search this blog

Saturday, November 17, 2018

What happened to Maykop?


The Maykop culture was probably the result of population movements from Transcaucasia and beyond into the Northwest Caucasus during the Eneolithic and Early Bronze Age. Its peak lasted for roughly 700 years, from about 3700 BC to 3000 BC, after which it seems to have vanished suddenly. Why? Are there any decent papers on the topic?

The currently rather popular idea that Maykop gave rise to the Yamnaya culture is likely false. It was probably somehow involved in the rise of the contemporaneous Steppe Maykop culture in the steppes abutting the North Caucasus. But, thanks to ancient DNA, we now know that the people associated with this culture were distinct from those associated with Yamnaya.

In fact, when Steppe Maykop disappeared, Yamnaya spread into much of its former territory, and this turnover registers clearly in the time transect of ancient genomic data from the North Caucasus steppes (see here).

My view is that Maykop was generally an alien entity to the indigenous peoples of the steppes. These natives may have emulated it in some ways, but there's no need, I'd say, to go as far as to assume that Maykop was the vector for the spread of Indo-European languages into the Pontic-Caspian steppe.

Indeed, it seems to me that when the technological and economic advantages of Maykop over the steppe peoples eventually eroded, it couldn't hold its ground on the edge of a vastly different and perhaps largely hostile world, and quickly disappeared.

Here's a quote from a recent paper by Trifonov et al. on Maykop jewellery that I found very enlightening in regards to these issues (emphasis is mine):

These deep-rooted Near East traditions of ritualization of the production and use of jewellery pieces made of gold, silver and gemstones in the Maykop culture, on the one hand, maintained familiar canons of ritual behaviour and, on the other, made perception of sophisticated symbolism of gemstones more difficult for neighbouring cultures with different living standards, levels of social development and value systems to understand. The jewellery traditions of the Maykop culture had no successors in the Caucasus or the adjacent steppes. In the third millennium BC , the goldsmiths of Europe and Asia had to reinvent the technique of making thin-walled jointless gold beads from scratch (Born et al. 2009).

I do wonder, in fact, if the language spoken by the Maykop people was even part of a still existing language group, let alone if it belonged to the Indo-European language family.

See also...

Big deal of 2018: Yamnaya not related to Maykop

138 comments:

Davidski said...

Another pertinent question: What happened to the Wang et al. paper on Maykop, eh?

https://eurogenes.blogspot.com/2018/05/on-genetic-prehistory-of-greater.html

Davidski said...

And the data?

Anonymous said...

Could it not be the case that this hypothetical Southern source of Yamnaya’s extra CHG was simply Eastern Caucasian rather than Western Caucasian? Given the terrain, you could easily have very different groups living in a (relatively) compact area, as today. At the moment, I’m leaning towards Leyla-Tepe as being this source, but regardless given L23 and Z2103 both look like broadly West Asian lineages in origin based on the phylogeny (as well as tonnes of other things, not least swastikas in the early Chalcolithic Near East - just google “Samarra bowl”), I think the paternal ancestors of those found in the Yamnaya kurgans would have come ultimately from the South.

Davidski said...

Nope, the CHG in Yamnaya is from people who lived in the steppes already before the Eneolithic. And so is the L23/Z2103, because its missing in all ancient West Asian samples except those with steppe admixture.

The Northeastern Caucasus was home to a population like Maykop. We have Kura-Araxes samples from Velikent, Dagestan, and they couldn't have been ancestral to Yamnaya.

Refer to the second map here...

Genetic borders are usually linguistic borders too

Bob Floy said...

What I really don't understand is this:

How can Maykop have "given rise" to Yamnaya, when we know that Yamnaya's forerunners(Khvalynsk, Sredny Stog, etc.) already had CHG ancestry well before Maykop existed? Isn't that alone enough to kill the idea? Or am I missing something?

Anonymous said...

Well the phylogeny in my opinion definitely points to a West Asian origin for both L23 and Z2103, but just to make it simpler (not that you can’t understand it of course, your knowledge eclipses mine) how can you explain the rather large presence of Z2103 in Iraq if it is of Steppe origin (using this map: https://cache.eupedia.com/images/content/Haplogroup-R1b-Z2103.png)?

Also, what about those bowls? How did supposedly completely unrelated to IE Middle Eastern folk pick up that symbol, because it surely isn’t coincidental...

Anonymous said...

Yamnaya has more CHG than its predecessors if I’m not mistaken.

Here’s an image of bowls from the Samarra (with two Rs, not one like the Russian culture) by the way:

https://www.eixdelmon.com/wp-content/uploads/2016/06/4vqvJk7.jpg

Davidski said...

@Bob Floy

I had some discussions on this topic with a couple of the scientists involved in this work, and my (hopefully not misguided) impression at the time was that they basically believed in the scenario that was illustrated in the qpAdm models schematic in the Narasimhan et al. 2018 preprint.

https://www.biorxiv.org/content/early/2018/03/31/292581

In other words, it seems like they thought that all CHG populations went extinct, and that the southern ancestry in Khvalynsk was actually ancestry from Neolithic Iran, and also that Yamnaya had additional ancestry from later migration waves from West Asia, like from Chalcolithic Iran (ie. Hajji_Firuz_ChL).

But I'm guessing that this was before the Steppe Eneolithic and also Caucasus Eneolithic samples were sequenced, because they make all of this very unlikely.

So now, as a result, those two Steppe Maykop outliers and the Yamnaya Ozera Ukraine outlier have been pegged as the likely vectors of the spread of Indo-European languages into the steppes, as per the Wang et al. conclusion, because that's really all that's left.

Davidski said...

@An Idiot

Yamnaya has more CHG than its predecessors if I’m not mistaken.

You're mistaken. Yamnaya has less CHG than the much earlier Steppe Eneolithic population from the North Caucasus steppes, and probably less than other, as yet unsampled, Steppe Eneolithic populations.

And the ancient phylogeny of R1b definitely points to an origin of L23 in Eastern Europe. Just have a look at all the Mesolithic and Neolithic R1b and R1 from Eastern Europe, and none at all from the Near East.

Don't know about the bowls. But things like that can be traded and imported from thousands of miles away.

Bob Floy said...

@Davidski
Seems very desperate and contrived to me.

@An Idiot
Yamnaya did have more CHG related ancestry than it's predecessors, but that was the product of a gradual increase which had begun during the Eneolithic. The CHG ancestry had been coming into the EHG community on the steppe long before Maykop existed, so the idea of Maykop somehow being the origin of Yamnaya is very messy, at best.

Anonymous said...

Oh and just as a note - I think these Middle Eastern L23 guys would have originally come from Eastern Europe/the Balkans (likely M269 homeland), and expanded from Vinca to Ubaid (Halaf and Samarra on the way) along with the spread of metallurgy, the swastika, distinctive figurines and pottery, proto-writing etc. These guys would be a metallurgical (and pastoral) caste elite amongst an otherwise farmer population (though of course not responsible for all innovations), and would also have brought the framework for what would become Anatolian and Euphratic (a pre-Sumerian Lower Mesopotamian language theorised to be of Samarra origin).

That’s just my (well, originally Tomenable’s) theory though, and mostly speculative, but it answers a decent number of problems I can come up with.

Bob Floy said...

@Davidski

"Yamnaya has less CHG than the much earlier Steppe Eneolithic population from the North Caucasus steppes, and probably less than other, as yet unsampled, Steppe Eneolithic populations."

Really? Not sure why now, but I had thought that Yamnaya and Repin had more CHG than Khvalynsk.

Davidski said...

@Bob Floy & An Idiot

The Khvalynsk samples aren't the only currently available Eneolithic samples from the steppes.

Have a look at the ADMIXTURE bar graph here...

Big deal of 2018: Yamnaya not related to Maykop

Anonymous said...

As you know L23 is far more recent than R1b and R1, and for example the “earliest” branches of Z2103 (those that split from the rest of the pack at an earlier stage) are all distributed around Armenia, which is totally inconsistent with a Steppe origin. L23* also seems to be distributed similarly.

Also, I’m sure you realise that it is unlikely that the bowl was traded with the Steppe folk instead of produced locally, not least the many many different varieties. Let me know what you think about the possibility of my reply to my own comment about M269->L23, it’s my pet theory at least and I think there’s not much wrong with it, other than the fact that aDNA from West Asia is relatively lacking from the Chalcolithic. We CAN infer, though, from phylogeny. It’s a useful tool that I think is really overlooked.

Davidski said...

@An Idiot

There's L23* in Yamnaya, but none in the Near East from the same period.

There won't be any L23 in Armenia before the Bronze Age no matter how many ancient Armenian samples are sequenced.

Forget phylogenies based on modern samples. They're misleading.

Davidski said...

@All

By the way, before anyone chimes in to say that there's R1b-Z2103 in Chalcolithic Iran, because of Hajji_Firuz_ChL I2327 from Narasimhan et al.

Well, yeah, but that sample isn't C14 dated and certainly has steppe ancestry...

Mbuti EHG Hajji_Firuz_ChL Hajji_Firuz_ChL_I2327 0.0141 3.25
Mbuti EHG Seh_Gabi_ChL Hajji_Firuz_ChL_I2327 0.0185 4.258

Bob Floy said...

@Davidski

Yeah I had seen those, they seem to have a ton of CHG if I'm reading them correctly. I'm a little confused here. There was a pretty big decrease in CHG from the Eneolithic to Yamnaya proper, then("Steppe Maykop" samples notwithstanding)? Makes the Maykop idea look even worse, in that case.

Davidski said...

@Bob Floy

Yamnaya overall does indeed have less CHG than the Eneolithic steppe samples from the North Caucasus steppes, due to WHG-rich Europe MN admixture, and probably a higher ratio of EHG ancestry too.

Steppe Maykop doesn't really fit into the picture because it has some sort of Botai or Siberian-related ancestry that is missing in Yamnaya.

Of course, someone might eventually find a few Steppe Maykop samples without this admixture, but even then it would be a lost cause to argue that Yamnaya derived from Maykop, because it's most likely that it moved into Steppe Maykop territory from the north at around 3,000 BC, hence its WHG-rich Europe MN admixture.

So it seems like, as per archaeological data, Yamnaya is the outcome of the amalgamation of various Sredny Stog, Repin and Khvalynsk groups, and not directly nor recently related to Maykop.

Bob Floy said...

@Davidski

"Yamnaya is the outcome of the amalgamation of various Sredny Stog, Repin and Khvalynsk groups"

This is my understanding, yes.

And I had forgotten about that Siberian ancestry in the steppe Maykop samples, that makes it even harder to see how they could have brought IE to the steppe, especially as late as 3000 BC. Corded Ware had already formed at that time, no? It just seems like a really desperate idea, agenda-driven and not based in reality.

Davidski said...

@Bob Floy

Corded Ware is still something a mystery.

It's obviously from the steppe, and closely related to Sredny Stog II, Yamnaya and Eneolithic steppe, although at this point it might be a sister group of Yamnaya or an early off shoot.

But in any case, like Yamnaya, it doesn't show any obvious ties to Maykop or Steppe Maykop.

Bob Floy said...

@Davidski

"Corded Ware is still something a mystery"

Speaking of this, an off-topic question, if I could. At the present time, which culture would you say probably represents the earliest proper Iranian speakers(as opposed to indo-Iranian)? Srubnaya?

Davidski said...

@Bob Floy

Yes, I think Srubnaya people were the earliest Iranians.

Bob Floy said...

@Davidski

Thanks.

George Okromchedlishvili said...

They got wiped out by incoming NWC speaking populations?

My pet theory is that NWC is actually something like quasi-Tripolye in its origins and came to Caucasus hoping around Blask Seas shore.

Grey said...

"My view is that Maykop was generally an alien entity to the indigenous peoples of the steppes...(snip)...when the technological and economic advantages of Maykop over the steppe peoples eventually eroded, it couldn't hold its ground on the edge of a vastly different and perhaps largely hostile world, and quickly disappeared...(snip)...production and use of jewellery pieces made of gold, silver and gemstones in the Maykop culture"

#

Georgia/Colchis was famous for gold panning (and still has some gold mines iirc)

if you look at the watershed for the caucasus you see one lot of rivers flowing south from the mountains through Georgia and then another set flowing north to the sea of Azov including through Maikop itself.

https://eurasiangeopolitics.files.wordpress.com/2016/04/north-caucasus-physical-map-chechnya-ingushetia-stavropol-kabardino-balkaria-adygea-dagestan-karachai-cherkessia.jpg

maybe some of those Caucasus gold deposits were washed north as well as south?

(but smaller so they eventually ran out?)

what you'd need is a geologist (or whatever specialist knows this sort of stuff) who could estimate where the original Caucasus gold deposits were and whether some of it jutted out into the north side of the watershed.

#

if so then i could imagine gold panners from south of the Caucasus crossing the mountains and setting up shop along the rivers in the Maikop region getting into conflict with the steppe dudes and then inviting some Botai as foederates to hold the steppe dudes off (steppe maikop) while they lived well in their terraced settlements trading gold - then when the gold ran out they disappeared back over the mountains and the steppe dudes reclaimed the territory?

people will go a long way for gold and silver

#

Samarra bowl

seems to me there might have been a chain all the way from sumer to maikop during the time of the akkadian empire so maybe metal workers and scary dudes to protect them lived all along the chain?

#

maybe "chalco" lithic should be changed to "soft metal" lithic to include gold, silver and copper?

Slumbery said...

Grey: "...seems to me there might have been a chain all the way from sumer to maikop during the time of the akkadian empire..."

Surely not during the time of the Akkadian Empire, because it was not contemporary with Maykop.

Davidski said...

Samarra is actually in Iraq, not on the steppe in Russia.

Samara is in Russia, but that has nothing to do with the Samarra in Iraq.

Grey said...

Slumbery

"not during the time of the Akkadian Empire"

yes, got mixed up with hittites - same basic idea though, chain of trading settlements from Samarra/Sumer/Uruk/Ubaid (whichever one it was at the time) to Maikop with intrusive metal workers paying for food and protection possibly from other groups either local or imported for the purpose.

#

Davidski

"Samarra is actually in Iraq, not on the steppe in Russia"

yes - soft metal age linking distant peoples along gold/silver/copper trade routes

Synome said...

Here's the issue I see with the stance of Harvard/Max Planck right now.

They're associating the initial spread of PIE with wheeled cart technology. But I don't think wheeled carts were the crucial factor. I think it was something more fundamental: Horse domestication. This is the David Anthony view of things outlined in "The Horse, the Wheel, and Language", and I agree with it. Early PIE was probably associated with horse riding. Not in battle, but as transport.

Wheeled carts may have been the trigger for the Yamnaya expansion, but I don't think they explain the origin of PIE well at all. The genetic evidence doesn't line up.

David Reich said the lack of wheeled vehicle terminology in Anatolian is evidence against a steppe origin of PIE. But that only counts as evidence if you assume that PIE wasn't already spoken on the steppe before the introduction of wheeled carts. I haven't seen much to support holding that assumption very strongly.

Grey said...

"Wheeled carts"

*if* travois* came before wheeled carts where might you find travois?

open, flat terrain?

(*or sleds)

(before they had horses American Indians used dogs to pull their travois)

http://www.native-languages.org/travois.htm

Matt said...

Very uncertain what happened to the Steppe Maykop group; assuming the model of Wang "Eneolithic Steppe" + Globular Amphora model is correct, then the vast majority of ancestry in Yamnaya is from the sampled groups from Wang from the southern parts of the Black-Sea Caspian Steppe subset of the Pontic-Caspian, and no further contribution from more heavily EHG cultures / Steppe_Maykop / Maykop really. (And most likely language is in steppe from at least eneolithic dates, quite probably earlier).

Things might be slightly different in the Eneolithic Ukrainian groups that contributed to Yamnaya were not quite like Globular Amphora, but rather more like the WHG heavy groups. There might be more room for some balancing EEF and ANE from other sources, though still probably little. It's less parsimonious, but we'll see.

Really we just want this data out so we can have much more proper discussions on this subject!

@Synome, no problem with horse domestication importance, but I think one question about it is Botai people had that earliest (apparently), and not so successful. So seems less useful as a why for expansion of IE at first blush.

epoch said...

As Maykop has some surprising Y-DNA L and Chechens have pretty large amount (10%) maybe Maykop is related to the N.E. Caucasian languages.

Romulus the I2a L233+ Proto Balto-Slav, layer of Corded Ware Women said...

It's interesting that Cucteni and Maykop decline almost at the same time as this:

https://en.wikipedia.org/wiki/Black_Sea_deluge_hypothesis

Amazing that the black sea used to be a fresh water lake.

Davidski said...

@bellbeakerblogger

The earliest Eneolithic steppe individual with more CHG than Yamnaya is from a pre-Maykop kurgan burial at Vonyuchka and dated to 4332-4238 calBC. The sample, from Wang et al., is VJ1001. Here's the site (44.019962"N 43.155538"E)...

Google Maps

But, importantly, already a little earlier than that (4594-4504 calBC) there was a population living in the North Caucasus Mountains (Eneolithic Caucasus) that could not have contributed ancestry to Eneolithic steppe or Yamnaya, because it had too much Anatolian-related admixture.

So I don't know where the CHG in VJ1001 came from exactly, but I'm pretty sure that it came from a population that was already present in the North Caucasus steppes well before 4332-4238 calBC, because it's highly unlikely that this was the very first CHG-heavy individual in the steppes, and, as per above, the Northwest Caucasus Mountains were already home to a very different group of people who aren't relevant in this context.

I guess that there was a native (Paleolithic-derived?) CHG or CHG-heavy population living in the North Caucasus Mountains and steppes during the Eneolithic, who were largely replaced by settlers from the south in the mountains and mixed increasingly with EHG in the steppes.

And that's ultimately who Yamnaya got its CHG from. Its European farmer admixture is probably from Sredny Stog.

Davidski said...

What's the bet that Alberto and Nirjhar think that the CHG in Eneolithic steppe and Yamnaya is from Central Asia? Or even India? LOL

Man, glad those sorts of pointless discussions are gone from here finally.

Ric Hern said...

@ An Idiot

The oldest Swastika was found in the Ukraine at Mezine, carved into Mammoth bone. Mesolithic, if I remember correctly.

Synome said...

@Matt

"I think one question about it is Botai people had that earliest (apparently), and not so successful. So seems less useful as a why for expansion of IE at first blush."

Here what I think the difference is. Botai was a hunter gatherer culture. Early PIE was Neolithic. This I think was the crucial factor that made PIE horse domestication spread rather than Botai's.

Ric Hern said...

As far as the CHG is concerned, this could be a probable source...

https://en.m.wikipedia.org/wiki/Kamennaya_Balka

Davidski said...

@JuanRivera

Keep in mind that one of the Khvalynsk samples already has about as much CHG as Yamnaya, so it's not yet clear where such people were living exactly during the Neolithic, but the steppes between the Black and Caspian seas are a good bet IMO.

Davidski said...

And when I say Neolithic, I mean Pottery Neolithic. Thus, these were probably foragers.

Matt said...

@Synome, when we say neolithic, are we talking about "wider set of domestic animals+horses enabling mass herding of those domesticates"?

If I remember rightly, Botai are neolithic in the Russian sense - pottery using - and more generally used definition - food producers subsisting mainly off their own domesticates (plants+animals)? Described as having a pastoral economy? Largely not hunting and gathering for subsistence. But they were a pastoral mono-culture only using the horse (horse milk, horse meat, etc.)? No woolly sheep, cattle, etc.

(E.g. Frachetti 2012 describes: "The first documented communities in Eurasia to have exploited domesticated animals are associated with the late Eneolithic/early Bronze Age “Botai culture” (Zaı˘bert 1993). At Botai, more than 99% of the total fauna was identified as horse (Levine 2005). According to recently published lipid analysis of ceramic pots from the type-site Botai (3600–2800BC), these north-central steppe communities raised domesticated horses for meat, milk, and probably for transport (Outram et al. 2009). Evidence of corralling and leather thong-smoothers imply that some horses at Botai were being controlled (Olsen 2006b) .... The mid-fourth-millennium-BC economy in the central steppe region differed strongly from that of contemporary herding groups in the western steppe, who were antelope hunters, fishers, and incipient herders of cattle and caprines. Although fish was also a likely supplement to the diet of Botai groups (O’Connell, Levine, and Hedges 2003), bones of cattle and sheep are unknown from Botai-type sites (Benecke and von den Driesch 2003:73).Distinct from the hunters’ campsites recorded in the Volga region and Caspian steppe, Botai type villages were large, with substantial pit houses, and were most likely occupied on a year-round basis (Olsen 2003)." - https://i.imgur.com/ta3ndpA.png)

Bob Floy said...

@Davidski

"What's the bet that Alberto and Nirjhar think that the CHG in Eneolithic steppe and Yamnaya is from Central Asia? Or even India? LOL

Man, glad those sorts of pointless discussions are gone from here finally."

I was thinking the exact same thing, a month ago my question about Srubnaya would have provoked an uninvited incoherent rant from Nirjhar about how we needed more samples from neolithic Iran. Do not miss that BS.

Anonymous said...

@Ric Hern Yes, but that was a long time ago, and speaks to the origin of the swastika in Eastern Europe (the main development area of R1). I'm not claiming it originated in the Samarra culture of Northern Iraq or the Vinca culture - just that it can be used to track a migration of people who identified with this symbol (from the Balkan Vinca culture to Halaf very near Anatolia, Samarra in Northern Mesopotamia and then finally Ubaid by the Persian Gulf). Even if it is superficial, it makes sense, especially as Vinca and Ubaid share a plethora of similarities (relatively similar proto-writing, extremely similar figurines and pottery), the fact that this spread matches the spread of copper metallurgy etc.

And I am extremely extremely skeptical that people in the Middle East were importing pottery from the Steppe lol - that's clearly BS, come on. Whether or not these swastikas in Samarra pottery can be seen as partial evidence for the southern origin of PIE is a different matter altogether, but this importation theory is rubbish. Is there any other evidence for the people of the Steppe making pottery and trading it with Mesopotamians? No - basically, that pottery is indigenous.

We know Lepenski Vir was R1b (yes, V88, which is pretty far removed from M269 phylogenetically, but that only serves to demonstrate an Epigravettian homeland for R1b - neatly explaining Villabruna's Y DNA too), so why is it so unlikely that a culture succeeding it (the Vinca culture) was somehow related to R1b (perhaps even M269)? And by the way, if I had to guess, I would predict the bulk of Vinca to be typical farmer ancestry, with the M269+ being part of a metallurgical elite. Just a guess, though.

Anonymous said...

@Ric Hern

Yes, but that only speaks to its origin. I think it's useful as a tool to track migrations, and from that we can see a migration from Vinca to Halaf, Samarra and finally Ubaid.

These aren't the only similarities though, you can just type in "Vinca Ubaid" into Google - all these similarities (for example, early copper metallurgy, very similar pottery and figurines, potentially similar proto-writing) point an ancestral relationship of one to another. Childe thought Vinca would have been derived from the more sophisticated Near Eastern cultures, but now we know Vinca is older, so chances are it is the other way round.

In terms of the relevance of this to the potential Southern origin story - I see R1b M269 men piggybacking on top of an agricultural society made up of typical farmers, putting themselves as the elite through their knowledge of metallurgy. So, Balkan/Ukrainian M269 -> West Asian L23, just as Vinca -> Ubaid.

Just a theory, though, of course, but there's other reasons for it too.

Ric Hern said...

Apparently the Kamennobalkovsky culture (Lower Don River) shows similarities with the Imereti Culture of the Caucasus in certain stages of its development. Could this be the source of CHG ?

Santosh Rajan said...

@Davidsky
"What's the bet that Alberto and Nirjhar think that the CHG in Eneolithic steppe and Yamnaya is from Central Asia? Or even India? LOL

Man, glad those sorts of pointless discussions are gone from here finally."

The latest argument from the OIT folks is BMAC. LOL

http://www.brownpundits.com/2018/10/17/a-tentative-out-of-india-model-to-explain-the-origin-spread-of-indo-european-languages/

Davidski said...

@An Idiot

Ancient DNA from Vinca and other closely related ancient Carpathian and Balkan groups is already available, and it doesn't suggest that there were migrations of such groups to the Near East.

In fact, there is no R1b in Vinca or in closely related populations. R1b appears in the Balkans much earlier, in forager groups, and then basically disappears when the Balkans are colonized by farmers from the Near East. See here...

Ancients list May 2018

Also, the ancient pottery from Samarra, Iraq, doesn't look like ancient steppe pottery to me. Maybe someone else with more knowledge in this area can chime in, but I'm skeptical that this Near Eastern pottery with swastikas has anything at all to do with the ancient steppes.

If the swastika symbol is your only argument, then that's a very weak argument not supported by any other data.

A said...

@An Idiot

Some Vinca figurines are very similar to later Cycladic culture figures. Marija Gimbutas also thought Linear A might have developed from the Vinca symbols.

Synome said...

@Matt

You're right, Neolithic was a poor choice of words. I think I must have been remembering an older stage of research about Botai when they were thought to be exclusively horse hunters. Domesticated animals is Neolithic by definition.

But I think you got the thrust of my point. I'm following Anthony, who says exposure of the Volga-Dnieper zone to a wide variety of domesticates and the eventual adoption of mixed farming was a necessary step that triggered the first mass expansion of early PIE speakers.

I speculate that Botai became a dead end because of their exclusive focus on the horse, and (maybe related) because of their isolation from cultural currents in Eastern Europe and Mesopotamia.

Anonymous said...

Not sure if I doubled-posted from one of my other many Google+ accounts (I'm new to Eurogenes regardless, not a sock), but I'll post again:

@davidski I'm not claiming Samarra pottery was from the Steppe - just the opposite. I'm using it as evidence for R1b people in that region during the Chalcolithic. It still needs answering, though - why would a bunch of farmers unrelated to anything R1 or IE-like randomly start putting it on a good deal of their pottery? Unless, as Tomenable has previously suggested a year or two back (idk if he still believes it), they had a metallurgical and pastoral elite (of R1b-M269+) amongst them - there is evidence for this at Ubaid for example, where there were clear-cut socio-ethnic and potentially caste-like divisions into three groups: native hunter-fishermen of the Persian gulf, the normal farmer types, and pastoral folk. And also as far as I've seen, the consensus does seem to be that metallurgists isolated themselves in their own separate group wherever they were found at the earlier stages of metallurgy, as their abilities lead to their social status and obviously that would be something they wouldn't want the rest of the population to know. Ubaid and Vinca btw, and I don't care what the aDNA at the moment says (not least as we don't have good aDNA from Chalcolithic Mesopotamia anyway), were so clearly related. Easier for me to just say google it instead of spending ages typing it all out, it would take less than 5 mins from a very quick search to see they're related (but I can rely on argument from authority - Childe himself thought Vinca was derived from something like Ubaid, but we now know Vinca is older, so it's probably the other way around. Same for copper smelted metallurgy, which was first in Vinca, and (arguably) proto-writing). Also, in favour of what I'm saying (though only hypothetical at this stage, does seem to be at least somewhat legit from what I've read online), is Euphratic, which is basically a proposal for an IE substrate among the language of the Ubaidians that would later have been replaced by Sumerian (which I see as related to Elamo-Dravidian from Iran, but that's a different story).

(1/2)

Anonymous said...

(2/2)

But on top of that, there are other questions - what about all that Iraqi Z2103, where did it all come from? There's an awful lot of it - it didn't come from the Hittites, Phrygians/proto-Armenians, or any Iranian groups (which would have been pred. Z93), so where did it come from unless it was there before the great migrational period out of the Steppe? Yes, there was one (maybe a few, idk) L23* samples in Yamnaya, but you would expect a small amount of that wherever you find Z2103 at that point in time given the age of Z2103. If you look at, for example, the earlier branched-off branches of Z2103, they point to West Asia (mainly around the Southern Caucasus). Similarly, if you look at the earlier branched-off branches of L51, they point to the West Med. I fail to see what is wrong with such a methodology - if you can tell me, though, I'd be happy to hear, but it really does make a lot of sense to me. Why would, for example with L51, the West Med region have the lion's share of "archaic" L51 subclades (archaic in that they split from the rest of the pack, so to speak, at an earlier stage), if it weren't the origin or at least an early seat of development? This is somewhat unrelated, but I think Los Millares is related to the spread of L51 across Europe (as an elite metallurgical group amongst mainly farmer types) - a quick Wiki shows the society was heavily caste-like and war-like, and of course was an early example of European metallurgy, of which R1b-types in particular seem to have been particular proficient. Related groups would have spread up to France, until the arrival of the CWC prompted their expansion (and mixing with Corded types). Coon's analysis of the Beaker folk is particularly interesting - he basically describes them as brachycephalic Dinaric metallurgists originating from West Asia (then going to Spain and from there Central Europe), and from the Rhineland and British examples there seems to have been an amalgamation of unmixed Dinaric Beaker and Nordid Corded types, indicating recent population contact. It's all on SNPA's TROE, under the Bronze Age section iirc. And on top of all of this, the CWC and Yamnaya (Danubian or not) seem to both be non-L51. AND also there's how you would have to explain the brachycephaly of L51 given a Steppe origin, as those guys were long-headed (Corded Ware was thin-faced and long-headed, Yamnaya was broad-faced and long-headed) - Vucedol was not Dinaric back then.

But yeah, that's the bulk of my argument for a Southern L23 origin, potentially related to PIE, but potentially PIE would just be a native Steppe R1a language (I haven't made up my mind yet). And sorry if there's typos/things obviously wrong, I'm super tired

Davidski said...

@An Idiot

I'v never heard of swastikas on Yamnaya pottery, nor that it was similar in any special way to the ancient pottery in Iraq.

I'm quite certain that there won't be any L23 in ancient DNA from Iraq unless in samples with steppe admixture, because L23 is from the steppe.

I think you're putting too much value into some random online comments that aren't valid.

Bob Floy said...

@An Idiot

I wouldn't associate swastikas too closely with IE cultures, you'll find swastikas everywhere, especially in native American cultures.

Anonymous said...

@Davidski No, I’m not saying anything about Steppe pottery at all. I just pointed out that the Samarra pottery in Iraq had swastikas on them, which indicates an ULTIMATE East Euro origin of at least part of the culture-bearer population (as as you’d expect, by far the oldest swastikas are in Eastern Europe, dating to 12000 years ago), but this is well before Yamnaya or anything remotely like that - as I’ve said, I see M269 originating in the Balkans/Eastern Europe and being part of the spread of copper smelting technology down from Vinca to West Asia (but as an elite group distinct from the farming population). Since these M269+ people would have had ultimate origins in Eastern Europe and the Balkans, in this theory, it’s potentially no surprise to see things like Armenia Chalcolithic having EHG and European-like admix, and for the Areni-1 sample to have red hair and blue eyes (from admix with this population).

Leyla-Tepe was iirc the first Caucasian copper smelting culture, and was theorised to have very likely been created by Ubaid migrants - I’d identify them with the pastoral guys in the tripartite Ubaidian caste system that I mentioned, who I would place as Z2103 (the presence of these individuals in Mesopotamia leaving the lasting imprint of a huge ~30%+ Z2103 in Northern-Central Iraq today). They would, then, through Leyla-Tepe, make their way to the Steppe via Azerbaijan/Dagestan - but this doesn’t necessarily mean to expect Z2103 in a handful of Leyla-Tepe sites, given they would have kept to themselves in their own caste probably in different areas (not really speculation - I can get a bunch of quotes from respected archaeologists and anthropologists about metallurgists being separate elites in whatever cultures they were in, because as mentioned metallurgy was a mysterious skill that gave you high social status, rather than something blue-collar like we imagine today).

And most of this has been developed by my own thought by the way, initially based on Tomenable’s connection of M269/L23 with the spread of copper smelting technology (though he advocated for a Near Eastern rather than Balkan/E. Euro origin of M269, probably due to the fact that variance of M269 is highest around Syria, but looking at variance isn’t all that useful), I connected the dots with Vinca and Ubaid, swastikas, phylogeny etcetera.

I would like to know what is wrong with the methodology of looking at the modern-day distribution of branches of a haplogroup that branched-off early from the rest of the branches, as it seems a watertight way of locating its approximate origin to me. That's why I believe and continue to believe L51* will not be found on the Steppe or anywhere East of Hungary - I'm sure there may be a few that made it as far as Hungary along with U152 though, for example, just as there was that L23* in Yamnaya. Los Millares is my candidate culture, as mentioned in my previous post, for L51's (or what would later become L51 perhaps) entry into Europe, because it also ticks the boxes with copper metallurgy, caste-systems and aggressive invasions. I would like to know why you think L23 is Steppe in origin, just because it was found in samples from the Steppe. Aliens from the future could look at early British settlements in America and make a similar conclusion...

Anonymous said...

@Bob Floy Yes, but I believe R1b made it to the Americas from West Eurasia aha, and no I'm not a crackpot. This is definitely more out-there, and goes against literally all aDNA, but I think it is far FAR easier to prove (and also raises questions about nefarious censorship due to political sensitivities, but I'm not going there). Firstly, I do want to just say it is very naïve to believe pre-Viking Native Americans randomly happened to use the swastika symbol - what a coincidence right? It's just a happy mistake, forget the fact that there are basically infinite simple geometric shapes they could have used. No, it was not a mistake. You can point at lots of things that don't make sense to have been Amerindian in origin, but the clincher is that some of the remains from mostly South American cultures (such as the Nazca cemetery) are so blindingly obviously West Eurasian it is actually concerning why it isn't talked about more. This is just copy and paste, but look at this, it is practically guaranteed to be enlightening: https://jpst.it/1tJ-h

Ric Hern said...

@ An Idiot

I think at this Blog we basically concentrate mostly on the DNA of Actual Ancient Samples. Have a look at the paper of Bell Beaker samples. That will also explain a lot....Davidski discussed this in other Posts...So I think you have some catching up to do.

Davidski said...

@An Idiot

There was no R1b in the Americas before Europeans got there ~400 years ago. I can assure you of that.

Also, unfortunately, I'm going to have to intervene here now and remind you that this blog, including the comments, doesn't stray very far from mainstream theories unless there are very good reasons for doing so.

So if you want to keep discussing ancient migrations from Iraq or thereabouts into the steppes and how swastikas fit into this, then you'll have to do it somewhere else, because I'm just not seeing this as a legitimate line of inquiry right now, and probably never.

In other words, please get back on topic.

Bob Floy said...

@An Idiot

"Firstly, I do want to just say it is very naïve to believe pre-Viking Native Americans randomly happened to use the swastika symbol - what a coincidence right?"

Not really. Thanks to ancient DNA we now know that Europeans and native Americans have deep shared ancestry from the Mammoth hunting days in Siberia, and the oldest known swastika on pottery that I'm aware of is from Mesolithic Ukraine, less than a thousand years after those shared ancestors went their separate ways. It seems self evident to me that it's probably a very ancient symbol that predates the split.

Also, the swastika is really just a glorified cross, like I said you'll see it basically everywhere, the ancient Egyptians used it too, and I'm pretty sure they didn't get it from the vikings(or even the Hittites, for that matter).

" I believe R1b made it to the Americas from West Eurasia"

Not calling you a crackpot as such, but I would agree with Ric Hern that you need to catch up, it sounds like you stopped following the new developments around 2010 or so.

Anonymous said...

If you’re talking about the Olalde paper(s), then I know about it. Basic point is introduction of R1b L51 to Western and Central Europe is associated with EHG and CHG ancestry (modelled as Steppe). I, however, still do not think L51 came from the Steppe, based on everything I’ve said earlier (in this instance mainly phylogeny, but plenty of other things as mentioned) - but instead either picked up this ancestry from contact with Steppe (mainly Corded Ware) folk or already had some of it to begin with, despite the Iberian Chalcolithic samples (which I expected, and is also in agreement with Coon, who based on his theories about the BB folk would also have believed in a Western origin of L51 (had he known about it)).

I would very much like Chalcolithic samples from the Rhone and Upper Rhine, from at latest 3000 BC onwards, where I’d expect a great deal of early L51.

Bob Floy said...

@myself

"less than a thousand years after those shared ancestors went their separate ways"

Meant to say 10,000, obviously.
Also, not "pottery", but swastika art.
Do pardon me, I haven't slept in two days. Sorry for cluttering up your comment section, David.

Anonymous said...

What about the phylogeny point? That seems perfectly genetics-based to me.

EastPole said...

@An Idiot

Re:swastikas

Globular Amphora Culture is very interesting. They were putting swastika signs not only on the outside but also on the inside of their pots which suggests that it was not a mere decoration but a religious symbol for them:

https://s2.postimg.cc/yyprjr7h5/screenshot_302.png

https://s2.postimg.cc/b7qe1rhll/screenshot_303.png

Davidski said...

@An Idiot

What about the phylogeny point? That seems perfectly genetics-based to me.

Yes, but it's a pointless exercise to try and work out where Y-haplogroups originated based on phylogenies created with modern samples, because you don't know where the ancestral lines of your samples were a couple of thousand years ago, let alone several thousand years ago.

The only way you can work out where Y-haplogroups originated and how they migrated is with ancient DNA.

Anonymous said...

10,000 years is a long time lol, and looking at my link (https://jpst.it/1tJ-h) is the conclusion not immensely obvious? There was clearly contact before the Vikings, any other conclusion after looking at that link is honestly cognitive dissonance. No aDNA even needed. Check also the text written at the end of that link. Also, does nobody not find it strange we know next to nothing about Native American R1(b)?

And I’ll stop posting about things like that now, on request of davidski, even though it is clearly at least somewhat relevant. But yeah, as mentioned, I really want Rhône-Rhine samples from the Late Neolithic to Early Chalcolithic.

Anonymous said...

Yes, but it’s the overall distribution that equalises factors like these.

So, for example, why would the “earlier” L51 branches be West Mediterranean in distribution if they were born on the Steppe/Eastern Europe? Did they just happen to all migrate there, leaving no trace in modern East Slavic lands, and not also migrate across the rest of Western Europe? It doesn’t make sense.

Davidski said...

@An Idiot

It makes no difference where they are today, the only thing that matters is where they used to be when you're trying to work that out.

Bob Floy said...

@An Idiot

"So, I ask you, read this post, and come to your own conclusions - if you do not conclude there was West Eurasian influence, you're wrong and have a distorted mindset."

That made my whole weekend.

Anonymous said...

But seeing Z2103 in Yamnaya shows nothing of its origin, just as seeing U106 in Jamestown shows nothing of its origin.

Davidski said...

@An Idiot

Really not sure what your point is about the Z2103 in Yamnaya?

Can you show me another ancient population from outside of the steppes from around that time period with even a few instances of Z2103?

If not, then the Z2103 in Yamnaya suggests a lot about its origin.

Anonymous said...

That’s off-topic now, but I don’t see what’s so funny. It’s all so obvious from the hair, which even if dyed, is clearly of the wavy variety.

Anonymous said...

No, but that’s likely because none have been sampled yet. The Middle East during the Chalcolithic is incredibly poorly sampled.

My point is that, in my hypothesis of a Near Eastern Z2103 moving up into the Steppe as the CHG link for Yamnaya, you would still see the exact same Yamnaya results. So the Y DNA results mean nothing for deciding the origin of the haplogroup. Just as taking an aDNA study of Jamestown means nothing for the origin of U106.

And just so I can ask again, how do you explain the really high Iraqi Z2103? Steppe just doesn’t cut it, imo.

Anonymous said...

And yeah I just got what you meant lol but the point still stands dammit!

Bob Floy said...

@An Idiot

"But seeing Z2103 in Yamnaya shows nothing of its origin, just as seeing U106 in Jamestown shows nothing of its origin."

We know that Z2103 probably originated in eastern Europe, not long before Yamnaya existed, while U106 definitely didn't originate in the Americas, even in your own scheme, so how is that a good analogy?

Davidski said...

@An Idiot

U106 in Jamestown means nothing for the origin of U106, because there's U106 in Bronze Age East Central Europe.

Z2103 in Yamnaya means everything, because there are no other populations with Z2103 in the ancient DNA record from around that time, and all of the later samples have steppe ancestry, including modern Iraqis.

Also, there's R1, R1b and L23 in Mesolithic, Neolithic and/or Chalcolithic Eastern Europe, and none in the Near East.

So why the hell would Z2103 be from the Near East and not the Eastern European steppe? Of course it's from the Eastern European steppe, unless you can turn all of this around with evidence from ancient DNA.

Anonymous said...

No lol obviously it means nothing but it’s an analogy. If Jamestown was the first aDNA study picking up U106, they might conclude it mutated there (don’t analyse it too much, but you get what I mean I hope). Like the example of Jamestown and Europe, it COULD be the case that Yamnayan Z2103 was originally West Asian, and that one day we’ll get a sample of West Asian Z2103 dating before Yamnaya (I agree that Hajji Firuz is unlikely to be accurate, but should still be chased up). That would make the conclusions of Z2103 being Steppe completely analogous to the conclusions of U106 being Northeast American.

Iraqis would have gotten Steppe ancestry from the Iranians, that doesn’t disagree with my point - it strengthens it. The Iranians would have been mostly Z93 (I’m sure you agree here), so how can that be used to explain the ridiculously high Iraqi Z2103? It is far higher than Z93 in that area - the Steppe invasions cannot explain it. It’s evidence for a local origin of Z2103.

As for L23 not being found outside of broad Eastern Europe, it’s the same point - the Chalcolithic Near East is essentially unsampled.

Anonymous said...

God you really don’t understand - the point is the earliest known does not equal origin.

Anonymous said...

And yes I’m aware that it’s best evidence until new evidence is found, but that’s not including the fact that we have circumstantial evidence from looking at modern phylogeny, which suggests a West Asian origin. In the hypothesis of a Southern origin of Z2103 to Yamnaya, we of course would still see Z2103 in Yamnaya.

Davidski said...

@An Idiot

Don't be ridiculous.

Of course there are very good reasons to conclude that Z2103 originated in the Eastern European steppes. The direct evidence from ancient DNA is unequivocal about that.

Forcing its origin in the Near East based on modern samples really isn't very convincing at all.

And there's no good reason to assume that U106 originated in the Americas, even if it's found in historic samples from Jamestown.

Bob Floy said...

@An Idiot

"the Chalcolithic Near East is essentially unsampled"

I don't think this is true. Again, you need to get caught up.

"No lol obviously it means nothing but it’s an analogy. If Jamestown was the first aDNA study picking up U106, they might conclude it mutated there (don’t analyse it too much, but you get what I mean I hope)"

I get what you mean, but what you mean is incorrect.

Anonymous said...

About Jamestown - you know what I mean.

But for Z2103, just say where all that Iraqi Z2103 came from? Because I can’t think of any other hypothesis other than that of a West Asian origin, unless there had been a huge Yamnaya invasion directly South across the Caucasus (and it would have to be huge and archaeologically significant to achieve around 1 in 3 lineages in modern Northern-Central Iraq). Such an event does not exist, your guess about the true origin of Hajji Firuz doesn’t outweigh that fact.
Hittites and Iranians also cannot explain it, because of the limited domain of influence of the former and the Y DNA makeup of the latter.

Matt said...

@Synome, ah cool. I can't really assess whether it's right (re: horses vs wagons), but yeah, if we're thinking about horses being used in a wider pastoralist system with more domesticates (and tools?) from West Asia vs the more isolated pastoralist system with a limited domesticate in Central Asia.

Actually it does seem that there are some still some disputes about wild / tamed / domesticated horses at Botai, but I think the arguments from residues on pottery, corralling, large settlements argue against really thinking of them as a hunter-gatherer subsistence culture.

(Really with Botai they seem often compared to Plains Indians - but reading about them, from the state of the art in 2012, it seems like they were quite different in terms of not being so mobile at all, and living in these large, fairly permanent settled villages. Not really very nomadic. Seems like more horse doesn't equal more mobile and there were different spurs to adopting the kinds of more mobile pastoralism that are probably associated with early IE and other later cultures?)

Btw, off topic, would recommend to anyone reading Frachetti's paper from 2012, as it offers an overview of lots of areas where we now have ancient dna - https://kuscholarworks.ku.edu/bitstream/handle/1808/21123/Olsen_2010.pdf?sequence=1.

Frachetti himself is not actually wholly persuasive in all points (particularly the main thrust of his argument is that Afanasievo was an independent inspiration representing an Eastern Steppe sphere, mediated by the Inner Asian Mountain Corridor, which now must have him facepalming). But in addition to the overviews he presents, there's a great deal of comment in the scholar "Comments" from the major players in the steppe archaeology scene, and he sort of forces them to respond in ways that try to characterise overarching phenomena, rather than the more circumscribed and careful comments that often come out in other papers.

Although Frachetti is not persuasive on the IAMC as the vector for Afanasievo (esp with hindsight), it does make me wonder about whether IAMC mediated the emergence of other forms of pastoralism on the Central and Eastern steppe, independently of the Western steppe related movements. There is some mention of "Tersek culture" of more westerly Kazakhstan that seems to have got a bit lost from discussion - some similarities to Botai and similar settlements, but less evidence of horse domestication and more of a mixed bag of remains including cattle. Perhaps we'll get some adna from them (or poss their burials are too scarce).

Anonymous said...

You don’t have to be so patronising without putting ideas or evidence behind your words - there really hasn’t been any proper studies of the Near Eastern Chalcolithic outside of peripheral regions such as parts of Anatolia and parts of the Caucasus. I’m pretty up to date with everything, there’s no reason to assume otherwise.

And I also get that your opinion is worthless if you limit yourself to unreasoned statements, that entire post contributed literally nothing to this discussion.

Davidski said...

@An Idiot

Yamnaya doesn't have any ancestry from Iraq.

But modern Iraqis do have ancestry from the Bronze Age steppe.

Ergo, Iraqi Z2103 is from the Bronze Age steppe.

Anonymous said...

That still doesn’t answer how it got there, and that logic is faulty as it assumes the source of Iraqi Steppe was also the source of its Z2103, which would be an okay assumption when just looking at Yamnaya, but we know the Iranians would have been overwhelmingly Z93+. So it doesn’t add up.

Also, I don’t know the finer details of Yamnaya’s auDNA, but I was under the impression from what Reich and the people at Planck have said that there was a CHG-related source that isn’t accounted for fully among Yamnaya’s predecessors. I know you disagree, though.

Davidski said...

@An Idiot

Max Planck are modeling Yamnaya now in the same way that I have been: without any direct, recent Near Eastern ancestry.

Ahead of the pack

And it's already rather obvious when and how Z2103 entered the Near East from the steppe: during the Bronze and Iron ages via Iran.

The Iron Age Iranian (?)

Bob Floy said...

@ An Idiot

"About Jamestown - you know what I mean"

Yes, and you're completely wrong, no one would assume that U106 had first appeared in the Americas, because they would know that R1b had come to the area very recently, with European colonists. So again, it's a nonsense analogy, no point is made with it.

"You don’t have to be so patronising without putting ideas or evidence behind your words - there really hasn’t been any proper studies of the Near Eastern Chalcolithic outside of peripheral regions such as parts of Anatolia and parts of the Caucasus. I’m pretty up to date with everything, there’s no reason to assume otherwise."

You're saying a lot of things that don't make any sense, and my tone, whether or not you find it patronizing, is appropriate.
More than a few chalcolithic genomes from the near east have been published over the last two years or so, and I can't see how Anatolia would be "peripheral" to the question? It's also a fact that a handful of those chalcolithic genomes are from Iran. You don't seem to be aware of that, and if not than you're lagging behind a bit, especially given your pet theory. In any case it dosen't really make sense to say that the chalcolithic near east is "essentially unsampled".

All of the factors here point to Z2103 having originated on the steppe, and Yamnaya is loaded with it, so it's not closed-minded to assume that Yamnaya related groups probably probably brought it to the near east. Is that well reasoned enough?

Jack Rusher said...

Regarding the Swastika, our visual cortex is highly sensitive to symmetry, including rotational symmetry, which leads this sort of glyph shows up everywhere. There's really no compelling reason to see every rotationally elaborated cross as part of a single cultural transmission.

https://en.wikipedia.org/wiki/Rotational_symmetry

Googling "vinca ubaid" mainly returns websites about prehistoric alien lizard people, which should give you pause regarding the plausibility of these ideas.

Bob Floy said...

@Davidski

"And it's already rather obvious when and how Z2103 entered the Near East from the steppe: during the Bronze and Iron ages via Iran. "

Or maybe with the Hittites, heh.

Ric Hern said...

@ Bob Floy

Gutians?

Anonymous said...

Learn to read or Ctrl+F for Hittite - it can’t explain it. As for your diarrhoea post - OBVIOUSLY Jamestown can’t be used to show U106 is NE American, stop being an idiot. It is clearly as I said a faulty example used only for analogy. You sound like a teacher or parent, and unless you contribute to discussion I’m just going to ignore you.

Anonymous said...

Yeah it’s similar figurines, I don’t think there were any aliens involved, obviously

Anonymous said...

If anything, that helps the hypothesis of a Southern origin, as there’s absolutely no evidence they came from the Steppe. The case of the Gutians is actually used by some as evidence of the Armenian hypothesis.

Grey said...

@Matt

"Really with Botai they seem often compared to Plains Indians - but reading about them, from the state of the art in 2012, it seems like they were quite different in terms of not being so mobile at all, and living in these large, fairly permanent settled villages."

just speculating but if Botai domesticated horses first (?) and yet didn't expand over the steppe that seems to imply they were restricted to a range where grass was plentiful enough to support their herds without needing to be mobile - hence maybe an adjacent people living in more arid regions outside that range developed the mobile life style - and then if at some point the climate changed such that the grass in the Botai range was no longer lush enough to support the sedentary form then the already nomadic population might be able to swoop in and take over that territory.

(something along those lines might be an explanation for modern horses not being descended from Botai horses, steppe maykop being genetically different from yamnaya, Botai language isolate etc)

i could imagine this assumed mobile group being similar to plains indians in some ways.

Grey said...

the reason i wonder about travois (or sleds) is

"David Reich said the lack of wheeled vehicle terminology in Anatolian is evidence against a steppe origin of PIE. But that only counts as evidence if you assume that PIE wasn't already spoken on the steppe before the introduction of wheeled carts."

although i guess that would mean they'd need to have a word for travois or sled?

Anonymous said...

But the Y DNA evidence is also relevant and much more precise than auDNA, and Khvalynsk afaik wasn’t even L23+ (though by all means correct me if I’m wrong). If I’m correct, the first sign of L23 on the Steppe is also the first sign of Z2103 (at Yamnaya).

Bob Floy said...

@Ric Hern

"Gutians?"

Maybe? From what I've heard next to nothing is known of their origins, but, could be.

Bob Floy said...

@An Idiot

"stop being an idiot"

I think I just heard a laugh track.

Davidski said...

@An Idiot

There's plenty of R1b of all sorts in Eastern European foragers with no Near Eastern ancestry.

So R1b is native to Eastern Europe. Hence, the R1b in Yamnaya is also from Eastern Europe.

You would need to prove with ancient DNA that it isn't.

Your claim that R1b migrated from Eastern Europe to the Near East, and then back to Eastern Europe as L23 sounds like special pleading.

Anonymous said...

@Davidski Firstly, all this talk about things as old as R1b are pointless - it’s only worth focusing on M269 onwards for this case, as of course R1b throughout most of its development was originally Eastern European.

And secondly - so you don’t believe there was any R1b of even minor significance in the Middle East before Yamnaya? Even the branches older than Yamnaya that are entirely limited to the Middle East today? That doesn’t really make sense - saying you can’t trust modern samples because of population migration absolutely does not explain it, because why would only these older branches migrate to the Middle East, leaving essentially no presence elsewhere? The conclusion is obvious.

But this is going round in circles now - we’ve both made our points and disagree with one another, so it isn’t productive. Let’s wait for more samples, then.

Anonymous said...

Firstly, I’m not advocating for Maykop to Yamnaya given the Y DNA (of only a few samples mind you). That “Yamnaya” ancestry in the Chalcolithic NW Iran sample from Hajji Firuz doesn’t make sense to have come from Yamnaya - it IS probably dated wrong, but there exists no known archaeological movements from Yamnaya direct to West Asia over the Caucasus, not least to leave such a huge imprint on modern Iraqi Y DNA. I’ve already mentioned Indo-Iranians were overwhelmingly Z93+, and the Anatolian IE folk never reached that far East across West Asia.

For PIE - I’m not claiming these Southern L23+ guys spoke it necessarily, I’m not sure about it.

But really the most important thing you mentioned is the lack of ANF in Yamnaya - how was this determined? Surely if it has EEF, it can just be said to have WHG+ANF, right?

Davidski said...

@An Idiot

But really the most important thing you mentioned is the lack of ANF in Yamnaya - how was this determined? Surely if it has EEF, it can just be said to have WHG+ANF, right?

This is easy to show, and I have a detailed post coming on the topic later today/early tomorrow.

But briefly, Yamnaya is a mixture of Eneolithic steppe and something very similar to Globular Amphora.

There's no room in this model for any ANF that isn't mediated via the Globular Amphora-like ancestry.

Bob Floy said...

@Davidski

"Yamnaya is a mixture of Eneolithic steppe and something very similar to Globular Amphora"

How much of that GAC-like ancestry would you say Yamnaya has at this point? 15-20%?

Davidski said...

@Bob Floy

My latest estimates put it at around 20%, and I'll show exactly how that's calculated in my new post.

It's Globular Amphora-like ancestry, but with a higher cut of WHG.

Anonymous said...

You are the auDNA wizard I suppose, but I’m still sceptical Globular Amphora has anything to do with Yamnaya. I’ll keep an open mind though.

Bob Floy said...

@Davidski

"My latest estimates put it at around 20%, and I'll show exactly how that's calculated in my new post.

It's Globular Amphora-like ancestry, but with a higher cut of WHG."

Looking forward to it. I'm still a little confused about why Yamnaya's EEF is like GAC rather than Trypolie.

Anonymous said...

@Davidski If you were forced to break down this Yamnaya EEF into its ancestral components of ANF and WHG, what percentage ANF and WHG would it have?

Matt said...

@Grey, re: Botai relative sedentism, that's a nice little ecological possibility. I don't know if the climate matches up though, or when changes happened, that said.

Davidski said...

@An Idiot

The European farmer ancestry in Yamnaya is something like 40% WHG.

It's definitely European farmer ancestry, because Yamnaya shows EEF mtDNA haplotypes, and lacks mtDNA haplotypes from south of the Caucasus.

Grey said...

@Matt

"I don't know if the climate matches up though"

yeah i vaguely recalled someone mentioning a dry phase but i can't remember when it was so it may be nothing.

Davidski said...

It's way more than 18%. The fits improve with more WHG.

Of course, it's imperative to have the correct steppe reference as well, but I just figured out that I do.

Anonymous said...

@Davidski So the reasoning behind Yamnaya's ANF being EEF in origin (as opposed to West Asian) is that Yamnaya has some EEF mtDNA? What's the prevalence of Yamnaya's EEF mtDNA - it would have to be far higher than its percentage of EEF ancestry if your model is to be correct, given this theoretical EEF ancestry in Yamnaya would surely be mostly female-mediated (as Yamnaya Y DNA is definitely not EEF-like).

It of course makes sense that Yamnaya would have extra EEF (perhaps WHG enriched, sure) from their neighbouring contacts in the Ukrainian Steppe, and that Yamnaya would have extra EEF mtDNA for the same reason. But proving that all of Yamnaya's ANF comes from this EEF-admix is a different story - it CAN, though, essentially be disproved by looking at the mtDNA frequencies, as mentioned. If % EEF mtDNA in Yamnaya is not >> % EEF ancestry, then there's a problem with your model.

Also, I thought Yamnaya did share some mtDNA with Caucasians, based on this Wang paper?

Davidski said...

@An Idiot

Yamnaya definitely lacks direct Anatolian ancestry.

There are no obvious signals of direct gene flow from Anatolia or the Caucasus in Yamnaya, except in the one female outlier from Ozera.

So that's one, young female in a grave with no inventory (no burial gifts).

None of the Yamnaya males with R1b show any Near Eastern input, and that's really obvious for I0429, the earliest sample with M269, because he has the highest ratio of EHG.

Anonymous said...

@Davidski None of that addresses the point that Yamnaya EEF mtDNA must exceed Yamnaya EEF auDNA as a percentage if your model is correct (that is, if I've understood you correctly, Yamnaya supposedly picking up this ANF ancestry from WHG-enriched EEF folk in Ukraine). I actually don't know if it is or isn't the case that the % EEF mtDNA significantly exceeds the % EEF auDNA, but I suspect it is not, which would call your model into question.

Also, it is surely expected that those Yamnaya samples (including I0429) from Samara show higher levels of EHG compared to the rest of Yamnaya simply due to location. As you go further South, you get (in general) more ANF and CHG-like ancestry, and vice versa. The point being, though, that it isn't necessarily the case that this was just a loss in "original" Yamnaya ancestry through admix with other types as they went further South - it could also very well be the other way round, where extra EHG is gained as they went further North. The core and birthplace of Yamnaya seems to be right around Stalingrad, so Samara is clearly at the Northern periphery of Yamnaya.

Davidski said...

@An Idiot

None of that addresses the point that Yamnaya EEF mtDNA must exceed Yamnaya EEF auDNA as a percentage if your model is correct.

No it doesn't have to.

Uniparental marker frequencies are easily skewed by founder effects and drift, especially in populations with few founders and small effective population sizes. So there's often no linear correlation between them and genome-wide ancestry.

Anonymous said...

@JuanRivera I'm not necessarily claiming that Ukrainian Yamnaya outlier did carry IE to the Steppe.

Also, do you have more information about this Balkan->Anatolia trail? Given the age of Anatolian, it would have to have been from before Yamnaya.

Anonymous said...

@Davidski True, but you just have to assume (without further evidence) that a lineage hasn't been helped or hindered by things like founder effects. There's no reason to assume either way given a potential founder effect, and there's no reason to even assume a founder effect in the first place. You can't just say "founder effect" to explain why uniparental markers don't match up with a hypothesis. Besides, does Yamnaya not have very diverse mtDNA? That isn't consistent with the possibility of a founder effect, at all.

Davidski said...

@An Idiot

Yamnaya mtDNA haplogroups more or less match its genome-wide ancestry: EHG, CHG, and WHG-rich EEF.

So there's no point forcing Near Eastern ancestry into the equation.

Anonymous said...

@JuanRivera interesting - I knew about Suvorovo and Varna but that's about it. Kumtepe is very interesting.

Where's the source for Kumptepe being Steppe admixed? And what date?

andrew said...

I explore the question of why this post and related issues matter in a blog post directed at a less knowledgable audience at https://dispatchesfromturtleisland.blogspot.com/2018/11/why-is-knowing-about-maykop-culture-and.html

Chad said...

Hello everyone,

I have a few things to go over here. Just some possibilities and models that work. I think that there is a good chance that things are more complicated than just straight Yamnaya+ whatever flavor of MN European you like. I think that there is a decent chance that things are going to end up closer to Rassamakin than we currently suspect. Once these samples are available, it will be easier to know for sure.

Firstly, Globular Amphora cannot possibly be the source of any MN ancestry in Yamnaya. They do not go into the steppes until the later half of the 3rd M BCE. Even then, they show no evidence of advancing beyond Western Ukraine. So, there goes that. I do get a decent fit with Trypillian samples, but it may not be them that the mixture comes from.

left pops:
Steppe_EMBA
EHG
CHG
Trypillian

right pops:
Mbuti_DG
Ust_Ishim
Karitiana
Onge
Kostenki14
Levant_N
Ganj_Dareh_N
Anatolia_N
Iron_Gates
West_Siberia_N
MA1

numsnps used: 181592

best coefficients: 0.492 0.425 0.083
std. errors: 0.015 0.023 0.022

fixed pat wt dof chisq tail prob
000 0 8 7.268 0.507983 0.492 0.425 0.083

I have a feeling that when the dust clears, and we have more samples, that Eneolithic Ukraine is going to be seen as much more important to the formation of the BA cultures in Central and Eastern Europe. I say this because it is easy to see that there is too much EHG to CHG for ancestry to be directly from Yamnaya into Bell Beaker and Corded Ware. We already have a Corded-like male in Stredny Stog well before Yamnaya existed.

Here is the average for Eneolithic Ukraine, at the best fit...
left pops:
Eneolithic_Ukraine
Ukraine_N
Armenia_ChL
Balkans_ChL

right pops:
Mbuti_DG
Ust_Ishim
Karitiana
Onge
Kostenki14
Levant_N
Ganj_Dareh_N
Anatolia_N
Iron_Gates
West_Siberia_N
MA1
EHG
CHG

numsnps used: 370431

best coefficients: 0.526 0.393 0.082
std. errors: 0.014 0.039 0.034

fixed pat wt dof chisq tail prob
000 0 10 9.907 0.448687 0.526 0.393 0.082

using an EHG/CHG or Khvalynsk sub didn't work. Too much EHG. So, here we may have that Meshoko mix, and influence from the Caucasus into Ukraine that Rassamakin mentioned. I can get references and such later. Just trying to rush this here as I am not sure how much time I can get with my son being tired and ready for bed.

Chad said...

Looking at Yamnaya, we can see the pop that resembles that Caucasus steppe pop..
In other runs than the Trypillian one, which I think could end up being the real deal, Ukraine Eneolithic is more important, with a later pulse from Maykop or Meshoko-related group(s).

left pops:
Steppe_EMBA
Eneolithic_Ukraine
EHG
CHG
Armenia_ChL

right pops:
Mbuti_DG
Ust_Ishim
Karitiana
Onge
Kostenki14
Levant_N
Ganj_Dareh_N
Anatolia_N
Iron_Gates
West_Siberia_N
MA1

numsnps used: 378952

best coefficients: 0.125 0.427 0.354 0.094
std. errors: 0.041 0.023 0.037 0.050

fixed pat wt dof chisq tail prob
0000 0 7 9.197 0.238842 0.125 0.427 0.354 0.094


Moving onto Bell Beaker, there was clearly too much EHG to CHG to come from Yamnaya. Using Khavalynsk also made little sense as there is no movement from this group to Western Europe. Eneolithic Ukraine worked great and would allow for the Balkan influence in ceramics, weapons, and burials that we see in BB.

That Yamnaya + Globular Amphora thing is as good as a failure.

left pops:
C_Beaker
Globular_Amphora
Steppe_EMBA

right pops:
Mbuti_DG
Ust_Ishim
Karitiana
Onge
Kostenki14
Levant_N
Ganj_Dareh_N
Anatolia_N
Iron_Gates
West_Siberia_N
MA1

numsnps used: 371254

best coefficients: 0.584 0.416
std. errors: 0.010 0.010

fixed pat wt dof chisq tail prob
00 0 9 40.674 5.7327e-06 0.584 0.416

Davidski said...

@Chad

You're missing the Eneolithic steppe samples from Wang et al.

But there's a way around that problem. I'll be posting about this later.

Chad said...

Here are much better fits with the Central European BB samples.

In this one, we see that Steppe EMBA, or Yamnaya admixture is as good as none. This is a common theme with BB models. Anything with Yamnaya as the sole source is a terrible fit. Ukraine Eneolithic is the way to go.

left pops:
C_Beaker
Balkan_ChL
Eneolithic_Ukraine
Steppe_EMBA

right pops:
Mbuti_DG
Ust_Ishim
Karitiana
Onge
MA1
West_Siberia_N
Iron_Gates
Kostenki14
Levant_N
Ganj_Dareh_N
Anatolia_N

numsnps used: 308130

best coefficients: 0.343 0.588 0.069
std. errors: 0.017 0.052 0.042

fixed pat wt dof chisq tail prob
000 0 8 7.564 0.477162 0.343 0.588 0.069

left pops:
C_Beaker
Balkan_ChL
Eneolithic_Ukraine

right pops:
Mbuti_DG
Ust_Ishim
Karitiana
Onge
MA1
West_Siberia_N
Iron_Gates
Kostenki14
Levant_N
Ganj_Dareh_N
Anatolia_N
CHG
EHG

numsnps used: 312273

best coefficients: 0.328 0.672
std. errors: 0.016 0.016

fixed pat wt dof chisq tail prob
00 0 11 13.223 0.278978 0.328 0.672

And the best fit, is some Anatolian BA pulse into the Balkans Chl group that mixes with Ukraine Eneolithic. This isn't hard to believe really, considering the movements from Anatolia into SE Europe in the Balkan EBA and just before.

left pops:
C_Beaker
Balkan_ChL
Anatolia_BA
Eneolithic_Ukraine

right pops:
Mbuti_DG
Ust_Ishim
Karitiana
Onge
MA1
West_Siberia_N
Iron_Gates
Kostenki14
Levant_N
Ganj_Dareh_N
Anatolia_N
CHG
EHG

numsnps used: 273464

best coefficients: 0.243 0.097 0.661
std. errors: 0.042 0.044 0.016

fixed pat wt dof chisq tail prob
000 0 10 8.330 0.596667 0.243 0.097 0.661

Chad said...

With Corded Ware, there are two populations it seems. One that may actually have Yamnaya admixture, and another that does not.

First, is the German Corded Ware..

They had too much EHG to CHG to be from Yamnaya, but some input from Khvalynsk does work well..No Steppe EMBA though..

left pops:
Corded_Ware
Eneolithic_Ukraine
EHG
CHG
Armenia_ChL

right pops:
Mbuti_DG
Ust_Ishim
Karitiana
Onge
Kostenki14
Levant_N
Ganj_Dareh_N
Anatolia_N
Iron_Gates
West_Siberia_N
MA1

numsnps used: 394879

best coefficients: 0.486 0.129 0.065 0.320
std. errors: 0.048 0.026 0.041 0.058

fixed pat wt dof chisq tail prob
0000 0 7 9.325 0.230127 0.486 0.129 0.065 0.320


left pops:
Corded_Ware
Eneolithic_Ukraine
Steppe_Eneolithic
Armenia_ChL

right pops:
Mbuti_DG
Ust_Ishim
Karitiana
Onge
Kostenki14
Levant_N
Ganj_Dareh_N
Anatolia_N
Iron_Gates
West_Siberia_N
MA1

numsnps used: 317190

best coefficients: 0.449 0.189 0.362
std. errors: 0.043 0.030 0.026

fixed pat wt dof chisq tail prob
000 0 8 7.100 0.525851 0.449 0.189 0.362

The best fit for Baltic Corded Ware showed some input from a source like Yamnaya. This would probably be female mediated flow though, considering the Ydna isn't any different here.

A couple extra percent from a Caucasus source increased the fit too..

left pops:
Corded_Baltic
Eneolithic_Ukraine
Steppe_EMBA
Armenia_ChL

right pops:
Mbuti_DG
Ust_Ishim
Karitiana
Onge
Kostenki14
Levant_N
Ganj_Dareh_N
Anatolia_N
Iron_Gates
West_Siberia_N
MA1
EHG
CHG

numsnps used: 388154

best coefficients: 0.474 0.473 0.053
std. errors: 0.041 0.045 0.028

fixed pat wt dof chisq tail prob
000 0 10 14.819 0.138803 0.474 0.473 0.053

Chad said...

@ David,

My work-around is EHG and CHG. It is pretty stable with a basic 55:45 split. They need extra Anatolian and WHG. Ukraine Eneolithic may be the winner, but I am guessing like anyone else until the samples are out.

Davidski said...

@Chad

Corded Ware didn't derive from two different populations.

The substructure in Corded Ware was driven by increasing east to west female-biased admixture from European farmers.

This was covered in a recent paper...

A closer look at the maternal origins of the Corded Ware people (Juras et al. 2018)

I'm pretty sure that genome-wide data from these samples are coming soon, and then we'll see that Corded Ware and eastern Bell Beakers are from western Yamnaya and post-Stog populations.

Chad said...

Not two different pops. Both Corded groups are rooted in Ukraine Eneolithic and mixed with two completely different pops. Not the same thing. The writing is on the wall already. R1a M417 is in Stog, but not Yamnaya or Afanasievo. L51 is not in Afanasievo or Yamnaya. They likely never were with how things look. There's just too much EHG to CHG too. It's all lining up this way. It's more complex and not just everyone is Yamnaya plus GAC. That just never jived with any archeo-cultural stuff.

Chad said...

Read Rassamakin on the complex interaction between the Caucasus, Balkans, and Ukraine. I think this makes more sense than some Samara>Western Europe. There's really nothing Yamnaya or GAC about Bell Beaker. Corded is nothing like Yamnaya either.

Chad said...

Gotta crash for now. I can pick this up tomorrow. Just take a look at his work. I think it has some good value. There's others too that aren't popping into mind at this late hour.

Anonymous said...

@Chad Rohlfsen Yes, I agree with Corded Ware being related to SS rather than Yamnaya - it makes more sense in terms of the difference in haplogroups (R1a vs R1b) too, but also physical anthropology etc. I think the formation of the Yamnaya culture displaced the precursors of the CWC to move West across the N Euro plain into Europe, mixing with the natives on the way.

Somewhat similarly, it is clear that all Yamnaya settlements are dominated by Y DNA Z2103, even as far West as Hungary. That, and Corded Ware being R1a dominated (and so both major routes to Germany (the Danube and the N Euro plain) being accounted for), combined with a likely West Mediterranean origin of Y DNA L51 based on the distribution of its "archaic" subclades, makes me think the Bell Beaker folk also are not descended from Yamnaya. How else could L51 have made it to the West, if both land routes are dominated by groups of other lineages? I actually go with Coon here, and think that L51 Bell Beakers, spreading copper smelting technology, reached Western Europe from the East Mediterranean via the old Megalithic sea route, for a whole bunch of reasons (so many I forget most of them, but the bulk would have been mentioned somewhere on this page by me).

The expansion of the BB folk across Western Europe, then, would be prompted by the arrival of the CWC at the Rhine. Coon predicted the route of the BB folk from Iberia to Germany in the first place was by rivers, through the Rhone and Upper Rhine - which, imo not by coincidence, is where L51 subclades are most "archaic". The same phylogenetic analysis can be applied to L23 and Z2103 to show a West Asian origin for both - and I don't think it is a coincidence the first signs of L23 on the Steppe are also the first of Z2103.

The question of the spread of Western IE is more complicated, however. My go-to theory at the moment is that L51 actually didn't speak IE, but adopted it when they began to dominate the Steppe groups they came across - when they dominated the CWC as U106, proto-Germanic was formed, and when they dominated Hungarian Yamnaya as U152, Italo-Celtic was formed. I'm not sure of this, though - but I am quietly confident on a Western origin of L51 and a West Asian origin of L23 and Z2103.

Davidski said...

@An Idiot

Your theories are sort of bordering on fantasy.

I don't really want the comments at my blog to be filled with this sort of stuff. You have to be more realistic and in tune with the data if you want to continue posting here.

Anonymous said...

@Davidski How have I not been reasonable? Do I not sound like a logical person to you, or do I just sound like a frothing Olympus Mons? If I have questions you can't answer, that doesn't mean you can ban me - besides, we more than likely agree on most things. We just happen to disagree on matters related to R1b L23.

epoch said...

@Chad

Would adding extra WHG make better fits for the Beakers. We know that during the entire pre-Beaker Neolithic almost pure WHG samples are found.

Samuel Andrews said...

@Chad,

But what's most important is Steppe Eneolithic, Ukraine Eneolithic, Yamnaya, Corded Ware, and (North) Bell Beaker are all descendants from the same ancestor population. The needy gritty details matter a lot but we can't forget they're all apart of the same phenomenon.

Davidski said...

@An Idiot

Yeah, you sort of sound like Olympus Mons before he went all bitter and twisted because his theories weren't really working out.

Dita said...

Given the mention of bell beakers being dinaric, balkan origins of m269, L23, Z2103, the vinca culture etc. What can we say about Albanians / Proto-Albanian part in this? They are the dinarics par excellence are they not? In the balkans, they have the highest R1b-Pf7562 (very rare), R1b-m269(xl51), and among the highest Z2103 also. The distributions because of this diversity cannot reasonably be argued to be inflated or founder effects. Neither does the distribution support ottoman related late entry. The lack of these specific Albanian R1b clades in south slavs also testifies to their presence since at least deep antiquity, or at the very confirmed least pre slav migrations. Kosovo Albanians also seem to have the higest concentration of these (oldest copper axe in europe found in Prokuplje near kosovo border)

Albanians speak the only surviving paleo-balkan language (which linguists like Hamp and Huld argue represent one surviving branch of a larger Albanoid language group encompassing a much larger territory in antiquity).

Dita said...

Link to relevant L23 distributions http://r1b-pf7562.blogspot.com/2017/02/r1b-pf7562.html?m=1

Ric Hern said...

Looking at MtDNA Haplogroup U5a1b it doesn't seem to have travelled along the Danube but rather a more Northern route...unless I missed something...