Tuesday, January 1, 2019

The PIE homeland controversy: January 2019 status report


Last year, the preprint that claimed to have presented archaeogenetic data that opened up the possibility of the Proto-Indo-European (PIE) homeland being located south of the Caucasus was, ironically, also the preprint that considerably strengthened my confidence that the said homeland was actually located north of the Caucasus.

Of course, I'm talking about the Wang et al. manuscript at bioRxiv, which is apparently soon to be published as a peer-reviewed paper in Nature Communications (see here).

It'll be fascinating to observe if and how the peer-review process has impacted on the preprint, and especially its conclusion. My impression was that the authors seemed pretty sure that the Maykop people gave rise to the Yamnaya culture, or at least Indo-Europeanized it. But, as far as I saw, the archaeogenetic data didn't bear this out at all, and instead showed a lack of any direct, recent and meaningful genetic relationship between Maykop and Yamnaya (see here). Was this also picked up by the peer reviewers? We shall see.

Moreover, there was some exceedingly interesting fine print in the manuscript's supplementary information:

Complementary to the southern [Darkveti-Meshoko] Eneolithic component, a northern component started to expand between 4300 and 4100 calBCE manifested in low burial mounds with inhumations densely packed in bright red ochre. Burial sites of this type, like the investigated sites of Progress and Vonyuchka, are found in the Don-Caspian steppe [10], but they are related to a much larger supra-regional network linking elites of the steppe zone between the Balkans and the Caspian Sea [16]. These groups introduced the so-called kurgan, a specific type of burial monument, which soon spread across the entire steppe zone.

Always read the fine print, they say. And they're right. Imagine if I only read the preprint's conclusion and missed this little gem; I'd probably think that the PIE homeland was located south of the Caucasus rather than on the Don-Caspian steppe.

Wow, proto-kurgans with inhumations densely packed in bright red ochre? A supra-regional network linking the elites of the steppe all way from the Balkans to the Caspian Sea? An expansionist culture? And, as evidenced by the ancient DNA from the Progress and Vonyuchka sites, a people who may well have been in large part ancestral to the Yamnaya, Corded Ware and Andronovo populations, that have been identified based on archeological and historical linguistics data as the main vectors for the spread of Indo-European languages as far as Iberia in the west and the Indian subcontinent in the east.

I wonder if the authors actually asked themselves who these people may have been, before so haphazardly turning to Maykop and, ultimately, the Near East, as the likely sources of the Yamnaya culture? To me they look like the Proto-Indo-Europeans and true antecedents of Yamnaya.

So as things stand, my pick for the PIE homeland is firmly the Don-Caspian steppe. And I genuinely thank Wang et al., and indeed the Max-Planck-Institut für Menschheitsgeschichte (aka MPI-SHH), for their assistance.

But, you might ask, what about the Hittites? Yes, I realize that no one apart from me and a few of my readers here can find any steppe ancestry in the so called Hittite genomes published to date. However, consider this: if the PIE homeland really was on the steppe, and a dense sampling strategy of Hittite era Anatolia fails to turn up any unambiguous steppe ancestry in at least a few individuals, then there has to be an explanation for it. But let's wait and see what a dense sampling strategy of Hittite era Anatolia actually reveals before we go that far.

See also...

The PIE homeland controversy: August 2019 status report

Yamnaya: home-grown

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

273 comments:

  1. ** Wow, lies back on couch, popcorns popping**

    ReplyDelete
  2. It would be very interesting if there were no steppe in post-2000 BC Anatolian samples because looking at the anthropological evidence clearly a new and quite large population invaded Anatolia at the time. (Brachycephaly suddenly goes from 16% to 50%)
    I also want to know if Anatolian names from 2500 BC Armi are really Anatolian. Because if they are, steppe needs to be found in Anatolia before that time.

    ReplyDelete
  3. Aside from the interesting question of PIE. Yamnaya kurgans and genes[ all ydna R] are almost all located in an ancient climate zone[looking at ice age climate- progression of Northern limits of human habitation]. It is interesting to look at the evolution/transition of the Mammoth Steppe climate to Continental climate zone on the areas of human and Neanderthal habitation successful interaction[producing offspring-when compared to sub-Saharan African gene pool] . What was once a vast region[Spain to the Wisconsin ice sheet] of rich biomass grass/shrubs and tundra[soil etc..] with evolutionary adapted animals[ wooly mammoths/rhinoceros giant sloths etc...], that lasted 100 000+ years and then collapsed with climate[continental] change causing extinction[ peak ice age--melting rising oceans-]. The evolution of proto-Yamnaya hunter gatherer and other mammoth steppe like populations [including all mammoth steppe ydna R/Q] were able to adapt/change both genetically and culturally to the transitioning continental type climate. In the back drop of the history I sometimes wonder why there is so much emphasis [by some-not all] to constantly having to mention/emphasize the agricultural[revolution] aspect of the middle east when trying to connect Yamana genes with various cultures.

    ReplyDelete
  4. Off topic: David, I have a request for some f3 stats, if you wouldn't mind (there are quite a lot):

    f3 stats #1: https://pastebin.com/czt1GCLd
    f3 stats #2: https://pastebin.com/6BK6BNig
    f3 stats #3: https://pastebin.com/waw7SrJp
    f3 stats #4: https://pastebin.com/yDay8bxj
    f3 stats #5: https://pastebin.com/ugQBj7Qs

    Reason: I've been looking at some old datasheets from last year and some of the patterns that show up using log(f3(x,y,z)), and I'm interested in doing so with a wider dataset. There are some patterns evident in the fine f3 stats that become more obvious under the log function. in particular there looks like some pattern in Mycenaean having more offset to relatedness to present-day European populations than expected from the general correlations with log(f3) for Barcin, Yamnaya, Iron_Gates, Levant_N, and this doesn't look like a CHG mediated signal. But I'd like to check this out and some other trends with a larger set of groups, including those which actually are very rich in CHG.

    ReplyDelete
  5. I hope it’s ok to post these videos below:

    https://m.youtube.com/watch?feature=youtu.be&v=Imj0_UhfMLs

    https://m.youtube.com/watch?feature=youtu.be&v=Imj0_UhfMLs

    @Davidski What do you think about this guy, “Survive the Jive” and his theories on YT? I like his one about the linkage between ANE and PIE


    ReplyDelete
  6. @Matt

    Yeah, I'll run those stats within a couple of days. I need to be on the right computer to do that.

    @Andrzejewski

    I'd have to watch those clips carefully to make a sound judgement about them, and I can't do that at the moment.

    ReplyDelete
  7. @Davidski 1. I remember reading either at Wang et. al preprint or in a related paper that not only the ethnogemetic links between Maykop and Yamnaya are rather weak, the linguistic ones are too. It’s hypothesized that Maykop and Kura Araxes spoke Proto-Hurrian-Urrarto and/or Proto-Kartvelian, not only remotely Indo-European.

    2. I think that following AG3, one branch (R1a1) ventured Westbound, and became ancestral to the Stredny Stog while R1b went down path from Lake Baikal, through Bactria and Iran and then via the Caucasus to the Steppes. This R1b branch perhaps packed CHG via the female bloodline while traversing through the Caucasus, but that maybe happened between 10,000-8,000 years ago, not later.

    3. I don’t think that it’s necessarily the Yamnaya Culture which gave rise to PIE; my guess is that Yamnaya spoke some late stage of a pre-Proto-IE, and that different languages came from different Steppe Cultures. It’s likely that Yamnaya spoke a Centum variety of PIE but Stredny Stog was a Satemized language.

    4. My other hypothesis is that the cultural links were stronger with the West rather than Caucasus, and that innovations such as the potter’s wheel originated in Cucuteni Tripoliye and GAC instead of the Caucasus mountain. The strong EEF (with some WHG element) in Yamnaya rather points to a stronger influences in that direction, and I wouldn’t be surprised if the assimilated Bug-Dniester/Dniester-Donets ones are the ones introducing Farmer ancestry.

    ReplyDelete
  8. @ Andrzejewski

    What about Villabruna R1b already in Northeast Italy 14 000 years ago ? And the Balkan Mesolithic samples ?

    ReplyDelete
  9. @Ric Hern I know about them, but I’m sure @Davidski is much more equipped to address this question :)

    ReplyDelete
  10. @ Andrzejewski

    I'm responding to your no. 2. How could R1b have passed from South of the Caucasus to the North between 10000 and 8000 years ago when they were already in Europe by and before that time ?

    ReplyDelete
  11. As far as I remember the only R in Iran at +- that time was R2.

    ReplyDelete
    Replies
    1. @Ric Hern it sure puts the last nail in the “Out of Iran” bogus theory’s coffin

      Delete
  12. On the Hittites? Why not R1b-L584 and R1b-L277? The old Cinnioglu data had R1b at about 15% in the country overall, peaking in central Anatolia (I believe 18% here), exactly where the Hittites were attested. Since then, I believe that data was SNP tested but I don't know the stats off hand. To the best of my knowledge, no R1b has popped up in aDNA of Neolithic or earlier and must be a more recent influx, possibly from several disparate groups. (That can be said about any YDNA though, J2 included)

    ReplyDelete
  13. By remarkable coincidence or not almost all[with exception 2 in North west Iran] R1b-Z2103[formed 6100 ybp, TMRCA 5500 ybpinfo] Yamnaya samples are located in Dfa/b under Köppen climate classification. It is highly likely that R1b-M269 -M269CTS12612/PF6530/FGC61 * PF6436 * PF6419/CTS623/FGC37+100 SNPsformed 13300 ybp, TMRCA 6400 ybpinfo] will also be found in this region.

    Dfa/b on the Steppe resemble the outline of Yamnaya culture.
    https://en.wikipedia.org/wiki/Humid_continental_climate

    "By definition, forests thrive within this climate. Biomes within this climate regime include temperate woodlands, temperate grasslands, temperate deciduous, temperate evergreen forests,[8] and coniferous forests.[10] Within wetter areas, spruce, pine, fir, and oak can be found. Fall foliage is noted during the autumn.[6]"

    ReplyDelete
  14. Does anyone have access to the papers referenced in the the supplement excerpt?

    10. Korenevskiy SN. "Rozhdenie kurgana: pogrebalnye pamjatniki eneoliticheskogo vremeni Predkavkazja i Volgo-Donskogo mezhdurechia." TAUS (2012).


    16. Govedarica B. Zepterträger, "Herrscher der Steppen: Die frühen Ockergräber des älteren Äneolithikums im karpatenbalkanischen Gebiet und im Steppenraum Südost- und Osteuropas." von Zabern (2004).

    ReplyDelete
  15. @ Juan
    '' As a note, horse domestication began before Suvorovo left the steppe.''

    Last i read, that hadn't even occurred by the Dereivka later Eneolithic period.
    Hvae you since come across a recently released scientific paper ?

    ReplyDelete
  16. @ Juan


    1) Samara culture had flat graves (e.g. Mallory's book p 206)
    2) Samara culture is not the precursor for Khvalynsk culture. Even the latter were predominantly simple earthen pits.
    3) It is abundantly clear that these cultures hunted wild horses.

    ReplyDelete
  17. @JuanRivera how would you characterize the WSHG ancestry in Khvalynsk? And what Haplogroup would it be?

    ReplyDelete
  18. @Andrzejewski

    In the frameworks of the various main steppe PIE homeland theories, Yamnaya is never considered PIE, but rather late PIE. In other words, the precursor to all of the Indo-European branches except Anatolian.

    But it may well be that Yamnaya was just one of the late PIE groups. Whatever the case, Yamnaya was certainly involved in the dispersals of at least some late PIE language groups. It's very difficult to argue against that nowadays.

    PIE is whatever gave rise to Yamnaya, which basically means Khvalynsk and Sredny Stog II. We could add Repin to that, but Repin is often lumped with Yamnaya under the Pit-Grave culture term.

    However, my view is that PIE is the population that I mentioned in the post above, that was already expanding out of the Don-Caspian steppe around 4300 calBCE and gave rise to Sredny Stog II, Khvalynsk, and those Eneolithic steppe groups from the North Caucasus steppe.

    It seems to me that you're focusing too much on the R1a/R1b dichotomy among the Bronze Age populations considered to have been early Indo-European. Keep in mind that Y-haplogroup frequencies are susceptible to drastic founder effects, so there's no reason why Bronze Age groups carrying practically 100% R1a and practically 100% R1b couldn't have come from the same Eneolithic parent population.

    And, by the way, R1b doesn't come from south of the Caucasus. It's obviously from the steppes and surrounds, and so is R1a.

    ReplyDelete
  19. Davidski, Sorry OT. I might have missed your older posts, but did you cover the paper by Fernandez in Polish Neolithic & CWC and incorporate their data into your G25 ?

    ReplyDelete
  20. @Dragos

    I don't have the data from that paper, and I often don't cover papers unless I have the relevant data, unless there's something really worth discussing.

    The results in that paper were fairly straightforward, with no great surprises. I already talked about Y-hg I2 in Corded Ware here.

    Awesome substructure within Czech Corded Ware

    Also, there are quite a few LNBA samples from Poland and surrounds in the Global25. But I will add the samples from that paper when I get them.

    ReplyDelete
    Replies
    1. I2 in CWC are likely remnants of the Hunter Gatherers.

      Delete
  21. @ Synome
    Youre requests are books, not papers.
    However, a paper by same author and same topic, in English is available here
    https://www.academia.edu/30815005/Conflict_or_Coexistence_Steppe_and_Agricultural_Societies_in_the_Early_Copper_Age_of_the_Northwest_Black_Sea_Area

    It's a saminal work and important to be understood, so that one's understanding can evolve/

    ''Within the framework of these cultural and economic relationships, a material prosperity arose among the local steppe communities and a social hierarchy gradually evolved. Imitating the structure of the east Balkan and east Carpathian social system, a leading group in society emerged in the northern Black Sea steppe area — the steppe elite — as expressed archaeologically in the group of sceptre graves. It is no coincidence that the oldest graves of this type appear in the territory of the Bolgrad-Aldeni culture (Giurgiuleşti, Cainari). And located in the same area are the graves in Suvorovo and Casimcea, among others, where leading personalities of the later Suvorovo phase were buried.''

    ReplyDelete
  22. @Dragos

    Thanks. I read the paper, very interesting.

    I wonder if the author realized they may very well have described the genesis of the PIE language community.

    I think I need to do more reading about this so called "Ochre Grave complex".

    ReplyDelete
  23. "It seems to me that you're focusing too much on the R1a/R1b dichotomy among the Bronze Age populations considered to have been early Indo-European. Keep in mind that Y-haplogroup frequencies are susceptible to drastic founder effects, so there's no reason why Bronze Age groups carrying practically 100% R1a and practically 100% R1b couldn't have come from the same Eneolithic parent population."

    Could you elaborate on that a little more for us less specialized folk? The differentiation between the two y-DNA heritage Park, I admit, over of the more confusing aspects of the PIE question. I wonder if the language actually originated in three Samara and Khvalynsk region and expanded westward more from cultural diffusion and the social dominance of the horse scepter cult. By the time Sredni Stog r1a individuals were displaced by Yamnaya r1b individuals that was already old news and their displacement was in part the primary cause of the generation of the CWC. Later displacement via reflux immigration on the steppes pushed later r1b cultures westward again where they became the Greeks, Armenians, Phrygians and the rest of the Paleo Balkan languages while the CWC derived from who pushed them out became the Sintashta/Andronovo, etc.

    But I admit that's based on painting with a pretty broad brush and mostly utilizing archeology and linguistics rather than genetics. Are you suggesting a different model where the r1b and r1a split is an artifact that we shouldn't pay too much attention to?

    ReplyDelete
  24. @Desdichado

    These sorts of theories about migrations being caused by cultures and haplogroups pushing each other out of the steppe sound kind of dumb.

    Judging by some of the Bronze Age mass graves in the forest steppe, there were some pretty heavy conflicts there, but it's likely that in these instances groups rich in R1a were fighting each other more often than anyone else.

    No one knows exactly yet how Corded Ware formed, where it's from precisely, and why it moved out of the steppe, but one way or another, just like Yamnaya, it's from those ochre grave kurgan groups from the Don-Caspian steppe and surrounds.

    Keep in mind that the earliest Corded Ware samples, one of which belongs to R1a-M417, from the East Baltic, actually form a clade with eastern Yamnaya, rather than with Sredny Stog or even other Corded Ware samples.

    We should find out soon whether, more proximally, Corded Ware is something like a sister group to Yamnaya, or actually descends directly from it.

    And I don't know much about those scepter carrying groups, but horses were far more important for the Eneolithic groups from the North Pontic steppe, including Sredny Stog, than for Khvalynsk and even Yamnaya.

    Interestingly, Sintashta is very similar to that one Sredny Stog sample available, and its use and worship of horses is on a totally different level to that of Yamnaya. See here...

    Major horse paper coming soon

    ReplyDelete
  25. Yamnaya formed just before 3000 BC, but the Pit-Grave or Repin culture, from which Yamnaya descends directly, formed as early as 3900 BC.

    Sometimes Repin and Yamnaya are both lumped into the Pit-Grave culture, especially by Eastern Euro archeologists.

    ReplyDelete
  26. ''Sometimes Repin and Yamnaya are both lumped into the Pit-Grave culture, especially by Eastern Euro archeologists.'

    Russian
    Not the rest of eastern Europe.;)

    ReplyDelete
  27. @juanrivera well, @davidski can correct me but according to you, if WSHG/Botai has R1b-M73 + Q1a then it’s very similar to Steppe IE groups. I mean Botai might be Europoid rather than East Eurasians looking and speaking a language close to an early PIE.

    Are you sure you are not conflating ANE, WSHG, EHG and Baikal HG by chance?

    ReplyDelete
  28. What is the current estimation for Steppe admixture in modern Northern Europeans? Is it still around 50%?

    ReplyDelete
  29. @Dragos

    W.r.t. horse domestication in Dereivka. There are two refutes to Anthony's idea Dereivka represented a site with domesticated:

    1) Marsha Levine compared the age distribution of Botai and Dereivka to different models and concluded the horse remains in both were the result of hunting. However, David Anthony already showed a age distribution of horse remains from a Roman site which showed remarkable similar pattern, and we can be pretty sure those were domesticated horses.

    https://www.researchgate.net/publication/265425122_Domestication_and_early_history_of_the_horse

    2) At Botai mare's milk lipid residue is recovered from pot shards. But milk is the definitive proof for domestication. If the Botai age curve matches Dereivka so well, and we know Botai milked mares, then we can be sure that Dereivka's age distribution is a sign of domestication rather than of hunting.


    https://www.researchgate.net/publication/24180858_The_Earliest_Horse_Harnessing_and_Milking

    There is one detail: Pot shards from Dereivka have been targeted for lipids too and no unambiguous dairy lipids have been found, with one possible exception.

    "evidence of ruminant dairy product exploitation was insignificant, with only one residue displaying a possible ruminant dairy fat origin. "

    https://research-information.bristol.ac.uk/en/publications/new-insights-into-the-subsistence-economy-of-the-eneolithic-dereivka-culture-of-the-ukrainian-northpontic-region-through-lipid-residues-analysis-of-pottery-vessels(f3908dd5-a1e2-4990-af15-ab9dfc1b966c).html

    I'd say Sredny Stog simply didn't milk their mares. But hey, no modern day horse keeper outside Central Asia does that either.

    ReplyDelete
  30. The fact that so very few Male Horse Lines survived could be a sign of selection. We see in the 3300 BCE Bulgaria horse remains that most were Male of different ages. This points towards early selection for certain traits...Did anyone consider this when looking at horse remains on the Steppe ?

    ReplyDelete
  31. @Davidski, cheers.

    @Tesmos, Davidski's qpAdm Tour of Europe is still pretty good I think - http://eurogenes.blogspot.com/2017/01/qpadm-tour-of-europe-bronze-age-invasion.html. Though he may believe there are sharper estimates he could make on more recent data.

    These are Steppe_EBA. If you want an estimate in terms of Steppe_Eneolithic from Wang's paper then we still have to guess, but about 90% as much of Steppe_EBA (e.g. based on Steppe_EBA, Steppe_Eneo - English 38%, German 35%, Belarusian 39%, Lithuanian 42%; then assuming Steppe_Eneolithic about 60% EHG, EHG would go, English 23%, German 21%, Belarusian 24%, Lithuanian 25%. Further - https://imgur.com/a/erQJkNn).

    ReplyDelete
  32. @All

    Happy new year, by the way.

    ReplyDelete
  33. One of the thing which can be linked with IE migration in NW Anatolia is burials with red ochre.

    ReplyDelete
  34. @Folker

    Could elaborate a bit on that?

    ReplyDelete
  35. A great YT presentation about the Proto-Indo-Europeans:

    https://www.youtube.com/watch?v=C4JPMYHTZis

    ReplyDelete
  36. @Davidski Do you think that David Anthony's book is too outdated today?

    ReplyDelete
  37. BTW, "Survive the Jive" on the YouTube video about the Proto-Indo-Europeans above mentions your blog at the end of his flick: "check your data on Eurogenes".

    ReplyDelete
  38. I'm watching another YT video about the Etruscan language. Would be interesting to know whether they are a Neolithic Farmer relic or a new wave of later invasion by CHG groups, non-IE speaking of course, akin to the ones who replaced the Anatolian, Aegean and to a large degree - the members of the LBK culture in Central Europe.

    ReplyDelete
  39. After all that's been said and done, I think early Khvalynsk/Samara is the best candidate for the original PIE community. To quote Anthony:

    "Immigrants from the Volga seem to have arrived in the Dnieper-Azov steppes at the beginning of the shift from DDII to Sredni Stog, instigating changes in both funeral customs and pottery making."

    I think this provides the connection we need to link Khvalynsk and SS II together to provide a single possible language community. Early Khvalynsk dates to the beginning of the 5th millennium, well before 4300 BCE, plenty of time to trigger an expansion.

    ReplyDelete
  40. @Davidski

    How can the commonalities between Balto-Slavic and Indo-Iranian be explained in case they originated from two different populations?

    ReplyDelete
  41. @Them meee: As I understand it, people who propose a IE tree structure where I-I languages branch off deeply from Slavic (at a depth too deep to be compatible with originating from Steppe_MLBA) tend to suggest that there is some contact and influence via the BMAC and Central Asian links in its chain anyway (one way or two way).

    It seems like it's not like the similarities are so clear that you can say that obviously originate from the same group either way (e.g. "few shared innovations" from PIE per Kortlandt and "not that many shared morphological or phonological features" between any branches as summarised by Armchair Prehistory inc. Slavic, I-II, when discussing how linguists have gone about constructing trees, and certainly there does not seem to be any especial lexical closeness between existing Slavic and II languages). Plus no one agrees about whether the lack / presence of shared M/P innovations that there are actually indicate descent, influence between branches, substrate influence, etc.

    Basically it's about weighing up the simpler hypothesis rather than that it can't be explained one way or the other (most informed linguists today still seem to say EBA Steppe).

    ReplyDelete
  42. Yamnaya formed just before 3000 BC, but the Pit-Grave or Repin culture, from which Yamnaya descends directly, formed as early as 3900 BC.

    The Varna outlier is dated to ~4500 BC so of course we're looking at something earlier for PIE, and even earlier than the date range of the expansion alluded to in the blog post above (4300-4100 BC). It's pretty clear it has to be an early date for things to make sense. Perhaps even *really* early, and the Bayesian linguistic analyses that keep finding a date around ~8000 BC aren't that crazy after all... doesn't mean it has anything to do with farmers though.

    Sometimes Repin and Yamnaya are both lumped into the Pit-Grave culture, especially by Eastern Euro archeologists.

    This is confusing for Russian speakers, as Yamnaya means something like "pitted" or "of the pit" so basically Yamnaya and Pit-Grave reference the same thing. Generally archeological naming seems to be quite a mess with a lot of overlap and not much consensus. :-/

    ReplyDelete
    Replies
    1. Not with Farmers. We know farmers spoke a different language family altogether. I assume a common source with other EHG/ANE -derived populations, like: Yenisseyan, Kamachatka-Chukchi, Nivkh etc. Linguitics point in the direction of PIE-Chukchi/Kamachatka similarities on lexical level.

      Delete
  43. The analysis in Narasimhan 2018 text gives WSHG as "~30% derived from EHG, ~50% from Ancestral North Eurasians (defined as being related deeply to 22000-15000 BCE Siberian (29, 30)), and ~20% related to present-day East Asians"

    Go to the proportions in their supplement (table S3.44) and those give out WSHG as 36% Karelia_HG, 10% Han, 54% ANE.

    If you assume EHG are 66% ANE, then WSHG ((36*0.66)+54) = 77%, or if EHG are assumed 80% ANE, then WSHG ((36*0.8)+54)=83%, or if EHG are assumed 50% ANE then WSHG, ((36*0.5)+54)=72%.

    If you follow the model in Sikora 2018 then EHG fits as 7.5% Basal Eurasian, 68.5% ANE, 24% WHG (https://i.imgur.com/0EKyXZY.png). (That is 17% CHG, 59% ANE, 24% WHG). Then WSHG goes to about 2.5% Basal, 79% ANE, 10% Han, 8.5% WHG.

    But that would probably better verify with an independent check on any CHG/Iranian related ancestry into WSHG. (Plus this is based on high Basal Eurasian estimates at about 45-50% into CHG and Anatolian, while some estimates go low, at about 20-25%).

    Botai probably has about 7% more East Eurasian ancestry than WSHG and proportional reduction in other components.

    ReplyDelete
  44. Oops, correction, actually Sikora models EHG as 7.5% Basal Eurasian, 8% Central HG, 61% ANE, 24% WHG. Then WSHG would go to about 2.5% Basal, 3% Central, 76% ANE, 10% Han, 8.5% WHG.

    (Central HG = sister to WHG and HG ancestor of CHG).

    ReplyDelete
  45. The book they cite is a 250 pages treatise on the Eneolithic roots of the Kurgan cultures on the North Caucasian Steppe in the Vth and IVth millennia BC. Here is the summary and the contents table of the publication for your auto-translation
    Кореневский С.Н. Рождение кургана (погребальные памятники энеолитического времени Предкавказья и Волго-Донского междуречья). М.: Таус, 2012.

    Монография посвящена изучению феномена кургана на самых ранних его проявлениях в эпоху энеолита в Предкавказье и на юге Восточной Европы. Тематика работы тесно связана с проблемой формирования культурного своеобразия восточноевропейского и предкавказского населения в обширном регионе западного фланга евразийской степи в V–IV тыс. до н.э. В книге показано, как в это время в результате сложных историко-культурных процессов, улавливаемых по археологическим источникам, развивались культы и верования, происходило обособление родоплеменной знати, а также акцентировалась роль личности, которую подчеркивали индивидуально курганные захоронения.
    Работа рассчитана на специалистов — археологов, краеведов, историков, искусствоведов, специалистов по религии, этнологов, лингвистов и на широкий круг читателей, поскольку значимость феномена кургана как символа связи времен интересует многих.
    Оглавление:
    Введение
    История изучения проблемы ранних курганов на юге Восточной Европы.
    Природный фон, термины и понятия.
    Каталог памятников
    Центральное Предкавказье.
    Волго-Донское междуречье, Калмыкия.
    Прикубанский регион.
    Предметы и традиции погребального обряда
    Предметы погребального обряда.
    Особенности погребальной практики. Поза скелета. Охра и угли в обряде, использование камня и дерева, формы погребальных сооружений.
    Хронология, ориентация подкурганных Предкавказских и Волго-Манычских памятников энеолита в климатологической схеме Блитта-Сернандера и культурном пространстве
    Хронология памятников.
    Культурная ориентация Предкавказских и Волго-Донских памятников.
    Анализ подкурганных энеолитических комплексов — социальный и культовый аспекты
    Эпоха раннего энеолита.
    Эпоха позднего энеолита.
    Комбинаторика вещей в комплексах.
    Особенности новых культов степных племен позднего энеолита понто-предкавказской общности.
    Сравнительная характеристика древнейших курганов Кавказа: энеолит — ранний период бронзового века
    Курганная практика майкопско-новосвободненской общности.
    Древнейшие курганы Южного Кавказа и северо-западного Ирана.
    Сравнительная характеристика древнейших курганов Кавказа в Урукский период Переднего Востока.
    Заключение. Феномен древнейших курганов Восточной Европы и Кавказа

    ReplyDelete
  46. @Matt

    I just tried this in G25 nMontes. Not the best method for deep ancestries like this, but some additional information.

    Global 25 nMontes EHG:Average
    AfontovaGora3 67.5%
    Barcin_N 2.5%
    CHG 3.33%
    Villabruna 26.67%
    Fit: 6.0082

    Global 25 nMontes West_Siberia_N:Average
    AfontovaGora3 74.17 %
    EHG 16.67 %
    Han_NChina 9.17%
    Fit: 4.2137

    Assuming Afontova Gora 3 can be taken as pure ANE (probably not, but whatever) and using your calculation this some ~80% ANE in WSHG. Less in Botai, because Botai has both BHG and Khvalinsk ancestry over WSHG.

    @Andrzejewski
    This does not mean that WSHG was very similar looking to EHG, that is more complicated than this, especially since there was also some evolution since ANE. And ANE itself was a divergent lineage between West Eurasians and East Asians, so they were neither.
    Also CHG had much less ANE than EHG.

    Also the ANE ancestry in EHG, CHG and WSHG had something like 6000+ years or probably more divergence by the time of the formation of Khvalynsk and that is way too much time for close linguistic cohesion. There is absolutely no reason to assume that Botai spoke a PIE related language.

    And on a general note: please stop conflating language and "look".

    ReplyDelete
  47. PS: Vonyuchka, sometimes spelled Vonjuchka in the preprint and its supplements (Russ. Вонючка, N 44.019962°, E 43.155538°) is an unbelievable appropriate name for the hypothesized cradle of the Indo-European culture. It has been excavated in 2010 at a sewage treatment facility of the city of Pyatigorsk, and the archaeologist appropriately called it Вонючка which just means "Nasty Stinker" in Russian. A 40 m dia, 2 meter high mound with a skeleton in a catacomb-like formation, densely packed in red ochre, with a partly healed trepanation of the skull. Up to 4300 BC.

    ReplyDelete
    Replies
    1. @MOCKBA So is it Voynuchka and not Samara or Khvalynsk that is the cradle of PIE?

      Delete
  48. Just listened through a radio program about the recent genetic discoveries on prehistoric Finns. Again they repeated the untruth that Indo-European languages were spread by Anatolian farmers. Not really surprised since the the Finnish public broadcasting company is pretty ultra-left and the idea of aryan warriors from the steppe is equal to supporting nazi eugenics to them. The steppe theory is far from being accepted in the current political climate.

    ReplyDelete
  49. @All

    No more physical anthropology on this blog.

    ReplyDelete
  50. @ Synome

    ''After all that's been said and done, I think early Khvalynsk/Samara is the best candidate for the original PIE community. To quote Anthony: ''

    I'd hedge that Khvalysnk & sasmara were non-IE.
    If IE is from the steppe, then its from Deereivka'/ West Sredny-Stog/ Karanovo VI. That's the only way it'll work for Balkans & Anatolia.
    Repin and R1b heavy groups were originally non-IE.
    More evidence will prove that. .....

    ReplyDelete
  51. @ Epoch

    ''W.r.t. horse domestication in Dereivka. There are two refutes to Anthony's idea Dereivka represented a site with domesticated:
    ''

    In fact, the refutes are a multitude. He's views are taken with extreme caution by specialists.
    Putting aside the Dereivka debarcle where the horse trappings returned dates of the Iron Age instead of Eneolithic, the arguement is essentially personal anecdotae & false analogy.
    ''I have no evidence, but I still think that horse were ridden in the Eneolithic because Numidians did it. And they did it with biodegradable material. Oh, and they just rode to battle, then jumped off ! ''
    It's one matter to be able to ride bareback (e.g. as the hoplites and Numidians) thousands of years after horses have been collectively domesticated, it's another thing to try do it on a wild animal.
    Even Yamnaya in Hungary, there is not much evidence for horse use.
    The earliest more convincing indirect comes from BB period - eg the noted acetabular pathologies in men, and the sheer rapidity of its expansion are suggestive.
    Even if the horse was beginning to be domesticated in the Eneolithic, then it would have been for food.
    In any case, it's just fictional speculation, because there was no invasion of ''old Europe'' yb horse-riding ''Indp-Europeans from the Volga.
    Pure Science-Fiction.

    ReplyDelete
  52. Like, I’m still confused. If Indo-Iranian and Baltic-Slavic are so closely related, how could we explain that if they originated in different communities?

    ReplyDelete
  53. @ Dragos

    Where do you think horse domestication happened ?

    ReplyDelete
  54. The main thing to keep in mind here is that, considering the major role of the horse in Indo-European religion, it's impossible that the Proto-Indo-Europeans came from a community that didn't make regular use of horses in some way and from a region where horses weren't initially very common.

    Horse riding and the use of horses in battle aren't necessarily directly related to the PIE homeland question, and don't even need to be discussed in this context.

    The earliest incidence in the horse ancient DNA record of a major Bronze and Iron Age domesticated horse Y-chromosome lineage (Y-HT-4) is this...

    May1, Mayaki, Eneolithic Ukraine, 3600-3100 BCE.

    Map HERE.

    So I'd wait for that major ancient DNA horse paper before arguing that the modern domesticated horse didn't come from the Eneolithic communities on the steppe.

    ReplyDelete
    Replies
    1. Yea, here’s the trick: I think it was either Gimbutas or Anthony who stayed the the transformation from Samara —> Khvalynsk was accompanied by horses turning from a source of wild game into a domesticated transportation vessel

      Delete
  55. @Them meee

    Like, I’m still confused. If Indo-Iranian and Baltic-Slavic are so closely related, how could we explain that if they originated in different communities?

    It depends what you mean by "different" communities. Balto-Slavic and Indo-Iranian probably both come from Corded Ware-derived communities.

    ReplyDelete
  56. Interesting that you bring up PIE religion, David. There is nothing in PIE religion that says the Divine Twins were domesticated horses but that doesn't matter because horses were common in the Caucasus anyways.

    Let me propose this: It is impossible that the Proto-Indo-Europeans came from a non-mountainous region where there was no proper metallurgy. I mean the most important PIE myth of Trito*(or Zeus/Thraetona/Indra) has him literally ascending a holy mountain after the smith hero/god make him a golden bolt/mace/vajra.

    Oh, and a cult of hearth, three castes...etc.

    Does that sound like a culture of egalitarian fishermen from Khvalynsk?

    ReplyDelete
  57. @Vara

    Here's my interpretation in this context of the Indo-European myth that you cite: holy things aren't usually things that are common or easy to find, so big mountains and metal were not general features of the place where the first Indo-Europeans lived.

    By the way, how you're still deluding yourself that the PIE homeland was in the Caucasus after seeing the data from Wang et al. is beyond me.

    ReplyDelete
  58. @ Ric

    ''Where do you think horse domestication happened ?''

    There's no reason why it did not happen somewhere on or near the steppe. Afterall, it was widley hunted there.
    What I do doubt is some of the reconstructions put forward. The key is in the details of the stages of how it happened, and the relevance at large.

    ReplyDelete
  59. @ Vara

    ''Does that sound like a culture of egalitarian fishermen from Khvalynsk?''

    I doubt PIE came from Khvalynsk. There's nothing there.
    Oh wait, but there's 0.0033% steppe in one of the MBA Anatolian females !

    ReplyDelete
  60. Someone at Anthrogencia already did a qpADM model using the Steppe Peidmont samples. Is the genotype data available?

    https://anthrogenica.com/showthread.php?14285-The-genetic-prehistory-of-the-Greater-Caucasus-preprint-Harvard-Jena&p=533068&viewfull=1#post533068

    ReplyDelete
  61. The BAM files are available, and they can be converted to genotypes, but I won't be doing that.

    ReplyDelete
  62. @Davidski

    Like if Corded Ware came from Yamnaya and Balto-Slavic came from Corded Ware while Indo-Iranian came from Sredny Stog, but if both are closely related then...?

    ReplyDelete
  63. @Them meee

    Too hypothetical to worry about at the moment.

    There are still lots of variants possible, with just one being that Corded Ware was a mixture of Yamnaya and post-Stog groups, and therefore showing a wide range of the level of farmer ancestry already on the steppe, with Indo-Iranians coming from the farmer rich groups.

    ReplyDelete
  64. @Davidski @Them meee The origins of The Tocharians is a mystery. I don’t believe that they are direct descendants of Afanasievo

    ReplyDelete
  65. @Matt

    Here are those f3-stats.

    https://drive.google.com/file/d/1T3d5LmcgytNi8K5YBS1O4y5dKfM9aBGp/view?usp=sharing

    FYI, there's no Iran_N in the dataset, and hasn't been for a while.

    ReplyDelete
  66. Interesting almost all the R1b samples including the 6k+/-BP are located in the ancient region of Alania, aka North Ossetia, an Eastern Iranian dialect region. Modern day Ossets have R1b1a2a2c1 Z2105+, CTS9219+, Y5586+, and DYS389a=12, DYS520=22, DYS456=16, DYS413a=22 that means R1b has been in this region for the past 6200+/- YBP
    PG2001 Progress 2 6207.0 Eneolithic steppe I3a R1b1
    PG2004 Progress 2 6090.0 Eneolithic steppe H2 R1b1



    BU2001.A0101 Beliy Ugol 2 North Caucasus R-Y20993
    GW1001.A0101 Goryachevodskiy 2 North Caucasus R-Z2103
    I1723 Goryachevodskiy 2 North Caucasus R-L23
    KBD001 Kabardinka Late North Caucasus R-Z2103
    LYG001.A0101 Lysogorskaya 6 North Caucasus R-Z2103
    MK5009.A0101 Marinskaya 5 North Caucasus R-Y20993
    PG2001 Progress 2 Eneolithic steppe R-V1636
    PG2004 Progress 2 Eneolithic steppe R-V1636
    RK1001.C0101 Rasshevatskiy 1 Yamnaya Caucasus R-Z2103
    RK4002.B0101 Rasshevatskiy 4 Catacomb R-Z2103
    RK4001.A0101 Rasshevatskiy 4 Catacomb R-Z2108
    SA6003.B0101 Sharakhalsun 6 Catacomb R-Z2103
    SA6013.B0101 Sharakhalsun 6 Steppe Maykop outlier R-M459 R-YP1272 ??
    SA6010.A0101 Sharakhalsun 6 Yamnaya Caucasus R-V1636

    ReplyDelete
  67. @Dragos How would PIE from Dereivka work for Balkans Anatolia, if PC steppe is urheimat? Only future Myceneans and Hittites will clarify, but if I had to guess we'll see more J2 and G and little R.

    ReplyDelete
  68. If I had to guess we'll see more J2 and G and little R.

    So what? No one said Mycenaeans and Hittites were paternally 100% from the steppe.

    Also, interesting factoid: the currently available Mycenaean J2 subclade doesn't look typically Anatolian, Caucasian or Iranian, but rather Neolithic Central European.

    So where did it come from, the Iranian PIE homeland? Haha.

    ReplyDelete
  69. A major paper which came out early last year claimed that Yamnaya and Botai had no contact nor genetic relatedness whatsoever. Botai domesticated Paszelski horses before Yamnaya, although the latter ones are the ones who sired all modern horses.

    I presume that Wang et al debunked that previous conclusion. Now, maybe the bonds between the cultures go deeper in time to Khvalynsk; I do on the other hand suspect that the Q1a1 sample with his head bashed in might’ve been a result of a hostile encounter between Steppe groups from Khvalynsk and a Botai trader.

    ReplyDelete
  70. @Andrzejewski

    Botai samples are very different from the Khvalynsk Q1a1 sample. They have significant East Asian ancestry and no or very little CHG, while the Khvalynsk sample looks basically like Yamnaya.

    So not only were the Botai horses different, but the people were too.

    ReplyDelete
  71. And is it almost totally likely Bell Beaker comes from Yamnaya, or not necessarily?

    ReplyDelete
  72. @ Davidski
    Can you please clarify..
    There is J2* in one of the LBK
    There is J2a1 in Anatolia central (curiously, not in the southwest which is J1).
    So how are you sure that Mycenean J2a1 is derived from central Europe ?

    ReplyDelete
  73. @Them meee Yes, most likely: Beaker is R1b-M479, same as Yamnaya. CWC is most likely derived from Sredny Stog (R1a1).

    ReplyDelete
  74. Am I the only one here finding some sort of correlation between CWC-offspring (Balto-Slav, Indo-Aryan) or R1a1 and the phenomenon of Satemization?

    ReplyDelete
  75. @Andrzejewski

    "Am I the only one here finding some sort of correlation between CWC-offspring (Balto-Slav, Indo-Aryan) or R1a1 and the phenomenon of Satemization? "

    You definitely aren't, it's not a new idea. But to imagine that we have the whole story at this point with respect to the linguistic situation is a little unrealistic, we really should be open to the possibility that it was more complex than that, because it may have been.

    ReplyDelete
  76. @Davidski On a general note I can't wrap my head around the fact that the construct of "Europe" has more or less corresponding similarities (without getting too profoundly into physical anthropology) when it comes to language (overwhelmingly IE), physical appearance, historic "Christianity", sort of part of a "Western" civilization, whereas as soon as one crosses the Ural Mountains they encounter East Eurasian looking people speaking non-IE speech; or venturing across the Bosporus straits into Turkey, all of the sudden you are in the "Middle East", with people's appearance, languages, customs all of a sudden change. Likewise the Caucasus Mountain Range, as soon as you depart "European Russia" and enter the Caucasus people mostly don't speak IE languages (except for Armenian); Republic of Georgia people as well as Armenia can easily and readily pass as "West Asian" rather than "Europoid".

    So I'm basically trying to understand how come so many people in places like Siberia and the Middle East/Levant share so many alleles and ancestros and yet - Europeans descended from Anatolia_N look very different than Turkish or Kurdish people who do, while the same goes towards descendants of CHG groups.

    So bottom line, my conclusion is that pre-BA Europe, Levant and Siberia were very similar in terms of ancestry and population genetics, but that Indo-Europeanization via a strong replacement event and demic diffusion originating in the Steppes has made a tremendous difference on the genetic, phenotypical, cultural and of course linguistic - landscape of Europe.

    ReplyDelete
  77. @Dragos & Epoch
    Several skeletons (11) with ochre have been found in Ikiztepe. This is seen as a cultural import from the Balkans or Pontic Steppe by somes (Zimmerman, Backofen). Even if Bilgi has some doubts about the intent (if ochre was not intentional), if it was intentional, he made a link with populations from Northern Black Coast.
    Obviously, ochre is not an unambiguous signal, but given the location (Ikiztepe) and the period (last quarter of the IIId millenium), it is something which could be significant.
    Ikiztepe has a long history of relations with Balkans, and presence of grave goods in arsenical copper is pointing to trade with other parts of the Black Sea (among other things). Also, Ikiztepe is at the door of the land of Pala. Presence of IE there should be expected. And at this time, not other burial with ochre has been found in Anatolia.

    ReplyDelete
  78. And while we are at it: there was a Lazaridis-authored preprint which came out back in 2013, and already then he concluded that the Anatolia_N EEF (which had been mixing with WHG groups as well) have been replaced by non-IE Caucasus HG groups swarming first Anatolia, then the Aegean and later on Central Europe.

    It was hypothesized that Etruscans and Minoan Greeks were not Anatolian but Caucasian in origin, and so were their extinct languages.

    Moreover, this CHG influx was responsible from increasing the ratio of mtDNA H from 19% (average in Middle East) to 45% today, starting in 6500ybp.

    Coincidentally or not, it was written in that paper that light skin frequency has also started to increase gradually since 4500 calBC.

    Has anyone else read this paper? Is there any conceivable affinity or correlation between 1. Increase in non-IE CHG across Europe; 2. gradual doubling of mtDNA H; and 3. lightening of the skin?

    ReplyDelete
  79. @Folker @Dragos & @Epoch red ochre was asserted by Renfrew and others to originate either on or South of the Caucasus, not unlike the Kurgan burial style, metallurgy and the potter wheel...

    Now we are more inclined to attribute all these inventions to more Westernly influences on fledgling Steppe cultures, for example the Cucuteni, possible via Bug-Dniester and Dniester-Donets. After all, it is now acknowledged that Yamnaya harbored a considerable EEF component, which could *ONLY* come from the west.

    ReplyDelete
  80. Explain “Europe, Levant and Siberia?

    ReplyDelete
    Replies
    1. Most of Europe and Middle East were populated by largely Anatolia_N groups (farmers), whereas Baltics and east were populated by same more or less EHG all the way to the Altai.

      Delete
  81. @Dragos

    The singleton Mycenaean Y-chromosome apparently belongs to J2a-L26(Z6057+), thus it shows a link with the Sopot and Lengyel cultures to the north of Greece, rather than anything east of it.

    ReplyDelete
  82. @Andrzejewski

    And while we are at it: there was a Lazaridis-authored preprint which came out back in 2013, and already then he concluded that the Anatolia_N EEF (which had been mixing with WHG groups as well) have been replaced by non-IE Caucasus HG groups swarming first Anatolia, then the Aegean and later on Central Europe.

    This, overall, doesn't sound right.

    ReplyDelete
  83. @ Folker
    The late M3 was put forth by the original excavators; same thing had happened with the Majkop originally, under the ''Mesopotamia first'' paradigm
    However, Lichter and many others date Ikiztepe from 4500 BC. This is based on numerous parallels with Karanovo VI, Varna, Tiszapolgar, etc. I guess if/ who is doing aDNA analysis on the site might also directly carbon date the skeletons. ...

    @ Davidski
    Interesting. Is there a link to a source for that , e.g. TK or CR ?

    ReplyDelete
  84. @Dragos Folker seems to have a habit of taking the academic consensus and then subtracting around 2000 years to fit his hoped for theories, remember the late arrival of Anatolian? Of course the academic consensus can turn out to be wrong, new data is always coming in, but it seems like the Damgaard paper provides yet more evidence for an early entry.

    ReplyDelete
  85. If I'm not mistaken it is the specific use of ochre within the burial that counts. Some covering the body and gravegoods with ochre and others using ochre as decorations on the walls etc....? So although ochre was used it was used in different ways by Steppe and Early Anatolian peoples....

    ReplyDelete
  86. @Dragos

    This might be of help...

    https://anthrogenica.com/showthread.php?12858-The-History-of-J2&p=531745#post531745


    ReplyDelete
  87. @Dragos & Al Bundy
    Ikiztepe was occupied from the Chalcolithic to the MBA. Even if there is some debate about datation, the last papers I’ve read (2014 to 2017) don’t challenge a datation of the last quarter of the last quarter of the IIId millenium for those burials.
    As I said, those results are not unambiguous.
    By the way, if you dare read papers from well respected Hittitologists, you’ll find that there is a large consensus among them to an arrival of various IE groups in Anatolia in the IIId millenium, not before.

    ReplyDelete
  88. What boggles my mind is, If Hittite split from PIE very early why isn't it closest to PIE. Why is Lithuanian proposed as the closest ? Doesn't this tell us that Hittite underwent a lot of changes and can not really be used as a benchmark for the characteristics of what PIE should have looked like ? If Hittite originated later than previously proposed it is even more troublesome since it points to very quick changes taking place which suggests a liberal linguistic adoption of anything new the Hittites came into contact with....

    ReplyDelete
  89. @Ric Hern
    To my understanding, Hittite was heavily influenced by Hattic.
    If you read the last papers by Melchert, you’ll find that he is challenging the idea that Hittite is far much more archaic. Some features could very well be innovations or influence from an unknown language. Perhaps from another extinct branch of PIE.
    If Hittite, as some have wrote, has influenced from other branches of PIE (like Italic) or Uralic, the ancestors Anatolian IE must have been in a relative close neighborhood to the Pontic Steppe for some times. Balkans are clearly on the top of the list in this prospect. And if the split is dated to the IVth millennium (which is rather consensual), we have plenty of candidates to be the ancestors of Hittites in the Balkans, as Steppe migrants can be traced to the Bosphorus.

    ReplyDelete
  90. Ric hern: PIE is only a theoretical construct. It’s the best linguists can approximate an ancient language with the available sources. And the theory would clearly have a bias towards existing and highly abundant languages over those we have little to no idea about.

    In reality the language we can call PIE would be significantly different from the theoretical models we have today. Even so, theory is extremely helpful. Lithuanian is not a fossilized version of PIE. It’s merely the most conservative language, quantitatively, relative to other modern languages after the Anatolian split.

    We can’t say for sure what happened between the split of Anatolian and the rest. Whether either went through more changes or received more external influences. A lot of things we will never know simply because no decipherable written sources were ever left.

    ReplyDelete
  91. @ Folker

    I'm not sure what you might count as a ''well-respected Hittotologist'' is, but word from the ground (Sari, Massa, Bachluber) have not detected any intrusive signs in the late third millenium, nor in fact, do those from preceding decades (Mellart, Darden).

    Needless to say, its a well known fact that current linguistic concensus favours the so-called Indo-Hittie model, although you mention Melchert specifically, who in his ''Luwians'' suggests ''There are no convincing examples of direct Hattic influence on Hittite morphology.''and ''Turning to the lexicon, we may begin by dispelling the persistent myth that Hittite has replaced much of its inherited Indo-European vocabulary.''

    As for the dating, Melchert suggests ''Darden (2001 204) presents arguments for the end of the fifth millennium'' and ''It is tempting to see the Demircihuyuk culture described by Korfmann (2001 36Iff) as associated with the arrival of Indo-Europeans in Anatolia (cf. the remarks of Yakar 1981 96 and Darden 2001 220). The dating to the second half of the fourth and first half of the third millennium would correlate reasonably with those proposed on
    linguistic criteria above..'' Not to mention Finkelstein, Garret-& Chang, Yakubovich, etc, etc all move between 4500-3500, some earlier.

    It could turn out that there was a migration in the 2200s, as you suggest, but I think you need to take some time to properly understand what current data points to.

    ReplyDelete
  92. So is in the OT book of Genesis, but they are based on Sumerian ones.

    ReplyDelete
  93. Anyone know if remains have been excavated from EBA Troy I and related sites?

    I've read multiple sources connecting the early Bronze age in the Balkans with the same period in NW Anatolia, including the first layers of Troy.

    We already have steppe ancestry in the EBA Balkans. Time to search Northwest Anatolia in the same time period.

    ReplyDelete
  94. @Dragos Folker How many Hittite remains are available, where and when are they being tested?

    ReplyDelete
  95. @Dragos
    The lack of intrusion means the lack of destruction. Because, « in the old days », migration of IE was equated to « migration of barbarians who destroyed everything on their path ». Obviously, we know better, and migration is not always easy to detect. DNA has proven this fact for sure.
    As said, there is very large consensus about the intrusiveness of IE in Anatolia. I know the Trevor Bryce is not on this line, but he is a bit alone (and there are not so many papers in English about Hittites, far more in French, Italian or German, for obvious reasons).
    Melchert has changed his point of view. He made it clear in his last papers, at least since 2012. He is now favoring a more recent split as said above.

    ReplyDelete
  96. If Anatolian came from S.Caucasus there must have been big migrations from West Asia.We have some EHG but it's been pointed out it's not, might not, be from the PC steppe.Is it female-mediated, sideflow?You can find steppe and EHG if you want but it's context that matters.



    ReplyDelete
  97. If Hittites only arrived in 3rd M BCE, it wasn't from anywhere in Europe. How does a pop between Anatolians ChL, Hajji Firuz, and Levant N equal steppe? Sorry, but there won't be any magical steppe fairy Hittites coming.

    ReplyDelete
  98. @Dragos
    I am not a linguist, but to my knowledge the Indo-Hittite model is out fashioned and defended by a small minorities nowadays.
    One point must be taken into account: the date of the split of the Anatolian branch is not the date of the entry of IE in Anatolia. Some features can be better explained if the ancestors of Anatolian IE were still in contact with some other PIE. This is the reason why Melchert has changed his view. And why he wrote about influence of Proto Italic on Hittite (among other things).

    ReplyDelete
  99. @ Chad
    Hittites were a small minority in Hatti. Even at Kanesh they were only « the largest » minority. But Hittite was the administrative language, and diffused in Central Anatolia with conquest. Luwian was a trade language.
    And the Hattic substrat in Hittite and Palaic is very telling about the size of their population. Anatolia during the EBA and MBA saw coexistence of various populations of different background in the same places. It is very clear from the tablets.
    If you don’t take into account those facts, I don’t see you could solve the origin of Hittites.

    ReplyDelete
  100. @Folker

    "As said, there is very large consensus about the intrusiveness of IE in Anatolia. I know the Trevor Bryce is not on this line, but he is a bit alone.

    I happened to come across Robert Drews' Militarism and the Indo-Europeanizing of Europe yersterday, which maintains a variant of the Indo-Hittite hypothesis while agreeing with the PC Steppe as the near-certain locus of Late PIE. You can read the gist in the previewed introduction:

    https://books.google.it/books?hl=it&lr=&id=SCvZDgAAQBAJ&oi=fnd&pg=PT8&dq=%22indo+european%22&ots=-BvyUM0Sj5&sig=Stm2tupwtwWpc54mESu6TJxSNvU#v=onepage&q&f=false

    Interestingly enough, he has since abandoned his support for the Armenian urheimat as outlined in his Coming of the Greeks from 1994. Just thought I'd share, keep up the good discussion.

    ReplyDelete
  101. @Chad

    Not sure why you're being so dogmatic about this, unless you know the results of around 100 samples from across space and time in Bronze Age Anatolia.

    The PIE from Armenia/Iran via Maykop thing didn't work out. It was never going to work out, because it was a theory based on the wrong assumption about the genome-wide ancestry of Yamnaya, rather than anything resembling a sound multidisciplinary approach.

    So now it's time to move on, or rather, reverse back to the time before someone or rather decided that Proto-Indo-European speaking Mesopotamians migrated to the steppe to give rise to Yamnaya and Corded Ware, and look at everything more realistically and objectively, especially now that there are very real signals of incursions from the steppe into the Near East during the Bronze Age.

    ReplyDelete
  102. Could it be that Satemization/Centumization, regardless of which was “the original state of IE”, occurred back in the steppe circa 4000-3000?

    ReplyDelete
  103. @Chad " If Hittites only arrived in 3rd M BCE, it wasn't from anywhere in Europe. How does a pop between Anatolians ChL, Hajji Firuz, and Levant N equal steppe? Sorry, but there won't be any magical steppe fairy Hittites coming. "

    You seem fairly certain with your idea while using less than 10 samples context and paper only on pre-print phase. Wich sounds very dishonest science. Also, that's very unlikely that the Steppe migration genes didn't irradiate in Anatolia, even without changing the local cultures. If on 100 or 200 samples from Chalcolithic and Bronze Age Anatolia we dont found any Steppe ancestry, i would way more think of a conspiracy than anything else.

    ReplyDelete
  104. I’m glad to hear that. But where do you supposed the CHG came from?

    ReplyDelete
  105. @all my theory is that some authentic PIE or pre-PIE started developing somewhere towards the Eastern boundaries of the Steppes, maybe as an offshoot of some ANE language. Then as they migrated westbound they packed original CHG natives via the maternal side because of exogamy

    ReplyDelete
  106. @All

    I don't want the comment threads here to degenerate into petty, childish ethnic arguments.

    If you have an original theory about your ethnic group being the most Indo-European that isn't actually supported by anyone except yourself and a few of your friends, please keep it to yourself.

    ReplyDelete
  107. I'm sure, no actually i'm fairly certain that nobody, not even Gimbutas ever stated that Yamnaya would have been the PIE people. The Dniepr, the Don and the Volga within differente cultures that we all know, were always the perfect candidates. The only exception, that even Gimbutas believed was Maikop, it was culturally so similar to Yamnaya, that it had to be related to it, but now we know that Yamnaya and Maikop were unrelated Genetically and that only a Cultural influence could explain the related features. So basically IE migrations for Harvard = " Steppe " Genes / " Caucasian / Iranian " Culture and " Caucasian / Iranian " Language. And i'm not sure how to resolve this.

    ReplyDelete
  108. @ Halfdap

    '' If on 100 or 200 samples from Chalcolithic and Bronze Age Anatolia we dont found any Steppe ancestry, i would way more think of a conspiracy than anything else.''

    LOL. You guys are insane.

    ReplyDelete
  109. @Halfalp

    So basically IE migrations for Harvard = " Steppe " Genes / " Caucasian / Iranian " Culture and " Caucasian / Iranian " Language. And i'm not sure how to resolve this.

    Language can be passed on without any accompanying genetic ancestry, so it's feasible that Yamnaya may have shifted languages from something now lost to Late (or rather PNIE) Proto-Indo-European along with cultural impulses from Maykop, perhaps indirectly via Steppe Maykop.

    But this is so convoluted and riddled with problems that right now I don't consider it possible. For one, the Pontic-Caspian steppe is apparently just about the only place where there's no sign of any non-Indo-European substrates.

    Also, where are all of the basal Indo-European languages in the Caucasus that would point to such a scenario? Conversely, why is there such a close genetic relationship between the Maykop population and present-day non-Indo-Europeans in the Caucasus?

    Moreover, Yamnaya may actually have got its wheels and wool from the farmers in the west, rather than Maykop, and interestingly, it shows admixture from the west and not from the south.

    It'll be fun to watch the MPI-SHH and Broad MIT/Harvard try and get around these issues, because I reckon they will try.

    But my humble advice is, instead of pretending that the PIE homeland was south of the Caucasus, because a few Hittites or maybe in fact Hattians or Assyrians don't show clear ancestry from the steppe, everyone should wait for some decent numbers of ancient samples from Bronze Age Anatolia. Then, and only then, a multidisciplinary study should be undertaken to try and work out where the ancestors of the Hittites came from, and how this might affect the PIE homeland debate.

    ReplyDelete
  110. By the way, let me just add that if anyone here thinks that out of ~200 samples from Chalco/Bronze Age Anatolia, none will show any unambiguous steppe ancestry, then you're not very good at this.

    Population movements from the steppe at that time were massive, and Anatolia isn't that far from the steppe.

    So steppe ancestry will turn up in Bronze Age Anatolian samples sooner or later. But then the question will be whether this actually represents gene flow from the Proto-Indo-European homeland, or something else altogether, like, say, the migration of a few horse whisperers from Sintashta.

    ReplyDelete
  111. I guess they just couldn’t handle that it now looks unlikely Maykop was ever IE.

    ReplyDelete
    Replies
    1. Maykop was either Hurrian or Kartvelian

      Delete
    2. Maykop was clearly NW caucasian.
      Kartvelian Urheimat located in Western Georgia in kolhis plain.
      Hurrian Urheimat in Karas river aka NW Turkey.

      Delete
  112. Yeah, the Wang et al. conclusion looks like a desperate as hell attempt to try and save a dud theory.

    I hope the peer reviewers were honest and straightforward enough to totally knock it out of their heads.

    It's bad enough that it was ever allowed to appear at bioRxiv.

    ReplyDelete
  113. And to possibly curry favor with MPI/Harvard/MIT. Possibly just following trends because now they’re saying PIE is from Iran.

    Even in spite of the genetic evidence saying otherwise.

    It’s mind-boggling.

    ReplyDelete
  114. @Davidski

    Bronze axes/adzes were common on the steppe 4200-3500BCE? It's not like they were literally using stone weapons.

    And yeah, I did see the data and so did Chad.

    I'm not sure why you're banking on Yamnaya that much when you know it's not needed anymore nor is it the right culture anyways.

    ReplyDelete
  115. @Vara

    I don't care what data you've seen and what you think it might be showing. Your data interpretation skills are crap.

    You were arguing with me until very recently that R1b-M269 in Yamnaya males came from their Caucasus ancestors, and they got their EHG ancestry from their mothers.

    That's how utterly clueless you are.

    ReplyDelete
  116. What would be your standard though@Davidski?If there's literally no steppe then maybe it came from Caucasus, but a few measly percent EHG would work for PC steppe?Everything else is J2 and Iran, and some EHG even came earlier? It's not my favorite outcome but if it happened it happened.

    ReplyDelete
  117. You sound a bit neurotic to be honest. You need to calm down.

    Let's relax and wait for some actual Hittite samples from Hittite military sites or even elite burials and then I'll have a look at them and decide for myself what the story is.

    ReplyDelete
  118. @Davidski

    Sure, just like your interpretation of copper workshops and bronze adzes being common on the steppe PIE homeland.

    IMO, it's very simple:

    1. Metallurgy
    2. Mountains
    3. Hounds of death figurines in burials
    5. Cult of hearth
    6. Horses

    ^ That doesn't sound IE to you but an egalitarian fisherman culture does? If Maykop doesn't explain IE religion I'm not sure what does.

    That isn't what I've argued. I thought that some EHG groups brought R1b down through the Caucasus where then it remigrated north with Maykop. Hey we all make mistakes and that includes your exotic bride theory. Either way R1b is not relevant ATM.

    @JuanRivera

    It's not about which is more advanced because there is no contest anyways. It's about which culture fits more with IE religion that we know a great deal about from Hittite to Celtic.

    ReplyDelete
  119. @Vara

    My exotic brides theory is still current.

    Have you noticed that the Eneolithic steppe samples are all R and Q, and yet they show excess CHG over EHG and...drum roll...Caucasus mtDNA?

    How did that happen, duh?

    ReplyDelete
  120. Hittites were hammered from all sides. Kaskians in the North, Egyptians in the South. Hurrians and Assyrians in the East. I think it will take a lot of very specific sampling in the right places to find a significantly Steppe rich population. Wonder why they did not start sampling in Northwest Anatolia+Troy and Bulgaria and from there work it Eastwards ?

    ReplyDelete
  121. It really dosen't make sense to expect some huge ocean of steppe ancestry from the Hittite areas if their society was established by a relatively small elite group, I've toyed with CHG based "indo-Hittite" scenarios myself in the past but really they're just fantasies. It's almost a cert that Hittite comes from the steppe, the only question is how it got to central Anatolia.

    @vara

    "R1b is not relevant ATM."

    Just keep telling yourself that, I'm sure you'll start to actually believe it after awhile.

    ReplyDelete
  122. Via a Caucasus route the only thing that maybe could make sense is that there were a Migration from the Mugan Plain (Connected to the Kugan Steppe) maybe related to the Gutians that formed the Hittites...however seeing that Palaic in Northwest Anatolia basically were blocked by Kaskians to their East and all kinds of other Languages to the East of Hittite the Caucasus route looks very doubtful to me. Wouldn't it be interesting if Maykop was Indo-Europeanized by Steppe people...? Heheheeh..

    ReplyDelete
  123. @Al Bundy

    But a South of the Caucasus PIE homeland has major issues. Early, deep contact with Uralic will become very problematic to explain. Why do Hittites have loan words for the basic flora if Transcaucasia has much the same flora as Anatolia. If Kura-Araxes spread IE, who spread Hurrian? Why don't we see jar burials on the steppe.

    You see, when it turns out that Anatolian samples had not a trace of steppe we have a situation that *every* homeland theory must explain a transfer without gene flow. A Transcaucasian homeland must explain a transfer to the steppe, while we have Wang et al to show a lack of gene flow.

    ReplyDelete
  124. @Ric Hern

    "Wouldn't it be interesting if Maykop was Indo-Europeanized by Steppe people...?"

    Wang et al had to use 4% EHG in Maykop to get their Admixture Graph model fit.

    ReplyDelete
  125. @ epoch

    I wonder who Maykops most major tradepartner was ? If there was some kind of hostilities which caused Maykop people to migrate from the Southern Caucasus to the North I can not see how that Hostile people would have made good trading partners afterwards. So maybe Maykop was indirectly forces to adopt Indo-European to establish trade around their new homeland....?

    ReplyDelete
    Replies
    1. Or at least part of the Maykop population. Proto-Kartvelians could have been at the very extreme edges of Proto-Indo-European influence into the Maykop territory...

      Delete
  126. When I get these samples I'll run some basic IBS stats using lots of markers. That usually picks up recent drift really well, much better than formal stats, like f3 stats.

    Then we should get a very good idea who existing today is closest genetically to these ancient Caucasus and steppe populations.

    ReplyDelete
  127. @ Folker

    ''I am not a linguist, ..''

    But one doesnt even have to be a linguist to realise that PA can't date to 2300 when already in 2000 BC differentiated branches like Hittite existed.

    @ Bob Floy

    ''Hittite areas if their society was established by a relatively small elite group,''

    Yes that is the narrative on this blog (because it explains away the lack of steppe ancestry in current and future Anatolia samples). But do you have any links for this view ?

    ReplyDelete
  128. @JuanRivera It was obvious irony from myself. Because it's not possible that any prehistoric Anatolian samples doesn't show Steppe. Genes are like a virus, Yamnaya, either it was IE or even Not, had a complexe civilization that would attract Anatolia, or being attracted to it. Now people keep talking about that Assyrian Colony with like only few samples like a fact against Steppe, it's ridiculous.

    ReplyDelete
  129. Neurotic is one of my best qualities, anyways @Floy how did you go from considering Caucasus as PIE homeland after Damgaard to now saying it's a certainty it came from the steppe?

    ReplyDelete
  130. @Davidski

    Yeah... All Maikop lineages found " but " L2, are modern Caucasian people who speak Caucasian Languages lineages. L2 is interesting tho, according to Wikipedia it is only found in Western Europe. I already discuss of this on Anthrogenica, but apparently the datas for it are a little bit confusing, L2 peak is said to be in Estonia!

    As far as cultural transmission of languages, you are right, but. Following the idea then, Maikop PIE that became LPIE " Centum " in Steppe and became LPIE " Satem " where it originally occured ( South Caucasus ). The Balto-Slavic and Scythians wich where clearly from Eastern Europe first were Centum speakers and got Satemized by ??? Then Thracians were themselves Satemized by Scythians? This is pretty much what the Indo-Hittite hypothesis gives you. Wich seems more complicate than anything else.

    ReplyDelete
  131. @Dragos

    "But do you have any links for this view ?"

    "The so called Hittite culture is a composite of the cultures of many language groups: Nesili from the city-state of Nesa, the Kussara dialect, Luwian, Hattic, Palaic and perhaps also of other cultures like Zalpuwa and the Upper Land. The culture of Nesa (Kanesh) was affected for centuries by merchants with an Assyrian, Anatolian, Amurru and Hurrian origin. The old picture of a group of Hittites or Nesili speakers settling in central Anatolia with a culture deviant from those around them is no longer credible. There is no evidence that a single people or language group developed a new kingdom of Hattusa over several centuries. ‘Hittite culture’ is probably to be distinguished just during the creation and later consolidation of the Old Kingdom of Hattusa."

    https://www.academia.edu/3483533/The_royal_family_of_the_early_Old_Kingdom_of_Hattusa_and_their_Tawananna_s._An_avuncular_system_

    ReplyDelete
  132. @Halfalp

    Y-haplogroup L isn't very common in Europe and unlikely to be native to Europe.

    Wikipedia entries about genetics are usually based on outdated and inaccurate sources, and written by people who don't know what they're writing about.

    ReplyDelete
  133. There's a lot of stuff on the Hittites at Google Scholar; about their culture, burials, religion and tablets. The obvious consensus is that the Hittites were newcomers to Anatolia, and heavily influenced by the earlier peoples there.

    Here's a potentially good one but I can't access it...

    The Hittite state is the oldest identifiable Indo-European state. The Hittites created the first large unified polity in Anatolia in the 17th century BCE. Their language, preserved on thousands of clay tablets written in cuneiform and discovered mostly at the site of their capital Hattusa, is the oldest attested Indo-European language. Anitta as the first 'literate' Indo-European gives us a unique look at how this group, newly established in Anatolia, interacted with 'native' Anatolian culture, which people call Haitian. The Anitta Text was clearly a text considered important by later Hittite kings or perhaps their scribes, since it exists in at least three different copies, ranging in date from the Old Hittite to the New Hittite period. Anitta therefore seems to have developed a pantheon headed by deities which exhibit both gender and cultural complementarity. The first identifiable Indo-Europeans, Pithana and his successor Anitta demonstrate the potential for cultural synthesis and an ability to adapt local cultural traditions and strategies.

    https://www.taylorfrancis.com/books/e/9781134945443/chapters/10.4324/9781315539256-12

    ReplyDelete
  134. @ Epoch
    Thanks for the link
    So, nothing about an elite conquest ?
    It implies a long-lasting presence of IE in central Anatolia, and as we've seen a dominant onee at that.
    In any case, that is more about political history than population history.

    ReplyDelete
  135. This looks like very solid background reading about the Hittites in general. It basically says what I said in the post above.

    The Hittites and their Geography: Problems of Hittite Historical Geography

    https://www.cambridge.org/core/journals/european-journal-of-archaeology/article/hittites-and-their-geography-problems-of-hittite-historical-geography/83F3F11C29A814C16E81C06C3FB514F7

    ReplyDelete
  136. In one paragraph, this is the consensus at the moment.

    It is assumed that the Hittites migrated to Anatolia in the late third to early second millennium BC. At that time, there was an organization of markets (kārum) and posts in Anatolia founded by the Assyrian merchants, who were trading with the local Anatolian chiefdoms. From the personal names and technical terms on the cuneiform tablets that these merchants used, we learnt that these Indo-Europeans were living in Anatolia long enough to find small chiefdoms. In pre-Hittite times, Anatolia was settled by the Hattians, who lent their name to the region.

    Same link as above...

    https://www.cambridge.org/core/journals/european-journal-of-archaeology/article/hittites-and-their-geography-problems-of-hittite-historical-geography/83F3F11C29A814C16E81C06C3FB514F7

    ReplyDelete
  137. @ Davidski
    Thanks. I had a quick skim within what Google Books happened to preview
    I think perhaps most here might agree that IE's were military prone, and they absrobed a lot of culture from their neighbours. And the langauge branching is more concistent an entry from the from the west, perhaps gradually expanding eastward, taking opportunance of turmoils etc.
    But i very much doubt they'd just arrived overnight from the steppe and a little bird tells me aDNA will prove me right

    ReplyDelete
  138. @Dragos

    I think you'll find that a lot of people, including myself, are aware of how the Hittites operated, in that they expanded gradually and adapted to local cultures and traditions, which suggests that they weren't isolated biologically from their non-Indo-European neighbors and mixed with them.

    That's why, as I keep saying, only a dense sampling of Chalcolithic and Bronze Age Anatolia will settle this issue, at least for most people. I won't be swayed by 5, 10, or even 20 samples.


    ReplyDelete
  139. We also have to be cognizant of not conflating the take -over of Nesa, e.g., by some militarized groups or elite clans with ''the Indo-Europeanizing of Anatolia''. The two are separate issues although one is of the other

    ReplyDelete
  140. @Ric Hern

    If you look at the oldest language contacts of PIE you see evidence both contacts with Uralic and Semitic languages. However, the Semitic cognates are some numbers, mead, donkeys, a word for paying and a word for a ritual. Whereas the Uralic cognates are "knee", "water", "(large) fish", "name". We can be pretty sure PIE cannot have been adjacent to both the proto-Uralic homeland and the proto-Semitic homeland so one of these sets must have arrived as Wanderwörter. I'd say that is Semitic and it seems trade related.

    http://loanwords.prehistoricmap.com/wp-content/uploads/2017/06/Bjørn-2017-Foreign-elements-in-the-Proto-Indo-European-vocabulary.pdf

    ReplyDelete
  141. IMO, too much work has gone into deciphering the meaning of the Anitta texts and there are way too many interpretations. The text basically says that a king of Kussara conquered Nesa and treated the people well. The text long has been used to push the Hittites to 2000BCE even though the Kussara group do not seem to be newcomers. There are many Hittitologists that disagree with that.

    Here is one interpretation:

    "If taken literally, it might indicate actual ethnic links between the Kussaran dynasts and the predominantly Indo-European population of Nesa, or more generally close ethnic affinities between he populations of Nesa and Kussara."

    https://www.academia.edu/5018455/Bryce_The_Kingdom_Of_The_Hittites

    ReplyDelete
  142. @ Davidski
    Slightly O.T., you've recently been critical of Harvard. Why is that, if I may ask ? Do they not support the steppe theory that you prefer ?

    ReplyDelete
  143. Re: Anatolian and Hittite languages, the unfortunate thing is that the other branches aren't much better attested. It's ostensibly this quite large family in terms of branches that are named, with a wide distribution across most of Anatolia (Hittite, Palaic, Luwic, Lydian and so forth). (I often forget this in the past). But most of the material is in the Hittite corpus as the group that were the head of this huge empire that used writing for accounting etc.

    If it were easy to tell if these languages were deeply divergent from each other in their lexicon / phonology / morphology, on a level with Italo-Celtic and Celtic-Germanic and so on, then we could dispense with Hittite itself being a late "intrusion" from the steppe or Balkans.

    While if the depth of divergence is very shallow, to a lesser degree that would instead support a lack of depth of Anatolian languages in Anatolia.
    In either case we could demphasise talking about Hittites and their story in isolation, or considering whether steppe ancestry should be shown in an elite or should be shown in the general Anatolian population.

    The Max Planck and their top philologist Scarborough seem to be looking at this corpus to some degree as part of their efforts: http://languagehat.com/etymological-dictionaries-for-anatolian-languages/ - blogpost on the excellent languagehat blog (not exactly my cuppa given its focus but a cool blog) about that. (Some other blog posts on Tocharian and so forth http://languagehat.com/tocharian/). Whether there's still enough material to be able to say anything much, I don't know.

    (The attempts to put the Anatolian languages on their lexical trees, based on Heggarty's presentation, gives these a divergence beginning about of 3000 years prior to the common attestation of Luvian-Lycian and Anatolian at around 3500ybp, in context of their date for 9000 kya for the whole family, but then there are big questions about those methods anyway!)

    @Davidski, thanks for the stats. I didn't get quite the immediate signs I was hoping for out of them resolved population relationships (not too surprising); the clarity of the signals from Beaker when under log function gave me too much optimism and some of the signs seem odd. Clear signals on the WHG vs Iron_Gates_HG and Iberia_Chl vs GAC stats for present day Iberians but not as much signal with these as I was hoping for with other Europeans (GAC vs Scotland_N doesn't give too much, and Sweden_MN is far too noisy), and some marginal Sarazm signal for Turan and Indian pops. There does still look like a marginal higher degree of sharing for the Mycenaean stats to present day NW and SW Europe relative to expected patterns based on Minoan+Yamnaya+Armenia_EBA and so forth etc though funnily enough.

    Anyway, I might keep looking and see if I find anything. Thanks again.

    ReplyDelete
  144. @Dragos
    You don’t seem very familiar with Hittitology. It is postulated that limited groups of IE migrated in Anatolia in different waves. So, differentiation between the different languages of the Anatolian branch began before the entry in Anatolia.
    You should read the papers I’ve linked previously on another thread, wrote by Freu and others. You’ll understand better why they think that IE were still intrusive in Anatolia during the colonies period. And how society was structured. The mariage contracts made by Assyrian merchants with local wives are very interesting by the way.

    ReplyDelete
  145. Reading between Dragos' none too subtle lines...

    No problem, I can't wait for some new Hittite samples. I'll take as many, or as few, as they're offering.

    But I couldn't care less what anyone says about them until I've got them here and can check them out myself.

    ReplyDelete
  146. @ Davidski
    But you are already aware of the MPI paper on Arslantepe right ?
    I thought you featured the abstracts & others here.
    My presumption is the linguists work with the aDNA people, hence their new theory

    ReplyDelete
  147. @Dragos

    I don't think the Arslantepe paper dataset includes any Hittites, because it covers the Late Chalcolithic to Early Bronze Age period, and the Hittites expanded into the region only during the Late Bronze Age.

    http://www.oxfordhandbooks.com/view/10.1093/oxfordhb/9780195376142.001.0001/oxfordhb-9780195376142-e-45

    And if it does include Hittite era samples, then I don't think they'll make much of an impact if they lack steppe ancestry, because this region was a late acquisition by the Hittite Empire.

    ReplyDelete
  148. @Davidski

    But there are Anatolian names from 2500 BC from a very close location.

    ReplyDelete
  149. @Master

    But there are Anatolian names from 2500 BC from a very close location.

    OK, so each sample in the Arslantepe paper has a name, and the authors were able to know that some of them were the Indo-Europeans mentioned on that 2500 BC tablet?

    Because I'm pretty sure that Arslantepe was populated by locals and Kura-Araxes people until after 1700 BC, when the Hittite army finally showed up.

    ReplyDelete
  150. @Master
    No. Kroonen has made the hypothesis that some names found in tablets in Ebla in relation with people of Armi (a city of unknown location) are from a population speaking an IE language closely related to the Anatolian branch.
    But it is not the main subject of his paper, and we are still waiting for a proper one with due demonstration.
    Mind that the corpus of those names is extremely limited (less than 20), and a large part of them could variations of the same names, meaning a more reduced corpus (less than 10).
    He argued for a presence of IE in Levant more early than thought before, and therefore disconnected to Yamnaya.
    So that the Anatolian IE are not the results of Yamanaya migrations.
    There are some problems with his paper, some difficult to address without a proper demonstration, others less difficult to test.
    His datation of Yamanaya could be argued against. The first Yamanaya sites are older than 3000 BC, not younger. And we find Yamnaya in Bulgaria already around 3000BC. So, obviously, Yamnaya are much older than he thinks.
    And even if he is right, it doesn’t tell us anything about Anatolian IE location at that time: it is not because some groups of IE have wandered in Levant in the middle of the IIId millenium that Anatolian IE did the same.
    My personal point of view is that several groups of people were present in Anatolia at that time, following the BA trade network between Balkans and Mesopotamia. Some were likely IE. But presence of some merchants or whatever is not the same as settlement of a structured group. We can see this in Thrace or Anatolia with Assyrian merchants.

    ReplyDelete
  151. Too much talk about Hittites. To make any correspondence between the linguistics and the genetics we need more Anatolian samples from ~4500 to 3000 BC.

    If the Balkan route theory is correct, which makes the most sense to me right now, there should be some people around NW Anatolia at that time speaking proto-Anatolian and at least somewhat Steppe-admixed.

    ReplyDelete
  152. The first Mycenean genomes carry hardly any Steppe ancestry. It wouldn't be a shocker Anatolian speakers had that little or even less.

    ReplyDelete
  153. @Matt

    (The attempts to put the Anatolian languages on their lexical trees, based on Heggarty's presentation, gives these a divergence beginning about of 3000 years prior to the common attestation of Luvian-Lycian and Anatolian at around 3500ybp, in context of their date for 9000 kya for the whole family, but then there are big questions about those methods anyway!)

    The crazy early dates for PIE that some linguistic methods are getting is really one of the only things saving PIE being a "CHG language." I guess it's not completely bonkers that it could be related to some specific CHG group that mixed with EHG to form the Steppe component AND the CHG group that moved into Anatolia towards the late Neolithic. Even today the Caucuses has a bunch of languages completely unrelated to each other living in close proximity, so it's even more possible that could have been the case at those early dates.

    Of course I find this quite unlikely but gotta keep an open mind...

    ReplyDelete
  154. @Samuel

    I wouldn't say ~20% is "hardly." But it's not an apples to apples comparison. A language change from an ethnic minority would come via elite dominance or a similar phenomenon, and that gets a bit harder to imagine at earlier dates and earlier levels of civilization.

    So far we've got the single Anatolia_Chl sample. Admittedly I've been kind of obsessed about it since it was released, and, contra Lazaridis, I did notice a few percent EHG-related ancestry then. If it's there though it's very minor, ~1-5%. Frankly I don't know if it's related to Steppe, Balkans, local EHGs, or if it's nothing at all... but fwiw here again is an nMonte run I posted a few weeks back trying to support a Balkan route.

    [1] "distance%=3.8462"

    Anatolia_ChL

    Tepecik_Ciftlik_N,69.4
    CHG,21.6
    Varna_o,9

    ReplyDelete
  155. @PF, JuanRiviera

    Anatolia_ChL is roughly contemporary (maybe a bit younger) that Armenia_ChL and those samples have very significant steppe/EHG ancestry, so no surprise here. However this connection means that Anatolia_ChL's steppe/EHG came via the Caucasus, not from the Balkan. Just because it can be technically modelled with Varna, that is not necessarily the correct model.

    ReplyDelete
  156. @PF, yep, I think that's about the size of it.

    Re: languages and steppe component, we'll probably find the Etruscans have a good whack of steppe ancestry, just as Basques do today and which Iberian paleolanguage contexts seem to in the ancient dna record if a quote by Inigo Olalde (I think) is to be believed. Likewise, who knows what the Swat Valley folk (who are different in level of steppe ancestry from present day North Indians) actually spoke?

    So correlation between steppe ancestry and IE speaking will never be 100%. It's quite clear that people with steppe ancestry, assuming they spoke IE, must at times have reproduced with people who did not speak IE with the descendants speaking the non-IE language, and that meanwhile serial dilution of ancestries must have led to ancestry becoming quite detached from original components fairly frequently.

    To make a kind of meta-point, the thing that was really interesting about Haak's paper in 2015 was that it refuted one of principal objections that the pro-Early Neolithic Anatolian crowd had to the steppe hypothesis, in their own terms.

    They had insisted that only mass genetic population movement could result in mass language shift, and so as they viewed populations to be founded by the first farmers and no subsequent mass population movement could happen, believed this refuted the steppe hypothesis.

    Haak showed that mass population movement from the steppe to Europe had happened, so refuted their objection.

    But there is the wider question of whether the pro-Early Neolithic Anatolian crowd's terms (language shift requires huge movements) was even the right terms to think of this in, and that their premise should have been accepted. Many historical examples that come to mind easily such as the movement of the English language without a vast amount of change in ancestry already refute it, and it's not certain whether it was at all sound to accept this premise at any time.

    ReplyDelete
  157. @SamuelAndrews That's exactly right, hardly any steppe ancestry.Is it good now for the Kurgan theory to find as little steppe as possible, because if it is it's going very well.

    ReplyDelete
  158. @Matt Things have moved on since 2015, the alternative to PC Steppe is not Anatolia but IranNWIran, unless I misread your comments.You make some really good points and seem to be unbiased.

    ReplyDelete
  159. @ PF
    Models are depended on how you set it up.
    Here, it'll depend which steppe source you might add, and which Aegean substrate you use (which vary in their CHG)
    e.g. if you use Tepecik as a trans-Aegean substrate, but also include Armenia (which you seem to not have done:

    Anatolia_ChL
    Armenia_EBA 35.35%
    Tepecik_Ciftlik_N 31.55%
    Armenia_ChL 18.6%
    Tiszapolgar_ECA 11.85%
    d 2.55

    If we use rather something like normal Greek Neol
    Anatolia_ChL
    Armenia_EBA 33.65%
    Greece_N 27.1%
    Armenia_ChL 26.25%
    Tepecik_Ciftlik_N 10.9%
    d 2.38

    In the latter, all the additional shifts are purely eastern/ south Caucasian

    ReplyDelete
  160. Or constructing a pan-Aegean Neolithic (avrg. Tepecik & Greek Neol.)

    Anatolia_ChL
    Aegean_N 40.5%
    Armenia_EBA 29.05%
    Armenia_ChL 27%
    d 2.4

    I think Chad has suggested whatever source of shift in Anatolia (at least Chalcolithic) exists is at archaic and contained within Caucasus stuff, at least for this individuals. Maybe others from west Anatolia will have northern steppe admixture.

    ReplyDelete
  161. @Slumberry @Dragos

    I'm well aware using Armenia_Chl produces good fits for Anatolia_Chl -- I said so myself, for example in the Hittite thread in September: http://eurogenes.blogspot.com/2018/08/focus-on-hittite-anatolia.html?showComment=1537147216860#c4517658841899934354

    Here is a run I posted then:

    [1] "distance%=3.132"

    Anatolia_ChL

    Armenia_ChL,55
    Tepecik_Ciftlik_N,41.2
    Seh_Gabi_LN,3.8

    Considering Balkans vs Caucuses route to Anatolia, from an ultimate Steppe derivation, I agree given the extremely limited samples we have now Caucuses looks better genetically. I was considering the Balkans because that's what seems to be better supported by archeology and linguistics.

    Like I said, I don't feel strongly about anything and I'm keeping an open mind...

    ReplyDelete
  162. @ PF
    Yes i agree, it's difficult to have a strong position currently, indeed we should expect both directions coming in, although it does appear that the south Caucasian one was predominant & pervasive.

    With regard to Myceneans, you quote a steppe admixture of 20%, however I think you mean 20% (rounded up) Yamnaya _Bulgaria.

    Mycenaean
    Aegean_N 64.7%
    Yamnaya_Bulgaria:Bul4 16.5%
    Armenia_EBA 16.1%
    d 1.9

    The interesting aspect here is that the I2a2 and R1a-Z93 found in Bulgaria EBA/ Yamnmaya are as rare as Hen's teeth in Greece. ANd I don't think it's due to any overwhelming post-BA turnovers.

    ReplyDelete
  163. @ Matt

    ''So correlation between steppe ancestry and IE speaking will never be 100%. It's quite clear that people with steppe ancestry, assuming they spoke IE''

    I think that's an important question. Did everything & anything coming out of the steppe between 4500 and 2000 speak IE -related languages ?

    Ths major difficulty in ascribing the obvious non-IE substratum in western Europe is that it needs to have come from somehwere.
    For a language to be preserved so long (until Roman times & beyond), it needs institutions, social-cohesion, a significant number of people, etc. This not going to come from the pre-BB populations, which suffered mass replacement, culturally & demographically. Quite literally, the Megalith civilization of Iberia & Britain was washed away.
    On the other hand, there are social processes which can explain a later introduction of IE into western Europe.
    I know many citizen-genealogists have unpleasant flash-backs to out-of-Iberia days whenever people mention this, but we're just explaining the evidence in a sober and comprehensive manner.

    ReplyDelete
    Replies
    1. The Daily Mail posted last year how the Bell Beaker invasion wiped out every single male in Spain; which explains why Basques have 92% R1b.

      Delete
  164. @Matt “Likewise, who knows what the Swat Valley folk (who are different in level of steppe ancestry from present day North Indians) actually spoke?”

    Not necessarily. It depends on the individual SPGT sample used. For example:

    Barikot_IA:I6548 (1000-800 BCE Swat Valley)

    Dzharkutan1_BA: 45.83%
    Shahr_I_Sokhta BA3: 27.5%
    Sintashta_MLBA: 23.33%
    Naxi: 3.33%

    fit: 3.3387

    Modern north Indians do not have higher steppe ancestry than that. This could easily be the sample that spread Indo-Aryan languages and steppe ancestry to south Asia.

    Compared to the Kalash (speakers of the closest language to early Sanskrit and practitioners of Rig Vedic derived religion)

    Kalash

    Dzharkutan1_BA: 50.83%
    Shahr_I_Sokhta BA3: 23.33%
    Sintashta_MLBA: 22.5%
    Naxi: 3.33%

    fit: 2.0368

    ReplyDelete
  165. @All

    It's now obvious that people from the steppe were dispersing into the south Caucasus and what is now Northwest Iran from at least the Middle Bronze Age.

    Hajji_Firuz_BA I4243 is dated to 2465-2286 calBCE and shows ~50% steppe ancestry, and it's extremely unlikely that this was the very first person with loads of steppe ancestry in the region.

    Also, importantly, this result makes sense in the context of archeological data, which suggest that people from the steppe were both attacking and merging with Kura-Araxes populations in that part of the Near East.

    So if anyone wants to posit that the Hittites and other Anatolian speakers came from the Caucasus/northwest Iran region, and were associated with Kura-Araxes pastoralists, which is what it seems like the MPI-SHH and Broad MIT/Harvard crowd are thinking, then it'd be pretty easy to accommodate an early PIE homeland as part of this theory, even if there's no unambiguous steppe ancestry in Hittite era samples.

    But I'm still a fan of the Balkan route for the entry of Anatolian languages into Anatolia, and I think that ancient DNA from early Hittite remains will back this up. The Kura-Araxes expansion just doesn't look right as the vector for the spread of Anatolian languages.

    ReplyDelete
  166. David said... Hajji_Firuz_BA I4243 is dated to 2465-2286 calBCE and shows ~50% steppe ancestry

    Yeah, I don't know how people can ignore that sample that was to the SOUTH-EAST (not north) of the Hittite lands 800 years before they rose to power and still claim a Southern Caucasus origin for PIE.

    ReplyDelete
  167. Oops, I was typing that last comment too fast. This is what I meant to say...

    So if anyone wants to posit that the Hittites and other Anatolian speakers came from the Caucasus/northwest Iran region, and were associated with Kura-Araxes pastoralists, which is what it seems like the MPI-SHH and Broad MIT/Harvard crowd are thinking, then it'd be pretty easy to accommodate an early PIE steppe homeland as part of this theory, even if there's no unambiguous steppe ancestry in Hittite era samples.

    Also, the point I was really trying to make was that if anyone still wants to counter the steppe PIE homeland theory with an Armenia/Iran one, then that avenue is now looking much more difficult to pursue, thanks to ancient DNA, because...

    - steppe ancestry was already in Armenia and Iran at least as early as the Middle Bronze Age

    - Yamnaya expanded from the north to the south (ie. into the North Caucasus region), not from the Caucasus, let alone Iran, into the steppe.

    ReplyDelete
  168. @PF

    We do not know if Anatolia ChL had anything to do with Hittites, who came much later (1500+ years). So the linguistic consideration is very speculative here. I personally think Hittites came from a later pulse, maybe the one that created Hajji_Firuz_BA I4243. (Davidski was quicker mentioning it.) Or not.
    What was the language of the Balkan 6500 years ago, anyway?

    ReplyDelete
  169. The fact most people fail to realize is that Hittite (Nesitic) is a late comer among the Anatolian languages. It’s sister branch, Luwian, was spoken at least from north Syria to Troy. They are mostly ignored because they never had an empire but that population is the most important piece to solving the Anatolian puzzle. The steppe in Hittites would have been completely flushed out by the time they engaged in Near East affairs. With all the intermarriages that occurred with Semitic groups, Hattians, Hurrians, Egyptians and whoever else along the way. Even in the stone depictions of their royals and gods they exhibited the so called “Armenoid” morphology (to borrow an outdated term). But meaning they were basically indistinguishable from their genetic brothers east of the Euphrates except preferring to shave their beards and wear their hair long. Could be also why Greeks liked to refer to the old Hittite areas as the realm of Amazons, female warriors.

    ReplyDelete
  170. @Davidski Would you accept a theory in which EHG groups in Samara expand westward towards Steppe Eneolithic CHG, kill the men, take the wives, and in the process their Indo-Uralic language overhauls its vocabulary towards its CHG’s lexicon?

    ReplyDelete
  171. @Andrzejewski

    "The Daily Mail"

    You're kidding, right?
    Obviously BB did not wipe out every single male in Spain, non IE languages were still being spoken in half of Iberia when the Romans got there and today they have more EEF ancestry than any European country except for Italy. Whatever "journalist" at that rag wrote that article probably meant to talk about the British Isles, and just fucked up, lol.

    ReplyDelete
  172. @Andrzejewski

    Would you accept a theory in which EHG groups in Samara expand westward towards Steppe Eneolithic CHG, kill the men, take the wives, and in the process their Indo-Uralic language overhauls its vocabulary towards its CHG’s lexicon?

    No I wouldn't.

    My theory is that the Don-Caspian steppe population that gave rise to the Eneolithic steppe groups of the North Caucasus, and eventually also Yamnaya and Corded Ware, came into being via Mesolithic EHG-rich groups from around the Don river swapping women with CHG-rich groups from the North Caucasus.

    This is called female exogamy, and it's especially common among small forager communities, which do it to avoid inbreeding.

    This eventually created a very stable blend on the steppe between the Black and Caspian Seas with almost totally EHG Y-chromosomes and mostly CHG mtDNA, but a higher ratio of genome-wide CHG to EHG ancestry.

    Meantime throughout the Caucasus, CHG populations were being absorbed and replaced by populations from the south with high ratios of Anatolian ancestry, and this is why by the time the Eneolithic rolled around, there was no unadmixed, or close to, CHG group anywhere.

    ReplyDelete
    Replies
    1. @Davidski “This eventually created a very stable blend on the steppe between the Black and Caspian Seas with almost totally EHG Y-chromosomes and mostly CHG mtDNA, but a higher ratio of genome-wide CHG to EHG ancestry.“

      I thought that ratio was 57:43 EHG:CHG

      Delete
  173. @ Bob
    No they did not ''fuck up''
    Have you been following the data ? Are you aware of the material culture changes in Iberia during this period ?
    How does Iberia go from a Chalcolithic dominated by G2, H2 and I2, to those lineages being rare today, and it just so happens that BB was almost exclusively R1b-L51, just like modern Iberians are.
    Now this is a bit advanced, but have you considered where a good chunk of the EEF in modern Iberians actually derives from ? (is it local Iberia, or brought with them from central Europe, or perhaps they got it from Sardinia :))

    ReplyDelete
  174. @Dragos

    "Are you aware of the material culture changes in Iberia during this period?"

    Yes.

    "Now this is a bit advanced..."

    Not really.

    ..."but have you considered where a good chunk of the EEF in modern Iberians actually derives from ? (is it local Iberia, or brought with them from central Europe, or perhaps they got it from Sardinia :))"

    Naturally a good chunk of the EEF in modern Iberians isn't original to Iberia, and of course R1b established a giant presence there, but every native male in Iberia was not "wiped out" and the Daily Mail is a piece of shit.

    ReplyDelete
  175. @Leron
    I think you are wrong. Yes, Luwian was spoken in a wide area in the late IId millenium. But we can’t say anything about this 1000 years earlier. Luwian around 1700 BC is only attested in Kanesh, as a minority language. We can guess it was spoken in the Konya Steppe, but nowhere else with enough clues. There are reasons to reckon that Luwian became dominant later in Arzawa, and probably also in Kizzuwatna. In both cases, Luwian is attested only after the end of the Old Kingdom, after those places were conquered by Hittites. The Luwian dialect spoke in Kizzuwatna has been influenced by Hurrian, but it is not the case in the West. Troy (Wilusa) was probably not IE before Hittite conquest as well.

    ReplyDelete
  176. @ Bobby

    Well I can't comment on the Daily Mail, but I'll take your word for it.
    Maybe it was only 95% & over few hundred years. But the pattern across Atlantic Europe is the same. Just variations of EEF gradients, not too significant culturally.
    Spain got more EEF-infused admixture post-Bronze Age, and a trickle of new Y-DNA, which affected Basques the least, acc. to the abstract.

    ReplyDelete
  177. Much better.
    You're a good egg, Dragos.

    ReplyDelete
  178. @Andrzejewski

    Sounds like you're talking about Yamnaya Samara.

    ReplyDelete
  179. So maybe EHG had many sons and CHG many daughters...Many sons good for hunting and many daughters good for gathering. Wonder if CHG focused more on veggies in their diet compared to EHG ?

    ReplyDelete
  180. Or the EHG sons kidnapped the CHG Daughters while they were out gathering...

    ReplyDelete
  181. C3 clades also probably from SE Asia into ANE bearing NE Asian pops today (though they don't have *that* much ANE ancestry, so that's perhaps more straight replacement than sex bias?). Based mainly on current distributions tho.

    ReplyDelete

Read the rules before posting.

Comments by people with the nick "Unknown" are no longer allowed.

See also...


New rules for comments

Banned commentators list