search this blog

Saturday, July 28, 2018

A Corded Ware-related Proto-Greek from the Pontic-Caspian steppe?

The recent Wang et al. preprint on the genetic prehistory of the Greater Caucasus features several supposedly already published ancient samples that, as far as I know, haven't yet appeared anywhere. These include five Yamnaya samples from Hungary and two Neolithic samples from Greece. I'm guessing that they're part of a paper that was scheduled to be released earlier this year, but was delayed, and will probably come out very soon.

Intriguingly, one of these new Greek samples, Greece_Neolithic I6423, appears to harbor an unusually high level of Yamnaya-related ancestry from the Pontic-Caspian (PC) steppe. So much, in fact, that in a Principal Component Analysis (PCA) he/she clusters amongst a pair of Corded Ware individuals from Northern Europe, and almost on top of a Varna Eneolithic outlier from Bulgaria, all of whom also pack a lot of this type of ancestry.

So if this isn't some sort of an error, then I6423 might turn out to be a very important sample in the context of the population history of Greece, including in the search for the Proto-Greeks. That's because the ancestors of the Corded Ware people are generally regarded to have been amongst the first Indo-European speakers to migrate out of the PC steppe, and ancient steppe ancestry is now widely accepted to be a signal of early Indo-European expansions across Europe (including those that took Proto-Greek to Greece).

But note that I6423 also clusters near several Eneolithic samples from the North Pontic part of the PC steppe (look for the inverted gray triangles in the PCA). One of these samples is the Corded Ware-like Ukraine_Eneolithic I6561 from a burial associated with the Sredny Stog II culture, which is often said to be a Proto-Indo-European archaeological culture. I've mentioned this sample on many occasions on this blog, including here.

Could it be, then, that the high level of ancient steppe admixture in I6423 is a signal of a surprisingly early Indo-European migration from the North Pontic region to the southern Balkans that led to the formation of the Proto-Greeks? I don't see why not, especially when looking at this map of the spread of corded ware pottery and other typically steppe cultural traits into the region around 4,000 BC (sourced from Bulatovic 2014 here). In any case, I'm really looking forward to getting my hands on I6423, hopefully soon.

See also...

A Mycenaean and an Iron Age Iranian walk into a bar...

Graeco-Aryan parallels

Main candidates for the precursors of the proto-Greeks in the ancient DNA record to date


EastPole said...

These two migrations of R1a dominated tribes from forest-steppe Vistula-Dnieper-Don area, ‘red’ to Balkans and Greece and ‘blue’ to India and Iran, if true, would explain many similarities between Slavic, Indo-Iranian and Greek languages and religions:

It would also explain how Eastern Europe was formed. We don’t need PIE from somewhere else.

Davidski said...


But how do you explain these three things?

1) the existence of Corded Ware individuals almost identical to Yamnaya, even with R1a-M417

2) the seemingly sudden appearance of an individual very similar to Corded Ware/Yamnaya with R1a-M417 in eastern Ukraine during Sredny Stog II

3) the need for Eneolithic steppe samples like those from the North Caucasus Piedmont steppe to model Corded Ware and Yamnaya ancestry

Don't these facts mean that there must have been a population ancestral to Corded Ware, Sredny Stog, Yamnaya and so on, living in a fairly small area of the steppe between the Black and Caspian Seas? And what prevents this population from being PIE?

Arch Hades said...

So he's dated to 4,000 BC? Sounds 1,500 years too early for proto Greek. But who knows.

old europe said...


"And what prevents this population from being PIE?"

The fact that PIE has agricultural vocabulary and there is no sign of agricolture east of the dneper/don before 2000 BC IIRC.
The zone from the black sea and caspian sea cannot be the homeland.

Mikkel said...

@old europe
The usual argument is that the agricultural vocabulary in Indo-European languages are often quite difficult to reconstruct for PIE, and that this vocabulary often seems to be early loanwords (e.g. only European branches and so on). Especially vocabulary for pulses is difficult to reconstruct for PIE, which at least speaks against PIE being in a culture who makes use of such crops.

Furthermore, one might expect especially agricultural vocabulary to be quite stabile since many of these crops have been used from their prehistoric appearance and up until historical times, unlike many wild plants which have fallen out of use with more systematic farming, and thus it is expected that the wild plant vocabulary was often forgotten as well.

old europe said...


quotes from a conference of Mallory (2013)

"All models cited above acknowledge that the Proto-Indo-Europeans possessed an economy based on domesticated livestock and domestic cereals."

"As Anthony remarks in this symposium, there is really no serious evidence for arable agriculture (domestic cereals) east of the Dnieper until after c 2000 BCE (see also Ryabo-gina & Ivanov 2011; Mallory, in press:a). This means that there is also no evidence for domestic cereals in the Asiatic steppe until the Late Bronze Age (Andronovo etc)."

Wastrel said...

I think Arch Hades is right - 4000BC looks way too early for anything that could meaningfully be called 'Proto-Greek'. Indeed, it's debatable whether PIH would have broken up by then. There's probably four viable scenarios for this individual (assuming he's not an outlier immigrant or whatever):

- perhaps at this time an undifferentiated PIH was spoken from the steppe, into the danube, and down into Greece. We know there was a migration from the steppe to the danube around this time, so they may still all have spoken the same language. If this particular dialect has descendents today, it could be anything - greek, sure, but perhaps even more likely celtic (via migration up the danube).

- or, perhaps Anatolian had broken off - late PIE was spoken on the steppe, but Anatolian in the Danubian/Greek area, as a result of the recent migration. The Anatolians later migrated through into Anatolia, and the Greeks followed on behind them.

- or, a different branch of PIH had broken off, but wasn't Proto-Anatolian (which in this scenario went the other way around the black sea, presumably). This different branch has since died out.

- or, PIH was still spoken on the steppe, and these people spoke a related language, which is unattested.

It's always important, when projecting languages back into the past, to remember first that languages move around a lot, and, second, that almost all languages die off without a trace. So historical people will often have spoken languages from 'dead-end' branches of the linguistic tree.

Synome said...

I'm not sure an entry to the Balkans this early would be a good candidate for proto-Greek. We know that elements of Mycenaean material culture show striking similarities to steppe cultures nearby in space and time, but 4,000BCE seems way early. Based on the typical models of IE language divergence, these early steppe migrants might be a better candidate for the ancestors of the Hittites.

Which makes me wonder if there's a new paper about the genomics of early steppe migrants in the Balkans and Danube basin coming soon...

Davidski said...

I don't know the date of this sample. It might be 4,000 BC, but then again it might something more recent.

One of the Peloponnese_Neolithic samples is dated to just 3933-3706 calBCE, so the Neolithic label for Greek samples doesn't necessarily mean they'll be older than Copper Age (Eneolithic) samples from elsewhere.

The Varna_Eneolithic_outlier, also with a lot of steppe ancestry, is dated to an amazing 4711-4542 calBCE, which means that this type of ancestry was already expanding out of the steppe 6,500 years ago. And it also means that it existed as much as 7,000 years ago, or even earlier.

EastPole said...

I don’t think we have enough data and I don’t think I know enough to answer all your questions. I don’t know for example if the formation of ‘Yamnaya component’, i.e. the mixing of EHG and CHG, was limited to “a fairly small area of the steppe between the Black and Caspian Seas” or it was a wider phenomenon and for example the mixing of EHG and CHG could also occur along rivers and in the forest-steppe area:

In case CHG wives were transported by rivers to their EHG husbands, ‘Yamnaya component’ children could also be born in Khvalynsk and Sredny Stog independently of what was happening on North Caucasus Piedmont steppe.

If all steppe component in Sredny Stog, Yamnay and Corded Ware comes from “the North Caucasus Piedmont steppe”, then I agree that there was a “population ancestral to Corded Ware, Sredny Stog, Yamnaya and so on, living in a fairly small area of the steppe between the Black and Caspian Seas.”
But we still don’t know the language, culture and religion of that population.
It could be PIE but we don’t see their continuation in any IE culture. Their Y-DNA for example should be dominant in all IE cultures, or at least significantly present, but as far as I know it isn’t.

It is all speculation and I don’t think we can prove or disprove anything regarding PIE at this stage. I can only say why I like fertile forest-steppe Sredny Stog PIE homeland theory and why I don’t like arid semi-desert North Caucasus steppe PIE theory based on what I know about elements of culture and religion common to Indo-Iranians, Slavs and Greeks.

Davidski said...


I have serious doubts that any CHG population still existed after 8,000 BC for CHG brides to be transported into the steppes.

If such brides were transported into the steppes from the North Caucasus foothills, then they were probably something like 50/50 CHG/EHG. But that didn't last much beyond 4,000 BC, because at that time new populations with ancestry from south of the Caucasus were migrating into the region, and, by and large, they didn't contribute ancestry to Sredny Stog, Khvlaynsk, Yamnaya, etc.

So after about 4,000 BC the bride exchanges basically only happened between the steppe groups, and they appear to have been generally cut off from these new North Caucasus populations, at least in terms of gene flow.

This is why the southern PIE homeland theory now looks like crap, which I think is a very interesting point on its own. But a PIE homeland on the steppe that encompasses Sredny Stog, Repin and Khvalynsk still looks the money.

Open Genomes said...

This is what I'm getting for ANI163 the outlier from the Varna Chalcolithic in Bulgaria, who appears to plot in a similar position to Greece_Neolithic I6423 from the unpublished Wang et al. Genetic Prehistory of the Greater Caucasus study, another Late Neolithic outlier:

Restricted nMonte3 Ward's distance-squared clustering ancestry composition of sample: ANI163 Population: Varna_o Chalcolithic Europe excluding Modern, Medieval, Iron Age, and Bronze Age Samples: Distance 3.51840%

ANI163 is female, with mtDNA H7a1.

With a distance of 3.51840% this isn't a very good fit. However, we can see that she has 38.8% Ukraine Eneolithic, from samples in 3 different clusters.

Including the Bronze Age samples gives us a much better fit, but of course with dates ranging between 4711-4542 calBCE and 4686-4542 calBCE this direct Bronze Age ancestry is impossible.

Restricted nMonte3 Ward's distance-squared clustering ancestry composition of sample: ANI163 Population: Varna_o Chalcolithic Europe excluding Modern, Medieval, and Iron Age, but including Bronze Age Samples: Distance 1.69980%

Here we have Baltic Bronze Age 29.4%, Petrovka (Sintashta) 11.0% and Early Baltic Corded Ware 5.8%.

Maybe there was a somewhat different population than the Ukraine Eneolithic (Sredny Stog and Repin) that were the (partial) ancestors of this woman.

Regardless, given that the Varna cemetery has very elite males (think golden "penis sheaths") it seems more likely that she was an "exotic import" from some trading partners on the northern side of the Black Sea, with ancestry from a group close to the PIE speakers, but who have not yet been sampled.

Samuel Andrews said...

"I have serious doubts that any CHG population still existed after 8,000 BC for CHG brides to be transported into the steppes."

Have you abandoned the Caucasian bribe theory? I'm continuing to look at mtDNA which should give some answers on the formation of 'Steppe.'

Ric Hern said...

Well I think the problem is how to establish if a language recieved Late Proto-Indo-European influence upon an earlier form of Proto-Indo-European since there are no written guidelines...the retention of Archaisms could point to an Earlier Proto-Indo-European substrate. Who knows ?

Davidski said...

@Samuel Andrews

Have you abandoned the Caucasian bribe theory?

I haven't, because it's obvious that all of those Yamnaya and Yamnaya-related populations are a mostly mixture of EHG males and southern females, one way or another.

I call bullsh*t on any claims that strong founder effects during the Eneolithic and Bronze Age got rid of the signals of male migrations from south of the Caucasus to the steppe, for the simple reason that during the Eneolithic we see a high diversity of EHG-related Y-chromosome haplotypes on the steppe belonging to R1, R1a, R1b and Q.

Actually, I'm shocked that this was totally ignored in the Wang et al. preprint. This and the fact that there was no migration from south of the Caucasus that led to the formation of Yamnaya. The phrase WTF comes to mind.

Samuel Andrews said...

"I haven't, because it's obvious that all of those Yamnaya and Yamnaya-related populations are a mostly mixture of EHG males and southern females, one way or another."

But when did this sex bias admixture event happen? Before, it seemed you argued it happened during and before the formation of Khvalynsk. Do you now think it happened earlier?

Davidski said...

@Samuel Andrews

But when did this sex bias admixture event happen? Before, it seemed you argued it happened during and before the formation of Khvalynsk. Do you now think it happened earlier?

In the southern part of the steppe it probably mostly dates back to the Mesolithic (aka pottery Neolithic) or the very early Neolithic.

But in other parts of the steppe it was still happening during the Eneolithic, because the Khvalynsk samples from Samara show something like 20% to 40% CHG admix. And then there was also that Globular Amphora-related gene flow from the west, plus the likely occasional migrant from the Caucasus like that Yamnaya_Ozera female.

It's actually remarkable how old and consistent this process appears to be. It's more than a cultural thing, it seems to have been a way of life on the steppe. But outside of the steppe we do see Y-haplogroup J in some of the EHG males, so they occasionally made it to Eastern Europe from the Caucasus, but couldn't make an impact on the steppe.

Ric Hern said...

Well apparently there were several aridification periods affecting populations within the Pontic Caspian Steppe. One such period started around 6300 BCE and another 4500 BCE. At this time (4500 BCE) there was apparently a migration of Lower Don Sredny Stog people towards the West. This event also hit the Balkans hard. So maybe the Earlier Steppe migration didn't encounter much resistance...?

Mikkel said...

@old europe
If you look at the "domesticated plants" chapter in Mallory/Adams 2006 it is quite clear that yes PIE did know of agriculture, but also that much of that part of the vocabulary either denoted very general concepts like "grain" or "husk", etc. or that the meaning of certain words appear as very different crops in the daughter languages. Since some agriculture has been found west of the Dnieper which could also have harboured some PIE dialects, the "instability" of this vocabulary fits quite well with agriculture being only a periferal or trade-related phenomenon, and that the more general meanings perhaps originally describing wild crops (e.g. Chenopodium and Amaranthus found almost everywhere in steppe sites), and the processing of these (like "grinding", etc.).

Here is also an excerpt of a summary of the problem in Guus Kroonen and Rune Iversens "Talking Neolithic" article from 2017:
"The Yamnaya pastoral-nomadic lifestyle also explains
the relative paucity of Indo-European vocabulary
for crops and land cultivation, which have proven to be
far more difficult to reconstruct. While such terms are
not completely absent (cf. *ieuo- “[unspecified] cereal”;
*ses(h1)-io- “[unspecified] cereal”; *g´rh2-no- “grain,
kernel”; *h2erh3- “to plow”; *k´ok-H- “[forked] branch,
plow”), names for key founder crops such as flax and
the primary legumes (lentils, peas, and chickpeas) typically
cannot be reconstructed for Proto-Indo-European
because the different Indo-European dialects use dissimilar
and unrelated terms, terms that are typically
also limited to specific regions in Europe or Asia. Such
regional words have pervaded the literature on Indo-
European linguistics, along with references to their
problematic origins. The Indo-Europeans almost certainly
had some knowledge of agricultural practices, in
other words, probably through trade with the farming
communities to the west of the Dnieper or the North
Caucasus, but they possessed nowhere near the full
crop assemblage that spread from the Fertile Crescent
during the earliest phase of the Neolithization."

epoch said...

@old europe

"The fact that PIE has agricultural vocabulary and there is no sign of agricolture east of the dneper/don before 2000 BC IIRC.
The zone from the black sea and caspian sea cannot be the homeland.

With the uptick in EEF/WHG related ancestry in Andronovo, possibly being a Corded Ware influx, we have a perfect solution for that.

Palacista said...

I think it is far more likely that you have found the route taken by proto anatolian speakers. The linguistic evidence points to a Greek-Armenian-IA late connection that excludes the other branches that is not consistent with the early date.

Dmytro said...

" a PIE homeland on the steppe that encompasses Sredny Stog, Repin and Khvalynsk still looks the money." (Davidski)

I think this is basically correct and in line with all available evidence. I would only add that the northernish boundaries of the homeland may well have included some small areas of the forest-steppe and even tiny "forest" links. Esp. since archaeologists have long known that elements of the formative Repin culture were absorbed from there. And another small point: some of the CHG could also have been sources from totally (so far) genetically non-analized cultures like Lower Mykhajlivka and Kemi-Oba (both of which had very strong archaeological links to Majkop). But the main point of the steppe PIE origin remains solid.

Matt said...

Re: the discussion on y-chromosomes and Wang's preprint, on this point about the Wang preprint, one interesting thing here is that the Steppe Maykop and the Lola individual, who all look to have ancient West Siberian (Botai-like) ancestry have the Q1a2, while all rest of the Steppe Cluster that lack this admixture are R1b1 or R1b1a2.

One exception being a Steppe_Maykop outlier SA6013 who has R1 (but they can't call it any more specifically) and probably the most Caucasus ancestry (though I don't think this indicates the R1 ame from Caucasus just to be clear, since the data doesn't go that way). (There's another Steppe Maykop outlier male who sadly they can't / don't call - sample id: IV3002.A0101).
Q1 is found in EHG, however seems much more frequent in the Botai culture, who judging by early domestication of horses, and the time of the culture (3700-3100 BCE) and lack of any input from the Western Steppe groups, and probably have more dense/advanced way of life than any EHG groups that are about. And this would fit with West Siberian component being introduced to Steppe_Maykop / Lola.

Richard Rocca said...

An important distinction should be made between "Corded Ware" and "corded ware". The former is a culture which spread into Central Europe around 3000 BC. The later is pottery that has a decorative technique applied by a cord. So, the 4000 BC date is not for the Corded Ware Culture.

Matt said...

Re: Eneolithic Caucasian ancestry spreading north of the Caucasus, I agree with Davidski's summary.

In a more longwinded way, the pattern is signals of Eneolithic Caucasus ancestry do show up north of the Caucasus, but crucially, like Davidski says, not seen in the earliest Eneolithic North Caucasian steppe samples (4000 BCE), and patchy thereafter, even if there is a general trend.

It shows up in the Steppe Maykop outliers (substantially; 45%), fairly north at around (3200BCE), and slightly later Yamnaya and North Caucasus culture samples from largely south of these sites (Marinskaya in the foothills of the Caucasus / Rasshevitsky 3000 BCE), at lower levels (10%). More substantially at Late North Caucasus (2200 BCE) at 20%.

We do see this slow bleed of Caucasian ancestry onto North Caucasian steppe over time, likely with punctuated pulses of long distance migration as indicated by Steppe Maykop outliers. But by we see substantial levels (even 10%?), looks like the Eneolithic North Caucasus steppe groups long since expanded north into North Pontic to form groups at eastern Ukraine, without acquiring Eneolithic Caucasian ancestry. Then they accumulate low ancestry from the expanding farmer complexes of Central-East / South-East Europe, and maybe some local EHG. While complexes with no Caucasian Eneolithic ancestry persist in the North Caucasian steppe as well by 3000BCE (e.g. Catacomb).

(Khvalynsk seems is intermediate Eneolithic North Caucasus steppe groups and EHG. Some individuals almost 100% derived from Eneolithic North Caucasus steppe. Just comparing the average dispersal, Khvalynsk, in contrast to the Eneolithic North Caucasus steppe and Yamnaya, much less a stable cluster with varied levels of admixture between EHG and Eneolithic North Caucasus steppe).

Probably going to be some transfer of domestic animals, ideas, technologies from the Eneolithic Caucasus to the Eneolithic North Caucasus steppe who go on to form the bulk of Yamnaya ancestry. At the same time, looks like it would be mediated by a few pioneers from the Caucasus who don't affect ancestry (or vastly most likely language) of Eneolithic North Caucasus steppe groups?

This is probably seeming like one of my normal longwinded, pedantic posts ;) (didn't start out intending to be this long). There's a lot of confusion about when Wang say the Caucasus was a corridor for migration, and yet the Yamnaya samples largely show no indication of this. The two things are not really in conflict; signs present of ongoing migration from the Caucasus over about 2000 years but this seems not not genetically important for the Yamnaya culture at Samara and Ukraine (or subsequently wider spread of Yamnaya like ancestry as far as the Altai through the Afanasievo). Though could have importance in transmission of culture and ideas.

Yamnaya culture, at Samara and Ukraine, seems to form from early dispersal of Eneolithic North Caucasian steppe groups, practicing pastoralism and probably early domestication of the horse (separately from Botai people), coming into contact with farmers expanding out of SE / CE Europe using wagons to practice agriculture in drier grasslands.

Likely low mutual exchange of female cross-cultural migrants between those groups (steppe women into Eastern European farmers and vice versa). However, looks like the tides turn to favour wagon+horse+pastoralist complex over wagon+horse+agriculture complex (e.g. climate change and "Dustbowl Theory" of the end of Cucuteni-Trypillian or better synergies). Then massive long distance, oft male bias expansion of pastoralist cluster.

Same process probably gave rise to Corded Ware culture, though maybe also introgression of local EHG at a low level? Which may be related to CW R1a as well? (Extra EHG such as Khvalynsk seems unnecessary for Yamnaya; so genetic dynamics of Khvalynsk look peripheral to formation of Ukraine Yamnaya cluster, even if culturally significant).

How it looks; lots of fine scale things to test, when data is released, see if this is all right.

old europe said...

As for lentil

type of annual leguminous plant, also its edible seed, mid-13c., from Old French lentille "lentil," also "a
freckle" (12c.), from Latin lenticula, diminutive of Latin lens (genitive lentis) "lentil plant, a lentil," cognate with Greek lathyros "pulse;" Old High German German linsa, German linse "a lentil;" Old Church Slavonic lęšta, Russian ljač.

as for chick pea

chick-pea, 1712, a false singular of chich-pease (1540s), earlier simply chich (late 14c.), cich, from Old French chiche "chick-pea" (13c.), from Latin cicer "pea," which is of uncertain origin, but with likely cognates in Greek kikerroi "pale," Armenian sisern "chick-pea," Albanian thjer "lentil." The Latin plural, cicera, is also the source of Italian cece and was borrowed into Old High German as chihhra (German Kichererbse).

As for bean

Old English bean "bean, pea, legume," from Proto-Germanic *bauno (source also of Old Norse baun, Middle Dutch bone, Dutch boon, Old High German bona, German Bohne), and related to Latin faba "bean;" Greek phakos "lentil;" Albanian bathë "horse-bean;" Old Prussian babo, Russian bob "bean," but the original form is obscure. Watkins suggests a PIE reduplicated root *bha-bhā- "broad bean;" de Vaan writes that the Italic, Slavic and Germanic "are probably independent loanwords from a European substratum word of the form *bab- (or similar) 'bean'."

These are all PIE word; the fact that the Indo-Iranian branch lacks it it is due to the particular history of that branch and its mixing with other cultures and the indian subcontinent.
Come on, how likely is that western IE language all borrowed these terms independently from a non IE source?

also PIE has words for village (*weik) and fortress ( *burg).
presumption stands for IE to be "settled" people and not nomadic.

Ric Hern said...

Looks like there was a ping-pong thing going on of Related and Semi-Related Peoples between Derievka and +-Samara and Derievka and the +-Kuban River since 8000 BCE. due to sporadic aridification....

epoch said...

@Open Genomes

"Regardless, given that the Varna cemetery has very elite males (think golden "penis sheaths") it seems more likely that she was an "exotic import" from some trading partners on the northern side of the Black Sea, with ancestry from a group close to the PIE speakers, but who have not yet been sampled."

Considering that the area was in the process of being populated by Suvorovo culture folks, who buried in Kurgans and gave zoomorphic scepters as grave gifts I am willing to bet a lot she was part of that. As was the nearby contemporaneous Smyadovo I2181 (Male, Y-DNA: R1).

Mikkel said...

@old europe
I think it's quite likely that they come from a single (widespread)
Non-IE source (but actually more like different related dialects that were spread with farming cultures and peoples) considering that steppe people mixed with farmers in most of Europe when they arrived, also considering that many of these words (show some sound irregularities which is why the dictionaries often write things like "uncertain etymology").

And when many dictionaries write PIE it is not necessarily the PIE spoken in the "homeland ", but rather a linguistic stage that could have been spoken only by for example corded ware and bell beakers.

Davidski said...


I agree that Greece_Neolithic I6423, if dated correctly, seems too old to be a Proto-Greek, and might actually be a Proto-Anatolian.

So as far as potential Proto-Greeks are concerned, I'm still betting on one of the possibilities discussed in this blog entry...

Main candidates for the precursors of the proto-Greeks in the ancient DNA record to date

Mikkel said...

For anyone interested in the question of Indo-European agricultural and domestic animal terminology (and more), I can recommend this searchable database (also with maps), which was/is a project headed by Guus Kroonen.

I was so lucky to work on it myself for a while as a student :)

As far as I know it is still incomplete, but it's already a great tool which I have found useful in many instances.

The most easy way to use it is to search for the meaning of a word (e.g. 'lentil', og use the "Filter by tags" function provided (or a combination of this).

Ric Hern said...

Thanks Mikkel.

Santosh said...

Slightly out of context. Razib Khan had suggested last week, that there is an Indian paper being published this week. Presumably Rakhigarh. Any news about that paper?

Bob Floy said...

That paper will never be published.

Davidski said...


If it's going to come out this week, then probably on Thursday.

Hopefully when it does come out there's nothing too crazy in the paper, like in that last press article in which it was claimed that the Aryan Invasion/Migration Theory was debunked because there was no steppe ancestry or R1a in the Rakhigarhi samples.

Obviously, according to the AIT, Harappans, like those at Rakhigarhi, weren't Indo-Aryans, so they shouldn't have any steppe ancestry, unless maybe during the final phase of the Harappan Civilization.

Steven said...

Does this mean that some of the genetic similarity that modern Greeks share with Slavs could be ancient in origin?

Onur Dincer said...


I am trying to make a Global 25-based nMonte test setting to test for the proportions of the Neolithic Anatolia/Greece, steppe, CHG/Neolithic Iran and extra WHG ancestries of various modern and ancient Greek populations (from the Greek mainland all the way to northeastern Anatolia). Which ancient populations would you recommend for this setting?

Davidski said...


You'll have to break that up into two or three different runs, each one focusing on different periods. I'm thinking...

- Mesolithic/Neolithic

- Chalcolithic/Bronze Age

- Iron Age/Modern

Onur Dincer said...


My test will focus on deep ancestry. I may include the Neolithic Levant as well in the references. I am especially undecided about which population(s) to use to represent the steppe ancestry.

Davidski said...

Try Yamnaya_Samara.

Onur Dincer said...


Try Yamnaya_Samara.

Thanks. That was what I had in mind initially. But after reading your posts about Sintashta- or Corded Ware-like steppe ancestry in Mycenaeans, I considered using a different strategy. Nevertheless, as my test will focus on deep ancestry, on the different poles of deep West Eurasian ancestry, it is probably a better idea to use Yamnaya_Samara to represent the deep steppe ancestry in Greeks. As a Neolithic Anatolia/Greece representative I may use Barcin, for WHG Loschbour, for CHG/Neolithic Iran Ganj Dareh, and finally the Neolithic Levant samples for deep level contribution from that part of the world. What do you think about my model? I may adjust my model based on your recommendations.

Davidski said...


Yeah, if you're testing deep ancestry, then for the western steppe you need something like those Eneolithic_steppe samples from the North Caucasus steppe. But they're not available yet, so the best alternative is probably Yamnaya_Samara, and in particular Yamnaya_Samara:I0429.

EHG, or even Khvalynsk, won't do, because they don't have enough of that CHG-like ancestry from the steppe.

But of course, the steppe ancestry in the Mycenaeans isn't really of that type, but of the Sredny Stog II/Srubnaya type. So it might be useful to see how that works as well.

Onur Dincer said...


Thanks for the recommendations.

But of course, the steppe ancestry in the Mycenaeans isn't really of that type, but of the Sredny Stog II/Srubnaya type. So it might be useful to see how that works as well.

I may do runs with Srubnaya instead of Yamnaya as well for precision in the calculation of steppe ancestry.

Arza said...

Re: Mycenaeans

page 43 (44 in pdf)
Modeling admixture from ghost populations
Table S2.25: Modeling Mycenaeans via cline intersection.

Comparison of the composition of real and ghost populations used to model Mycenaeans shows that the source of steppe ancestry was more Europe_MNChL-shifted than Steppe_MLBA or even Europe_LNBA cluster. In other words it was more like Western CWC, BBC, Czech_EBA etc.

68.5% Steppe_EMBA 31.5% Europe_MNChL

53.1% Steppe_EMBA 46.9% Europe_MNChL

48.5% Steppe_EMBA 51.5% Europe_MNChL

IMHO this rules out direct migration of Srubnayans and it leaves us with either 2-stage migration with an intermediate population in the Central Europe/Balkans or a Central European source of migration from the very beginning.

Davidski said...


I never claimed that there actually was a migration by the Srubnaya population to Greece that gave rise to the Mycenaeans. I always used Srubnaya as a proxy for this.

K33 said...

Any other details? Date and location? Was this in northern Greece or the Peloponnese?

If this genome is dated pre-Yamnaya (pre- 3,300 BC?) my first guess would be that he was a pre-Anatolian speaker. The Mycenaean Greeks likely came from a later steppe-related group in the north Balkans/Carpathians and possibly there was a "Dorian" wave from the same region a few centuries later.

Groo Salugg said...


In a comment some time ago you said that the least admixed Sintashta samples are closest to Czech CWC.
A migration from Central Europe to Sintashta, and another independent migration from that place to Greece, would explain the Sintashta-like artefacts in shaft graves in a simpler way than a migration from Central Europe to Sintashta and then all the way back to Greece.

Davidski said...

@Groo Salugg

That's correct, the population most similar to the main Sintashta cluster are the two CWC_Czech samples (not including the third one, the obvious outlier with farmer ancestry).

But this doesn't necessarily mean that Sintashta derives from any population in Central Europe, or even from just east of Czechia, like the forest steppe of western Ukraine.

The most likely scenario still is that the population ancestral to the Sintashta people came from the forest steppe north of the Black Sea in Ukraine, and the Sintashta culture formed in the Trans-Ural steppe east of Europe. See here...

The mystery of the Sintashta people

The Mycenaeans are related to this in a profound way, but it's still not exactly clear how. As I see it, there are now two main options:

- the proto-Mycenaeans migrated to the Aegean region from the steppe during the Late Neolithic/Early Bronze Age, and then were influenced in a big way later by Sintashta-related peoples who brought with them chariot technology and horse know how

- the proto-Mycenaeans were migrants from the Middle Bronze Age steppe and basically a sister group to the Sintashta and Srubnaya peoples. The Multi-cordoned ware culture (KMK) comes to mind.

Davidski said...

Actually, what I meant to say is that the non-steppe population most similar to the main Sintashta cluster are the two CWC_Czech samples.

Steven said...

From where in Greece does I6423 originate?

Davidski said...

We don't know yet. Details about this sample haven't been released.