE strand if there is more of it in Greeks than Slavs and Germanic speakers. Greeks are notably low in the major Y-DNA haplogroups that are well known to correlate with IE speakers. In fact they are a stand out exception, seeming to have assimilated much more in the way of previous Neolithic people. Nor can it have much to do with IE speakers if it is strong in the Caucasus.

    One big problem for this analysis is that on archaeological and aDNA evidence (from Andronovo and Bell Beaker sites) IE speakers appear to have arisen from mesolithic hunter-gatherers carrying mtDNA U5 and U4, just like the people of the rest of Europe in the Mesolithic. Yes they apparently had mixed with dairy farmers of an Anatolian origin, but the massive supposedly Mesolithic component is liable to have actually arrived in most places in Europe in the Copper Age.

  • http://dienekes.blogspot.com Dienekes

    So true. Whatever the West Asian component here is, it can’t be a major IE strand if there is more of it in Greeks than Slavs and Germanic speakers.

    There is no reason to think that Slavs and Germanic speakers first attested ~2 millennia after the Greeks would have preserved the IE gene pool more faithfully.

    Greeks are notably low in the major Y-DNA haplogroups that are well known to correlate with IE speakers.

    The only haplogroup known to correlate with IE speakers is J2a since it is the one that occurs in Hindu upper castes.

    In fact they are a stand out exception, seeming to have assimilated much more in the way of previous Neolithic people.

    The previous Neolithic people belonged as far as we know so far to haplogroups G and I, and these make up (combined) no more than, say, ~25% of the Greek gene pool.

    Nor can it have much to do with IE speakers if it is strong in the Caucasus.

    Why not? That is like saying that R1a in North America can’t have much to do with English and German speakers, because it is strong in Russia and Poland.

    The fact that the PIE community possessed a gene pool that was related to the West_Asian component does not mean that they had to top the “West_Asian” charts in their homeland. If we sample Mestizos today we’ll find a lot of the “Atlantic_Med” component in them, but we won’t conclude -based on your flawed logic- that it “can’t have much to do with Spanish speakers if it is strong in Basques and Sardinians”.

    One big problem for this analysis is that on archaeological and aDNA evidence (from Andronovo and Bell Beaker sites) IE speakers appear to have arisen from mesolithic hunter-gatherers carrying mtDNA U5 and U4, just like the people of the rest of Europe in the Mesolithic.

    We don’t know what language was spoken by the Andronovo people; assuming that they spoke is IE is to accept a steppe origin, and hence leads to a circular argument.

    As for the Bell Beaker people, the novel finding is that they possessed haplogroup R-M269, and this ties them to the Transcaucasus area. I see absolutely no reason to think that the Bell Beaker mtDNA gene pool travelled with the Bell Beaker males that spread over a large area in a very short period of time.

    Actually it is irrational to maintain that the mtDNA detected in Bell Beaker sites is anything but local and to simultaneously maintain that R-M269 entered Europe late with Bell Beakers: to account for its present-day success one _has to_ assume that the R-M269 males took a lot of local mates, and hence their gene pool is more likely local than descended from their ultimate West Asian homeland.

  • toto

    Sorry to ask the dumb question again, but…

    We find a West Asian component that happens to be less frequent in the margins of Europe and/or the glacial refuges.

    How do we know whether we’ve found TEH INDOYUROPEENS, or some other wave of advance – like, say, the Neolithic wave of expansion of the first farmers, riding on the awesome multiplying power of agriculture, Bantu-style?

    (Same question for the “ANI” component in India).

  • http://www.buildinghistory.org/distantpast/ Jean M

    @ Dienekes

    The ancient DNA evidence indicates that incoming farmers replaced Mesolithic hunter-gatherers over much of Europe. You know that. And it is supported by craniometric evidence, as you also know. There is precious little evidence of said farmers absorbing any DNA from hunter-gatherers as they went in most parts of Europe. In fact the hunters seem to have vanished from many places before the farmers arrived. (Greece happens to be a prime example of that.) The idea of massive local continuity of the DNA of Mesolithic hunters has already been stabbed in the heart, though it is bleeding to death very messily across DNA forums and blogs.

    The only exceptions are places where hunters had an advantage and/or the environment was not suitable for farming (generally the same thing in effect.) Again the craniometric and aDNA evidence match up. It is the eastern fringes of Europe in the main where hunters survived well enough to compete with farmers and take on what elements of the farming package suited them, at their own pace.

    These people still carrying a lot of mtDNA U5 and U4 after the advent of farming included Proto-Uralic forest hunters and Proto-IE speaking steppe pastoralists at the dawn of the Copper Age. There also seem to have been hunting/fishing survivors in the far north, as we would expect, speaking an unknown and now vanished language or languages. Fenno-Scandia and the Baltic then received further injections of DNA that in origin was Mesolithic via the migration into the area of Uralic speakers (represented by the Pitted Ware individuals in your analysis), and various waves of IE speakers. The outcome is a remarkably high level of mtDNA U5. It is no surprise at all that the highest level of your Mesolithic component corresponds roughly to that.

    If mtDNA U5 had pretty well vanished from the territory now Germany during the LBK, it is scarcely irrational to suppose that the U5 in Bell Beaker arrived with newcomers. I’m not going to attempt to precis my whole book here, but the craniometrics are surely known to you. The BB people did not look like the previous farmers.

  • http://dienekes.blogspot.com Dienekes

    If mtDNA U5 had pretty well vanished from the territory now Germany during the LBK, it is scarcely irrational to suppose that the U5 in Bell Beaker arrived with newcomers. I’m not going to attempt to precis my whole book here, but the craniometrics are surely known to you. The BB people did not look like the previous farmers.

    There is no reason to think that mtDNA U5 vanished from the territory of Germany during the LBK, or anywhere else in Europe from that matter. That presupposes that the archaeologically visible LBK farmers were the only inhabitants of Germany at that time, so the re-appearance of mtDNA haplogroup U5 would require a fresh infusion from elsewhere, rather than being a local re-emergence.

    The craniometrics are surely known to me, and Bell Beakers were distinctive Dinaroids with parallels in the Balkans and the West Asian highlands and no resemblance whatsoever to the Proto-Europoid aboriginals of the European steppe. If you’re imagining an infusion of steppe populations into Europe, you’re out of lack because nothing of the sort is evident in the paleoanthropological record of Europe, except perhaps in Hungary and Romania.

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    #10, i’m alluding to actually finding commonalities across all these populations. not *the* west asian element as observed by dienekes, as such. you can substitute this for “indo-european”:

    -a demographic pulse which impacted *both* western europe and northern india, and excludes basques, finns, south indians, people of mediterranean islands

    this might not be indo-european. but would seem to be the best candidate.

  • http://www.buildinghistory.org/distantpast/ Jean M

    @ Dienekes

    It would be a minor miracle for Mesolithic DNA to “re-emerge” in Copper Age Germany from a population that had been totally invisible there for millennia. Let us be practical.

    As for the BB craniometrics, I do not suppose for one moment that the well-known BB brachycephaly emerged on the steppe. I don’t even suppose that it predominated in the trek up the Danube from the steppe, since it doesn’t even appear in all BB people, only those of the eastern/northern branches. Exactly where it was picked up I don’t know. That is one little mystery I have not solved. Romania you think? Sounds reasonable. It is irrelevant to this particular issue however. The BB people are obviously different from their Neolithic predecessors in Germany as demonstrated by ancient DNA (Y-DNA R1b) as well as craniometrics.

    One rare mtDNA haplogroup may be an Indo-European marker – U2e. It appears in Bell Beaker and Andronovo sites and its distribution has a steppe element http://eng.molgen.org/viewtopic.php?f=46&t=322&start=10 . Naturally it cannot have been carried by all or even most IE speakers, but its possible specificity makes it interesting.

  • http://dienekes.blogspot.com Dienekes

    Romania you think? Sounds reasonable.

    I don’t. Nor do I think there was a trek of steppe people from the east that gave birth to the Bell Beaker phenomenon. All of that is baseless speculation as far as I am concerned. The only strong pieces of evidence about Bell Beakers are their physical type and their R1b, and both these link them with the West Asian highlands.

    It is irrelevant to this particular issue however. The BB people are obviously different from their Neolithic predecessors in Germany as demonstrated by ancient DNA (Y-DNA R1b) as well as craniometrics.

    The BB people also different from the people of the steppe as evidenced by both craniometrics and ancient DNA (lack of mtDNA C and Y-haplogroup R1a). And, they are most similar based on both craniometrics and Y-DNA to people like the present-day Armenians and other R1b-bearing brachycephals of the West Asian highlands.

  • Raimo Kangasniemi

    Steppe peoples have often formed heterogenous confederations in later times, made up of groups of different backgrounds, even when they are often treated as far more homogenous in the sources of the settled people that encountered them.

    The Indo-Europeans would have basically been the first in a long line of Central Asian and Pontic steppe peoples expanding into neighbouring areas.

    If they would have been similarly composed as the later steppe confederations, a mix of peoples from the very beginning, then chasing some special, uniting feature in modern people that would tell that the “Indo-Europeans were here”, might mean chasing something that doesn’t exist.

    Also, some of those Western Asian genetic signs could come from much later groups than those that participated in the first waves of Indo-European expansion. Scythians, for example.

  • http://www.buildinghistory.org/distantpast/ Jean M

    @ Raimo Kangasniemi

    I agree entirely. Just looking at Y-DNA you get a picture of Indo-Europeans carrying both R1a1a and R1b1a2 as well as a few other haplogroups that seem to have travelled with those.

  • http://dienekes.blogspot.com Dienekes

    A little tidbit for all those who imagine hordes of Indo-Europeans inundating Europe with their R1s and their mtDNA from the steppe:

    “The presence of steppe tribes in the Carpathian Basin is well established
    but other than an occasional exception such as the Jamnaja-like burial at Bleckendorf
    in eastern Germany, clear evidence of steppe expansions any further
    west of the Tisza remains elusive. Unless the steppe hypothesis can demonstrate
    that a steppe culture crossed the Tisza line, it is incapable of providing
    an attractive solution to the Indo-Europeans of central, northern and western
    Europe.

    Written by the principal advocate for the steppe hypothesis in the West, J.P. Mallory, in 2010.

  • http://www.buildinghistory.org/distantpast/ Jean M

    “Unless the steppe hypothesis can demonstrate that a steppe culture crossed the Tisza line… ”

    That is the challenge (among others) which I feel genetics has the potential to take up. My text attempts it. Naturally I don’t ignore the latest archaeological thinking either. The crucial paper by Harrison and Heyd 2007 makes the archaeological connection between Yamnaya and later European cultures, principally Bell Beaker, though others are covered. Yamnaya may go no further than the Carpathian Basin, but its archaeological and genetic offspring spread far and wide.

  • Matt

    I would think this West Asian component pattern is less due to a people who actually went anywhere (demic diffusion, sort of) rather than an increase in ongoing geneflow in the Bronze Age up to the present which was sort of bidirectional (between West Asia (whatever composite of Near Eastern farmers and local hgs that was at the time) and the farmer-hg composite of Europe), but probably with the Europe as a net sink, rather than source. The groups identified with the linguistic isolates of Europe without a recent attested extra-European origin probably participate in this dynamic less.

    But there’s not a great deal of distance between that interpretation and this. I think the idea of the origin of the earliest Indo-European languages in West Asia seems plausible.

    Do we have the prospect of obtaining West Asian ancient autosomal dna any time soon?

    Of course, the Sardinians speak an Indo-European language. There are groups in the Levant who are not Indo-European speakers with major (presumably post Neolithic) West Asian geneflow. There is an Indo-European group with no significant West Asian component. How much linkage between IE languages and West Asian component has to exist before the theory becomes plausible/implausible, is what seems like the tricky question here.

    Re: Bell Beakers, Wikipedia seems to suggest a Bell Beaker impact in Sardinia, as amongst the oldest area to demonstrate such. How does this work out with an absence/minima of West Asian components there?

    And besides which, are there actually any signs of the Bell Beaker material culture in West Asia (e.g. the West Asian highlands)?

  • https://plus.google.com/109962494182694679780/posts Razib Khan

    Of course, the Sardinians speak an Indo-European language.

    this is a function of the historical period. there is debate whether the sardinian language of the roman period was indo-european (cicero alludes to sardinian has being particular uintelligible i believe).

    i too wonder about the possibility of a diverse origin for the indo-europeans as explaining some of the patterns we see.

  • http://www.buildinghistory.org/distantpast/ Jean M

    “And besides which, are there actually any signs of the Bell Beaker material culture in West Asia”

    No. Bell Beaker is a culture of Central and Western Europe (with some overflow into Morocco from Iberia). It correlates pretty well with the distribution of Celtic and Italic languages at the time that these language emerge into history. In between the Copper Age arrival of Bell Beaker and the historical Celts and Italics there is a fair degree of archaeological continuity in those regions, so Bell Beaker (and its origin cultures) seem the best bet as the breeding ground of the Celtic and Italic branches of Indo-European. Since we now have the archaeological link between Bell Beaker and its parent culture on the steppe, we can track these people back to the steppe. We can also track people archaeologically and in ancient DNA eastwards from the steppe into Indo-European speaking cultures to the east.

    Anatolia only comes into this story because of an earlier branch of R1b there, and because certain IE speaking people entered Anatolia, including the Armenians, who settled in Armenia in the 6th century BC.

  • Justin Giancola

    Another thing that makes an IE home in Anatolia, or the immediate vicinity of the Caucasus difficult, is that I’ve seen a couple times linguists claiming that IE is closer to Altaic and Uralic languages than any other Eurasian language, as well as these three being closer to themselves than any other. I have never once seen IE being compared to Caucasian languages, and if IE did start near by said region wouldn’t that imply the other two would also have to been birthed around the same spot? And now you are packing in a lot of languages in the region to not have an overarching family relationship, which gets murky when three are seemingly close and others aren’t??

  • Wim Van Dijk

    @ Dienekes:

    “Not sure what you are arguing here. Actually, the Keyser samples also have East Eurasian mtDNA, so if they’re supposed to be an example inconsistent with the Ukrainian results, they are not a very good one.”

    As I mentionned, the earliest samples of these south Siberian remains are almost all west eurasian (just as their morphology seems almost completely dominated by the proto-europoid element on the Ural (and the core element of the earliest “Kurgan” populations (Khvalynsk/Samara), the east Asian element increase in the early iron age (just like in Kazakstan in the same time frame (Lalueza-fox et al, 2004)). It seems pretty well established, even the Derenko that was part of a study you mentionned says it in one of his own work.

    “Your contention that Afanasevo is derived from east Yamna is flawed, and certainly not as certain as you seem to think.”

    You’re fighting a lost battle. There is no way it can be another way.
    The culture, the type of sepulture, the cultual objects, the similarity of their pottery with similar east Yamnaya ones, their europoid morphological type (the same as the one from the Volga and Pontic steppes), the use of pastoralism (obviously coming from the west, especially as we find European cattle genes in Siberia (and beyond)), the time of apparition of Afanasevo, the use of copper and _silver_, their haplogroups both male and female (with matches deep in Europe), the clear presence of a north European autosomal component in all these Asian regions (and beyond)…

    “There is also apparently new evidence that Europe was a destination rather than a source vis a vis Central Asia.”

    Get real. These Europoid R1a1a, mtDNA U5, H, T, I4 (etc…) and their ways and technology, couldn’t arrive from elsewhere in this time, especially in this place. Especially carrying such a north European autosomal component.

    “A few males into Eastern Europe do not Indo-Europeanize an entire continent.”

    Indeed, but the populations they had indo-europeanized, yes.

    “There is also the added difficulty that a few males into Eastern Europe do not turn a G/I gene pool of Western Europe into a predominantly R1b one either.”

    The actual chronology of the R1b spreading and their nature remains to be clarified.

    As for the Mongoloid element in Dnepr-Donetsk, I had never heard of it before. Fine. If it took so much time to be noticed or at least mentionned it probably wasn’t much important in this general population anyway. Whatever, I consider my other points holding.

    And BTW personnaly I don’t see mass migration a requirement for slow language switching.
    After all, Anatolia was once one of the main craddle of civilization on earth and many languages of very different language families were spoken there. Now almost the whole population of Anatolia is speaking the language of a bunch of warlike nomadic horse-riders arrived from the depth of Asia (apparently south Siberia) about 1000 yrs ago, that ended up building an empire in a few centuries. These newcomers are between 7 to 10 percent _tops_ of the Turkish gene pool.

  • Wim Van Dijk

    “the proto-europoid element on the Ural “

    Sorry, here I meant at the Volga.

  • http://www.buildinghistory.org/distantpast/ Jean M

    “There is also apparently new evidence that Europe was a destination rather than a source vis a vis Central Asia.”

    “Get real. These Europoid R1a1a, mtDNA U5, H, T, I4 (etc…) and their ways and technology, couldn’t arrive from elsewhere in this time, especially in this place. Especially carrying such a north European autosomal component.”

    Agree entirely with your perspective on the eastward movement of IE speakers which is very clear in the record. I suspect that Dienekes is referring to a scattering of aDNA results which do indeed indicate some westwards movement as well, or more probably the westwards movement of people from a mixed population in Central Asia.

    There must have been some Mesolithic link between the Volga region and Lake Baikal for pottery to have reached the former from the latter. The exact nature of that link is unclear. Could be forest/steppe hunting bands meeting and mixing between the two rather than a person moving all the way. Later there was some mixing between Afanasevo people and local Central Asian hunters. Once the route was established between the European steppe and Afanasevo, people were moving between the two, as shown in the archaeology. So that may be the explanation for the single Y-DNA haplogroup C in among the clump of R1a1a tested in Andronovo results. There are also three Central Asian mtDNA haplogroup C results from Dnieper-Donets, including one C4a2, which would fit with the C4 in Xiaohe. See my online table.

  • Eurologist

    As for Tripolye and the Balkans, of course I think it’s highly likely that the West_Asian component may have been present there earlier than in the rest of Europe. This makes geographical sense. How early, remains to be seen.

    Dienekes, I agree with this (from a language viewpoint), but it also emphasizes the fallacy of searching for a geographically confined homeland. There have been sound shifts in Germanic within the past 2,000 years that affected Northern Sweden and Holland as much as Austria – while the people there were speaking different Germanic languages. Diffusion is a powerful force. I wouldn’t be surprised if PIE at one point stretched from the early, core LBK area and Ukraine to Anatolia and the Armenian highlands – at a much earlier time than some suspect, but consistent with computational linguistic tree research. Languages can become somewhat marginalized, and then, when they blossom again, they seem way too homogenous for the large area covered (while making quick progress in areas that still have a memory of the old form).

    There is no reason to think that mtDNA U5 vanished from the territory of Germany during the LBK, or anywhere else in Europe from that matter.

    I agree.

    The craniometrics are surely known to me, and Bell Beakers were distinctive Dinaroids with parallels in the Balkans and the West Asian highlands and no resemblance whatsoever to the Proto-Europoid aboriginals of the European steppe.

    This is only true for a very small subset of crania studied. In general, BBs do not appear to be much different from Europeans of the general local area – although they often were not locals in the sense of 10-100 km, or so, from isotope data.

    “Unless the steppe hypothesis can demonstrate that a steppe culture crossed the Tisza line, it is incapable of providing an attractive solution to the Indo-Europeans of central, northern and western Europe.”

    There are multiple additional arguments against a steppe origin. (i) Northern Germany and Western Poland show almost perfect cultural continuity, during this time. (ii) These regions always received cultural influence from the South, during this time and later – not from the East. (iii) Horses are extremely expensive and useless in the swamps and forests and cold/ snow-covered winters of the North, and for that very reason, were completely unimportant until Roman times.

    Finally, we should be very careful not to confuse the spread of the bronze age with IE. The former has been clearly shown to be an extremely extended process in Europe, with many local variations. It does not fit any model of a spread of language, in general, or IE in particular.

  • http://www.buildinghistory.org/distantpast/ Jean M

    “Horses are extremely expensive and useless in the swamps and forests and cold/ snow-covered winters of the North, and for that very reason, were completely unimportant until Roman times.”

    Whether horses were or were not important in Poland is completely irrelevant to the issue of the origin of PIE. Certainly the horse was domesticated on the steppe, which is where this species had best survived. Domesticated horses do radiate from the steppe c. 3000 BC, along with wheeled vehicles. These were just part of a cultural package encompassing all the items of the Secondary Products Revolution, plus metallurgy, which had spread across the whole of Europe and parts of Asia by the Bronze Age. The chariot entered the Near East from the steppe later.

    The arguments for a steppe origin for PIE are more complex. Whole books have been written on the topic. One crucial point has already been made above. PIE clearly evolved in close contact with Proto-Uralic. That did not happen in Poland. It did not happen in Germany. It did not happen in the Indus Valley or Anatolia or the Caucasus or any of the other countless places suggested by those dissatisfied with an origin for PIE that does not include their home town. Like it or not, prehistoric languages had to evolve in a small area, because they were limited to the regularly communicating group.

    It is rather late in the day to fight the standard thinking on the PIE homeland, when the latest SNPs in Y-DNA R1a1a are proving it right. Of course we need more aDNA, but the writing is on the wall. http://www.familytreedna.com/public/r1a/default.aspx

  • Karl Zimmerman

    While it of course seems likely now that Indo-Europeans followed a “first wave” of farmers (although it’s still unclear if they were farmers or herders themselves), I have to say I’m a bit flummoxed at how they managed almost total language replacement.

    Recorded history suggests that there are two main methods of language shift in a given territory. One is through population replacement, which can involve colonization, displacement, and extermination. The other is having some sort of governmental system which sets up the groundwork for a shift to the “official language” – which can include both opening up opportunities to those who learn the “mother tongue,” as well as active suppression.

    The idea of PIE moving into essentially virgin land and replacing the original inhabitants has historical analogues in Southern Africa with the Bantu, and Southeast Asia with various language families. But it’s pretty clear that it wasn’t virgin territory, and they didn’t replace the indigenous population.

    On the other hand, elite dominance is problematic, because history shows it’s a spotty process. look at Britain just in recorded history. On one hand, the Anglo Saxons succeeded in England, the English in Ireland, and somehow Scotland lost all it’s indigenous languages just prior to recorded history. On the other hand, The Romans and Normans influenced the native language, but did not succeed in replacing the tongue of the commoners, and Welsh was pretty stable until the Industrial Revolution brought in English settlers.

    History is full of examples like this – where either the ruling class fails to shift a language, in total or in part, or the ruling class itself experiences a language shift. But for IE to spread on elite dominance, we would need an unrecorded civilization holding down territory almost as wide as the Mongols, with the centralizing power of the Romans, and probably maintaining itself in kingdom-sized chunks with little flux for 500 years or more. Otherwise, you’d end up with a patchwork of successor tribes, some of which would incorporate more of the first-wavers culture and language, others less. In some ways it’s less understandable even in Europe than India, as at least the Indus had an established civilization to decapitate and co-opt the structures of, and a geography which lent itself to a large empire straddling the Indo-Gangetic plain.

    Regardless of being flummoxed, however, it’s pretty clear that is what happened.

  • James

    @ Jean M

    Said :
    “The ancient DNA evidence indicates that incoming farmers replaced Mesolithic hunter-gatherers over much of Europe. You know that. ”

    My reply :

    But then where do you think all the high Northern-Euro found in places like Iberia comes from ?? Don’t you see that when mixing the Mesolithic types like Braña with neolithic types like Oetzis and Gok4 you have a pretty similar results to modern Spaniards ? And actually there is 16% of mtDNA U5 in parts of Northern Spain (Cardoso et al. 2011a). As a matter of fact, the most commont mtDNA found today, the mtdna H, was already there in the Upper Paleolithic and Mesolithic (Hervella et al. 2012)

  • http://www.buildinghistory.org/distantpast/ Jean M

    @ James

    There certainly is a high level of U5 in the Basques. I suspect that most of this does reflect continuity from the Mesolithic. That doesn’t mean that the Basques are 100% descended from the artists of Lascaux. That was rather a romantic idea, which we are now having to reassess, given their strong genetic similarity to their neighbours. I think that the Basques are a mixture, like most modern Europeans, of genetic inputs at various times: Mesolithic, Neolithic (Cardial) and Copper Age pastoralists from the Balkans. We need more aDNA to be certain. So this is just a tentative idea at the moment.

    The Iberians overall are a different story, since their genetic mix includes IE Celtic speakers, but I’d say there is indeed a pretty big Neolithic component, both Cardial and North African. I am dubious about the claims of Hervella 2012 on mtDNA H, which seems more likely now to be Neolithic. They did not publish all the supporting test results. But if H was really there so early, no doubt another lab will be able to replicate their results and prove me wrong. :)

  • Raimo Kangasniemi

    When it comes to replacement of previous languages by Indo-Europeans, one might benefit from looking where there are archaeological signs considered typical of Indo-Europeans – Christopher Beckwith in his “Empires of the Silk Road” uses chariot as the defining example of this and ends up claiming, among others, at the minimum Indo-European contact and perhaps a ruling class even in the proto-Chinese Shang dynasty – but where language replacement never happened and compare these with the cases where the Indo-European language took over. Another place where Indo-European linguistic takeover failed in the end failed was Syria and Mesopotamia, where the Indo-European ruling class in the Hurrian state of Mitanni in the end lost, even when they were on the borders of the larger “IE world” of the time .

    The early waves of Indo-Europeans seem to have had greatest success on areas where there had been no advanced civilizations (meaning ones with cities etc) or where these had failed, like in northwest India. When they came against civilisations still on a higher level of organization and cultural unity and strength, their role changed from linguistic assimilators and founders of a new ruling class to those who were more likely to be assimilated and merely pass on their innovations to existing cultures and states. So, their mark was the greatest where the “resistance” – and I don’t mean as much force of arms as culture, although military power would have counted too – was the weakest.

    When it comes to replacement of languages from above, from a ruling class, to below, the ruling class doesn’t necessarily have to be so well-organized that we would have to speak of real state formation on a large scale. If we leave proper genocides – even of those akin to the Moriori at the hands of the Maori – aside, if the new ruling class figuratively “decapitates” the previously existing ruling class and through it the existing traditions and the prestige connected to them, then the existing culture and languages will be severely weakened, and if you add the new innovations of the new ruling class as items that would hold great prestige among the now oppressed majority population, a linguistic replacement could happen if not smoothly, then partly on it’s own weight.

    If IE languages would have been part of a prestigious cultural package whose adaptation would have brought real benefits, both material and social, the replacement might have soon enough advanced on it’s own weight without need of a strong state to push it forward. At the lowest level these might have meant influence from master craftsmen etc.

    But in places like northern China and Syria-Mesopotamia the existing cultural packages would have been so strong on their own that the IE languages would have held no great prestige compared to them and the natives could have picked from what they wanted from the IE’s.

  • Eurologist

    One crucial point has already been made above. PIE clearly evolved in close contact with Proto-Uralic. That did not happen in Poland. It did not happen in Germany. It did not happen in the Indus Valley or Anatolia or the Caucasus or any of the other countless places suggested by those dissatisfied with an origin for PIE that does not include their home town. Like it or not, prehistoric languages had to evolve in a small area, because they were limited to the regularly communicating group.

    Jean, I have made the same Proto-Uralic arguments, before. However, this still leads to diverging homeland theories, and I completely disagree that ancient languages for some (otherwise unobserved) reason were strongly localized.

  • princenuadha

    “Actually it is irrational to maintain that the mtDNA detected in Bell Beaker sites is anything but local and to simultaneously maintain that R-M269 entered Europe late with Bell Beakers: to account for its present-day success one _has to_ assume that the R-M269 males took a lot of local mates, and hence their gene pool is more likely local than descended from their ultimate West Asian homeland.”

    There was not a continuity of mtdna in cental Europe around the bronze age.

    François-Xavier Ricaut, Murray P. Cox, Marie Lacan,… “A Time Series of Prehistoric Mitochondrial DNA Reveals Western European Genetic Diversity Was Largely Established by the Bronze Age”. 2012

    And a good chunk of the meso_European-like mtdna in central Europe came after the neolithic.

    Esther J. Lee 1 , Cheryl Makarewicz 2 , Rebecca Renneberg 1, … “Emerging genetic patterns of the european neolithic: Perspectives from a late neolithic bell beaker burial site in Germany”. 2012

    Due to the significant pattern of migrants not mixing with locals in these ancient studies (ie migrating families, not just men), the ydna and mtdna replacements, the mtdna changes being linked to meso Europe, and the dramatic increase in “North European” for central Europe, I would say that the migration to central Europe came from the periphery of Europe, rich in “meso European”.

    “Jean M Says: July 2nd, 2012 at 11:47 am”

    You are so holding back.

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at http://www.razib.com

ADVERTISEMENT

See More

ADVERTISEMENT

RSS Razib’s Pinboard

Edifying books

Collapse bottom bar
+

Login to your Account

X
E-mail address:
Password:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »