Substack cometh, and lo it is good. (Pricing)

The “Yamnaya” were not the ancestors of the “Corded Ware” and “Bell Beakers”


In 2015 Massive migration from the steppe was a source for Indo-European languages in Europe and Population genomics of Bronze Age Eurasia were published. These two papers were game-changers. They established that the Yamnaya pastoralists of the Pontic steppe contributed a substantial proportion of ancestry to modern Europeans (later, the same was found to be the case in Indians and other Asians). As I’ve been reading and thinking about the expansion of Indo-Europeans ~3000 BC for my Substack series on the steppe, I have come to the conclusion that the populations sampled in these two papers were actually marginal to the primary expansion.

The first thing to note is that the Yamnaya samples were R1b, but of a haplogroup distinctive from that of the R1b common in Western Europe, and brought by the Bell Beakers. The Yamnaya R1b is the same as that in the Afanesievo culture of western Mongolia though. The early Corded Ware tended to be R1a. How to resolve this issue? On my podcast about Indo-Europeans with David Anthony, he posits that the elite during the early period was R1b, but that later on R1a (and Bell Beaker R1b) came to the fore due to social convulsions. Perhaps.

I think the other option, that there’s unsampled paternal diversity, is more plausible. I labeled where the 2015 Yamnaya were sampled from. It seems like they’re on the eastern end of the Yamnaya range. Anthony in The Horse, The Wheel, and Language, seems to lean toward the position that these eastern Yamnaya were culturally more significant than the less nomadic western Yamnaya. That’s fine, but I think it was the western Yamnaya that were the precursors to the Bell Beaker and Corded Ware. In my conversation with Nick Patterson he mentions that the Reich lab has detected Corded Ware who descend from the Yamnaya samples genealogically. How to square this with what I’m saying above?

The people of the Yamnaya horizon were patrilineal and exogamous. If Corded Ware men took Neolithic wives, they almost certainly took Yamnaya wives. The western and eastern Yamnaya may have had different paternal lineages, and even been different ethnolinguistically, but still shared similar gods and folkways so that intermarriage occurred. Their autosomal genome was very similar because exchanging wives across these patrilineal kindreds was common and prevent whole-genome distinctiveness from building up.

It needs to be noted that it turns out direct descendants of the Yamnaya R1b variant are present in Eastern Europe and the steppe to this day. A Russian group has found Yamnaya R1b in Crimean Tatars, and this lineage is also found in Chuvash. Basically, the eastern Yamnaya ancestry has been sloshing around the steppe for thousands of years. After 2000 BC they were absorbed into Indo-Iranians, but their far eastern outliers, the Afanesievo maintained some cultural continuity in the form of the Tocharian languages.

Finally, an issue in regards to time depth. The R1a division between Asian and European variants seems to date to 3500 BC. The paper above suggests that the division between Yamnaya R1b and Bell Beaker R1b dates to 4000 BC. The Yamnaya horizon people underwent a cultural revolution in the century or so prior to 3000 BC. But, they have differentiated already. On Clubhouse, I was talking to Jack V. of the Ancient Greece Decoded podcast, and he has a hard time believing that Indo-European diversified around 3000 BC. He says he can already read and understand Mycenanaen Greek from 1500 BC.

I put “Yamnaya”, “Corded Ware” and “Bell Beakers” in quotes. This isn’t what they called themselves. We’ll never know their ethnolinguistic divisions. They were probably part of a broad array of related peoples, more united by lifestyle and religion than language.

Note: the main issue I wonder about is the new samples that the Reich lab has and what they know.

28 thoughts on “The “Yamnaya” were not the ancestors of the “Corded Ware” and “Bell Beakers”

  1. Western yamnaya where descendants of Sredny Stog. They have R1a and even R1a-Z93. They also had some EEF, unlike eastern yamnaya. So western yamnaya where the real “yamnaya” so to speak. They seem to have originated around the lower Don/Azov sea.

  2. @Santosh, all Yamnaya published have some EEF/WHG ancestry compared to immediately preceding samples on the steppe (who have some variability in EHG/CHG proportions). Yamnaya have probably about 10-15% of this ancestry which increases to about 40% in later Steppe groups who seem to descend f back migration from Europe (Sintashta). It doesn’t seem to be too continuous admixture as the Poltavka and Catacomb who postdate Yamnaya on the steppe dont have more.

    There are actually no Sredny Stog samples published so far; the one published by Iain Mathieson’s paper was reclassified in their latest datasets to be a mislabelled / misdated sample from a Steppe_MLBA (Sintashta type) culture. (So all models which used it as ancestor of Yamnaya are unfortunately extremely wrong!).

    A presentation by David Anthony earlier this year showed that the Reich lab have a ton of Yamnaya samples and they’re all pretty similar, but with European farmer admixture increasing as they move deeper into Central Europe, like Yamnaya Hungary and Slovakia (except Don river samples with slightly more EHG/WHG it seems). Unfortunately the institute got YouTube to take the upload down as a copyright breach but a thread here on it- https://www.reddit.com/r/IndoEuropean/comments/mdzwir/yamnaya_genetics_societal_organization_david_w/. This includes some samples of “Sredny Stog” who were apparently similar to Yamnaya, but we don’t know when they dated from.

    I expect these are all Yamnaya R1b-Z2103, as Anthony seems like he’d have said during the presentation if they were not. It seems like the Yamnaya kurgan rite was only for R1b-Z2103s and to some extent their female relatives. I will upload a copy of the slide showing their PCA if that’s ok.

  3. It needs to mentioned that Corded Ware comes from people identical to Yamnaya. As identical as Bosnians & Croatians. Their similarity is striking.

    If Corded Ware doesn’t come from Yamnaya, it comes from Yamnaya’s brother. We know they come from Sredny Stog. But are they brothers, or father-son.

  4. @Santosh, PCAs from that presentation: https://imgur.com/a/MWX38hE

    The first two are taken directly and the next has some annotation where I tried to add some labels onto the PCA to correspond with the legend (commenter Forgetful on here had some suggestions on that as well).

    (One label I added was “Yamnaya Hungary cline connected to CWC” but just to qualify this was to comment that there seems some overlap between Yamnaya in Central Europe with EEF, and the CWC in Europe with EEF, not to say that one is ancestral with the other or not. And the more Yamnaya like character of the earliest Baltic Corded Ware samples can suggest this would happen independently)

  5. Razib, do you conclude that the dominant haplogroup on the Western horizon of the Yamnaya are Bell Beaker R1b, Baltic R1a and Indian R1a ?

    So the only legacy of the eastern Yamnaya R1b is on some extant populations in the same region as well as Afanasievo (Tocharians) ?

    And why do you say that after 2000 BC they were absorbed into Indo-Iranians ? Do we see minority Yamnaya R1b in Sintashta samples ?

  6. “As identical as Bosnians & Croatians. Their similarity is striking.”

    It is striking because they are identical. Sc. ‘Bosnians’ are Serbs converted to Islam and Croats are Serbs converted to Catholicism (plus few percentages of Avars), so, the comparison is pretty good, almost identical twins (like Schwarzenegger and DeVito :-).

  7. CLARIFYING TERMINOLOGY

    From the abstract of the linked March 2017 paper:

    “In view of the contribution of West Asians to the autosomal gene pool of the steppe Yamnaya archaeological culture, we sequenced a large portion of the Y-chromosome in haplogroup R1b samples from present-day East European steppe populations. The ancient Yamnaya samples are located on the “eastern” R-GG400 branch of haplogroup R1b-L23, showing that the paternal descendants of the Yamnaya still live in the Pontic steppe and that the ancient Yamnaya population was not an important source of paternal lineages in present-day West Europeans.”

    Apparently (I’m spelling this out to allow myself to be corrected if I’m misunderstanding it) the R-CG400 branch identified in the article more or less corresponds to R1b-Z2103 while Bell Beaker more or less corresponds to R1b-L51 (the main sister clade of R1b-Z2103 that the same level of the tree within R1b-L23), while Indo-Iranian corresponds to R1a-Z93, and Corded Ware to R1a-Z283 (these are two of the more notable sister clades within R1a-M417).

    Also, it is worth noting that it seems likely that Bell Beaker Indo-Europeans adopted the religious/cultural Bell Beaker culture that originated in Iberia, seems to be derived from the megalithic culture that preceded it in the region, and probably preceded the arrival of Indo-Europeans in Western Europe, as early Iberian Bell Beaker ancient DNA lacks steppe ancestry and Y-DNA R1b-2103. There is a plausible case to be made that the Bell Beaker culture was adopted by male dominated Indo-European migrants to the region as a result of the cultural impact of their widespread marriages to local wives.

    THE YAMNAYA DIES AND FLED EAST, RATHER THAN FLEEING WEST

    I wrote a blog post once entitled: “Did the Yamnaya die or run?” suggesting that the Bell Beaker people were derived from the mass emigration of R-1b Yamnaya people in favor of proto-Indo-Iranian R-1a successors. It looks like this conjecture was mostly wrong.

    Instead, it seems like there was probably mostly a genocidal slaughter of Yamnaya men by their war-like proto-Indo-Iranian successors (who were mostly Y-DNA R1a-Z93).

    What Yamnaya mass migration/flight did happen seems to have been a Tocharian flight to the east (implying a somewhat different route to the Tarim basin than is conventional for the source Tocharian population) along with some absorption of the Yamnaya men (perhaps young boys were were spared slaughter and integrated into the new ruling class) into the society of the conquering Indo-Iranians.

    IMPLICATIONS OF A LATE NEOLITHIC-ENEOLITHIC INDO-EUROPEAN POPULATION STRUCTURE MODEL

    To the extent that Europeans are derived from populations further north and west that were less nomadic than the eastern Yamnaya, that helps explain why narratives beginning from the assumption that the main wave of Indo-Europeans in Europe were nomadic herders don’t seem to work very well.

    The model makes testable predictions about loan word patterns we would expect to see.

    For example, we’d expect Uralic loan words in Corded Ware derived populations, but not so much in Indo-Iranian or Bell Beaker Indo-European derived languages, as the loan word borrowing would probably occur only once proto-Indo-European had developed its pre-expansion population structure.

    Following the genetic evidence to conclude that the Indo-European language family probably had significant structure even before expansion out of the steppe is likewise a helpful observation in understanding later cultural dynamics.

    A probable linguistic divide between the Bell Beaker Indo-Europeans and Corded Ware people within a larger Indo-European language family would help explain why there was a standoff between cultures derived from these culture for a thousand years or so in Europe, rather than an easy blending of the two.

    And, don’t forget the tyranny of small differences. Even if they shared a basic religious context, surely the different populations disagreed on some religious doctrines that seemed important to them as well, much like Catholic-Protestant and Sunni-Shia divides that wars have been fought over.

    For example, Corded Ware derived pagans and the Greeks appear to have had one fairly analogous highest god (Odin/Zeus), but in Celtic mythology plausibly associated with Bell Beaker people, Lugh, a god that the Romans saw as a counterpart to Apollo (which was a secondary god in Greco-Roman mythology), was the highest of the gods. Disputes of which polytheistic God is primal are just the sort of things that one could imagine these societies fighting over and being divided by.

    The notion that Tocharian, a top level division of the Indo-European languages, might have been associated with a fourth branch of top level Y-DNA structure also makes sense.

    GREEK LINGUISTIC TIME DEPTH

    “The R1a division between Asian and European variants seems to date to 3500 BC. The paper above suggests that the division between Yamnaya R1b and Bell Beaker R1b dates to 4000 BC. The Yamnaya horizon people underwent a cultural revolution in the century or so prior to 3000 BC. But, they have differentiated already. On Clubhouse, I was talking to Jack V. of the Ancient Greece Decoded podcast, and he has a hard time believing that Indo-European diversified around 3000 BC. He says he can already read and understand Mycenanaen Greek from 1500 BC.

    I think that Jack V. is wrong when it comes to time depth. The evidence that language diversification tends to be punctuated rather than primarily involving gradual random evolutionary drift, and that it is heavily driven by language contact (especially substrate influences) would make it entirely plausible that Mycenanaen Greek grew distinctive from a Southwestern Steppe component primarily in the time period from about 2300 BCE to 1500 BCE when Indo-Europeans start to expand into the Balkans and Greece triggered by the 4.2 kya climate event, after having been more or less static from ca. 3500 BCE ± 500 years, until around 2300 BCE. This is entirely consistent with heavily punctuated and language contact driven theory of language evolution that a lot of fairly recent linguistic evidence supports.

    EARLY GENETIC STRUCTURE AND ANATOLIAN LANGUAGE TIME DEPTH

    Finally, it is worth noting the dog that didn’t bark. There are fairly basal Y-DNA divisions in the Y-DNA R1a and R1b phylogenetic trees corresponding to Corded Ware, to Bell Beaker, to Indo-Iranian, and to Tocharian, each associated Indo-European populations that emerged 500-1500 years before the main waves of Indo-European expansion to Europe in the West (in two distinct branches), to South and West Asia, and to the East. Across all of these expanding populations patrilineal extended family/clan/tribal organization and exogamy is a highly consistent norm.

    What there isn’t, is a distinctively Anatolian branch of Y-DNA R1a or R1b of comparable or greater time depth. As one paper notes: “among ancient YNDA data gained from 26 samples of the Anatolian Neolithic, R1a (as well as R1b) is entirely absent, and it is also absent from Neolithic cultures of the Balkans, such as the Stračevo, Vinča, and Lengyel cultures.” Skourtanioti (2020) likewise notes that “there is no direct evidence for an early incursion from the Pontic steppe during the main era of Arslantepe” (an ancient DNA site from South Central Turkey with ancient DNA from the Early Bronze Age where Anatolian languages were spoken at least from Hittite conquest in 1350 BCE to the 600s BCE), despite the presence of one man with Y-DNA from sister clade to early Neolithic R1b-V88 found mostly in Africans. The major basal split from R1b or R1a you would expect in a model with a very early Indo-European linguistic/genetic structure in which Anatolian language speakers break off long before the main Anatolian expansion isn’t there.

    No ancient DNA evidence (and no written records in a place near the cultures with written languages in Mesopotamia, the Levant and Egypt, and no iconic burial practices or human made goods) point to significant levels of Indo-European associated clades of Y-DNA R1a or R1b, or to significant autosomal steppe ancestry, or to Indo-European religious, cultural or linguistic influences, in Anatolia significantly prior to the 4.2 kya climate event.

    In modern Turkey, Y-DNA R1b is from the R1b-M269 clade (about two-thirds of R1 in Turkey; more detailed subtyping is hard to find But not R1b-U106) associated with Bell Beaker derived people, and Y-DNA R1a (about one-third of R1 in Turkey) with a predominantly Slavic subclade of the Corded Ware associated clades of Y-DNA R1a appearing to be most numerous (although again detailed subtyping is hard to find). There is no evidence of a significant proportion of Y-DNA in modern Anatolia that doesn’t fit in one of the two main European Y-DNA R1 clades.

  8. “Y-DNA R1a (about one-third of R1 in Turkey) with a predominantly Slavic subclade of the Corded Ware associated clades of Y-DNA R1a appearing to be most numerous (although again detailed subtyping is hard to find).”

    Europedia says lots of R1a in Turkey is R1a-Z93 which is broadly distributed but most intense in Southeast Turkey, with another clade mostly concentrated in NE Turkey with much narrower distribution (probably the Slavic one). This would suggest Indo-Iranian sources for R1a-Z93 perhaps also from Turks admixing with Indo-Iranians before admixing with Anatolians (i.e. a comparatively recent origin for R1a in Turkey not associated with the Anatolian languages), and a fairly recent origin of the Slavic clades as well (perhaps with Anatolian Greeks who had Slavic admixture in Medieval times).

    Turkish R1b is not Yamnaya R1b which is notable since that would be the most plausible old R1b to find in Anatolia since the Yamnaya were adjacent to Anatolia in the time frame of old Anatolian divergence linguistic theories.

  9. @MilanTodorovic

    There is recent direct evidence on Bosnian v. Serbian genetics.

    Zsolt Bánfai1 (2019) looks at genetic evidence of Muslim Ottoman occupation in the Balkans. The big picture story is that the paper finds it (as well as Roma introgression).

    In particular, as related to the comments above, it finds that while Bosnians (Muslim) and Serbs (Orthodox Christian) are similar and adjacent, with some overlap on a PCA plot of Principal Components 3 and 4 (of autosomal genetics), the populations as a whole are distinguishable and distinct on a PCA plot of Principal Components 1 and 2. (The Ancestry plot also shows that many individuals in both groups also have non-trivial Roma ancestry.)

    The two populations have common origins, a lot in common, and the same primary source population. But they aren’t actually identical and indistinguishable genetically, although the distinctions are subtle.

  10. Considering the distances between river basins, the inhospitable nature of the sea of grass without horses and wagons, and the general linguistic patterns of hunter-gatherers (cf. native Californian languages), I think it’s safe to assume that each river on the pre-Neolithic Pontic-Caspian steppe had its own unique and deeply diverged language (if not several). As far as I know, the local Neolithic was more a matter of adaptation than population replacement, so it shouldn’t have significantly affected the ethnolinguistic makeup of the region.

    So… if there was ethnolinguistic diversity within the early Yamnaya horizon, it could have included completely different language families. Not just internal Indo-European splits. Theories that Basque or Uralic come from the steppe aren’t so crazy (though we have a far better candidate for Uralic in Iron Age N). You could even throw in Tyrsenian languages as a candidate.

    But it’s also plausible that the expanding Yamnaya cultural package went along with linguistic replacement on the steppe – in which case, deep divergence within Indo-European could be the result of contact-induced language change. Who knows? Maybe satemization is the result of Dnieper peoples imperfectly learning a Volga language.

  11. @ohwilleke

    I believe most R1b in Turkey – well, certainly among Western Armenians at least – is R-Z2103 in its different varieties. I’ve also seen it proposed that even though it’s highest frequency in the Balkans (particularly among Albanians) these days, R-PF7562 might be Anatolian-affiliated.

    Phylogenetically, it’s a perfect match – parting ways from R-L23 ca. 6400 ybp, according to YFull, essentially just a few hundred years before the R-L51–R-Z2103 split.

    Also leaves open the possibility of an early Balkans -> Anatolia route, which I’ve heard suggested before (and which IIRC, there’s some autosomal aDNA evidence for).

  12. @ohwilleke

    Europedia says lots of R1a in Turkey is R1a-Z93 which is broadly distributed but most intense in Southeast Turkey, with another clade mostly concentrated in NE Turkey with much narrower distribution (probably the Slavic one). This would suggest Indo-Iranian sources for R1a-Z93 perhaps also from Turks admixing with Indo-Iranians before admixing with Anatolians (i.e. a comparatively recent origin for R1a in Turkey not associated with the Anatolian languages), and a fairly recent origin of the Slavic clades as well (perhaps with Anatolian Greeks who had Slavic admixture in Medieval times).

    The R1a-Z93 concentration in Southeast Turkey is due to Kurds. As you know, Southeast Turkey is overwhelmingly inhabited by Kurds. The Cinnioglu et al. paper, from which our Y-DNA data on Turkey come from, does not have an ethnicity-based sampling, so Kurds are included too. The Slavic R1a clades in Turkey mostly come from Balkan Turkey or recent Balkan immigrants (Balkan Turks, Bosniaks, Albanians, Pomaks etc.).

  13. @ben-canaan

    I believe most R1b in Turkey – well, certainly among Western Armenians at least – is R-Z2103 in its different varieties. I’ve also seen it proposed that even though it’s highest frequency in the Balkans (particularly among Albanians) these days, R-PF7562 might be Anatolian-affiliated.

    Yes, most of R1b in Turkey is R1b-Z2103. In fact, most of R1b in the Northern West Asia and large parts of the Balkans and Caucasus is R1b-Z2103 and R1b-Z2103 peaks among Armenians and Assyrians in those lands.

  14. @oh
    What does it mean identical? Even in today’s Serbia, people are not absolutely identical, every 100 km is slightly different language, folklore, history, physiognomy (Dinaric vs Pannonians), etc. Bosnia was a Serbian land since the beginning of time and did not exist as a separate state until communists split Serbian corpus in several republics for its easier control. People in Bosnia (Christians and Muslims) live in mixed villages since Turkish occupation when some under pressure converted to Islam. Very often, one brother converted to Islam and the other remained Christian to help each other in uncertain times. So, as there is also some ‘non-trivial Roma ancestry’ it could mean that OIT proponents, with a slight delay, were right (-: ?

  15. About 150-250 K (according to various sources) Turkish warriors came to the East Roman Empire where 6 million people lived. It means that many today’s Turks are converted Greeks but also Serbs, so because Turks looks more like as ‘European’ than as Kazakhs. Brigians (Phrygians) were Serbs, so as Trojans and many other. They were sc. ‘Hellenised’ (although this is a Serbian word so as the word Greek) in the 4th c.BC. Later, during Turkish occupation, many Serbian boys were abducted from their families (sc. ‘tax in blood’) and sent to Turkey to be trained to become Janissary, the elite Turkish soldiers based on their physic and genetics (several of them became Ottoman viziers, i.e. prime ministers). Osman I, (born c. 1258—died 1324), ruler of a Turkmen principality in nw Anatolia who is regarded as the founder of the Ottoman Turkish state was R1a. This dynasty lasted until the 20th century.

  16. @ oh
    “For example, Corded Ware derived pagans and the Greeks appear to have had one fairly analogous highest god (Odin/Zeus), but in Celtic mythology plausibly associated with Bell Beaker people, Lugh, a god that the Romans saw as a counterpart to Apollo (which was a secondary god in Greco-Roman mythology), was the highest of the gods. Disputes of which polytheistic God is primal are just the sort of things that one could imagine these societies fighting over and being divided by.”

    All above mentioned gods originated in Vinca. Zeus has nothing to do with Greeks. He appears in Iliad, that is almost 1000 years before Greeks reached Mt Olympus. Instant Greek and later Roman mythology are replicas of ancient Serbian mythology which evolved over thousands of years. For e.g. Serbian goddess of love – Priya, was transferred by Aryans to India where is now frequent female name. Its Greek replica is Aphrodite and Romans’ – Venus. Lugh is also a Serbian word. Above mentioned megaliths were built by I2 people.

  17. @ohwilleke

    Lugh & Odin/Wotan are identified with Mercury (Wednesday, Miercoles). Thor is identified with Zeus and Jupiter (Thursday, Jueves).

    Otherwise it’s certainly true that pagans can have major religious disputes. Didn’t the last of the Babylonian kings get into hot water for worshipping the moon god Sin instead of Marduk? Capturing the enemy’s cultic statues also seems to have been a major coup in the wars of Ancient Mesopotamia and Levant.

  18. “He says he can already read and understand Mycenaean Greek from 1500 BC.”

    Languages can change a lot in 1500 years. 1500 years ago nothing resembling any form of English existed. 1,000 years ago Old English existed but it is incomprehensible to modern English speakers. 500 years ago was Middle English, still pretty much like a foreign language. Not until the end of the 16th Century 400 years ago do we get to an language that modern readers over the age of 30 can read. (under the age of 30 they are massively ignorant, illiterate, and proud of it).

    One of the key factors in driving the evolution of English was a series of invasions and cultural collisions. The creation of old English resulted from the mixing of Scandinavians, Germanic peoples, and indigenous Celts. Middle English arose from the Norman conquest. But, we seem to be looking at similar events here.

    Isolated languages may evolve slowly, but languages evolve quickly when events mix disparate peoples.

  19. ‘Mycenean (so as Minoan) Greek’ is an oxymoron. The term ‘Greek’ was first used by Aristotle but widely is used after Roman conquering in the 2nd c.BC. Neither ‘Greek’ nor ‘Hellenic’ are of Greek origin. Sc. ‘ancient (future) Greeks’ came from Middle East and Egypt and pretty much accepted the language of indigenous people who lived in today’s Greece. There are so many falsifications related to ‘ancient Greeks’ including the year of the first sc. Olympic games in 776 BC, when they from nowhere officially entered the history, what is an absolute joke, because Greeks in the following several hundreds of years haven’t seen Mt. Olympus, which was within ancient Macedonia. This stretching of Greek history back in time and its appropriation by the West was done with an intention to present the western history much older than it really was. Languages change over time, but there are still many identical or highly similar words, for e.g. in Sanskrit and modern Serbian in spite of 4000 years distance, including complex family relationships (e.g. one word for husband’s brother’s wife, husband’s mother, etc).

  20. Guess my current notional model for all this is:

    1. Around 3500-3300BCE, movements of “Steppe Maykop” or other wagon users who are traders from the south spread wagons into steppe. (DA recently said in Q&A on presentation that around 3500-3300BCE is time of a mobility surge where people with unusual Maykop DNA on Western Steppe and other outliers showing up in different places, consistent with sudden mobility increase). Then post Sredny-Stog, Repin and Khvalynsk “Steppe” Culture (in unknown mixture) who are the numerous early proto-pastoral groups on the steppe get a hold of it, and they use this with the horse to create much more mobile, productive economies which can use more steppe land.

    2. … Which are also dominated by small numbers of male founders who are successful in realising or capturing the potential of the new economy and the ability that controlling more wealth and territory gives you to boost lineages.

    This is strong patrilocality that is not the case as far as I know in hunter-gatherers; for them it’s perhaps more about hunting skill and they can’t “afford” to marry daughters off preferentially to clan members, if it endangers the survival of their band and engenders some level of inbreeding and such. I think strong patrilocality is a thing which early agriculturalists have more incentive and ability to do, even if HGs still pretty “patriarchal” otherwise. Strong patrilocality seems common across Copper Age Europe to varying degrees.

    3. … But on the steppe, some clans more successful than others. The R1b-z2103 clearly seems to dominate the kurgan rite, and is probably the first mover in the pastoral-wagon economy, and spreads far to the east in the form of Afanasievo.

    4. Nascent R1a-z93 and R1b-l23 clans who did not practice the kurgan rite (or at least not to the visible degree of “Yamnaya”) are then driven to emigrate to the west from the homeland by this expansion of the “Yamnaya” clan and by competition… (They may have been helped in Europe by disease or military advantage from social structure or from the horse, but it is hard to know?)

    5. In the end though though this is to their advantage; the Yamnaya (by which I include post Yamnaya cultures like Catacomb) are eventually hit hard by the aridification of the steppe, then rather than migrate to west (perhaps because it’s “full”) tend to go more directly south, where they’re absorbed by large populations.

    (This may explain the finding Onur indicates where R1b-z2103 relatively common in West Asia, in fact maybe more common in Western Iran than R1a?).

    6. In time, the regions where the early CWC / pre-Beaker clans migrated to eventually experienced much more population growth, being much further from carrying capacity frontier. It may be that even in 2000BCE, there were still more R1b-Z2103s about (though this is a very low confidence hot take!) but low population Northern Europe, and North Central Asia and India were all to boom later in history. Particularly as Indo-European cultures fused benefits of mobile pastoral and sedentary economies and switched around according to what worked best.

    One thing in defense of this model is that the latest Yamnaya-like samples we have, around the time of the aridification event, are from a culture called Kubano-Tersk, are down south by the Caucasus and show genetic signs of beginning to merge in with Caucasus cultures. Then after which or around the same time the steppe proper first sees these highly heterogenous “Potapovka” samples (which include some Yamnaya like people and highly WSHG like people, such as a young male who was the first high status light chariot driver we have DNA from I think; Reich lab divide Potapovka into outliers and main, but think from memory there are about as many or more outliers than main, published!). Then proper Sintashta people.

    Time will tell how this bears out…

  21. Re my last comment, Anthony’s Q&A with Kristiansen and others is still up on YT: https://m.youtube.com/watch?v=1O1zDrW7SvE

    Comments on mobility and Maykop:

    0:33 – direct relationship between a Steppe Maykop person and Usatovo person at other end of Black Sea.
    2:30 – does believe Yamnaya got wagons from Maykop
    3:00 – “movement of Maykop people up to the Lower Don River – that far – in the steppes” (the Steppe Maykop from Wang’s paper published 2019 are from around the Manych River which is a southern tributary of the Don, so maybe this is them or not).
    6:45 – “More mobility in that period, right when you get to 3300-3500 BC, all of a sudden, mobility goes (up) through the roof, people you don’t expect to be showing up in a particular region are there … There’s actually a fair amount of Maykop ancestry in some Usatovo individuals, who are at the other end of the Black Sea. So I think there’s maritime trade going on around the Black Sea at 3500 BCE, and possibly in response to the Uruk expansion, the expansion of trade from Mesopotamia”.

  22. @if

    You don’t have the knowledge, but your hubris stops you from asking the questions. You are not the first who tries to be provocative. However, none found one mistake or the lack of logic. Also, none could challenge the naïve story about sc. ‘Indo-European’ language, allegedly spoken by nomads on a twice distance of LA- NY, formed without Proto-phase period and without any civilizational discipline vocabulary. And this language without any sophisticated word influenced all modern Euro and many Indic languages, maybe even created Rg Veda (you may know the meaning of these words)?

    So, considering that you are interested in Christianity (and its relationship with Serbs), there are, in a nutshell, things that are maybe interesting to you. Please, announce loudly here if you find one incorrectness. The first Christian diocese was established in today’s Serbia, in 29 AC in a Roman capital Sirmium, 50 km from Belgrade (and 60 km from Vinca). St Peter first visited Serbia before he went to Rome. St Paul was hiding from Nero and baptizing people in Serbian caves (river Trebisnjica). Later, both finished hanged in Nero’s garden. The following Roman-Serbian Emperors are relevant to the Christianity – Diocletian, who persecuted Christians, Constantine who won the war against his co-Emperor Licinius and legalised Christianity, Julian and Jovian, one tried to return to paganisms the other restored Christianity, Justin and Justinian who built Sveta (Hagia) Sophia, etc. Hope, it is enough for you to start your study and pls, if you come with ‘rg’ and ‘veda’ (you may try ‘Indo-European’ language?) that would be a globally significant contribution. Enjoy your journey and stay cool.

  23. I read the text: patrickwyman – indo-europeans-and-the-yamnaya-culture. I could not find the classification, but it seems that it is classified: NPG-15 (under 15, no parent guidance necessary).

  24. @Walter, Wyman’s take is pretty good… Obviously it comes down in favour of Yamnaya culture as mediator of steppe ancestry which is a bit doubtful or a simplification as we discuss here, but otherwise pretty OK as an entry I think…

    On a wider thing, as he notes, some things aren’t totally straightforward in the archaeological trail of steppe ancestry and kurgan burial practices in Europe though. Like sometimes pre-steppe ancestry people were using burial mounds (https://www.persee.fr/doc/mom_2259-4884_2012_act_58_1_3455 – Baden) while the Corded Ware are most often associated with flat graves. It’s more of a suite of characteristics that revealed migration and the adna is important in confirming things. (I do think this is something where Wyman sort of implies that kurgan rites always spread out from the steppe and was emulation outside that region, but it was possibly more of a back and forth dynamic where the earliest kurgans may have spread into steppe from the West during early Chalcolithic, and then later elaborations from Maykop culture in Caucasus hills also influenced Yamnaya).

    In the case of Corded Ware, I’m not actually sure that we’d know without using adna and skull measuring phys anthropology, just from the material culture and burials! Comparing them to the preceding Globular Amphora/TRB and you’re matching similar pottery, stone battle-axes that are not too particularly like anything from the steppe but are like TRB caches found in or near their settlements, flat graves with single inhumation burial like Globular Amphora. No red ochre, no kurgans and few small tumuli, no wagon burials. It’s not vastly a very close match to the Yamnaya steppe rite, over preceding North-Central European cultures? When it was in vogue for Western archaeologists outside the Soviet Union to reject phys anth (“because Boas has proven it was bunk and the Nazis were nuts” or whatever reasons they had 😉 ), I’m not sure it was then irrational “anti-migrationism” for archaeologists to come down on the side of local emergence… If you ignore the skull shapes and you have no adna, you probably are gonna go “This looks a lot like GAC/TRB in material culture, and my strontium isotopes tell me they’re not first generation migrants, so…”.

  25. Harvardians have failed to prove male genetic continuity among Khvalynsk-Eneolithic Progress (R1b-V1636, Q, R1a), Maykop, Yamnaya (R1b-Z2103-I2a-L699), CWC (overwhelmingly R1a-M417) and BBC (overwhelmingly R1b-P312). Reich, Anthony and colleagues have dozens of samples to publish and will be able to demonstrate the genetic relationship between Yamnaya and CWC, but do not forget the widespread practice of exogamy and female migrations. In Sredni Stog there is no R1b-L23>L51>P312 but Z2013. This lineage reached the Balkans (Vucedol, Hungary, Poland BBs, Mokrin and even Italy) but never reached Western Europe (neither did R1a-M417). The conclusion is obvious, Yamnaya has been debunked as the origin of R1b-L51 and this lineage for the moment cannot be linked to the expansion of IE languages in western Europe, ergo the BBC never spoke an IE language. Haak and colleagues have told us a fairy tale without having checked the R1b subclades typical of the steppes. When Villabruna appeared they were surprised and the nail in the coffin of their theory was the appearance of three Swiss Neolithic farmers buried in dolmens (2,750 BC) with no archaeological, anthropological or even genetic relationship (in two of the cases) with Yamnaya or the CWC. In some North-American blogs I have read surprising conclusions about how we Basques R1b-P312 (87%) have our paternal origin in the steppes, but we lost our IE language because surprisingly south of the Pyrenees we decided to adopt the language of our women (Basque).Harvard needs much more convincing evidence for his theory to be accepted by the European scientific community.

Comments are closed.