Saturday, March 18, 2017

Greek confirmation bias


A new paper at the EJHG claims that Slavic admixture in Peloponnesean Greeks averages a few per cent at best (see abstract below). However, I'd say the authors are making two potentially erroneous assumptions: 1) that Slavic invaders arrived in Greece straight from the Slavic homeland, probably located somewhere in East Central or Eastern Europe, and 2) modern-day Northern Slavs (Belarusians, Poles, Russians and Ukrainians) are accurate proxies for these ancient invaders.

Keep in mind that when the Slavs moved into the Balkans during the Early Middle Ages, they routinely absorbed the natives into their bands as free men and women (excellent paper on the topic here). So their numbers swelled thanks to this more southerly, local input, and, at the same time, their genetic structure shifted in a big way, probably from more or less Northern Slavic to modern-day Southern Slavic. Indeed, it's likely that by the time they arrived in the Peloponnese, they were less like this and more like this, or even this.

So was Fallmerayer correct when he theorized that the Peloponnese was totally re-populated by Slavs during the Medieval period? Probably not, but the population shift may still have been profound, and totaling much more than a few per cent.

I can't wait for more ancient DNA from Greece and Italy, especially from the Bronze and Iron Ages. Based on my experiences with many Greeks and Italians, it's sure to be a big eye opener for them, and a beautiful thing.

Abstract: Peloponnese has been one of the cradles of the Classical European civilization and an important contributor to the ancient European history. It has also been the subject of a controversy about the ancestry of its population. In a theory hotly debated by scholars for over 170 years, the German historian Jacob Philipp Fallmerayer proposed that the medieval Peloponneseans were totally extinguished by Slavic and Avar invaders and replaced by Slavic settlers during the 6th century CE. Here we use 2.5 million single-nucleotide polymorphisms to investigate the genetic structure of Peloponnesean populations in a sample of 241 individuals originating from all districts of the peninsula and to examine predictions of the theory of replacement of the medieval Peloponneseans by Slavs. We find considerable heterogeneity of Peloponnesean populations exemplified by genetically distinct subpopulations and by gene flow gradients within Peloponnese. By principal component analysis (PCA) and ADMIXTURE analysis the Peloponneseans are clearly distinguishable from the populations of the Slavic homeland and are very similar to Sicilians and Italians. Using a novel method of quantitative analysis of ADMIXTURE output we find that the Slavic ancestry of Peloponnesean subpopulations ranges from 0.2 to 14.4%. Subpopulations considered by Fallmerayer to be Slavic tribes or to have Near Eastern origin, have no significant ancestry of either. This study rejects the theory of extinction of medieval Peloponneseans and illustrates how genetics can clarify important aspects of the history of a human population.

Stamatoyannopoulos et al., Genetics of the peloponnesean populations and the theory of extinction of the medieval peloponnesean Greeks, European Journal of Human Genetics advance online publication 8 March 2017; doi: 10.1038/ejhg.2017.18

See also...

Greeks in a Longobard cemetery

146 comments:

  1. Maybe proto Slavs were like Corded Ware or maybe like Hungary BA. There's a lot of possibilities.

    ReplyDelete
  2. Proto-Slavs were probably very similar to the Baltic_BA genomes from this paper.

    http://eurogenes.blogspot.com.au/2017/03/the-genetic-history-of-northern-europe.html

    But that doesn't tell us what the Slavs in Middle Ages Greece were like.

    ReplyDelete
  3. How is Greek confirmation bias though? All they can do is guess what Slavs in Greece were like.

    ReplyDelete
  4. Did they consider the possibility that they weren't like modern Poles?

    ReplyDelete
  5. @Davidski:

    Indeed, the "Slavs" who moved into Greece were most likely just Slavicized Balkaners whose "Slavic" component---the admixture associated with the first Slavic speaking invaders----was not the dominant one. Analogously, we can say that the Kurds who expanded their settlements into Eastern Anatolia in the Middle Ages were Iranian speakers but surely quite distinct from the "original" steppe-rich Iranic speakers roaming Central Asia.

    So, yes, in that sense, "Slavic" admixture in Greece---especially northern Greece----has got to be higher than 20%.

    ReplyDelete
  6. @Davidski:

    The thing is, a lot of South Slavs are CONVINCED, in fact, that they are indistinguishable from Poles, Ukranians and the like. Literally. They seem to opine that no mixing took place and all/most of the local "Illyrians" and "Thracians" were somehow wiped out. Anytime I point out that the average Serb/Croat with no recent German, Hungarian or Ruthenian ancestry resembles Dime Jankulovski or Marin Cilic, many angry comments get hurled back my direction.

    I have never taken this sentiment seriously and for good measure: pan-Slavicism is a recent phenomenon in the Balkans and Serbo-Croats did in fact refer to themselves as "Illyrians" back in the day. The complex history of this ideology in the Balkans dates back to Austro-Hungarian times.

    Anyway, my question to you is this: What do you predict is the proportion of "native" Balkan ancestry in South Slavs? Let's assume the migratory Slavs coming to Illyricum and Moesia resembled modern-day Ukranians and Belorussians.

    (Omit Pannonian Croats and Vojvodina Serbs from consideration since they are otherwise not representative of the rest in multiple ways, not just genetically.)

    ReplyDelete
  7. I do think there was a massive migration (pre-Slavic) to the Peloponnese circa 2200-2000 BCE bringing the Mycenaean-like language to the area. Best guess is that E-V13 originated north in the Balkans (see E-l618 in Sopot/Lengyel Hungary) and with the adoption of the IE languages through Steppe migrations, moved south, so that V13 may be the best Y-chromosomal index of this movement. I-M423/R1a perhaps moved into the Peloponnese later. R1a is quite rare in modern Greek Peloponnese. V13 is very common in the Peloponnese circa 40%, but has low YSTR variance and likely expanded there after the Neolithic during the Early Bronze Age.

    ReplyDelete
  8. Anyway, my question to you is this: What do you predict is the proportion of "native" Balkan ancestry in South Slavs? Let's assume the migratory Slavs coming to Illyricum and Moesia resembled modern-day Ukranians and Belorussians.

    I don't really know, but probably considerable. We might find out how much exactly later this year or next year when Medieval Slavic and Balkan genomes are published.

    ReplyDelete
  9. @ Roy

    Today, Balkan EV13 looks "Epirotic" and in no small part is from from medieval movements
    Some might even be from resettlement from Byzantine south Italy
    There'll be a lot of change between Bronze Age and modern Greece

    ReplyDelete
  10. @Rob,
    Of course there have been many recent movements (since the collapse of the LBA) to Greece. When we've studied Greek Y chromosomes, we were very careful not to sample Albanian communities or exchange communities from Asia Minor now living in Greece. Yes, Epirus and Albania, have a high frequency of V13, but they also has very low variance. The sample of L618 (immediate precursor of V13 circa 5000 BCE Lengyel is a game changer for understanding the origin, migration patterns and demography of V13.

    ReplyDelete
  11. After the movements of 'Sclavenes' in Greece there are other movements towards Balkans, though. That makes things even more complex.

    Imo the Slavs in S. Greece were SW Slavic like. On the other hand, in N. Greece there was more Bulgarian/Romanian-like admixture.

    ReplyDelete
  12. @ Roy

    Yes I know we already have E in Neolithic Balkans. However, I first of all question the utility of the concept of "STR variance".
    Secondly, we can't ignore the massive depopulation then repopulation which occurred thoughout the Greek countryside in the 7-9th century.

    ReplyDelete
  13. @Davidski:

    Indeed. It's pretty hard to guess since we do not have any Illyrian, Thracian, Dacian, Paeonian, etc. genomes on hand to compare with, to my best knowledge.

    For all we know, the steppe-ward shift of Montenegrins and Bulgarians on your PCAs (in relation to Albanians and Northern Greeks) could be the result of mostly just that: pre-Slavic migrations rich in steppe DNA.

    I am willing to make a bet, but gonna keep it conservative: >50% of South Slavic ancestry is of pre-Slavic "Balkanic" origins.

    ReplyDelete
  14. Ratna ,

    Which types of M-417 mutations exist in Italy and Greece ?.

    ReplyDelete
  15. @Roy King:

    Roy, how knowledgeable are you about the modal subclade of I2a in the Western Balkans?

    This is what I've always wondered about:
    1)Where and when it first originated
    2)When it first arrived in the Western Balkans(assuming it did not originate there) and from where
    3)Is its high frequency there attributable to the scale of the migration or to a series of bottleneck scenarios in the centuries that followed(think Basques and R1b)?

    ReplyDelete
  16. Old Yugoslavia was inhabited by Illyrians, who where absorbed into Celtic society prior to the coming of the Romans, then the Romans butchered the illyrians after the great Illyrian revolt , next the ostrogoths settled in these areas ( Old Yugoslavia ) for hundreds of years after the Roman empire collapsed .............clearly any slav migration into old Yugoslavia absorbed many different races.
    In regards to Greece, the slavs settled mostly in Thessaly and where called Servians , I cannot recall many slavs in the pelopenese but if there were then it was not as great a percentage as Thessaly

    ReplyDelete
  17. There isn't enough information available to call this bias. It is perhaps a poor hypothesis. But science is built by proving those right or wrong with additional data.

    ReplyDelete
  18. It is by definition confirmation bias because they didn't consider the possibility that the Slavs who settled in the Peloponnese were like modern-day Balkan Slavs.

    Confirmation bias, also called confirmatory bias or myside bias, is the tendency to search for, interpret, favor, and recall information in a way that confirms one's preexisting beliefs or hypotheses, while giving disproportionately less consideration to alternative possibilities.

    ReplyDelete
  19. No, my impression is that many Greeks believe that they're not much from ancient Greeks in terms of ancestry. But I reckon the genetic shifts that we'll see from the Neolithic to the Mycenaeans to modern Greeks will be pretty big.

    ReplyDelete
  20. Slavs got Orthodox Christianity from the Greeks but maybe the Greek language ultimately came from the steppe.Hopefully we'll find out this year.

    ReplyDelete
  21. @Anthro Survey
    Your analogy is probably close to the truth. I can say according to David's K7 ADMIXTURE run Early Iron Age Hasanlu sample represents 96% of modern Kurdish ancestry, while the number for other modern west Iranics was much much lower. I wish David would test it further via D-stats et al, but that's what we have now. I guess the number would be somewhere around %70 to %90 from one extreme to another regarding how much ancestry Kurds have from pre-Median populations.

    http://eurogenes.blogspot.com/2016/07/sneak-peek-basal-eurasian-k7.html

    ReplyDelete
  22. @Davidski "It is by definition confirmation bias because they didn't consider the possibility that the Slavs who settled in the Peloponnese were like modern-day Balkan Slavs."
    In fact this might have been the most likely scenario. For example we can see today E V13 (which is a neolithic haplogroup) comprises about 35% of population in Greece and Albania, but also around 20 % among Slavs in Serbia and Macedonia so it seems that a good part of the old people on the Balkan were not totally annihilated but most probably were assimilated. But this leaves the question of Falmerayer open. If there was a total annihilation of the population in Greece, is it possible that this haplogroup was reintroduced by more northern populations in early Middle Ages? Would have been interesting if the study questioned this possibility. This point becomes more critical for those who have read Falmerayer know well that he also spoke about a second massive wave of migration of Albanians in Greece in Middle Ages distancing even more the current Greek population from the Ancient one. As far as I see there were no Albanian samples in the study although they can be easily found in several databases (as far as I have seen Albanians and Greeks are very close almost identical in results being very close in every graphic) This is a second bias of the study. If they wanted to fully tested Falmerayer hypothesis they chosed the wrong method, unless their bias was part of a political agenda.

    ReplyDelete
  23. They are discussing Fallmerayer’s theory and they mentioned in the article that he was a slavophobe. Slavophobic theories claim that Slavs expanded in 5th century from Pripyat marshes and flooded Eastern Europe including Greece.
    This theory is obviously false and they showed it.
    There are very old links between Slavs and Greek, for example in poetry, similar to the links between Slavs and Vedic Aryans.
    In the Harvad publication “The Common Heritage of Greek and Indic Meter” we read about some similarities between ancient Greek poetic meter and Slavic meter:

    http://s22.postimg.org/a5wbsz7xt/screenshot_151.png

    http://chs.harvard.edu/CHS/article/display/6442

    The same archaic meter is also noticed in Rigveda:

    http://s22.postimg.org/illwahasx/screenshot_149.png

    http://s22.postimg.org/jp60mfvg1/screenshot_150.png

    http://chs.harvard.edu/CHS/article/display/6429


    We know that Sintashta was R1a and autosomaly very similar to Eastern Slavs. The question is how Slavic were Thracians and Illyrians and who migrated to Greece. Some Slavic scholars claim that Thracians and Illyrians were Slavs and Southern Slavs are their descendants. Let’s wait for aDNA.

    We are talking about links in poetry, religion and language between Slavs, Greeks and Indo-Iranians which were much later than PIE.

    http://s18.postimg.org/fmh8v8dpl/screenshot_78.png

    ReplyDelete
  24. Yes bias runs both ways
    Some argue Slavs were everywhere

    ReplyDelete
  25. Seems like Coop 2013 - http://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.1001555 - can already set our priors to quite a large extent.

    From Coop we see for Greece, IBD above 1cm (recent) shows somewhat of an impact of sharing with the Polish population - somewhat - above the Dutch and Italians, at about the same clip as in East Germany or Denmark. Italians seem to have very little IBD outside the peninsula, but regional subsamples may be useful there.

    (Alternatively, more recent than Coop 2013 is Busby 2015 - https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4714572/, based on TVD copying which is also giving - 35% Lithuanian cluster, 65% Cypriot cluster for Greece, 500 AD. Though the models are quite complex and non-transparent to me. Also this is in the context of output modelling Hungarians as 43% Lithuanian cluster, 57% Sicilian cluster, 1100 AD, the CEU cluster as 11% Sicilian cluster, 89% Norwegian cluster, 700 AD and the Ukraine48 (Poland, Belarus, Ukraine) as 17% Turkish cluster, 83% Lithuanian cluster, 900 AD. So take that for whatever it's worth.)

    (OT: With Coop, there is an interesting puzzle in Eastern European ancestry for me in that:

    "One of the striking patterns we see is the relatively high level of sharing of IBD between pairs of individuals across eastern Europe, as high or higher than that observed within other, much smaller populations. This is consistent with these individuals having a comparatively large proportion of ancestry drawn from a relatively small population that expanded over a large geographic area."

    While Anagnostou 2016 - http://www.nature.com/articles/srep41614 - has found that the highest autosomal effective population size in their panel was in Poland).

    ReplyDelete
  26. @EastPole

    Slavs had nothing to do with Sintashta. They most likely originate from the Dnejpr-Pripyat region. It seems that Proto-Slavs were much influenced by East Germanic tribes (Bastarnae, Goths..) and there are a lot of loanwords from this languages. Genetically they probably resembled modern day Polesians. They were in the beginning a very small population and Slavic languages were spoken in a very small area but the population increased dramatically somehow and they quickly expanded in all directions where they quickly absorbed locals. In most of East Europe the locals themselves were mainly descendants of Corded Ware so the genetic profile not very much changed but in the Balkan mmuch of the populations was probably Albanian-like before Slavs arrived there

    ReplyDelete
  27. @Matt

    Busby used Globetrotter which is one of the three haplotype-based mixture tests used in recent papers (Broushaki et al. method and the recent MixFit from Haller et al. being the other two). Looks like it has a tendency to search for more distant source populations if more proximate ones are admixed themselves, hence Armenian+Lithuanian modeling around the Balkans.

    Broushaki's method, as I've previously noted, does not proxy Slavs with Lithuanians. Here are the models for some Balkan populations using the Busby et al. dataset (allowing all modern and ancient donors):

    Greek
    59.5% Bulgarian
    11.1% NE1
    8.6% LBK
    8.4% Cypriot
    5.5% Armenian
    3.8% Lebanese
    3.1% TurkishN

    Bulgarian
    33.6% Romanian
    30% Greek
    20.1% Hungarian
    8.7% Croatian
    6.4% Ukrainian
    1% German
    0.2% Turkish

    Croatian
    46.7% Bulgarian
    37.2% Ukrainian
    16.1% Hungarian

    Lazaridis dataset had two different sets of Greeks and Albanians, them allowing to act as donors to each other eliminated the need for Lebanese, Turkish and Neolithic farmers, and greatly reduced Bulgarian.


    About the IBD issue, an effectively large population having some recent ancestry from a broadly genetically similar but founder effected population could perhaps result in that scenario?

    ReplyDelete
  28. @Al Bundy,
    "Slavs got Orthodox Christianity from the Greeks but maybe the Greek language ultimately came from the steppe.Hopefully we'll find out this year."

    Slavic and Steppe aren't interchangable.

    ReplyDelete
  29. It's super confusing how Ashkenazi Jews, Southern Italians, and Greeks became so similar.

    ReplyDelete
  30. I can't wait to see Bronze Age and Iron Age ancient DNA from Italy (and from Greece) either, indeed it would be a beautiful thing. Technically I don't qualify for Italian, but a big piece of my ancestry is from there, yet I don't need an eye opener, nor do I fear one, I just want to know when and with which cultures the Steppe_EMBA admixture entered, and when and with which cultures the Armenia_EBA admixture arrived and how and when it spread northwards. It would also be useful to know what the Etruscans were like and if there are genetic hints regarding their origin. And moreover it would be super cool to see a comparison between the pre-Roman inhabitants of Gallia Cispadana and the Roman ones.

    ReplyDelete
  31. Some Greeks want to think that modern Greeks are exactly the same as ancient ones. That can't be true off course.

    I agree with the first two paragraphs of this article, although I am not sure what type of big shifts he thinks that took place here.

    Imo Ancient Greeks could have been more Southern shifted, almost South Italian to Cypriot-like. The myths indicate that too (see the myths about Danaus and Cadmus which point to movements from Levant and Egypt and the myth of Pelopas which points to a movement from Anatolia) -- the languages they spoke is unknown but all of them became part of the elites. Also the name of the mythical founder of Mycenae was Perseus and Greeks believed Persians descended from his son Perses.

    @ Al Bundy
    Ancient Greece isn't the birthplace of Western culture really. It's one of the influences but through modern interpretation.

    ReplyDelete
  32. I'm interested in Slavic or Germanic migrations into the Balkans and any new data as my mtDNA matches in 700 AD are Serbia, Macedonia,Croatia & Slovakia. But.......there really is no way to determine if my lineage arrived in 600 AD or 2200 BCE in the Balkans.Could be a Yamnaya, Catacomb, Corded Ware,Proto-Slavic, Slavic or Gothic migration. Some time in that 3000 year period anyway!

    ReplyDelete
  33. Mr. Davidski, every time you call someone out on "confirmation bias", you are supposed to disprove them, not leave snarky comments and insults for the authors.

    I'm not seeing any evidence here (or in any of your posts regarding "confirmation biases"). Not even the biggest scientists on the field expect us to take them on their word, do you?

    Also, how can you be so sure that you aren't exhibiting your own confirmation bias by disagreeing with everything you don't like and applauding everything you do?

    Finally, it seems to me that you are doing a disservice to science by letting ethnic chauvinism creep in.

    ReplyDelete
  34. There is archaeological evidence for a migration of north Anatolian people to Greece about 2200 BC. From a theoretical point of view they might have taken Greek to Greece. The best argument against this theory and for a northern, European origin of Greek is perhaps the Dorian migration which is thought to have taken place in the late Bronze Age: Dorian and Northwestern Greeks migrated in and pushed off earlier Greek inhabitants, thereby giving rise to the Ionian and Aeolian tribes. Apparently the Dorians and northwestern Greeks came from a more northerly place in the Balkans where the Anatolian wave didn't reach to. In more recent times there have been sceptics doubting that there had ever been such a thing like the Dorian migration. I'm not an expert on this topic, but suspect this is part of the more general fashion of doubting all long held views in the archaeological domain. In any case, ancient DNA from Greek would be very useful to answer these questios.

    ReplyDelete
  35. @Simon For PIE origin in general Anatolian and Greek DNA is huge.Europe Central and Northern are great and all but it seems like we know where those languages came from.

    ReplyDelete
  36. @Ezekiel

    When I call someone out on confirmation bias I'm not obliged to prove them wrong, I'm just expected to explain how they're showing confirmation bias.

    I've done that several times here, in the post and comments.

    ReplyDelete
  37. i have looked at ftDNA data sets of ppl with 4 grandparents born in greece. there are two clusters. a major cluster that is shifted toward other balkan ppl. and a minor (but not trivial) cluster shifted more toward sicilians and west asians.

    so i'm pretty sure that david is correct. post-6th century slavic admixture after the collapse of the balkan limes probably on order of 10-20% or something.

    why the two clusters? i suspect that the minor cluster are from greeks in the islands and those who escaped turkey in he 1920s.

    ReplyDelete
  38. Greek Cypriots can be presumed to have no Slavic admixture and Broushaki 2016 models them in an interesting way: Busby set (which has one set of mainland Greeks) models them using only modern Anatolian and Levantine populations with additional Neolithic Farmer accounting for their western shift when all modern and ancient donors are allowed. The other dataset, from Lazaridis, has two groups of mainland Greeks, and the more southern of them (Greek_Coriell) appears as a minor source in the Cypriot fit which is otherwise similar to the Busby set's model.

    When ancient donors aren't allowed, Greek_Coriell fully replaces the Neolithic when fitting Cypriots. Could be that an Iron Age Greek sample might remove the need for Neolithic Farmer entirely. Additionally Greeks or Cypriots don't get Sicilian or South Italian in those fits, instead Greek_Coriell is a donor for Sicilians which is likely the most expected result.

    ReplyDelete
  39. @ Al

    "heard some claim that means those languages came to their respective locations more directly from the Caucasus and surrounds rather than the steppe.It's all speculation of course and the ancient DNA so far doesn't seem to be on their side."

    Can you elaborate which aDNA evidence you are regerring to ?

    @ Simon
    The greatest cultural shift arguable occurred in 4000 BC. Also, in 2200 BC there *is* evidence of a migration from the Balkans in addition to Anatolia. It comes from the post-Vucedol Cetina culture . I think the "Dorian migrations" were mostly internal realignments, periphery conquering the old palaces (kinda like the La Tene - Hallstatt shifts).

    ReplyDelete
  40. Please observe that:

    1. Slavic is a language that formed AFTER the expension of the (East-)Roman empire, parallel to the formation of an another, common language-standard that formed to become the administrive common-language of the western occupations and colonies.

    2. As the western and eastern empires (re-)established trade and civil inter-actions with their neighbours the "lingua franca" spread to the neighbours and trade-partners of the west-Roman empire, becomming the 'soldiers latin' today called 'franco-roman' or 'french' language-form. At the end of the symbiosis between the native, german language-form and the incomming latin/french we got a 'latinified' form within the Brittish Isles that ultimately became the English language.

    3. A similar symbiosis happened around the Black Sea, where a mix of old, indigenious languages of the area had to melt into new sound-forms and sound-tracks, that could serve the entire trade-zone, from Greece and Anatolia to the town along Donau, Djepr and Don. Later also Dniestr-Bug, west to Wizla/Vistula - at which end "the largest town and market in Northern Europe" used to lay, according to old, Roman sources.

    4. Before the spread of the Greek-Roman language to the northern shores of the Black Sea, we had a "Achamenedid Empire" that arose in the east, on the ruins of Old Babylon and the sumerian Persegard. Based in the new Persepolis this mix of Akkadian and (conqered) Medians
    were able to invade and occupy an enormous empire, stretching from the border of India to mainland Grece, inclusing the shores of the Arabian Sea and the Red Sea, as well as Egypt and the Middle-East, the Black Sea, the Caspian Sea and the Aral Sea - taking control of the ancient spice-route to India (Karakorum) and the first silk-route (Tarim) to China.

    5. At the end of the Persian and Semnitic influence across Anatolia, Armenia and Bactria there was a few centuries of Greek dipålomacy and a renesance of the native language - amongst surviving Massagetae, Thyssagetae, Bulgars, Scyti/Sarmati and Venedae. Thus we would have uralic and gotonic ('proto-germanic') natives amongst them. Especially since the "Gotic Corridor" was established during the first East-Roman expensions, to support the natives of the 'european' trade-routes of Eastern Balkan as well as the Black Sea from Roman occupation and exploitation. Moreover, the Ostro-Goths, Heruls and Varangi would even help the Bulgars secure the trade-routes from the Caspian Sea, from intruding slave-hunters and violent plundering from the old, Elamo-Perisan area.

    6. The term "slav" was definitly coined during this time - based on the greek-persian-semitic terms 'saka' (saqqua-saka-sceat-scyt-skyti ('shooter')) and "liba" (lib-leb-lev-lav (limb)). Thus the persian-semittic term "saqqua-liba", while the persian-greek wrote "saka-liba" and "saka-lava", describing the "limbs (children) of the sakas", also known as "sarmatians".

    In the elamo-persian world the term "saqqua-liba" became synonymous with terms like "barbarian" and "cattle" - catched and bound to obidiance by a "herder", as in "slave-driver" and "slave-owner". Thus the expression "saqqua-liba", "saka-lava" become a description of the old Thyssagetae, Massagetae and Bulgars alike - ending with comprised verson Sklav-Sklavin and Slav/Slavan/Slavon.


    ReplyDelete
  41. 7. As the Eastern Roman Empire organizes a state and church separate from Rome we find the process of (re-)aligning the trade-zones around the Black Sea as essential. Thus we find the standardization of the "sarmatian languages" to be finalized in a process between the Bulgarian nobility and their Greek counsels - to spread along all the trade-routes connected to the Black Sea. Include the "Gothic corridor" as a result of the northern inter-action between the Vends/Venedae and the Goths/Getae - and we may understand why the slavonic trade-language - as a "lingua franca" of the east - could extend via Dnieper-Dvina and Bug-Vistula to the main ports of the Baltic/Atlantic trade-zone.

    8. North of the scytian nobility and their 'limbs' (common population) the Greek and Roman geographers place the "venedae", who apparently used to occupy the boreal woodlands of Russia - from the Baltic to the Caspian.) It's been suggested that this 'culture' - due to this geographical trait - was responsible for the spread of the Fenno-Ugrian ("Uralic") languages.

    It seems like the East-Roman church were counting the Vends into the category of "slavic-speaking" - as in "slavons".

    As Russia united with the Bulgarian empire (862) the slavonic languages started to influence their new mates up north, sepaking 'vendic' (fenno-ugric). When Russia-Bulgaria later subjected to the Greek-Roman Empire (987) the official language had to be slavon.

    500 years after Cyril and Method created the first slavonic bible, the slavonic language became truly 'imperial' - which is the basic reason for the spread of the slavonic languages. Thus we HAVE påresume that the 'slavic language' - just as the franco-roman - was a creole language - normalized and standarized to create wide alliances after wide-reaching and long-time wars. Consequently we find elements from Sarmatian/Samogaetic, Greek and Indo-Iranian at the base of the slavon language.

    9. Seemingly, we have to settle for the 5th century BC as the "birth" of the Slavonic basics - when a normalized standard was formed under the auspices of the Eastern Roman Empire. The formation was a result of a trade-alliance with the Bulgarian Empire, that grew out of the Volga-Don - to occupy both Suzdal and Ouana/Vana-gard, Ashov, Konugard (Kiev) and Krim.

    10. Connecting the slavonic language to a specific etnicity - originating from an old, indigenous dynasty - is NOT possible.

    Still the slavonic speakers can be identified on the premises of historical research as a whole, due to the facts and insigths of modern science, such as carbon-dating and genetical spread-sheets.

    ReplyDelete
  42. 11. The spread of R1a-clads along Eastern Europe seem to hold substantial clous to the origin and spread of the slavonic languages. Understanding the multi-cultural and multi-etnic aspects of the process we may explain why todays slovonic populations can have lot of y-dna N's, I's, J's and R1b's - too.

    12. Looking at the indigenous languages of NW Eurasia we may find some links between the mesolithic/neolithic dynasties of y-dna G, H, I, J and K2/R1 - to point back to a common genetic as well as cultural ancestry. We may call these cultural commodities, including the stem of these languages, "proto-IE".

    Consequently we may adress the ancient dynasties of y-dna hg I as "proto-Germanic", populating northern Europe, while hg G populated southern Europe - to become the 'proto-Greek' or 'proto-Hellenic'. From which the Minoan culture grew to connect the entire Med, before split into Greek AND Roman.

    Moerover we find an early outlier (G1) between Eufrat and Tigris, indicating that the Sumerian agricutlrualist were of "Greek descent" - bordering an area dominated by y-dna J (proto-Indo-Iranian).

    Much later we had the experts of cattle-farming spreading - with R1a in the colder climates and R1b in the milder. Pr. consequence we see "Scandinavian farming" different from "NW European" farming already during LNE. By the same standards we may likewise discriminate between TBK/CWC/Srubnaya-Sintasta-Tarim and BB/GW/Yamna.

    13. The cattle-farmers that one split in R1a and R1b did obviously share a very basic, 'proto-IE' toungue.

    This I-E baseline is still reflected in their slavonic, franco-roman or german-gottonic language(s) - eventhough they have been overran by "foreign masters" and subdued to economical and cultural slavery during the last two millenias. To get the most close to this "proto-IE" language we should have another look at the northern tiers, where the indignous dialects still remains - amongst indigenous populations with genetic roots in the early, Eursaian Mesolithic.

    ReplyDelete
  43. @bat How do you have the time to fight crime in Gotham City and know so much about this stuff? @Rob No I can't because I take the word of Davidski and others who know a lot about admixture and expect sizable enough impact in Greeks to affect language change during the Bronze Age.They could be wrong of course so we need to see finally some stuff from the Myceneans and others in SE Europe as we've talked about before.I know you have some interesting theories as well let's see how it pans out.

    ReplyDelete
  44. @ Al
    Thanks for clarifying. I thought when you said the evidence points to X, Y or Z, I thought you were actually referring to evidence rather than random heresay
    My mistake

    ReplyDelete
  45. @Rob Is there anything he wrote that makes sense?

    ReplyDelete
  46. @ AL
    I prefer to obtain my information directly from the Don's twitter account.

    ReplyDelete
  47. I think Batman has some good & interesting input about Roman & Medieval Periods, but the careless back projection of medieval structures to the Holocene, let alone his theory that macrofamily C-F expanded from Holocene Antalntic coast deserves no further comment.

    ReplyDelete
  48. @ Razib:

    "so i'm pretty sure that david is correct. post-6th century slavic admixture after the collapse of the balkan limes probably on order of 10-20% or something."

    This is probably an underestimate. Northern Greece is probably underrepresented, and modelling "Slavic admixture" on commercial companies' 'east European' cluster would also significantly underestimate the impact. From someone who does know something about 6-7th century events; I wouldn't be surprised if the impact is in the order of 40-50% .

    ReplyDelete
  49. Just don't tell them they're Turks Slavs are ok though

    ReplyDelete
  50. @AlBundy,
    "It was all haha but in some ways Ancient Greece is a modern construct bathing them in this shimmering light for all to admire.Modern classical scholarship is in many ways a German invention."

    It wasn't invented but it was exaggerated.

    "but this whole Greece Rome equals the West is obviously not true.Simple narratives are usually wrong."

    In reality Western Rome did for the most part become Medieval Western Europe and Western Rome looked at ancient Greece as an important ancient civilization. But to say Romans and Greeks are the ONLY cultural contributors to Medieval Europe is definitely wrong. I new complex narrative which admits we don't know a lot more than we do know and that ancient people other than the famous ancient civilizations(Rome, Egypt, etc.) contributed to the modern world.

    That's the impression I get as well. There's also a narrative for all of human history which experts have been pushing for generations that's mostly true but definitely not completely true.

    Stupid Ape like Cave man(who lived with dinosaurs)>Glorious Civilizations>Greece>Rome>Dark Ages>Enlightenment>Modern Times

    A philosophy, I guess from 19th century European historians, about human progress is attached to that timeline which is really just an opinion not fact but is treated as fact in the mainstream.

    ReplyDelete
  51. @Al Bundy,
    "Classics as a discipline was invented"

    The study of the Neolithic was invented in the same sense. It doesn't mean the Neolithic never happened and it doesn't mean the Classical era never happened.

    ReplyDelete
  52. @Al Bundy
    The development of historiography among Greeks made their achievements survive through time. Also the role Greek language had in Eastern Roman Empire played its role.

    Other cultures may have had important achievements that are lost. For example, a culture in the forest steppe may have had impressive temples and art made from wood that are lost or more advanced political systems than we think etc.

    In terms of material wealth other cultures were definitely in a better position, for example Lydians or Persians.

    @Rob
    50% seems too much to me. I wouldn't try to refute it because I don't care (Slavs are ok in my book) but it's a statement without arguments. The hardest thing is to separate Slavic admixture from other types of admixture from the North. That's difficult even if we call 'Slavic' people all those who came to what is now modern Greece speaking Slavic languages irrespective of their origins.

    I believe it's possible to measure the post-classical NE or NW admixture as whole only with ancient samples from multiple sources (Spartan, Athenian, other Ionian, Macedonian and Epirotan) while taking into account if there are other possible influences from an opposite direction.

    I would also like to note that Classical Greeks could have been partially different from Myceneans and also different between them (the different lifestyle of Ionians compared to Dorians may have been result of admixture etc)

    Also concerning Macedonians and Epirotans, those are sometimes labeled barbarians in Greek sources. Practically they were different ethnē and weren't considered Hellenes originally. Calling them Hellenic and/or Greek makes sense though as long as we accept that the language they spoke should be classified as Hellenic and/or Greek. Non-Greeks (Thracians) were plenty in Macedonia since the times of Strabo but things were quite complex already (There were Macedonians proper, Thracians, Epirotans, Illyrians and Ionians in Macedonia then).

    Ionians and Dorians were different ethnē too. Herodotus who was partly Dorian, partly Karian considered the Ionians cowards for example.

    ReplyDelete
  53. @ Apostolos

    "50% seems too much to me. I wouldn't try to refute it because I don't care (Slavs are ok in my book) but it's a statement without arguments. "

    Sure its hypothetical but it is founded. If R & C calculated 35% Slavic-like admixture for all of Greece, ~ 50% in northern-most parts seems to follow. Moreover, I guess you might have not considered the extent of depopulation in Greece after 620 ? W.r.t to Macedonia, you have Phillipi, Thessaloniki, and few towns in Halkidiki showing continuity. The rest was Sklavinias - even if already 'Palaeo-Balkan' admixed . Of course, we should not forget other migrations later on, as Aigesti pointed out.

    "The hardest thing is to separate Slavic admixture from other types of admixture from the North. That's difficult even if we call 'Slavic' people all those who came to what is now modern Greece speaking Slavic languages irrespective of their origins"

    Not really. Goths, Huns, or Sarmatians, will look different to proto-Slavs, autosomally & from haploid perspective. That's already rather clear, but of course aDNA will fine tune everything.

    "I would also like to note that Classical Greeks could have been partially different from Myceneans and also different between them (the different lifestyle of Ionians compared to Dorians may have been result of admixture etc)"

    I think mainland Iron Age Greeks will all be fairly similar. Dorians, Ionians were just ethno-political constructs from the 7th century onwards, when pan-Hellenism began to be constructed.

    "Also concerning Macedonians and Epirotans, those are sometimes labeled barbarians in Greek sources"

    Although I had not mentioned this, I agree. The ethno-politics of 5th century Hellenism is a distinct & complex subject of its own. But it's entirely reasonable to conclude that Epirotes & Macedonians were Greek speaking by historic times.

    But my main interest for the present is what happened in 4000 BC, and 2200 BC ;)

    ReplyDelete
  54. Please read again:

    "12. Looking at the indigenous languages of NW Eurasia we may find some links between the mesolithic/neolithic dynasties of y-dna G, H, I, J and K2/R1 - to point back to a common genetic as well as cultural ancestry. We may call these cultural commodities, including the stem of these languages, "proto-IE"."

    Q1: Doesn't y-dna GHIJK have a common ancestor F, with roots in the european Paleolithic?
    Q2: Does y-dna R1 come out of one of his decendant lines, such as K2 - or not?

    Q3: Doesn't the language-groups classified as "Indo-European" show some common traits of culture, apart from a yet unidentified "proto-IE"?
    Q4: In case they do - and the age of these traits are mesolithic - what would that imply regarding the age of the I-E languages?

    ---

    Regarding the I-E culture(s) it's obvious that they have "Eurasian" roots. Upheldm, developed and spread by the Eurasian "caucasians" - from the LP to the late IA.

    Here's a 11.000 year old example of the very basic symbols-elements of Mesolithic and Neolithic Eurasia - known from the 9.000 yrs old Volga-Ural-pottery, as well as the later Sperrings - and thus the PCC, EBK, TBK, CWC, LBK, BB and CAW:

    http://siberiantimes.com/science/casestudy/features/n0379-revelations-on-shigir-idol-change-our-understanding-of-ancient-civilisations/

    Please note that the "sacred symbols" decorating the Shigir-idol were continued pretty much intact throughout the Neolithic and following metal-ages. In NW Europe the very same key-elements are repeated on most art of bronze, silver or gold - as well as numberless clay-pots. Until it became a mission of the Holy Roman Empire to conquer the Baltics too - plundering its treasuries and erasing the last remaining traditions from Antiquity, also known as "Old Europe". Here's one of the last examples of "the Shigir Idol", molded in "celtic bronze-age art" - in viking-time Scandianvia.

    https://s-media-cache-ak0.pinimg.com/736x/aa/e1/fe/aae1fe8b102b64216a13b4537830a0cc.jpg

    Regarding the origin of the I-E languages there are various theories already well etsablished, side-by-side. Picking one of them before the other is NOT an excerize in Popular Genetics. Ancient DNA can't reveal languages - UMLESS they can be coined to known high-cultures and civilizations. Such as the Mionan/Greek, Egyptian, Sumerian, Celtic-Germanic, Vendic, Indian, Chineese, Melaneese and Meso-American and the African. (To "identify" objects and haplogroups one need "historical identities" to make sense.)

    https://commons.wikimedia.org/wiki/File:Urheimat_ipotesi.png

    ReplyDelete
  55. Following the distributions of y-dna GHIJK from the very start of Holocene makes it obvious that there is, indeed, a co-relation between the known history of the I-E culture - and the genetical traits (mutations) and the various 'mega-structures' (haplogroups) of the Human Genome.

    If, again, you look into the time and space where the brother-lines from forefather F spread, we end up in Europe - just after the Younger Dryas. Moreover we find that they were all invoilved in the various, regionally distributed dynasties that started the very re-population of northern Eurasia - some 11.800 years ago.

    As the mesolithic cultures spread we find their aDNA to form various y-lines that reflect dynastical, patrilineal structures - distributed in respective parts of the continent.

    Which is why y-dna I is in gross majority in mesolithic NW Europe, while their cousins of y-dna G populates SW Eurasia. East of the Baltics we find mesolithic y-dna J, spreading south-east unto India. It may seem somewhat funny that the major dynasty in southern India, where the 'proto-Indians' had survived the Late Paleolithic, should end up with a dynasty from y-dna H, who is classed as a brotherline to G, I and J. at the helm of their dynasty - along with the first Vedic verses and idols. It seems to be a very early mix between "arcaic Indians" and "archaic Caucasians". Based on a mingle with a tropic refugia that had sheltered the "proto-indians" and their "proto-dravidian" language through the Late Paleolithic.


    Meanwhile we find a y-dna E to be continous - from the LP to ME and NE - along the tropical side of the Meds.

    The coldest parts of Eurasia - east of Ural - got repopulated last. Here we find a certain string from y-dna K, named NO, from which hg N starts populating the boreal forests between Carelia and Tocharia, while hg O makes it all the way to China. Preparing the grounds for a later spread of farmers bound to hg Q, later also R1a.

    Around the start of the Holocene Optimum, some 8.200 BP, there's a (new) branching of F->K2, branching into R1a and R1b. It just so happens that the spread of these TWO lines co-relates quite contionusly with the spread of the laborious cow-dungers ready to oxchart themselves around and fertilize ALL the open grasslands of moist and temperate Eurasia.

    As these cowboys learned to horseback-ride themselves they made headways across Eurasia, as soon as the wide, open fields of this vast continent became fertile and vegetative - by lush meadows and juicy graslands.

    How do the cattle-farmers producing Dniester-Bug and EBK - relate to the CWC, BB, Srubna, Yamna and Maikop? Where can their common origin, genetically as well as linguistically, have been?

    If I where you I would return to Davidskis blogg-page concerning the "Bi-furication" of R1a/R1b. There's NO way to start discussing the branching of R1a and R1b without mentioning the area of Pommeria and the rivers of Weichsel, Oder und Elbe.

    ReplyDelete
  56. HI Batman

    1) No; GHIJK/ F* isn't from Palaeolithic EUrope

    2) Yes R1 is linked to macrogroup K, but the latter isn't from Europe

    3) Yes, but that culture seems to have galvanized c. 4000 BC, not in the Younger-Dryas

    4) as per #3, No.

    ReplyDelete
  57. @ Rob

    A major off-line of the human genome existed in the northern parts of Eurasia - from Iberia to Siberia - during the entire period called LP.

    Thus we need a C/F->GHIJK to be roaming the arctic tears of Eurasia - throughout the entire 'ice-time'. Producing off-shoots like K2 (xLT) - first known from Ust-Isthim and Pestera Oase, later to be reproduced during the early Holocene and thus spread throughout the entire Eurasia - as 'derivatives' of K2.

    Hard to imagine that such parallels could exist due to anything but direct, genetical ties. Thus you have to explain how and why hg F->GHIJK did NOT exist in northern Eurasia during the last part of ice-time.

    Moreover you have to dig up some substantial evidence of WHERE K2 existed to claim it NOT to be "European".

    As for 3 & 4: Please explain why and how the symbol-elements of the Shirig-idol (above) is ANYTHING but Indo-Euroepan.

    ReplyDelete
  58. @apostolos: "Ancient Greece isn't the birthplace of Western culture really. It's one of the influences but through modern interpretation."

    I agree. That is indeed a simplistic paradigm many people hold and peddle today. It is true, of course, that Hellenistic philosophy, reasoning forms an important underpinning of western intellectual tradition in a way it doesn't elsewhere.

    Nevertheless, I think of Western culture proper as developing from an initial synthesis of Hellenistic and LaTenic material cultures that first occured in places like Veneto/eastern Padania, Rhone Delta, and Tuscany-Latium. This is symbolized nicely in Roman legionaires, whose equipment combined various LaTenic technologies(chainmail, montefortino-style helmets, etc) with Hellenic discipline.

    It continued to evolve and culminated into the "Gllo-Roman" set of traditions, the culture of the common people in the Western Roman and later Carolingan empires. There was some Germanic cultural influence but it tends to be overstated, as does their genetic contribution.

    The surge of interest in and subsequent translation of Hellenistic texts on mathematics, etc. starting in the 10th century or so was the final nail, imo. (Greek was not spoken in the old western empire and most of those works, after translation, were circulating in "Frankish lands" for the first time. They were never "lost".)

    ReplyDelete
  59. I would be shocked if modern Greeks aren't mostly descended from Classical Greeks, they are Greeks after all, no ethnic replacement occurred. What I expect is some parts have heavy Slavic admixture while most don't.

    ReplyDelete
  60. In the wake of the Byzantine Empire turmoils, and to break up the confluent Slavic settlement of mainland Greece, emperors imported both Greek speakers (from Italy and Asia Minor) and non-Greeks (such as Armenians) back to Greece.
    Thus Slavic admixture will be individually variable; but still will show a definite north to south gradient .
    So to fully elucidate, you'd need propper regional sample, and investigators who know what they're doing

    ReplyDelete
  61. @Rob:

    What part of my "analysis" of Bulgarians and Montenegrins is completely ignorant? Maybe you didn't understand the reasoning and assumptions behind them?

    Are you familiar with the article published here several months back on South Asia? Let me know if not. Anyway, it beatifully demonstrated, with arrows and such, how ancient population events would have produced the "shifts" on the PCA over time. I'm making assumptions and using a similar approach in the Balkans.

    If you look at modern Balkan populations, they form a roughly linear line, going from Greeks to Albanians(Kosovars), to Bulgarians, Romanians, Serbs and Moldovans.

    1. Let's assume that only a single post-neolithic West Asian population(or at least can be modeled as single) in pre/post-Roman times influenced the region---including Greece. 2. Let's also assume that ancient populations similar to French, Germans, Brits and other W. Euros did not influence the region appreciably.

    In this case, then, the staggering along the cline is mostly attributable to varying levels of steppe-related admixture.

    The question is, was this gradient established in pre-Slavic times upon arrival of IE speakers? Or is most of it better attributed to Slavic migrations instead? Or perhaps it varies significantly from case to case?

    I know this is a rather informal approach but until we get some aDNA from the region, we're kind of in the dark.

    ReplyDelete
  62. @ Anthro Survey

    Your analysis isn't rooted on what is already known.
    Specifically, you argued that "For all we know, the steppe-ward shift of Montenegrins and Bulgarians on your PCAs (in relation to Albanians and Northern Greeks) could be the result of mostly just that: pre-Slavic migrations rich in steppe DNA"

    To repeat something I've pointed out several times: there was a large population replacement in the 7th century in the Balkans. Vast tracts of the Balkans suffered demoghaphic decline due to centuries of warfare, especially the Gothic Wars, then famine, plague, earthquakes. The final event was early in the 7th century when Heraclius evactuated the remaining 'Byzantine' population to Asia Minor & coastal regions of Dalmatia, Albania / Epirus & the Aegean coast.

    This left the entire Balkans north of the Rhodopes (& east of the Dinarides) empty. As far as Bulgaria was concerned, this was settled by Bulgar horsemen from the steppe and Slavs from Ukraine.

    So where do you think the majority of the Balkan 'steppe -like admixture' comes from ?

    Of course, I am not suggesting that Bronze Age or Iron Age Balkans had no steppe admixture, but how there originally was and how much of this remained by the 500s AD remains to be seen ..Either way, I'd contend that it forms a small fraction in modern Balkan populations, I'm willing to hedge even without aDNA

    ReplyDelete
  63. Also, we should envisage potentially 2 or 3 significant post-Neolithic movements into Balkans

    ReplyDelete
  64. @Rob
    "no, i understood. You just don't seem to know the historical processes, rendering your entire analyses useless. But don't feel offended. "

    No offense taken, but this is where I have to correct you. See, I'm well aware the major recorded historical events associated with the turmoil of the 4th-8th centuries(or thereabout). Whether they were Avar raids, Slav incursions or resettlements of Anatolian/Syrian populations into the Balkans----I've read it all.

    The difference is that you seem to place a bit too much weight on ancient records of migrations & devastations (many of which don't even attempt to be quantitative) without enough critical oversight.

    "Vast tracts of the Balkans suffered demoghaphic decline due to centuries of warfare, especially the Gothic Wars, then famine, plague, earthquakes. The final event was early in the 7th century when Heraclius evactuated the remaining 'Byzantine' population to Asia Minor & coastal regions of Dalmatia, Albania / Epirus & the Aegean coast."

    "This left the entire future Bulgaria north of the Rhodopes empty."

    Hmm.... I am not disputing the turmoil, but we have nowhere near the exact numbers on hand. In any case, what you seem to ignore is that the Balkan population did not merely consist of those cities situated on the Danube and Morava rivers at the time. Sirmium and Naissus were not all there was to it. In fact, large sections of the Balkans were comprised of "primitive" non-Romanized rural folks inhabiting the hills and mountains of the region. Moreover, many city-dwellers took to hills, not just the aforementioned coastal regions. And there is good reason to think hill and mountain dwellers were better "insulated".

    Of course, if we're talking about Pannonian/Danubian regions, it's a different story and possibility for major population turnovers is higher---proximity to the source and wide, flat terrain. But this was not the subject of my inquiry. My questions concern populations of Herzegovina, mountains of Western Serbia, Dalmatian Zagora, and the like.

    Are we to believe to believe, for example, that the Langobard influx into parts of the Po Valley(after it was supposedly depopulated by Attila and Byzantine wars) left a huge imprint? There seems to be little indication of that, save for Friuli, maybe.

    "This was settled by Bulgar horsemen from the steppe and Slavs from Ukraine. So where do you think the majority of the 'steppe -like admixture' comes from?"

    Once again, I don't know where to begin guessing. How can I? Until we gain a better understanding of the relative magnitudes at work here, this is all but speculation. How demographically significant was the influx of first IE speakers into the Balkans? To what extent were the events of late Antiquity military raids vs migrations, on a sliding scale? What were the ratios of migrants to locals? So, all in all, your guess is as good as mine here.

    ReplyDelete
  65. As far as envisaging 2-3 post Neolithic movements: You mean from the Near East?

    I did consider the possibility of a West Asian influx when Byzantine emperors were busy resetting the supposedly depopulated region.
    I've wondered about the impact of this for quite some time. So much so that in my quest to obtain more info, I've stumbled on some clowns trolling each other about this on apricity forum and the like. :D

    I've even taken into account the influx of Levantine and Armenian merchants into Moldova and Wallachia during the Ottoman era.

    The latter must have been trivial, for sure.

    As for the resettlements(if consequential) and a putative wave from the vicinity of the West Asian highlands earlier: somehow I don't think they were THAT different. So in theory it shouldn't affect my "math" drastically. What is the likelihood of one being Copt-like and the other Georgian-like? Not very high. I cannot say the same for southern Italy, however. Things seem to look more interesting there.

    ReplyDelete
  66. Anthro

    You're implying that I have only relied on literary sources- which is disingenuous of you.
    You also seem to be under the romantic impression that all these rustic mountain folk took to the hills, or lived in the clouds perhaps ?

    No, extensive landscape surveys show that it was the very opposite is true- it was the rural population that was hit hardest. The towns lasted the longest, before they too were evacuated.
    No villae, no peasant huts, no evidence of arable farming .
    By 620 all inland towns and hilltops were deserted. Again this is well known because it's been researched & written about for decades.

    _______
    We are fairly well informed about the extend of early Bulgar and Slav settlement. Even if You won't trust Theoohanes figure of 10000 Bulgar horsemen; the rather well investigated burials on the lower Danube corroborate the broad scale at least.
    But I think you're quite right , it's probably best not to guess if you don't really have a grasp of the material at hand ; which was my main point against your initial statement.

    ReplyDelete
  67. @ Anthro

    "I did consider the possibility of a West Asian influx when Byzantine emperors were busy resetting the supposedly depopulated region. "

    No, I was referring to a 2nd Neolithic migraion, a 4000 BC migration and a 2220 BC migration . Perhaps you haven't heard of these postulated phenomena ?

    These would have made a much more significant impact than the settlement of a few medieval Armenian communities

    ReplyDelete
  68. Davidski

    Montenegro BA is close in time and space to Greece BA
    How much Yamna scores Montenegro BA?

    ReplyDelete
  69. Arame
    It wasn't good quality, but Dave pulled off a decent PCA

    https://drive.google.com/file/d/0B9o3EYTdM8lQXy1feWVFYWp0WlE/view

    It plots with BA Hungary; so imaginable BA Greece would plot between that circle and LN Greece (maybe a bit more EEF & CHG than Vatya or Mako, for example)

    These BA Montenegro and Hungary individuals plot along modern French and Spanish, not Serbs or Montenegrins. Which again goes to show a sh*#tload has happened in the Balkans since then.

    ReplyDelete
  70. @Roy

    The tMRCA of the "Dinaric" I-S17250 only dates to around the time of the Roman Emperor Marcus Aurelius (c. 160 CE) who campaigned north of the Balkans:

    YFull I-S17250 tMRCA 1950 ybp (from 2010)

    Notice the extremely rapid population expansion that took place in I-S17250 c. 160 CE.

    This is still the time of the Early Late Proto-Slavic language. Specifically, the Sclaveni, the people who the Byzantines identified as the South Slavs, were first mentioned by Procopius in the year 518 CE.

    The interesting thing is that the tMRCA of "Dinaric" I-S17250 seems to correspond to the Antonine Plague of 165–180 CE, and the first mention of the expansion of the Sclaveni seems to correspond to the Plague of Justinian of 541-542 CE, continuing to the 8th century.
    Both plagues had a very high mortality (25% in the Antonine Plague) and would have profoundly changed the demographics of the Balkans.

    Who are the specific candidates for the ancestors of I-S17250, aside from the later 6th century Sclaveni?

    Some presumably elite individual belonging to one of the groups on this map of the Roman Balkans in the 2nd century:
    Map showing Roman Dacia and surrounding peoples in AD 125

    This extremely recent tMRCA of "Dinaric I2a" is both unexpected, and a complete mystery, but it does help explain the very late breakup of the South Slavic languages and the Slavic languages in general. This means that we can rule out I-S17250 as being present at all in the Classical Era Balkans, even though today it is up to 50% of all Y-DNA haplotypes of some areas of the Balkans.

    ReplyDelete
  71. Rob

    Thanks. Yes it is really shocking to see where it plots.
    Your mention of Vucedol Cetina makes sense. It could be that Vucedol is to blame for that strong shift toward Crown Eurasians.

    ReplyDelete
  72. ? crown Eurasians . As in Ust-Ishm?

    ReplyDelete
  73. None :) I mean shift toward WHG/SHG/EHG.

    Illyrian languages are considered as centum right? If so then it would make sense that E-V13 was Indo-Europeanised in West Balkans.

    ReplyDelete
  74. Ok. I was impressed by Balkan IA.
    Balkan LBA is in more or less predictable place.

    ReplyDelete
  75. Davidski

    Is it possible to place one of good quality Bulgaria IA samples into that PCA?

    ReplyDelete
  76. I've done as much as I can do with the Montenegrin and Bulgarian samples.

    We just need better sequences to analyze them properly.

    ReplyDelete
  77. Early Mycenaeans seemed to have had connections to the later phase of the Vucedol Culture, Wietenburg Culture and even as far as the Wessex Culture and Helversum Culture....

    ReplyDelete
    Replies
    1. The Vucedol Culture seemed to have had close connections to Cernavoda/Cotofeni, Baden, Ezero,and Globular Amphora Cultures.Basically the Bronze Age Danubian Complex Cultures. This was the transition phase of Pre-Indo-European to Indo-European within the Balkans starting with the Cernavoda Culture who replaced the Karanova Culture at +-4000 bC.with the major tipping point between 3000 bC.and 2500 bC.

      Delete
  78. @ Arame

    Yes, Illyrian is centum, although whether there was one distinct Illyrian language is not likely. Anyhow, I have the feeling that it is closer to Celtic & Italic due to Halstatt & Urnfield influences. Thracians could be quite different.

    Indeed, in Dave's plot, the Iron Age MOntenegrin looks central European, whilst the Iron age Thracian in the original paper plotted near Italians IIRC.

    ReplyDelete
  79. Rob, going by that PCA, it seems that Hungary_BA plots along much of the European cline (from Tuscany to the UK?). Perhaps similar variation existed in a non-homogenized Bronze Age Montenegrin population too? Is the Iron Age guy a recent intrusion in the area of Velika Gruda or a representative of more or less the same population that hadn't yet homogenized?

    Going just by this, contemporary Montenegrins seem like even more steppe and even less farmer shifted than the Bronze Age one (who seems like a farmer/steppe mix, more or less, no?) or is there something else going on too?

    Based on just this limited sampling, it seems that Albanians must have changed quite a bit since pre/proto-Illyrian times as well, unless something like what I mentioned in the first paragraph was going on with the elements that constitute the modern South Balkans already more or less in place, even if reinforced/replaced by the later, historical migrations from East, West and North.

    I do find value in the OP study though, even if the conclusion about the Slavic intrusions is very tentative (and it depends on what one means with "Slavic", the Slavicization of Greece continues today after all with quite a few mixed marriages, but I assume they had proto and early pre-Balkan Slavs in mind), since it gave us samples of the Doricizing Tsakonians (the south ones more representative) and even Arvanites in the Argolid and Corinthia samples.


    Anyway, I can't wait for more and better pre-Hellenistic South Balkan samples. Are any early medieval Slavic samples being tested, since David mentioned something to that effect?

    ReplyDelete
  80. @Coldmountains
    “Slavs had nothing to do with Sintashta.”

    Coldmountains, read Mittnik et al. 2017

    “The presence of ancestry from the Pontic Steppe among Baltic CWC individuals without the Anatolian farming component must be due to a direct migration of steppe pastoralists that did not pick up this ancestry in Central Europe. This could lend support to a linguistic model that sees a branching of Balto-Slavic from a Proto-IndoEuropean homeland in the west Eurasian steppe.”

    They link R1a-Z645 from the steppe with Balto-Slavs. So R1a-Z645 Corded-Ware originally was Balto-Slavic. Balto-Slavs migrated from the steppe west and north and in Vistula-Dnieper area were mixing with Neolithic Farmers of Tripolye and TRB cultures and there pre-Slavs formed. In the north, around Baltic, pre-Balts were forming.
    After that from Vistula-Dnieper area admixed with Neolithic Farmers Slavic Corded Ware was migrating East. They were mixing with Yamnaya tribes on the steppe and as a result of that mixing new Cultures like Sintashta, Srubnay etc originated.
    This is why there are so many common things between Slavic and Indo-Iranians languages, religions, poetry meters etc.

    Slavs also influenced Greek culture. Greek poetry, religion also shows links to Slavic and Indo-Iranian common features. You probably have heard about Orpheus , Hyperborean Apollo etc. It came to Greece from the North. It not an accident that Rigvedic poet with a name which Slavs can understand is using the same metaphors as Pindar and those metaphors can be explained by Slavic folklore from XVI century. It proves that such links really existed.

    ReplyDelete
  81. Some comments show up and are vanishing but others stay. Nothing about the content, it's some kind of bug ... I give up.

    ReplyDelete
  82. Rob, in spite of the Slavic expansion, many areas retained their earlier languages, Greek was retained in Greece, Albanian in Albania/Kosovo and Romanian in Romania. It would be interesting to know the position of modern Albanians, Romanians and Hungarians on the PC map "Balkans-Baltic_LBA-IA_Europe_PCA.png".



    ReplyDelete
  83. #1 "Dinaric I2a", I-S17250, has a remarkably recent tMRCA according to YFull, of only 160 CE.

    I-S17250 YFull tree, tMRCA 160 CE (1,850 ybp)

    I-S17250 shows signs of an extremely rapid population expansion at the very start. It seems that the closest group to the ethinic origin of the I-S17250 group are the Sclaveni. The tMRCA is right at the start of the Late Proto-Slavic linguistic period. It also coincides with the Roman Emperor Marcus Aurelius' Marcommani Wars, a campaign against the Marcomanni, Quadi and Sarmatians, when there was a movement southward by various allied tribes into the Roman Empire.

    A map of the archaeological cultures of Eastern Europe in the 1st Century

    It also coincides with the Antonine Plague which had a mortality of 25%. The Antonine Plague may have been either smallpox, acquired from trade with India, or measles, but measles in its modern form is thought to have originated a bit later. The combination of invasions from the north, including peoples who may have been associated with the Proto-Slavs, and the high mortality from the Antonine Plague, seems to have produced a strong founder effect followed by a rapid population expansion

    ReplyDelete
  84. #1a "Dinaric I2a", I-S17250, has a remarkably recent tMRCA according to YFull, of only 160 CE.

    I-S17250 YFull tree, tMRCA 160 CE (1,850 ybp)

    ReplyDelete
  85. #1b I-S17250 shows signs of an extremely rapid population expansion at the very start. It seems that the closest group to the ethinic origin of the I-S17250 group are the Sclaveni. The tMRCA is right at the start of the Late Proto-Slavic linguistic period. It also coincides with the Roman Emperor Marcus Aurelius' Marcommani Wars, a campaign against the Marcomanni, Quadi and Sarmatians, when there was a movement southward by various allied tribes into the Roman Empire.

    A map of the archaeological cultures of Eastern Europe in the 1st Century

    ReplyDelete
  86. It also coincides with the Antonine Plague which had a mortality of 25%. The Antonine Plague may have been either smallpox, acquired from trade with India, or measles, but measles in its modern form is thought to have originated a bit later. The combination of invasions from the north, including peoples who may have been associated with the Proto-Slavs, and the high mortality from the Antonine Plague, seems to have produced a strong founder effect followed by a rapid population expansion.

    ReplyDelete
  87. I-CTS10228, the immediate ancestor of South Slavic I-S17250, according to YFull has a tMRCA of c. 200 BCE, and has members from Greece, Poland, Ukraine, Belarus, North Germany, Lithuania, and Finland.

    This seems to correspond to some degree to peoples associated with the other two proto-Slavic tribes, the Antes and the Vistula Veneti who Pliny the Elder calls the "Sarmatae Venedi". Later Ptolemy names the "Vouenedai" along the Baltic coast.

    Jordanes in his Getica (c. 550 CE) says that the Veneti are the ancestors of the Sclaveni and Antes. I think then we may want to look for the origin of generally "Slavic" I-CTS10228 as being in the far north, along the Baltic coast, rather than in regions further south and east where it is found today, particularly in high percentages in the Balkans. The late Neolithic precursor of I-CTS10228, I-L621 is also found in Britain and Ireland. While I-L621 existed at the time of the Corded Ware Culture, I-CTS10228 did not expand at all until c. 200 BCE. so we can't say that formed a part of the Corded Ware expansion.

    ReplyDelete
  88. Interestingly, I-Y23118 is an Ashkenazi Jewish clade within I-A2512 which has a tMRCA of 1950 years, c. 40 CE and this I-CTS10228 has Greeks from the Peloponnese as its earliest members. This also includes a very small number of Hispanics. The question is whether the ancestor of the Ashkenazi Jews in I-Y23118 joined the Jewish population (or fathered a Jewish son) at the time of the First Jewish War and the exile of many Judeans throughout the Roman Empire, when this could have happened in Greece, or whether the ancestor of these I-A2515* Greeks moved south later. Based on the tMRCAs and the phylogeny, it seems that these Greeks, who are clearly of northern (Baltic?) origin, were present in the Peloponnese from the first century onward, because separate Greek Y I-A2512 lineages branch off right at the root, starting in the 1st century CE. If this wasn't uniquely Greek and ancient, we would expect to see other Greeks scattered throughout I-CTS10228, which we don't.

    So here, we can perhaps use one event (the First Jewish War of of 67-70 CE) to help validate the presence of a particular group of northern (Baltic) origin (the Vistula Veneti?) in the 1st century in a far distant place, Greece. It could be that the common link is that these were both enslaved by the Roman Empire.

    ReplyDelete
  89. The origins of I-CTS10228 appear to be in the Baltic region among the ancestral (Vistula) Veneti ("Sarmatae Venedi"), the "Winden" or "Wends", but they appear to have made a rapid migration to the far south within 250 years of their origin, perhaps with a few captives or mercenaries living in Greece. We should also be able to trace this Baltic-originated migration southward through the mtDNA and the autosomal DNA. WHG haplogroup I-M423, as represented by Loschbour, remained relatively rare outside of Sardinia until this rapid expansion, but within a few centuries, it began to dominate certain South Slavic regions like Bosnia and expand throughout the Slavic-populated areas. This shows how an essentially non-existent Y-DNA clade from one distant region can overtake and replace native Y-DNA clades as the result of a migration and expansion of a new group, and that the Y-DNA makeup of historic populations as late as the Classical Era may have been very different from the present day Y haplogroup profile in these same regions.

    Clearly, it's not only migrations that change the population structure of a region, but also pandemics with high mortality. Random variations in immune system alleles can cause a new group to dominate and partially replace a previous population lacking immunity. This can be a model for what happened at the start of the Early Bronze Age, with the Proto-Indo-European steppe migrations replacing earlier Neolithic populations, as a result of their previously acquired immunity to disease like the bubonic plague contracted through contact with steppe rodent colonies that were missing in regions like Europe, the Near East, and South Asia.

    ReplyDelete
  90. @Rob
    I see that there are still some misconceptions about Illyrians.

    The first one about the character of their language. It is still repeated that it was a Centum language while this is not the case actually. The centum character hypothesis for Illyrian was proposed in the beginning of XXth century when Venetic language was considered as Illyrian. It was later proved that Venetic is part of Italic group and was not an Illyrian language. There are no sufficient data on Illyrian and if we want to be very picky on what remains it should be grouped with Balto-Slavic, Albanian and Armenian as partly satemized language.

    The second misconception is about the relation with Urnfield culture. As the famous Yugoslavian archaeologist Benac has stated Urnfield people and Proto Illyrians were two different people (Glasinac-Mat culture). The dominant form of burial practice among Illyrians was inhumation in tumulis. This practice appeared around 2200 BC and continued even during Roman Empire up to Early Middle Ages in some cases. Urns appeared late around 1300 BC and were not widespread among Illyrians. They were very few in numbers although they also were placed in Tumulis. An eg. during excavations in Kamenica Tumulis in Albania (which lasted from 13th century BC to Vth century BC) were found the remains of some 430 inhumated individuals while there was only one case of urn burial.

    ReplyDelete
  91. "Samuel Andrews said...
    It's super confusing how Ashkenazi Jews, Southern Italians, and Greeks became so similar."

    The similarity is surely due to multiple layers, but it's interesting that it can already be seen in the Anatolia_Chalcolithic sample from 4000 BC.

    I think at some point A_Ch type people infused some European farmers, like Remedello_BA, and then merged with Levant_BA type people to form the "East_Med" component we all know and love. (The even small amount of Remedello is important because it explains the extra bit of WHG seen uniformly across the Levant and North Africa.) I got decent fits on nMonte awhile back:

    Druze
    “Jordan_EBA” 44.55
    “Anatolia_Chalcolithic” 37.2
    “Remedello_BA” 11.25
    “Munda” 6.95
    “Masai_Kinyawa” 0.05
    distance=0.014307

    Ashkenazi_Jew
    “Remedello_BA” 31.8
    “Anatolia_Chalcolithic” 31.2
    “Jordan_EBA” 29.6
    “Nganasan” 3.85
    “Munda” 2.85
    “Yoruba” 0.7
    distance=0.01097

    Cypriot
    “Anatolia_Chalcolithic” 38
    “Jordan_EBA” 31
    “Remedello_BA” 24.85
    “Munda” 4.7
    “Nganasan” 1.45
    distance=0.013694

    Italian_EastSicilian
    “Remedello_BA” 38.45
    “Anatolia_Chalcolithic” 28.8
    “Jordan_EBA” 25.9
    “Munda” 3.65
    “Nganasan” 2.5
    “Yoruba” 0.7
    distance=0.013346

    Makes me think that population movements from western Anatolia across the Med were massive, and some of the Luwians / Sea Peoples / Bronze Age collapse stuff might be onto something. The question is when/how the extra Levantine was added...

    ReplyDelete
  92. Did Jewish men marry a lot of Italian and Greek women? Would that account for some European admixture in Ashkenazi?

    ReplyDelete
  93. @Rob

    The evidence of strong cultural continuity in the western Balkans(esp in the Dinaric region) in pre and post-Slavic, be it in construction of houses to burials, is substantial. The rural areas devastated most were in proximity to major commercial routes(i.e Morava and Lower Danube corridors). It is also in those corridors that we find the highest density of artifacts indicating penetration of Slavic material culture. This devastatation and pattern of Slavic settlement tends to be patchy, too, and often these Slavs tended to settle together with the locals. Various scholars like Florin Curta, Dan Dzino, and John Fine have discussed these points thoroughly.

    And if it was so "empty" north of the Rhodopes as you say, there wouldn't be such a strong autosomal "south euro" and west asian pull in folks from places like Trebinje and Cetinje. Well, barring any kind of later bottlenecks in which autochtone genes regained ascendancy, of course. Not that I'm a fan of the "phenotype" crowd, but to observe Herzegovans and to state they have minor autochtonous ancestry is to observe Uzbeks and state they have minor ENA.

    P.S: It would be interesting to analyze the Yugoslav language family for tell-tale signs of a pre-Slavic substratum. Given our limited set of Illyrian and Thracian words, this will not be an easy task. The strong ideology of Slavicism in the region complicates matters further.

    ReplyDelete
  94. @Rob

    "No, I was referring to a 2nd Neolithic migraion, a 4000 BC migration and a 2220 BC migration . Perhaps you haven't heard of these postulated phenomena?"

    There was a paper published by a Turkish team in the summer of 2016 suggesting such events, if I recall correctly.

    "These would have made a much more significant impact than the settlement of a few medieval Armenian communities"

    Well, yeah. I think here is where you're spot on. I also don't believe that any major post-Roman influx from West Asia had much of a demographic impact.

    So, it's safe to assume we are left with a, more or less, crystallized population of native Balkaners until the Slavic migrations. That's where I'm wondering about the nature of any gradients that can be ascertained.

    ReplyDelete
  95. @Kristiina

    In fact, Slavic was spoken in modern-day Albania, too(alongside greek). This explains heavy Slavic lexicon in Albanian. For instance the word for south---jug. The ancestor of Albanian language has been traced to the region of Dibra-Mat by some. In the middle ages, the language expanded outward: first into Albania proper and then into Kosovo during Ottoman times(massive founder effect there associated with E-v13 as described by Dienekes some years back).

    As for Romanian----HUGE Slavic layer in that language. So much so that even certain phrases using Romance words still display Slavic construction. This is apparent in Romanian two-digit cardinal numbers, for instance.

    ReplyDelete
  96. @Open

    The wends only appear on the western side of the Vistula river and are associated only with the slavic tribe called the Veleti . They migrated from the upper vistula to Mecklenburg and where later had wars with the saxons...Wendish crusades...........called Wendish , because Veleti where the only wends.
    The Vistula Venedi where from west-baltic cairns culture , with most disappearing into Goth society ~200AD, the remained later appeared as the Warmians ( proto-prussian 12th century tribe )

    The Illyrians where all different tribes , speaking many different languages and the term Illyrian is only a geographical term and not an ethnic term

    ReplyDelete
  97. @MFA said

    @Anthro Survey
    "Your analogy is probably close to the truth. I can say according to David's K7 ADMIXTURE run Early Iron Age Hasanlu sample represents 96% of modern Kurdish ancestry, while the number for other modern west Iranics was much much lower. I wish David would test it further via D-stats et al, but that's what we have now. I guess the number would be somewhere around %70 to %90 from one extreme to another regarding how much ancestry Kurds have from pre-Median populations."

    The unwise mistake you make here, is that you call this population "Pre_Median" as if the Iranic tribes that possibly moved into the Iranian Plateau, were already the Medes as if the Medes or West Iranic tribes as a group already existed before they are recorded in Western Asia. That is in historic, linguistic and archeological point of view incorrect to impossible. Reminds me of the past when some people tried to meassure Scythian ancestry solely based on Andronovo samples.

    How the Medes or West Iranic tribes came to existence is when Yaz like Iranic speakers (who should have differed already significantly from Srubna, Andronovo or Sintashta by themselves) moved further West into the Zagros mountains. From sources we know these people did not call themselves yet Medes but simply Aryans. When they got caught under Assyrian oppression these Aryan tribes went into alliance with the local Kura Araxes like tribes (who going by historic sources should have made up ~30% of the population). And a new nation was born. The Medes were the product of six/seven tribes that went into an alliance. And not a group of Iranic speakers that came straight out of the Steppes.

    The same is the case with the Persians many people (especially those not very knowledged in history) assume the Persians and Medes moved into the Iranian Plateau as two seperate groups. The reality is the Sw Iranic branch itself is more like an offspring of the Proto West Iranic group that became the Medes and it was only one major movement of Yaz like Iranic speakers into the Zagros mountains. The Persians are the product of a group within the Median confederation that was rather forced into the Southwest of the Iranian Plateau and merged with the local Elamites. This is the moment when the Persians were born.

    The Iranic contribution in modern West Iranics looks so little because we only have Steppe Iranic samples to hand. With Yaz samples at hand things will look drastically different. Nevermind that the Medes themselves could not even be represented only by Yaz ancestry but something in line of Yaz + Kura Araxes. The Kurds look like 90% derived from Hasanlu because the Medes themselves would have been almost identical to the Hasanlu sample to begin with. Even their Yaz like ancestry would have been practically very similar with 75% CHG_IranCHL and only difference of ~25% EHG admixture. How else are you going to explain that between Hasanlu sample and modern West Iranics is only a difference of like 10-20% Steppe like admixture? It can't be that the Medes were so Steppe like when suddenly thousand years later the modern West Iranic tribes look again almost identical to the local Hasanlu samples. Obviously the Median period did not bring allot of change in genetics from Hasanlu to modern era.

    ReplyDelete
  98. @Kurti

    Nice analysis of the progressive dilution. Mirrors what I've always believed regarding movement of Indo-Aryan speech into Iranian plateau.

    Now, can we extend this sort of reasoning into the Indian subcontinent?

    How steppe-like were the Indic speakers moving into India, presumably a mixture of ASI and Iran_Neo-like components at the time? & What was the ratio of ASI/IranNeo in pre-Aryan India?

    See, I had always been under the impression that Indo-Aryan speakers, prior to crossing the Khyber pass, were mixed considerably with IranNeo-rich locals of BMAC/Yaz area to begin with.

    But recent qpAdm workups here as well as data from Lazaridis et al. threw a curveball. A typical breakdown was like this: 40% steppe, 35% IranNeo and the rest being ASI. If this is true, then Indo-Aryan invaders would have been fairly steppe-like, or at least >>50% steppe. That is, of course, unless the locals were pred. ASI with little endogenous IranNeo.

    This was a typical high caste result. Perhaps the invaders mixed directly with hunter gathering ASI-rich tribes and this formed the backbone of Hindu high caste society? (Basically, invaders of 50-50 steppe:Iran mixing with 90%+ ASI)

    But these scenarios of ASI HG refugiums seem implausible on many levels..... This, for example, does not seem to have been the case in Europe upon the arrival of steppics. Europe is a fairly analogous system of 3-part mixture consisting of West-Eurasian HGs, basal-rich Anatolian farmer types and steppe. & Unlike Northern India, Europe has plenty of hilly terrain in which HGs could have taken refuge more easily.

    Another reason why these results were a curveball is that they don't seem to intuitively align with previous admixture studies and the multi-sample workup by Kurd on anthrogenica in which there were many(but not all) South Asian members. Member Jesus was on that spread if you recall or know what I'm talking about. Steppe values were far smaller and people tend to accuse him of spiking them.

    ReplyDelete
  99. Anthro, good comments, though I'd point more towards Slavic toponymy in Albania and Greece (lower density in North-Central Albania and in coastal Greece) than Slavic lexical borrowings for that sort of thing. For example, Greek seems to have much less lexical Slavic influence than Albanians and instead seems to have influenced its neigbhoring Slavic dialects much more than they did. Similarly Ottoman Turkish has influenced the lexicon of both Greek and Albanian way more than Balkan Slavic did, despite Anatolian Turkish ancestry being seemingly non-existent in the Balkans.

    In general, it seems that the languages of the high/elite cultures (Byzantine Greek, Ottoman Turkish) influenced the rest more than vice versa but with toponyms that factor can be often (but not always, some Ottoman Turkish toponyms named after local rules come to mind for example) controlled for. Those have their own problems too but in general they seem like a good placeholder until a lot of ancient samples come in. Ghegs and Aegean Greeks do seem to be less Slavic influenced than Tosks and inland Greeks, after all.

    ReplyDelete
  100. Slavs in the Balkans were quite mixed with Mongoloid Avars and Scythians. If anything modern Greeks are mostly descended from Turks and Albanians.

    ReplyDelete
  101. @Alogo

    Indeed, there is a high density of Slavic toponyms in Toskland, Greek Macedonia and Greek epirus. Two readily jump to mind: Pogradec and Korce(alt. Koritsa).

    I think languages associated with elite culture tend to borrow less in general. Farsi has far fewer Turkic and Indian words than north indian dialects and Turkish languages(they say majority of Ottoman Turkish lexicon was Farsi and most important words are).

    Ottoman Turkish influenced speech of Muslims to a greater extent, tho trade-specific vocabulary entered duscourse of Christians like Serbs, too. Speaking of Ottoman Turkish, Albanian seems to have even borrowed izofa structure(a Farsi feature, ultimately derived from Arabic)! Ex: qarku i korces(meaning county of korce).

    What about the traditional dialects of inland Greece? Do they show Slavic influence to the extent Tosk or perhaps even Romanian do?

    ReplyDelete
  102. @ Aigest

    (1) I take your point about perhaps the southerly Illyrian languages which might of been The forerunner to Proto Albanian bing partly satemized due to long contact with Thracian, but I very much doubt central and northern dialects such as Pannonian, Liburnian and Japodian having any Satem features. Of course, the evidence at disposal is bleakly meagre

    (2) Urn cremations might be rather rare in the Balkans , but if that one iron age Montenegro Sample in David's plot is anything to go by -there was a significant migration of central European like people after the Bronze Age as far down as Montenegro. And of course this individual came from an inhumation burial, not cremation

    ReplyDelete
    Replies
    1. 1) I beg to differ on this issue. First Liburnian is not part of Illyrian languages but of Venetic branch of Italic group. For example take Liburnian personal names which were different even from Dalmatians their closest neighbors. I might add that the supporters of Satem character for illyrian languages took many of their examples from northern illyrians onomastics like Osseriates (in Pannonia) derived from /*eghero/ (lake) or Birziminium (old name of Podgorica in Montenegro) from PIE /*bherǵh/ or Asamum (in Dalmatia) from PIE /*aḱ-mo/ (sharp) showing all these examples Satem characteristics.

      2) I was only pointing out some wrong old assumptions about Illyrians and Urnfield culture. According to archeologists (Beach, Stipcevic, Korkuti etc) Proto Illyrians and Urnfield people were different. Urnfield culture never became a mainstream in Illyrian territory. There were people who were cremated and put in urns and they appear around 1100 bc but the overwhelming majority continued old traditions of inhumation in tumulis. From all later Illyrian territories there is no sign of influence from Urnfield culture except in parts of Slovenia. This means that existing western balkan bronze age people kept their culture. Since tumulis among Illyrian areas appear around 2200-2100 BC I am very interested of your comments on migrations in the Balkan around that time. Who were those people. Can you tell us more?

      Delete
    2. Multi-Cordoned Ware Culture maybe ?

      Delete
  103. @alogo

    I forgot to add that Toskland has been one of the core territories of the Bulgarian Empire and in the vicinity of a major cultural center at the time: Ohrid.

    ReplyDelete
  104. Anthro, from what I recall the mainland Greek dialects (before levelling due to Standard Modern Greek) with the most Slavic lexical borrowings were those of Epirus. Similarly, Slavic toponymy seems more dense in Epirus compared to most of Greece.

    ReplyDelete
  105. We have no a-dna from south-eastern Europe or southern Italy, just nothing at all. Until we have something concrete you all should talk about something else...

    ReplyDelete
  106. @ Antrho Survey

    "The evidence of strong cultural continuity in the western Balkans(esp in the Dinaric region) in pre and post-Slavic, be it in construction of houses to burials, is substantial. The rural areas devastated most were in proximity to major commercial routes(i.e Morava and Lower Danube corridors). It is also in those corridors that we find the highest density of artifacts indicating penetration of Slavic material culture. This devastatation and pattern of Slavic settlement tends to be patchy, too, "

    To be specific, the continuity is in the littoral strip - Istria , Dalmatia, Epirus. The actualmountain region was never densly populated at anby period. East of this, between the Dinaridies and Danube -Sava, there is a complete cultural shift (which began even before the Slavic settlement, but goes back to Goths, Lombards, etc)



    "often these Slavs tended to settle together with the locals. Various scholars like Florin Curta, Dan Dzino, and John Fine have discussed these points thoroughly"

    No that is not what has been written (we can exclude Fine becuase, although very good, he is a general historian). As per above, outside Dalmatia., Ionia, & parts of Macedonia / Greece, we are looking at a landnahme, after a hiatus of 50-100 years.


    * "And if it was so "empty" north of the Rhodopes as you say, there wouldn't be such a strong autosomal "south euro" and west asian pull in folks from places like Trebinje and Cetinje. Well, barring any kind of later bottlenecks in which autochtone genes regained ascendancy, of course. Not that I'm a fan of the "phenotype" crowd, but to observe Herzegovans and to state they have minor autochtonous ancestry is to observe Uzbeks and state they have minor ENA"

    -> well you need to think in 4D :)
    How do you know where admixture with 'south Euro' or Palaeo-Balkan admixture took place & when ? Sure there might have been some early episodes, even before the Slavs crossed the Danube to settle permanently (through POWs and impressed East Romans), but the presence of 'South Euro' phenotype (& indeed observable genetic admixture) doesn;t negate that the Sava-Drava - Danube- Morava plains were depopulated to an extent that arhcaeological remains are non-existent, and -essentially- settlement occurred on a 'terra deserta'. This means that admixture occurred after the Slavs had settled the major plains and reached the limits of their expansion - Dalmatia, Greece, southern Thrace. The soon formed multi-ethnic duchies, which led to 'assimilation' & population mobility within the newly formed, and more stable, territories. When the Bulgars expanded into macedonia & Thrace, they engaged in large scale population transfers, - entire communities were moved from Thrace to the Danube plain (!)

    Secondly, we have other, non-Slavic groups re-expanding back a little later: so we have Albanians expanding north again from Albania, Vlachs moving about territories after the Vasil's campaigns & during the Bulgarian Empire, and Greeks being settled strategically in major towns (Belgrade, Naisssu, Skopje, Serdica, etc) after the Byzantine re-conquista of 1018. Even if the local Slav princes then booted out Byzantine Strategoi in the recurring cycle of vassalage / rebellion, the regular civillians - whatever their language- remained. Indeed, this is what the physical anthropology shows, the early Slavs in central Balkans are said to have typical Nordid features, and only after a couple of centuries do they start becoming 'Dinaricized'.

    ReplyDelete
  107. @Continued ;

    * " would be interesting to analyze the Yugoslav language family for tell-tale signs of a pre-Slavic substratum. Given our limited set of Illyrian and Thracian words, this will not be an easy task. The strong ideology of Slavicism in the region complicates matters further."

    That isn't an accurate assessment. There is no pro-Slavic bias in the studies of scholars from ex-Yugoslavia. The pre-Slavic stratum has long been recognized, as it is plain as day. Entire volumes of of works are dedicated to Illyrian & Thracian roots, not to mention Gothic, of modern day south Slavs. If anything, it was over-emphasized.
    The evidenc of substrata is certainy there- but, as outlined before, one needs to be specific as to where and when it was. Blanket staements like Slavs came across natives "living in the Mountains" are quite wrong, and miss the point.

    ReplyDelete
  108. @Open Genomes

    Your comments are going into the spam folder.

    Try changing the structure of your posts, maybe by breaking up and shortening your paragraphs?

    ReplyDelete
  109. @Ariel
    I think that part of the problem is ignorance about archaeology, both from the geneticists who conduct the studies and from posters here. In many ways academic genetics bears the brunt of the problem--too often hypotheses originate from antiquated culture-historical archaeology ( I know Luca Cavalli-Sforza's models often suffer from this old perspective). I recommend going directly to excavation reports and, if you must, recent archaeological syntheses of such reports. So much excavation has been conducted in mainland Greece, Crete and Cyprus and the summaries often afford us new and valuable information for interpreting genetic data. Ancient demography is a very valuable field and can yield estimates of population immigration, elite dominance effects, etc...Forget about Wikipedia--it will lead you down the rabbit hole! That said, what we know about Greece at this time is relatively straightforward: 1) Neolithic Greece resemble Barcin and the Balkan Neolithic 2) Modern insular Greece--Cyprus and Crete--affiliate with Southern Italy, Sicily and, somewhat to the Levant. Modern Mainland Greece has an excess of Eastern European/Steppe Ancestry. What we don't know is when Mainland Greece diverged from Mediterranean Greece. Was is during the EBA III (circa 2200 BCE) or during the coming of the Greek language, or later--Slavic incursions, Roman, Venetian thalassocracy, etc...Open Genomes suggests that I-M423 is later via a recent TMRCA for the Dinaric Modal Haplotype and, given the lack of R1a and I2a (xM26) in Sicily and Southern Italy, I would wager that Mycenaean Greece lacked I2a and much R1a. That leaves E-V13, various G lineages, and J2a/J2b lineages as representative of Neolithic and LBA Mycenaean Greece on the Y chromosome.

    ReplyDelete
  110. "I think that part of the problem is ignorance about archaeology"
    'What we don't know is when Mainland Greece diverged from Mediterranean Greece."

    Looks like archeology alone can't solve a central/basic question, while aDNA can and will. ;)

    I agree to your greater point though, and admit that I'm one of those who approaches this from a genetics/math perspective, and only has a very surface knowledge of archeology. So... what can archaeology tell us about the spread of Levantine ancestry across Greece and Southern Italy? Is it simply a specific wave of Near-East shifted Neolithic farmers? Or a later significant source/event? Or slow coalescence through millennia old contact networks? Or all or none of the above?

    Also, has anyone looked really closely as the common Y chromosomes of Greeks, southern Italians, Jews, etc., especially G2a and J2 as you mention? A high-res look at that could tell us something while we wait for more aDNA.

    ReplyDelete
  111. How much R1A the Myceneans have,or don't have, will be very interesting.Presumably the R1A they do have came from the steppe originally.

    ReplyDelete
  112. I say that Anatolia Chalcolithic kind of proves that there were Cyprus-like populations even back then. The question about "when" they start moving west is being trivialized in many ways. If you think about it populations as far a central-northern Italy show east-med admixture.
    -Italian_Tuscan
    "Sardinian" 35.6
    "Yamnaya_Samara:I0441" 26.15
    "Barcin_Neolithic:I0708" 22.3
    "Anatolia_Chalcolithic:I1584" 11.15
    "Jordan_EBA:I1706" 4.8
    I can imagine hundreds of theories and explanations, none with "true data" to back them, because there is no data. Because appartently "they" have to test every single ancient sample in Spain before we get something from Greece or Italy.
    But what do I know? Too me even neolithic Greece had some east-med already.
    -Greece_LN:Klei10
    "Barcin_Neolithic:I0708" 52.55
    "Sardinian" 27.25
    "Anatolia_Chalcolithic:I1584" 12.9
    "Jordan_EBA:I1706" 7.3
    "Yamnaya_Samara:I0441" 0
    And that lack of WHG in south-eastern Europe goes a long way back. Maybe there were never sardinian-like people in those areas. At the same time the east-med component never truly reached the Iberian peninsula. And it is reasonable to think that if that wave of east-med people is very ancient the rest of southern Europe should have felt the impact. So it's totaly possible that the east-med is fairly recent. Again, how should I know? How should anybody know?
    Portuguese
    "Sardinian" 71.7
    "Yamnaya_Samara:I0441" 25.15
    "Somali" 3.15
    "Jordan_EBA:I1706" 0
    "Barcin_Neolithic:I0708" 0
    "Anatolia_Chalcolithic:I1584" 0
    "Greece_LN:Klei10" 0

    Italian_South
    "Anatolia_Chalcolithic:I1584" 31.1
    "Sardinian" 23.15
    "Jordan_EBA:I1706" 16.45
    "Yamnaya_Samara:I0441" 13.3
    "Greece_LN:Klei10" 12.3
    "Barcin_Neolithic:I0708" 3
    "Somali" 0.7

    (0% east med against +50% is a huge difference)

    And it's not true that the levant has any kind of southern European admixture, a fact that complicates the jewish question even further... (IMO euro-jews are partially roman convert, thus the similarity with some europeans is incidental)

    Lebanese_Muslim
    "Anatolia_Chalcolithic:I1584" 55.35
    "Jordan_EBA:I1706" 32.55
    "Yamnaya_Samara:I0441" 8.15
    "Somali" 3.95
    "Sardinian" 0
    "Barcin_Neolithic:I0708" 0
    "Greece_LN:Klei10" 0

    ReplyDelete
  113. Dave, I am still interested in the position of non-slavic continental Greeks, Albanians, Romanians and Hungarians on your PC map Balkans-Baltic_LBA-IA_Europe_PCA.png (https://drive.google.com/file/d/0B9o3EYTdM8lQXy1feWVFYWp0WlE/view)

    Could you tell me where their position is with respect to Montenegrins?

    ReplyDelete
  114. Recently R1a-Z280 was found in the most western area of Lusatian culture. I also expect plenty of R1a-M458 in this culture and of course Slavic languages.

    Archeologists tell us about many migrations from the north to the Balkans and Greece. There were Corded Ware migrations, there were also Lusatian culture migrations.
    Look at the maps of very old and purely Slavonic toponyms of the types: ‘sopot’, ‘bardo’, ‘krak’, ‘nakel’:

    http://www.historycy.org/index.php?act=Attach&type=post&id=11773

    http://www.historycy.org/index.php?act=Attach&type=post&id=11772

    http://www.historycy.org/index.php?act=Attach&type=post&id=11771

    The names of gods like Greek ‘Keraunos’, Phrygian ’Bagaios’, Phrygian and Thracian ‘Sabazios’ have Slavic etymology.

    I expect Ancient Greeks from some regions of Greece to be genetically not very different from modern Greeks because the links between Ancient Slavs living in Vistula-Dnieper area and Ancient Greeks existed in old times. Slavic influence in Greece has a long history.

    ReplyDelete
  115. @RK

    The two outlier Karakalpaks appear to have western ancestry from the cline that runs from Tuva to eastern Turkey/Armenia, not to the European Mediterranean. So I reckon what you're picking up is Anatolian farmer admixture.

    https://drive.google.com/file/d/0B9o3EYTdM8lQSFA2c0NmMWpyXzg/view?usp=sharing

    @Kristiina

    I can show you how they cluster on a very similar plot.

    https://drive.google.com/file/d/0B9o3EYTdM8lQWGlpWWFnVVBoT00/view?usp=sharing

    Albanians - black dots
    Greeks - black pluses
    Hungarians - black squares
    Macedonians - black Xs
    Montenegrins - black diamonds
    Romanians - black triangles

    ReplyDelete
  116. Thanks Dave! A great map! I see that some Greeks are quite close to Anatolia Chalcolithic. In general, Greeks look quite heterogeneous. Some Greeks are like Albanians and some close to Romanians, but there is not any Greek “plus” close to Montenegrin “diamond”. It is interesting that two Albanian dots are very close to the most Anatolian like Hungary Bronze Age! Hungarians are interesting, as compared to Bronze Age Hungarians, they are Yamnaya and Bell Beaker-shifted. Macedonians are quite heterogeneous. One is very close to Albanians, the second is very close to Montenegrins, the third close to Hungarians and the rest cluster with Romanians. In general, while Hungarians are Yamnaya and Bell Beaker-shifted, Montenegrins are Bell Beaker and Hungary Bronze Age-shifted. Romanians are also interesting as they are Armenia BA and Armenia Chalcolithic-shifted.

    On the basis of this PC map, I would not say that Greeks were replaced by South Slavs.

    ReplyDelete
  117. New Big Study by Reich, Haak, Lipson et al to come soon:

    http://biorxiv.org/content/early/2017/03/06/114488

    ReplyDelete
  118. @ Kristiina
    The Balkans is extremely complex, you can't analyse its history from a PCA designed largely to capture major gradients like EEF, WHG and ANE. And I can tell you that Montenegrins (& S in general) aren't BA Hungary shifted, for example. You're confusing convergent evolution with ancestry.

    And, nobody said Greeks were "replaced" by anyone - how could they be when they still exist ?
    The question is how much Slavic admixture they have
    AS a start, why don;t you analyse the Y haplogroups, if you're interested ?

    ReplyDelete
  119. We can measure ancestry in different ways, and PC map is one method and my analysis was based on it. I do not understand why we could not use it in this case. However, I would be very interested to see nMonte analysis etc of the relevant populations, so you are free to offer us more analyses of the Greek population with reference to Albanian, Sicilian, Romanian, Hungarian and Slavic populations!

    YDNAs are very prone to drift, and mtDNA is usually more local and very important for the autosomal makeup. Do you know what is the percentage of the Slavic I2 haplotype in Greeks or neighbouring populations?

    So, the paper itself proposes that the Slavic ancestry of Peloponnesean subpopulations ranges from 0.2 to 14.4%. You say that in Northern Greece it is in the order of 40-50%. According to Anthro Survey it has got to be higher than 20%. According to Razib Khan, order of 10-20% or something. On the basis of the PC map, I agree with Razib and think that 10-20% is more probable, in particular for the continental Greece. However, I am ready to change this conclusion if you provide me with more data.

    ReplyDelete
  120. @ Kristiina

    "We can measure ancestry in different ways, and PC map is one method and my analysis was based on it. I do not understand why we could not use it in this case. However, I would be very interested to see nMonte analysis etc of the relevant populations, so you are free to offer us more analyses of the Greek population with reference to Albanian, Sicilian, Romanian, Hungarian and Slavic populations! "

    PC map will give a good overlay, but it isn't particularly useful for the origins of post -3000 BC populations because there are too many permutations of possible origin.

    BUt here is some runs using 4000 BC source list. That is i have limited it to 4000 BC, becuase we have nothing more recent from Anatolia or Greece, so it ould skew analysis

    Ukrainian_West
    LBK_EN:I0056 35.55 %
    Latvia_HG:ZVEJ32 32.85 %
    Samara_Eneolithic:I0434 19.25 %
    Kotias:KK1 11.6 %
    Paniya 0.55 %

    Slovenian
    LBK_EN:I0056 33.9 %
    Latvia_HG:ZVEJ32 33.1 %
    Kotias:KK1 12.75 %
    Armenia_Chalcolithic:I1631 10.65 %
    Samara_Eneolithic:I0434 9.6 %
    Villabruna:I9030 0 %

    Serbian
    LBK_EN:I0056 43.75 %
    Latvia_HG:ZVEJ32 20.6 %
    Samara_Eneolithic:I0434 16.25 %
    Kotias:KK1 9.75 %
    Anatolia_Chalcolithic:I1584 8.95 %

    Bulgarian
    LBK_EN:I0056 43.4 %
    Latvia_HG:ZVEJ32 20.7 %
    Kotias:KK1 17.15 %
    Samara_Eneolithic:I0434 10.6 %
    Anatolia_Chalcolithic:I1584 6.4 %
    Paniya 1.15 %


    ReplyDelete
  121. Albanian
    LBK_EN:I0056 36.45 %
    Anatolia_Chalcolithic:I1584 18.5 %
    Latvia_HG:ZVEJ32 12.15 %
    Kotias:KK1 11 %
    Barcin_Neolithic:I1099 10.35 %
    Samara_Eneolithic:I0434 10 %
    Paniya 1.55 %

    Greek
    LBK_EN:I0056 48.3 %
    Anatolia_Chalcolithic:I1584 18.8 %
    Kotias:KK1 11.85 %
    Samara_Eneolithic:I0434 11.8 %
    Latvia_HG:ZVEJ32 6.65 %
    Barcin_Neolithic:I1099 2.2 %

    Cypriot
    Barcin_Neolithic:I1099 33.45 %
    Armenia_Chalcolithic:I1631 28.15 %
    Iran_Chalcolithic:I1665 20.9 %
    Jordan_EBA:I1706 11.6 %
    Greece_LN:Klei10 3.3 %


    Italian_WestSicilian
    Armenia_Chalcolithic:I1631 30.55 %
    Greece_LN:Klei10 24.95 %
    Barcin_Neolithic:I1099 17.7 %
    Jordan_EBA:I1706 14.3 %
    Ukraine_HG1:StPet2 4.95 %
    Latvia_HG:ZVEJ32 4.05 %


    As it can be seen these GReeks in Dave samples are very much part of continuum from Balkans to Ukraine.

    Cypriots are quite distinct; as are Sicilians

    But these results don't mean much ultimately. Where does this "LBK" ancestry actually come from ? From the original FEF ? I don't think so
    Where does the steppe (Samara / Kotais) come from ? The original Yamnaya ? Maybe a small amount only ? It'll make more sense we we start doing runs with Iron Age sources. Then it'll get exciting & revealing

    You can see the Mediterranean (Sicily, Cyprus) has its on thing going on



    ReplyDelete
  122. That Samara Eneolithic is interesting.

    ReplyDelete
  123. @ Kristiina

    "YDNAs are very prone to drift, and mtDNA is usually more local and very important for the autosomal makeup. Do you know what is the percentage of the Slavic I2 haplotype in Greeks or neighbouring populations?'

    Maybe if we are talking about Palaeolithic Siberia, Y DNA can drift. But how much drift do you expect from events which occurred 1000 years ago ?
    In northern GReece I2a1 + R1a = 50%

    "On the basis of the PC map, I agree with Razib and think that 10-20% is more probable, in particular for the continental Greece."

    I think that is a reasonable figure, nowadays, after significant attenutation of original Slavic ancestry (due to "homecoming" of Anatolian Greeks, etc). But the ultimate / original figure in the north of Greece is closer to 50%. And the interesting thing is - that is the very same figure for (Slav) Macedonians, Bulgarians, Serbs, Romanians.

    ReplyDelete
  124. A compelling post, but there's no concrete indication of any "Greek confirmation bias" since the authors of the genetic study only appear to be testing the validity of an old historiographical model involving medieval population dynamics (typical of anyone adhering to the basic tenets of the scientific method). So despite whatever shortcomings it may have, the peer-reviewed genetic study is generally sound (robust samples, decent methods of data-analysis, no conflicts of interest) since it confirms the historical fact that Slavic and/or Slavicized admixture in the majority Greek population was and still remains negligible. Moreover, the Stamatoyannopoulos et al. study complements, to an extent, the findings of a 2015 genetic study by Kushniarevich et al. involving virtually all Slavic and/or Slavicized populations in Europe (unless anyone would care to dismiss Kushniarevich and her team as having a "Slavic confirmation bias").

    Kushniarevich A, Utevska O, Chuhryaeva M, Agdzhoyan A, Dibirova K, Uktveryte I, et al. (2015) Genetic Heritage of the Balto-Slavic Speaking Populations: A Synthesis of Autosomal, Mitochondrial and Y-Chromosomal Data. PLoS ONE 10(9): e0135820, page 10: "Notably, the 'south-east European' component does not extend to the whole Balkan Peninsula, as South Slavs are differentiated from Greek sub-populations except Macedonian Greeks."

    There's also some interesting commentary in the link below.

    http://forwhattheywereweare.blogspot.com/2015/09/negligible-genetic-flow-in-slavic.html

    Cheers.

    P.S. Fallmerayer's theory was already refuted on historical grounds in the 19th century (see Papparigopoulos, Hopf, etc.) and again in the 20th century (see Vryonis, Constantelos, etc.) when Romilly Jenkins, despite his erudition, was foolishly beating what was already a very dead horse.

    ReplyDelete
  125. On Eupedia, Archaiocapilos posted the following figures:
    Northern Greeks (Thrace & Macedonia) (296 samples)
    I : 21.6
    R1a : 18.2
    R1b : 13.2
    E1b : 20.6
    G2 : 4.7
    J2 : 14.9
    J1 : 3.4
    LT : 2.7

    All Greeks (1038 samples)
    I : 15.1
    R1a : 12.0
    R1b : 16.9
    E1b : 21.0
    G2 : 6.3
    J2 : 20.1
    J1 : 4.3
    LT : 3.2

    So the percentage of I and R1a is c. 40% in northern Greece and c. 15% in all Greece.

    However, Sparkey on Eupedia noted that “haplogroup I on Crete is of a different subclade than in the rest of Greece (I2*-B like Armenia instead of I2a-Din like the Balkans)”. Archaiocapilos proposed that yDNA I (I-M170) would be distributed as follows:
    (739 samples):
    I1 : 3.7
    I2a1b-Din : 9.1
    I2*-B : 1.4
    I2a2 : 1.6

    ReplyDelete
  126. Correction: So the percentage of I and R1a is c. 40% in northern Greece and c. 27% in all Greece.

    ReplyDelete
  127. @EastPole

    From what I remember, Lausitz material indeed has been found in various parts of the Balkans and who knows, they might actually end up remarkably similar to contemporary West Slavs genetically, but I highly doubt they were "Slavic speakers". :) And there's no need to confuse cognates in Indo-European languages with one being the origin of the other.

    @Rob

    Where do you get the figure for "In northern GReece I2a1 + R1a = 50%" from? It's <40% from what I've seen based on the attempts of some people to calculate regional data. Not that I think we necessarily have great regional data, either way, so it might be higher or lower.

    @Kristiina

    Though the I isn't solely Din (as you mention) so not as relevant to this particular point. See the attempt of "rafc" in the same thread you posted, as well. ~33% for Northern Greece is the highest combined overall. His average for all is ~18%

    ReplyDelete
  128. @Alogo
    “they might actually end up remarkably similar to contemporary West Slavs genetically, but I highly doubt they were "Slavic speakers". And there's no need to confuse cognates in Indo-European languages with one being the origin of the other.”

    Why would you think that genetically Slavic people were not very likely Slavic speakers?

    Analogically, if some words found in the Balkans have cognates in Slavic and Indo-Iranian only then based on genetics with whom would you rather link the origin of these words?

    ReplyDelete
  129. My Y data from my publications on modern Greek DNA:
    Nea Nikomedeia--R1a (21.1) I (12.4---M423-8.8) in Macedonia
    Sesklo/Dimini--R1a (10.5) I (14.0---M423-7.0) in Thessaly
    Lerna/Franchthi Cave R1a (1.8) I (21.2---M423-12.3) in Peloponnese
    Crete---R1a (8.8) I (13.0----M423-1.6)
    Cyprus---R1a (2.4) I (3.5---M423-1.9)

    Note that all these studies excluded Albanian groups in Greece and 1920 exchange groups from Anatolia. I-M423 frequencies are dissociated from R1a frequencies. I-M423 is high in the Peloponnese while R1a is high near Nea Nikomedeia in the north.

    ReplyDelete
  130. Macedonian Greeks (n=114): 13% I2a1, 16% R1a1a = 29%
    Thessalians (n=57): 7% I2a1, 11% R1a1a = 18%
    Athenians (n=92): 6% I2a1, 16% R1a1a = 23%
    NE Peleponnesians (n=57): 12% I2a1, 2% R1a1a = 14%
    Ionians (Smyrna and Phokaia) (n=89): 4% I2a1, 6% R1a1a = 10%
    Cretans (general) (n=320): 2% I2a1, 8% R1a1a = 10%
    Lasithi Plateau Cretans (n=41): 5% I2a1, 20% R1a1a = 24%

    Greece FTDNA Project (n=280): ~14% I2a1, ~10% R1a1a = 24% - but most of this is just predicted from STR clusters as usual.

    From Underhill's sample most but not all of the Greek R1a1a falls under M458 and M558, which are typical of Slavic R1a, though both are much too old to tie specifically to the Slavic expansion.

    ReplyDelete
  131. @ Alogo

    I think I used the study by Semino which features Macedonian Greeks . IIRC the total sum was forty something percent; so similar to what Kristiina calculated.
    Of course I'm not suggesting it straightforwardly correlates with nowadays overall admixture

    @ Kristiina
    I1 is also found in Slavs though, although it is considered "Germanic".

    ReplyDelete
  132. @ Roy K

    "I would wager that Mycenaean Greece lacked I2a and much R1a. That leaves E-V13, various G lineages, and J2a/J2b lineages as representative of Neolithic and LBA Mycenaean Greece on the Y chromosome."

    I have the same feeling
    But surely there'll be some R1b-Z2103 also ?

    ReplyDelete
  133. @ Aigesti

    #1) Fair enough. As I said, I wish more Illyrian toponyms and inscriptions existed..

    # 2) Yes cremation never gained mainstream currency. We cannot say much more about any supposed Iron Age migrations, because all we have are 2 rather poor quality individuals.
    But as can be seen on the PCA, the LBA individual & the IA individual plot quite contrastingly. Not even the archaeological context was explained much in the Mathieson paper. So all we can say is some movements occurred from central Europe to W.B. during the Iron Age. Was this individual outlier , or part of a large population movement ? We don't know.
    If not Urnfield, archaeologists have long described Lausitz material appearing as far as northern Greece in the post-Bronze Age collapse Era (since the days of Huxley), but recent archaeologists were more and more digging into atmosphere of 'immobilism', until aDNA came along and created another revolution.

    ReplyDelete
  134. Davidski I have a question. What Haplogroups are Indo European? Only R1a and R1b? No other Haplogroup? thanks!

    ReplyDelete
  135. Some people say J is Indo European but I doubt it.

    ReplyDelete
  136. Rob

    ''But surely there'll be some R1b-Z2103 also ?''

    Off course there will be. And maybe a lot.
    Today R1b is relatively high in those countries that are not affected by Slavic expansion.
    Also if we move further into pre E V13 expansion we can expect even more R1b.
    Not surprising that 2 out of 3 samples from Vucedol are R1b.

    ReplyDelete
  137. @RK

    Yeah, I came across something perhaps related to this weirdness when looking at my own Y chromosome, G2a2b1 (G-M406). As expected it's mostly Italians, Greeks, (Armenian) Turks, Levantines... but also Kazakhs. According to Yfull TMRCA is 8600 years ago.

    Kazakhs also show up in the pretty rare G1 group, which again according to Yfull split from G2 ~26,700 years ago.

    Could be anything -- Neolithic contacts between steppe and farmers, or something in historical times. And I wouldn't discount the idea that there were early G2 people over a wide range of land around the Black and Caspian seas for a very long time.

    On a related note, but from other lines of reasoning, I believe there should be a less WHG-shifted but basically similar to other Anatolian Neolithics, G2a-heavy population around eastern Anatolia. Some aDNA from that region is most definitely in order.

    ReplyDelete
  138. @PF,

    According to Balanovsky et al 2015, Y haplogroup G1 is supposed to be a very recent founder effect among Kazhaks, which was originally brought into the region by migrations from Iran.

    I think the paper did not work work with ancient DNA though, so it can still do with confirmation from aDNA.

    ReplyDelete
  139. "I can't wait for more ancient DNA from Greece and Italy, especially from the Bronze and Iron Ages. Based on my experiences with many Greeks and Italians, it's sure to be a big eye opener for them, and a beautiful thing."
    Ah yes, spoken like a truly dispassionate and objective observer. No confirmation bias here folks! Nothing to see here! Move along, move along...

    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