search this blog

Wednesday, April 11, 2018

Bronze Age Central Asia: terra incognita no longer

I've updated my Global25 datasheets with the samples from the Narasimhan et al. 2018 preprint (look for these labels). Feel free to use this output for anything you like, and please show us the results in the comments below.

Global 25 datasheet

Global 25 datasheet (scaled)

Global 25 pop averages

Global 25 pop averages (scaled)

Also, here's my Principal Component Analysis (PCA) of ancient West Eurasia featuring most of the new samples. Note the cline made up of ancient and present-day South Asians running from the likely Indus Valley diaspora individuals (from the Gonur Tepe and Shahr-i Sokhta archaeological sites, in present-day Turkmenistan and Iran, respectively) towards the Bronze Age steppe. The relevant datasheet is available here.

I have little doubt that these are indeed migrants from the Indus Valley Civilization (IVC). Their relatively unusual genetic structure - which includes ancestry from an West Eurasian ghost population that is inferred to have been exceedingly poor in Anatolian-related ancestry, as well as significant indigenous South Asian ancestry - leaves little scope for plausible alternatives. If you're wondering what they may have been doing so far north of the IVC, Frenez 2018 has a detailed discussion on the topic. From the paper:

An alternative and intriguing hypothesis is instead supported by significant archaeological and textual data from comparable socio-economic or geographical contexts, which suggest that the likely high commercial and ideological value of ivory and of the expertise required to carve it made also possible and economically profitable the presence in Central Asia of independent itinerant ivory carvers native to or trained in the Indus Valley. These itinerant artisans might have provided at the same time both the raw material and the unique skills to transform it into finished objects.


Moreover, the existence of itinerant ivory workers in ancient South Asia is also described in a few literary sources. The Guttila Jātaka mentions a group of ivory carvers who traveled from Benares to Ujjain to offer their products and skills to the local elites (Pal, 1978: 46), while a Buddhist Sanskrit Vinaya tells the story of an Indian master ivory carver who traveled “up to the land of the Yavanas”, most likely the Hellenistic Bactria, to put his superior expertise at the service of a renown local artist (Dwivedi, 1976: 19).

Citation: Frenez, D., Manufacturing and trade of Asian elephant ivory in Bronze Age Middle Asia. Evidence from Gonur Depe (Margiana, Turkmenistan), Archaeological Research in Asia (2017),

See also...

On the doorstep of India


«Oldest   ‹Older   201 – 255 of 255
Samuel Andrews said...


Just like with the Myceneans, you chose to tout the lowest Steppe ancestry estimate in Iron age Pakistan. You have either not tested them with nMonte or using horrible models that include ANE-rich non-Steppe sources which eat up their Steppe ancestry such as ancient genomes from Armenia.

There's eight Iron age Pakistan subgroups. Here are their scores.

West Siberia N-3.6%

Barikot_IA. Yamnaya-18%, West Siberia N-3.7%
Butkara_IA. Yamnaya-8.8%, West Siberia N-5.5%
Katelai_IA. Yamnaya-15.3%, West Siberia N-3%
Loebanr_IA. Yamnaya-16.2%, West Siberia N-5%
Saidu_Sharif_IA. Yamnaya-15.4%, West SIberia N-5.2%
Saidu_Sharif_IA_o. Yamnaya-3%, West Siberia N-4.5%
Udegram_IA. Yamnaya-14.2%, West Siberia N-4.7%.

"Thirly, I am right to castigate you guys. With a few exceptions you seem to be steppe ideologues light on knowledge and heavy on the talk"

I'm not a Steppe fanboy as you have called me before. I consider other possibilities for the origin of IE. I consider an origin south of the the Caucasus because Armenians seem to almost completly lack Steppe admixture.

You tend to make inaccurate assessments of people.

"There is evidence of steppe admixture in SA, but its far from the reality you masterminds predicted. Anyone genuinely interested in prehistory will now reconsider things and come up with newer or adapted models to explain the expansion of PIE because your whizz -bang scenario has just been falsified. "

Yes, the admixture in Swat is not what we expected. But Swat doesn't represent all of South Asia between 1500 and 1000 BC. Let's wait for more ancient DNA. I never expected pure Andronovo in South Asia considering how little Andronovo ancestry modern Indians have.

The evidence that Indo iranian languages are from the Steppe (ultimatly Sintashta-Andronovo) doesn't come from India. Y DNA, mtDNA, Scythian DNA, Tajik, Kalash, Pathan is really what confirms where Indo Iranian languages are from. Hindi is not the only Indo Iranian language bro. If Iranian is from the Steppe then Hindi is too.

With confirmation, Steppe people moved into India & Greece at the right time the Kurgan hypothesis is alive and well. You are too picky. Would 30% instead of 15% Steppe admixture in Swat and Myceneans make the Kurgan hypothesis more likely?

The exact percent of Steppe admixture doesn't matter so much. Just because everywhere early Steppe folk went they didn't cause massive population replacement doesn't mean they didn't spread IE languages. 15%, 10% admixture it doesn't matter.

Look at modern Turkey or Azeri. They don't have much ancestry from the original Turkic speakers. I wouldn't be surprised if the typical Turkish individual in the 14th century only had like 15% actual Turkic ancestry from Central Asia. None the less, Turkic languages did originate Asia. The same is true for Steppe ancestry in Myceneans and Swat.

Davidski said...


Armenians do have steppe ancestry and Andronovo does work for Indo-European-speaking South Asians, because they have Andronovo ancestry.

So I have no idea WTF you're on about. Nor do I care. But I felt obliged to call you out on your bullshit for the benefit of others.

Rob said...

@ Sam

There you go again. When did I propose OIT?
And your estimates are way off. 20% Yamnaya in Swat ? You're dreaming boy
My estimates of 5-15% MBA steppe match Alberto's & Seins

And yes Myceneans have 5-15% Yamnaya
Could it be proof of steppe hypothesis ? Yes it could be but im not going to jump up and down about it just yet

Anthro Survey said...


Beggars might be a bit of a strong word, but, yes, it makes sense for them to have been on the edge of civilization for some time.

As for Swat----the alignment between pots and people isn't always flush and there were other opinions about that, as well. To be honest, I was never really holding my breath on it. Then, ofc, there's the issue of castes.
In general, there is not a lot of consensus in terms of material evidence as to which archaeological horizons in India were a result of "Aryan" movements. That's because they're just as discontinuous with steppic traditions as the bronze age cultures of Central Europe, and the latter actually packed a lot of well-distributed steppe ancestry.


Your Yamnaya for Swat is too high. Best to use Sintashta/Andronovo for south asians and iranians.

Seinundzeit said...


"Alright thanks. I suppose we will have to patiently wait for Neolithic and Mesolithic samples from South Asia. I hear that there are ongoing attempts to get aDNA from such samples. God willing they will succeed and we may see the paper in a couple of years."


"But does the use of Sarazm + Iran N + AASI obviate the need to use any steppe ?"

Steppe is still needed for Brahmins, and southern Central Asians.

"Can the Afanasievo be modelled as EHG/WSHG + Sarazm_EN + ANF ?"

I can give that a spin.


"Please do."

Here you go brother (I scaled this manually):


It's a rough piece of work. Far from perfect, but much better than the Onge.


"I'm getting very similar results with the scaled sheet by using just Onge as AASI stand-in. Have you tried your zombie on the scaled sheet?"

Yes sir.

In truth, I was into scaling by the sqrt of each eigenvalue before we had the scaled sheets.

So, since I was too lazy to manually scale, I would use (and continue to use) a modified version of nMonte which scales the data.


"The non-ASI part of Indus P is like 85% Iran_N + 15% AG3/ANE (so approx 37-38% ANE proportioned), and the non-CHG part of Sarazm is like 75% Iran_N, 17% AG3/ANE, 8% West Siberia N (that's something like 40-43% ANE proportioned) from the stats you posted above.

I think Indus_P is really just Neolithic South Central Asia (Sarazm percursor) plus extra Iran_N of some sort.

It would also explain the Baloch. Why a population like the Baloch, who seem similar enough to everyone else, generate components that are really high in ANE (zombies of the component usually have 35% ANE or more). That component is always labeled as South Central Asia. I think that's a shadow of the real South Central Asia which was, obviously, in South Central Asia. For whatever reason, the algorithm couldn't reconstruct it from modern Tajiks/Afghans as well as it could from Baloch. It could be due to the influx of CHG and Chalcholithic-era Caucasian (i.e, Iran_Chl) type admixture into the region. Even Sarazm_EN is chock full of CHG."

Very interesting thoughts.

Anthro Survey,

"I'll post models later where Pamiris and Pashtuns are successfully modeled as a combo of Steppe_MLBA and Scythian-related Zevakinski_LBA. This makes a good historical sense, imo. I would have modeled Jats this way, too, but Dave removed them from the datasheet. :-("

True; Pamiris and Pashtuns show a mix of Steppe_MLBA and Scythian.

Speaking of which, the Pashtun average in Global_25 isn't optimal. It's essentially the most Pakistani/northwest Indian HGDP Pashtuns, with 3 of the 4 Afghan Pashtuns.

So, you guys might profit from using this average.


I was too lazy to manually scale this (again, I have a modified version of nMonte which scales everything by the sqrt of the eigenvalue).

4 Pakistani Pashtuns, and 5 Afghan Pashtuns, went into this average. Very diverse bunch, so the average is concealing things, but I can't share the individual results.

Regardless, much more representative than the average in Global_25. Give it a spin.


Today was really hectic, so I didn't get a chance to post the output I mentioned yesterday. If things are slower tomorrow, I'll make sure to post the results. It's a very solid setup.

Mr. Kulkarni said...

@anthro survey
Yes IVC proper ancient samples, from near sindhu and saraswati should be good.

"Also, it won't be easy to find remains from proto-Brahmins: cremation."
Imagine me using same argument for IVC, where cremation seems to have been most common.

mzp1 said...

Dave's theory that EEF in MLBA was just confined to the elites doesn't work because there is sufficient time for it to propogate down and diffuse into the population.

There is EEF in the swap sample, no? And in modern pashtun.

How do we know the emba in South Asia is truly from the steppe?

Davidski said...


I don't have that theory any more, because the new data from South and Central Asia clearly show that South Asians have Andronovo ancestry.

See that's the point of getting new data, to improve theories and work out what really happened.

epoch2013 said...


If there are no Kurgans outside the steppe yet the Rigveda and Iranian sources mention them, then we have proof of steppe culture in both sources.

"Borrowings into Uralic are Iranian, if somewhat Indo Aryan shifted"

No. Some borrowings in Uralic are Iranian. Not all.

Kristiina said...

@ Kulkarni
Yes, it is possile that there are more Old Indian words than Iranian words but could this be due to Old Indian being better documented? In any case, I have not made any calculations. Parpola could maybe answer to this question. However, there may even be more common words than expected. IMO, the amount is very significant, in particular in Finnish, but this may be due to my much wider knowledge of Finnic roots. It is also possible that part of the common roots go to PIE level.

I have tried to make a more comprehensive picture of the distribution of roots, but it is very time-consuming. IMO, It is problematic that PIE keeps hoarding roots. In Encyclopedia of Indo-European culture by Mallory and Adams there are 18 roots for ’bend’, twelve for ’bind’, seven for ’branch’, twelve for ’burn’, six for ’fear’, seven for ’field’, seven for ’goat’, seven for ’grain’’, eight for ’grow’, five for ’axe’, five for ’water’ etc. Modern languages usually have one or two.

I expect to see R1a1-Z93 and N1c in Abashevo (ca. 2500–1900 BCE); should this be proven, there is still surprisingly little Z93 in Uralic speakers. If there is no Z93 or N1c in Abashevo, I am lost. In that case, the only possibility seems to be the Corded Ware related Fatyanovo (3200 BC–2300 BC).

epoch2013 said...

@ Kristiina

Am I correct that you are a linguist? I have been thinking about the possibility that PIE was a mixed language. I know that there are mixed languages, i.e. Michif, that maintain a complex structure and have kept a lot of the vocabulary of both languages that constructed it. It would explain a number of things: The abundance of PIE roots you describe, the strange likeness of the most basic words between Proto-Uralic and PIE. It would fit the origin of Yamnaya as a merger of several cultures.

Anthro Survey said...


Cemetary H would be even better.

Well, you can use the same argument for IVC, but I guarantee you there was no steppe_mlba-like affinity in early IVC elites. Perhaps differential amounts of Iran_N vs AASI across social strata would make more sense.

Anthro Survey said...


Thanks for that. I'll incorporate it.

Kristiina said...


Yes, I have been thinking about the same and I have often used the expression "Bronze Age modernization process" which has formed the modern populations.

I found Open Genomes's analysis of Hajji_Firuz_ChL_o:I2327 very interesting. I have been thinking that there was a bridge between the Balkans and Caucasus/Iran and a new modern culture radiated to the steppe and to South Central Asia and South Asia.

Davidski said...

@Open Genomes

Your model and analysis aren't realistic.

The R1b Hajji Firuz individual is just a mix of Hajji Firuz ChL and something from the steppe.

Davidski said...


Something definitely radiated from the steppe to Iran during the Bronze Age.

When more samples come in from Early to Middle Bronze Age Hajji Firuz I'm betting we'll see some 100% Yamnaya clones among them.

Anthro Survey said...

Alright everyone,

Gave xMix a try and modeled virtually all of the South, Central and South-Central Asian populations in G25.

Here it is. Target and input populations are color-coded by region/macro-ancestry.

You may notice that I've used two ASI-rich proxies and two Iran_N-ANF populations, but I allow these degrees of freedom so that:
1.populations are given correct drift
2.deep ancestry(Iran_N, ANF, AASI) is properly accounted for

Somewhat over-fitted for certain populations, but the object was to get a sense of preferred ancestry. In fact, some inputs were hardly ever preferred by most populations.

Nothing too unexpected. I highlighted some key numbers--expected and unexpected alike---in red.

-In this run, Scythian(Zevakinski) affinity to Pashtuns and Pamiris is reduced a bit. Could be that it got "decoupled" into Han and Sintashta_MLBA somehow. Han was not used in my previous models.

-Baloch/Brahui: Seem to derive their Iran_N and Iran_Chl ancestry from a slightly different "branch" of migrants than the rest or these are really drifted populations. If neither Iran_N nor Iran_Chl are included, the fits fail. In this model, I've included Iran_N, so they are OK but still a bit borderline. Passing fit distance should be no more than ~2.5 for Brahui and no more than ~3 for Baloch.

-Paniya is preferentially used by some Indian groups while none at all by other Indians nor SA at large.

-Sindhis get the most Near Eastern affinity among Indians.

Rob said...

@ Anthro

"Beggars might be a bit of a strong word, but, yes, it makes sense for them to have been on the edge of civilization for some time. "

That wasn't a serious explanation, just a joke.

Matt said...

@Davidski, oh well thanks for trying.

The test topology was meant to be this basically:

It may be too complex to decide, though I also note I got one node wrong so that might be it.

Can you try the following variations, a corrected version of the last tree and some simplified trees:

1: Last tree (corrected that one node error from the last qpGraph file):

2: Simplified1: (only Yamnaya, Barcin, Abdul_Hosein_N, Gonur2_BA)

3: Simplified2: (only Barcin, Abdul_Hosein_N, Gonur2_BA)

4: Simplified3: (only Yamnaya, Abdul_Hosein_N, CHG, Gonur2_BA

If these don't work, I'll probably call it a day on these, at least for now!

Kristiina said...


When we get the carbon dating for Hajji Firuz, we will see if Yamnaya is possible or not.

Davidski said...


It's not only possible, it's a fact. The genome-wide and Y-chromosome data say so.

Chetan said...

"no Kurgans anywhere outside the Steppe"

You should do more reading about this. There are kurgans all over south-east and central Europe during the third millenium BC.

"How do you go from PIE in Western Steppe to IIr in Sintasta. What caused such a big change? Dont say Uralic influence because those borrowings are all one-way. And how do you go from IIr to Vedic and Avestan?

I have said this before but it looks like there was substrate influence in the case of Indo-Iranian. And it would be tempting to identify the substrate as Uralic/ Finno Ugric because of some neat sound correspondences - palatization and vowel convergence to A.

It seems a really coherent theory to me that Poltavka herders introduced their late PIE language to people from the north and thus Indo-Iranian was born.

"The Steppe theory cannot explain the divergence of IA and Iranian from IIr. I have yet to hear a single coherent attempt."

How do you expect it to explain this? Indo-Iranians were one linguistic-cultural group, perhaps even into the early second millenium BC. And the period of unity is represented by Sintashta-early Andronovo. The early Indo-Aryans who migrated south from the steppes preserved the phonology of the Proto- language more than the later Iranians.

"Steppe cultures are closely related to Iranian speakers, and to no other branch of IE"

What you are trying to do is push back late PIE migrations farther in time than is warranted. The linguistic consensus says 4000 -3000 BCE. Refer any standard work on IE linguistics

"Borrowings into Uralic are Iranian, if somewhat Indo Aryan shifted"

3 strata. Late PIE like (probably from Poltavka), early I-Ir like (from Sintashta-early Andronovo) and Iranian (from later Iranian tribes on the steppe). Indo-Aryan phonology shows far less divergence from Proto-Indo Iranian and specifically Indo-Aryan loanwords may be just Proto Indo Iranian.

Rob said...

"Not everything is steppe. I'll finish by saying that in addition to the 8.7% steppe admixture I see in those SPGT samples, they also have an extra 12% Iran_ChL, or an extra 21.3% SC Asian admixture when compared to the Indus diaspora samples (though part of the latter could be local, given the location). We might have to start looking into those things too. And SC Asian/Iran_ChL admixture in the steppe, etc..."

God point.
It seems Krause's model will prevail. Eventualy I suspect Reich will move to that position too
SCA was indo-Europeanised via an intricate network including Turan.
Andronovo was perhaps non-IE or later became indo-Europeanise / Iranicized, as seen by the southern admixture. These then moved to the Black Sea, and brought about the secondary convergence with Balto-Slavic. Actually not really Baltic, just partially for Slavic. The IA - Slavic similarities have always been overestiamted.
Balto-Slavic in turn probably developed in Trziniec culture, which was a pagan, post-CWC which came into the IE fold of the Carpathians.

Chetan said...

"It seems a really coherent theory to me that Poltavka herders introduced their late PIE language to people from the north and thus Indo-Iranian was born."

This BTW is borne out by that extra pulse of steppe_EMBA ancestry into Andronovo that we saw and the presence of R1b-Z2103 in Sintashta

Chetan said...

@David Isn't Sintashta Steppe_MLBA related and Poltavka Steppe_EMBA?

Davidski said...


Isn't Sintashta Steppe_MLBA related and Poltavka Steppe_EMBA?

Sure, they're related, but the population that created Sintashta came from somewhere in the west.

It carried a lot of Europe MN farmer ancestry, and was purely R1a-Z93.

So not sure what the options are exactly, but you can scratch a northern or East Baltic origin. It was somewhere near farmer country and in close contact with very advanced societies for their time in the region, in terms of military technology, metallurgy, mobility and, heck, even town planning.

Does the town planning bit sound like early Uralics to you?

This is probably why Sintashta overwhelmed Poltavka so quickly, and then did the same across much of Asia as Andronovo.

Chetan said...

@David This is predicted in the Carlosian model. I'm not linking because you have asked me not to, but you can look it up.

Davidski said...


Carlos was claiming that Corded Ware was derived from Cucuteni-Trypillia, so apparently Cucuteni-Trypillia was proto-Uralic or something. Very funny.

But of course Corded Ware wasn't derived from Cucuteni-Trypillia. It was derived either from Yamnaya or a sister group from the steppe.

And like I said, there's nothing Uralic about Sintashta per se. It didn't come from Karelia or the Baltics. Its cultural package looks like a hybrid between steppe semi-nomad and something very advanced derived from Old Europe.

This is also what its overall genetic structure looks like, except for some outliers, who may have come from the northern forests and spoken Uralic. Take a look at the PCA datasheet I posted. There's some considerable heterogeneity in Sintashta, but the main pulse obviously came from somewhere near former Old Europe.

Now quit acting crazy.

epoch2013 said...


"It seems Krause's model will prevail."

Do you have a link to where that model has been published?

Davidski said...



Kristiina said...


There is not much available on Internet, but check this "dead-cat-bounce" post by David on October 8, 2016, and the video:

Matt said...

@Davidski, cheers mate, thanks very much for running these.

I've summarized the main results from the main tree (once I sorted my dumbwitted self out on the right qpGraph), here:

It basically looks like, if we consider the level of Basal Eurasian, Gonur2_BA without AASI is more like CHG than it is Iran_N/Abdul_Hosein_N.

So I'd guess Basal Eurasian looks to decrease on a cline from a maximum somewhere between Natufian/Iran_N towards both the north east and north west. Though Natufian/Levant_N may have a bit more than Iran_N? Anyway, it doesn't seem likely that the non-AASI part of Gonur2 had higher Basal Eurasian than Abdul_Hosein_N, probably less Basal Eurasian.

At the same time, looks like the most ANE:UHG (or WHG) ratio of the different Neolithic populations is maximized for Gonur2_BA, though not significantly different from Iran_N.

Overall Gonur2_BA's ANE:UHG ratio is very like Iran_N, *but*... it has much less ratio of Basal_Eurasian:(ANE:UHG), much more like CHG.

The non-AASI part of Gonur2_BA is almost exactly like a population with less Basal Eurasian than Iran_N, but more ANE relative to UHG than CHG. Gratifyingly, what we'd simply have predicted assuming clines where that ANE peaks in North-Central Eurasia and Basal Eurasian peaks in the South-Central Middle East!

This may explain why the unclear signals of CHG vs Iran_N relatedness in modern South Asians; the main real admixing population with AASI is like CHG in terms of Basal, but like Iran_N in terms of ANE:Other West Eurasian ratio.

(The Basal Eurasian may be a little high, since the real BEu might be more deeply diverged and less prominent as a % in the ME populations than these ratios suppose. If we're following Kamm's paper - - multiply the Basal Eurasian level by about 0.2 which means roughly so 9% in Barcin, 5% in Yamnaya, 15% in Abdul_Hosein, 11% in CHG, 11.5% in Gonur2_BA without AASI).

Have to look at the secondary models, but this should be most informative of them.

Alberto said...

@Anthro Survey

Nice to see that Xmix works for you. And that we're getting more or less equivalent results (though this should not be surprising, really). You already commented the results, so not much to add about it.

Just a tip: you can place just one population as the source and the Global_25 datasheet as the target and it will show you the distance between the source and all other populations. Then you can order it in descending order and see how they rank:

(Though maybe this is already easy to do with other programs like Past3, don't know).


Quite informative graphs regarding deep ancestry, thanks. I guess it's within expectations, but still good to have it corroborated this way and explained in an easy way to grasp.

Vara said...

Here is my comment again:

As for the Indo-Iranian question itself, the reason my favored hypothesis was the Armenian hypothesis is that because it could explain Indo-Iranian and Hittite. Before Kuzmina there was Viktor Sarianidi the OG of archaeology, and he himself believed in a South East Iranian homeland for Indo-Iranians basing it on traditions of metallurgy and religion found in the later Indo-Iranian regions. The only argument that the Andronovo guys have is: "Indra was a city destroyer", ignoring the fact that the Asura rebel gods lived in those cities and Indra himself lived in one.

So here are the arguments:

1. According to the Rigveda Indo-Aryans were cereal farming, making rice cakes...etc. The Gathas, which are the most Archaic of the Indo-Iranian texts despite the s>h shift, also mentioned farmers and camels. This doesn't fit with a northern homeland for the earliest books as some claim.

2. While it's true that some people were nomadic in those texts it still does not mean that all of them were. We have evidence of sedentary settlements in Bronze Age Luristan, yet from Sumerian and Akkadian tablets we know most of those Gutians were nomadic herders same with later people surrounding Persepolis and Ecbatana.

3. Iranian Plateau two wheeled chariots were depicted around 2000BCE. However, spoke wheeled chariots appear in the Late Rigveda that is dated to around 1200 BCE, basically when steppe ancestry reached the south, while no mention of spokes or battle chariots before that. Coincidence?

3. Evidence of riding both camels and horses in the south while none in the steppes before the Iron Age. No bone trauma, no depiction and no gear.

4. I can accept the Indo-Iranians showing up 1500 BCE and burrowing symbols of the vara from Jiroft somehow, fire worship, camel veneration and using an IE word for it (ushtra). But there is no way in hell they could have burrowed the dragonslayer myth, an IE myth all Indo-Europeans share that was exclusive to them before the Iron Age, from non Indo Europeans. It is depicted in a cylinder seal the same way it is mentioned in the Avesta with Zoroastrian symbols all over that seal. Mind you fire worship reached the steppes around 1500 BCE Federovo probably with the BMAC to Andronovo movement this paper suggests, and yes fire worship was important in the Graeco-Aryan religion eg. Hestia and Tahiti.

So if these southlanders weren't Indo-Iranian then what were they? The non-Indo-Iranian Indo-Iranians?

PS. I'm not sure where epoch found kurgans in the Rigveda and Avesta.

Rob said...

@ Epoch
You’ll know about it when it is. Naturally, I can’t say much , but if your observe around you might be able to figure out for yourself what might be coming to help add to the puzzles (eg see rozenfelds posts)

Rob said...

@ Vara
Very interesting points
Quite refreshing to get views from informed individuals

Davidski said...

Krause's model was never published because it was a total fail for South Asia. I pointed this out in the comments here...

After Krause did his talk in Moscow, Nick Patterson was invited to Max Planck to give them a lecture on the peopling of South Asia. And it looks like they dropped their model.

Their new model will take into account the fact that Indo-European languages arrived in South Asia from the steppe, not from West Asia.

Rob said...

Yes dave I know it irritated you .
Seems like you’re also not completely satisfied with the South Asia paper, because it didn’t bring to fruition your nuanced predictions of “crashing into Asia, bigly ” which is why you and the usual stooges are now trying to claim that HF sample is infact a Bronze age migrant from the steppe
Keep the comedy coming mate.

Davidski said...


More samples are on the way. And they won't bring good news for you.

Reza said...

How would more Eastern populations from the subcontinent with probable Austroasiatic admixture, like Bengalis,fare with the three way population mix? What other proxies might you add?

@Sein, would you be kind enough to run my mother's results, these are the updated coordinates.


Rob said...

@ Dave
I think it’s great more data is coming
Given that I have not been making an bold proclamations, and am not above believing that migrations have occurred to South Asia as any other region in the world, more data can only make me happier.

Chetan said...

"As for the Indo-Iranian question itself, the reason my favored hypothesis was the Armenian hypothesis is that because it could explain Indo-Iranian and Hittite."

This is ridiculous. Indo-Iranian and Hittite separated millennia apart from PIE. Look at Hittite and Tocharian and their attested positions in the world. These are the two most archaic IE languages and their positions are in Anatolia and Xinjiang. Says a lot to the unbiased learner.

If you actually picked up an introductory textbook on Indo-European before playing with the genetic data, half of the uninformed comments here won't be made.

epoch2013 said...


So it's a tad premature to talk of a model.

Similar with the R1b HF sample. Even if it's totally legitimate it would mean little in the language debate. Would one R1b sample, pretty downstream, in an area where there isn't any other R1b, let alone a high R1b diversity pointing to a pool from which this sample could have emerged, in an area where there is no serious trace of PIE in any of the known languages, point to the Urheimat of early PIE? Really?

Now, in discussions like these someone is bound to come up with something like: "You just wait till the next batch comes out". But I rarely hear substantiated what exactly is coming and how that is going to change the evidence. So, are we going to find massive amounts of pre-Bronze Age Iranian R1b? Because it looks to me that is needed for such a change of mind. If not, what will change the location of the IE Urheimat?

Matt said...

@Alberto, thank you, also had a look at proportions from some more iterations of the trees David ran for me, as I think looking at them together with the main graph could us some things, though I think the main graph should probably be best (despite a worse final Z) because of the number of populations present.


It looks like:

a) In model without CHG / EHG, the level of AASI in Gonur2BA goes down. Theory: model less well able to distinguish between AASI and ANE.

b) In model without CHG / EHG / Yamnaya, Basal in Gonur2BA goes down as well as AASI. Theory: model less well able to distinguish between Gonur2BA having higher levels of AASI+Basal vs higher ANE, as distinguishing factor from Iran_N and Barcin.

c) model without Barcin drives ANE levels higher and UHG levels lower than in the main model with Barcin. Theory: There's more complex unmodelled gene flow in these trees between Neolithic populations to exclusion of ancients which my main model forces to be best expressed by UHG (as the only "HG" ancestor shared with Barcin and all other populations in that model). So these models probably only provide good relative and not good absolute references of amount of "UHG":ANE.

Ric Hern said...

Something interesting about Horses.

Looks like Domesticated Horses spread from the Steppe during the Chalcolithic to Azerbaijan, Armenia and Northwestern Iran. Did Steppe people exchange horses for women ?

Did Steppe people migrate Southwards with their horses and create a Pseudo or Proto Yamnaya population in Azerbaijan ?

Just a bit of speculation so do not crucify me yet...

epoch2013 said...

@Ric Hern

How about an incursion into Maykop and even Leila-Tepe where the first Kurgans were found? They are contemporaneous to Khvalynsk and an interesting feature is that the Leila-Tepe yielded a horse head sceptre, just like in some of the Khvalynsk non-kurgan or proto-kurgan burials.

Ric Hern said...


Yes it doesn't seem to be too far fetched especially now that we have seen how Steppe people adopted the Bell Beaker Culture...

Ric Hern said...


Yes it doesn't seem to be too far fetched especially now that we have seen how Steppe people adopted the Bell Beaker Culture...

Could it be that the Maykop was a backmigration of this Southward exploring Steppe people ? Maybe due to desertification or pressure from other Southern Populations...

epoch2013 said...

@Ric Hern

Something like that. Or Leyla-Tepe was a "first pulse" or whatever. Hell, maybe even the first pulse that caused Anatolian, even if Agamemnon on Anthrogenica once made compelling arguments against a Caucasian route.

Frankly, I don't have a clue. But why not speculate? Say: Leyla-Tepe were the first Steppe people migrating and their new feature, the tumulus, spread back to the steppe.

Ric Hern said...

@ epoch2013

Maybe a colony of Steppe traders ? Traders did not need to arrive with one bulky migration ? How can traders be traced if they carried all kinds of tradable goods from different cultures and trading these goods on their for new local goods ?

Vara said...

"Indo-Iranian and Hittite separated millennia apart from PIE."

Where did I say otherwise? Of course you ignored all my points and went for a strawman.

Sorry, but current evidence supports Greco-Aryan from the south.

Davidski said...


Sorry, but current evidence supports Greco-Aryan from the south.

Well obviously it doesn't, so don't pretend otherwise.

Andronovo pastoralists brought steppe ancestry to South Asia (Narasimhan et al. 2018 preprint)

Steppe admixture in Mycenaeans, lots of Caucasus admixture already in Minoans (Lazaridis et al. 2017)

Vara said...


They brought steppe ancestry in the Iron Age. So that disproves Andronovo bringing Indo-Iranian languages. Mycenaeans picked their steppe ancestry from Thrace on their way to Greece.

I'd like an answer to my question: who were the non-Indo-Iranian Indo-Iranians?

Davidski said...


But, obviously, steppe ancestry is already present in early Iron Age Swat. So it arrived during the Bronze Age, not the Iron Age.

So, as I said, you're not making any sense.

Seinundzeit said...


I nearly missed you:

44.65% Shahr_I_Sokhta_BA2 (IVC_Diaspora_West_Eurasian)
33.95% AASI
13.30% Kashkarchi_BA (Steppe_MLBA_East)
4.65% Cambodian
2.55% Mongola
0.90% West_Siberia_N

Shahr_I_Sokhta_BA2 is the IVC diaspora sample which is essentially West Eurasian (only 10%-20% AASI), so only a small percentage of AASI is covered by it.

«Oldest ‹Older   201 – 255 of 255   Newer› Newest»