search this blog

Saturday, March 31, 2018

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


Over at bioRxiv at this LINK. Note that the Andronovo samples that are shown to be the best fit for the steppe ancestry in South Asians are labeled Steppe_MLBA_East (ie. Middle to Late Bronze Age eastern steppe). Below is the abstract and a couple of key quotes from the paper and its supp info PDF. Emphasis is mine:

The genetic formation of Central and South Asian populations has been unclear because of an absence of ancient DNA. To address this gap, we generated genome-wide data from 362 ancient individuals, including the first from eastern Iran, Turan (Uzbekistan, Turkmenistan, and Tajikistan), Bronze Age Kazakhstan, and South Asia. Our data reveal a complex set of genetic sources that ultimately combined to form the ancestry of South Asians today. We document a southward spread of genetic ancestry from the Eurasian Steppe, correlating with the archaeologically known expansion of pastoralist sites from the Steppe to Turan in the Middle Bronze Age (2300-1500 BCE). These Steppe communities mixed genetically with peoples of the Bactria Margiana Archaeological Complex (BMAC) whom they encountered in Turan (primarily descendants of earlier agriculturalists of Iran), but there is no evidence that the main BMAC population contributed genetically to later South Asians. Instead, Steppe communities integrated farther south throughout the 2nd millennium BCE, and we show that they mixed with a more southern population that we document at multiple sites as outlier individuals exhibiting a distinctive mixture of ancestry related to Iranian agriculturalists and South Asian hunter-gathers. We call this group Indus Periphery because they were found at sites in cultural contact with the Indus Valley Civilization (IVC) and along its northern fringe, and also because they were genetically similar to post-IVC groups in the Swat Valley of Pakistan. By co-analyzing ancient DNA and genomic data from diverse present-day South Asians, we show that Indus Periphery-related people are the single most important source of ancestry in South Asia — consistent with the idea that the Indus Periphery individuals are providing us with the first direct look at the ancestry of peoples of the IVC — and we develop a model for the formation of present-day South Asians in terms of the temporally and geographically proximate sources of Indus Periphery-related, Steppe, and local South Asian hunter-gatherer-related ancestry. Our results show how ancestry from the Steppe genetically linked Europe and South Asia in the Bronze Age, and identifies the populations that almost certainly were responsible for spreading Indo-European languages across much of Eurasia.

...

Third, between 3100-2200 BCE we observe an outlier at the BMAC site of Gonur, as well as two outliers from the eastern Iranian site of Shahr-i-Sokhta, all with an ancestry profile similar to 41 ancient individuals from northern Pakistan who lived approximately a millennium later in the isolated Swat region of the northern Indus Valley (1200-800 BCE). These individuals had between 14-42% of their ancestry related to the AASI and the rest related to early Iranian agriculturalists and West_Siberian_HG. Like contemporary and earlier samples from Iran/Turan we find no evidence of Steppe-pastoralist-related ancestry in these samples. In contrast to all other Iran/Turan samples, we find that these individuals also had negligible Anatolian agriculturalist-related admixture, suggesting that they might be migrants from a population further east along the cline of decreasing Anatolian agriculturalist ancestry. While we do not have access to any DNA directly sampled from the Indus Valley Civilization (IVC), based on (a) archaeological evidence of material culture exchange between the IVC and both BMAC to its north and Shahr-i-Sokhta to its east (27), (b) the similarity of these outlier individuals to post-IVC Swat Valley individuals described in the next section (27), (c) the presence of substantial AASI admixture in these samples suggesting that they are migrants from South Asia, and (d) the fact that these individuals fit as ancestral populations for present-day Indian groups in qpAdm modeling, we hypothesize that these outliers were recent migrants from the IVC. Without ancient DNA from individuals buried in IVC cultural contexts, we cannot rule out the possibility that the group represented by these outlier individuals, which we call Indus_Periphery, was limited to the northern fringe and not representative of the ancestry of the entire Indus Valley Civilization population. In fact, it was certainly the case that the peoples of the Indus Valley were genetically heterogeneous as we observe one of the Indus_Periphery individuals having ~42% AASI ancestry and the other two individuals having ~14-18% AASI ancestry (but always mixes of the same two proximal sources of AASI and Iranian agriculturalist-related ancestry). Nevertheless, these results show that Indus_Periphery were part of an important ancestry cline in the wider Indus region in the 3 rd millennium and early 2 nd millennium BCE. As we show in what follows, peoples related to this group had a pivotal role in the formation of subsequent populations in South Asia.

...

These results—leveraging our rich data from ancient samples closer in time to the Bronze Age—show that the group(s) that contributed Iranian agriculturalist-related ancestry to South Asia shared more genetic drift with the Iranian agriculturalist-related groups in our dataset that are temporally and geographically closest, compared to Caucasus HGs (CHG) or early Zagros related agriculturalists previously shown to be related to source populations for South Asians (11, 81). We are not only able to exclude these early farming and hunter-gathering groups, but also Copper and Bronze Age groups in western Iran (Seh_Gabi_C and Hajji_Firuz_C), and even in eastern Iran and Turan (Tepe_Hissar_C, Gioksiur_EN, and BMAC). Our detailed analyses in Text S3 indicate that what is driving the failure of these models is an excess of Anatolian agriculturalist-related ancestry in all of these groups, suggesting that the Iranian agriculturalist-related population that mixed into South Asia had less Anatolian agriculturalist-related ancestry than all of these. However, we find that mixtures using the Indus_Periphery sample (a pool of three outlier individuals from the BMAC site of Gonur and from Shahr-i-Sokhta), provides an excellent source population for the Iranian agriculturalist-related ancestry in South Asia when combined with any individuals in the Steppe_MLBA cluster (Srubnaya, Sintashta_MLBA, Steppe_MLBA_West or Steppe_MLBA_East).


Narasimhan et al, The Genomic Formation of South and Central Asia, Posted March 31, 2018, doi: https://doi.org/10.1101/292581

Update 12/04/2018: The dataset from the prerprint has been made available early at the Reich Lab website here. I've already started analyzing it. You can see the results in several new threads, for instance here, here and here.


See also...

Late PIE ground zero now obvious; location of PIE homeland still uncertain, but...

Central Asia as the PIE urheimat? Forget it

Ancient herders from the Pontic-Caspian steppe crashed into India: no ifs or buts

877 comments:

«Oldest   ‹Older   201 – 400 of 877   Newer›   Newest»
music lover said...

Hi all, please see: https://twitter.com/vagheesh/status/980646170648285185

Davidski said...

@music lover

Thanks, let's see if that helps.

But I reckon we won't really know what's what until the BAM files are released.

Seinundzeit said...

Matt,

Those simulations are pretty cool, and quite effective! I mean, Proto_IVC peaking in Gujrat is nicely in sync with the data we've been seeing for a couple of years.

Also, your simulations do seem to demonstrate that populations like the Kalasha and the HGDP Pashtuns are somewhere between 10% and 15% AASI.

In addition, despite the inclusion of a Proto_IVC simulation, the Kalash do display additional Srubnaya_outlier (our closest current proxy to West_Siberian_HG). It does seem that this population may have an excess of ANE (so, that old paper arguing for ANE continuity via the Kalasha wasn't totally off-base; just wrong about the isolation, and there is a Bronze Age steppe contribution of around 25%-30% ancestry).

Rob said...

Would be great to see some kelteminar genomes - to see how they relate to “Siberian HGs”, Hotu etc , and if they left any local genetic legacy

Anthro Survey said...

@Rob

Well, I was quite upset not to see Kelteminar genomes. Was really hoping they'd be in this paper.

Hmm, so you think there was direct demic impact from Northern Iran, as opposed to Caucasus, followed by founder effects? Where do you place L51's origin, then?

& not so fast at taking credit for "pseudo-steppe", I meant. :D I was pretty vocal about the potential effects of some ANE-rich, Kelteminar-related(?) ancestry spiking steppe scores.

Anthro Survey said...

@Matt

Do you still have the PAST3 PCAs for Iran_Chl by chance? :-)

If not, it's cool. So, it didn't tend toward CHG at all and was dead on the ANF-Iran_N cline? Interesting, because it would imply a nearly "pure" ANF-like spread into Iran OR one mediated by ANF-Iran migrants from Mesopotamia.

The latter would make sense for SOUTHERN Mesopotamian migrants, of course, but ANF-CHG would be expected for Halaf. So, perhaps both types of people migrated eastward, with CHG-ANF following a more northerly route. Hajji Firuz is from Azerbaijan. Iran_Chl is from around central Iran, a part that continues to have intimate ties with Southern Iraq to this very day. In fact, it's also called Irak....Irak-i-Ajam.

So, let's see where HF ends up plotting and if there was any sort of geographic structure.

Rob said...

@ Anthro

I was not claiming any exclusivity on an idea :)

Hhm not sure about these calls anymore. Lets see what the revised shows
L51 - could have arisen in central Europe itself (i.e. Hungary or Romania from a steppe-arrived L23*) ?

old europe said...

http://www.treefrog.ru/images/stories/Nedoluzhko/Rezepkin.pdf

Here a study of the russian archeologists rezepkin about the strong input of central european ( TRB) culture upon the maykop novosbodnaya cultural complex.

Just talking about farmers input on the eurasian steppe. We don't have only the input of cucuteni upon sredny stog ( see the study of manzura).
We don't have only the middle neolithic input in andronovo ( both cultural, religious and genetical)

we have also this ( perhaps the most stunning of all):

Signs of a direct migration in early III millennium from Atlantic France to ALTAI; EASTERN KAZAKSTAN and....WESTERN CHINA.
Read carefully this:

https://www.google.it/url?sa=t&rct=j&q=&esrc=s&source=web&cd=3&cad=rja&uact=8&ved=0ahUKEwiz0LftjpvaAhVLcRQKHW4eDe8QFgg3MAI&url=http%3A%2F%2Fsuyun.info%2Findex.php%3Fp%3D4_10012017_1_1%26LANG%3DENG&usg=AOvVaw1piZPuuwrFLPbscGjEMZOz

Aram said...

Kristiina

Thanks for that link and map.
Now after more thinking yes it is possible that this African Y dna are correct. This could be a relic of ancient Out of Africa migration that rushed toward South Asia from Horn of Africa avoiding Near East.
And Onge people are after all Africans. Their high level of Y dna D could be later drift. Initialy they could have African Y dnas.

EastPole said...

Looks like we have two distinct steppe groups:

1. Sintashta, Petrovka, Srubnaya, Poltavka outlier, Andronovo etc.: R1a dominated, mixing with Tripolye Anatolian farmers, related to CWC

2. Yamnaya, Afanasievo etc.: R1b dominated, mixing with Iranian farmers, related to BB.

https://s31.postimg.org/qm8vxyhez/screenshot_358.png


The interesting cases are Polatavka outlier and Yamnaya_Ukraine_Ozera:

https://s31.postimg.org/latzcrq23/screenshot_357.png

https://s31.postimg.org/k759neg4r/screenshot_356.png


It would be interesting to determine the languages of these two groups.
Maybe PIE never existed but from pre-PIE languages of EHG and CHG two distinct PIE were formed PIE 1 and PIE 2 which later converged so that mistakenly one PIE is postulated.
In such a mode PIE1 should be linked with Group 1 : Balto-Slavic, Indo-Iranian and Greek .
PIE2 with Group 2;Italo-Celtic, Germanic and Anatolian:

https://s31.postimg.org/dn72ufh3f/tree7.png

old europe said...

East Pole

How can Yamna be the source of the bell beaker?

Steppe people were all dolicochephals while the bell beakers were one of the most brachycephals race ever existed ( they were of the round head alpine race which has its closest prototype in the remedello skulls). As I, and many others said remedello culture looks like a proto bell beaker one ( burials, metope decoration, bows daggers etc etc.).

EastPole said...

@old europe

“How can Yamna be the source of the bell beaker?”

It is shown on the map:

https://s31.postimg.org/c1rt2ujsr/screenshot_359.png

Davidski said...

@EastPole

Early Baltic CWC is closer genetically to Yamnaya than to most late CWC or Sintashta.

@old europe

Northern, Central and Eastern Bell Beakers do derive in large part genetically from Yamnaya.

Chetan said...

@East Pole IE languages (other than perhaps Anatolian) are too closely related to be anything other than daughter languages of a single proto language which expanded from a core region (most likely the region of Yamna culture) roughly 5000-4000 YBP

Anthro Survey said...

@Rob

Just messing with ya.

Ah, ok. So, where do you think L23 itself arose? I mean, how did it get to the steppe?
Do you see the "axe bearing" Caucasus movement just carrying Z-2103 or also L23* destined to give rise to L51? Assuming Z-2103 was born ~5700BC, L51 being born ~2800 BC would make it a pretty late bloomer, so to speak.

old europe said...

yes dave

I know genetics as for now it's the strongest point in favor of the steppe. And correspondingly it is the weakest part of the old europe theory. But you agree the all the R1b question has not yet be settled .
Many question that steppe R1b is ancestral to the western clades but we'll see.

Aram said...

Imho there was a real influx of EHG into South Caucasus and NW Iran that started 8200 years ago.
The traces of this influx are visible in Areni, in Hajji R1b and maybe even in Anatolia Chl.

EastPole said...

Davidski

„Early Baltic CWC is closer genetically to Yamnaya than to to most late CWC or Sintashta.”

Yes, they are closer genetically to Yamnaya but are not derived from Yamnaya culture and have different Y-DNA. It may mean that their ancestors in Late Sredny Stog culture had similar composition of EHG-CHG components as the ancestors of Yamnaya culture.

Mittnik et al. 2018 postulates Balto-Slavic steppe language separated from steppe PIE, but we have to remember that we know nothing about this language because pure steppe Balto-Slavic populations do not exist anymore.
The Baltic and Slavic languages that we know originated from Balto-Slavic-Tripolye populations.
I think we can talk about two Balto-Slavic languages B-S_1 which is pure steppe language and which is unknown and B-S_2 which is mixed with Tripolye and related to Indo-Iranian.

@Chetan
“IE languages (other than perhaps Anatolian) are too closely related”

Because of convergence. Yamnaya interacted with CWC, Bell Beakers formed from Yamnaya mixed whith LBK_EN derived groups which may have spoken a languages related to Tripolye languages, moreover BB interacted with CWC and later Germanic and Celtic interactd with Slavs, with Romans who were influenced by Greeks etc…

Anthro Survey said...

@Old Europe

Nice Camunian rose, man! I hope we'll get some DNA from Camonica someday. Bona pasqua.

As for brachycephaly and craniometry: weren't EEF groups across Europe generally doliocephalics with gracile bone structure while steppics tended to be more mesocephalic and robust?

I do stress the word "generally" here. Perhaps there was regional variation with GAC or TRB being an "exception" to the "rule".

Alberto said...

@Davidski

Apparently, the genotype data might be available within a day.

Really? That would be awesome.

@Aram

Maybe, though one interesting thing would be to figure out if Armenia_Chl had recent input from the steppe or they were migrants from Turan. We have the L1a connection and those Geoksiur_EN samples have some 15-20% ANE/EHG admixture (though Armenia_ChL seems to have quite more ANF).

Davidski said...

@EastPole

There's a clear progression in the eastern Baltic region from early CWC to late CWC to Baltic_BA, which has R1a subclades typical of Balto-Slavs.

I'm not sure how Tripolye fits into this, except perhaps as a source of some of the EEF admixture in late CWC, mostly female mediated by the looks of it.

I haven't looked in any great detail at the mtDNA lineages of Tripolye, but if they don't appear fairly typically Balto-Slavic then that's a problem.

Anthro Survey said...

@Aram

So, do you think L23* came to the Caucasus with this EHG wave or do you think EHGs brought M269 and L23 was born there?

Alberto said...

Also no mention so far about the Dali_EBA sample. That's an amazing one, representing a nomadic Central Asian population that was pioneer in the formation of the Silk Road. They're probably the ones that can be credited for taking ovicaprids to China and Mmllet to West Eurasia. They practised cremation, so I didn't expect to get DNA from them, but here it came as a bonus.

https://wustl.academia.edu/MichaelFrachetti/Papers

@old europe

Archaeology is sometimes amazing. You can look at some of the papers by Alexey Kovalev about the Chemurchek culture of Western Mongolia (or thereabouts) connecting it with Atlantic megalithic. I think he's the one who first found out that Afanasievo was R1b-m269, which he thought confirmed his hypothesis.

Rob said...

@ Anthro

Let's put aside this controversial Iran Eneolithic sample for now
So far, Ukraine Meso & Neolithic appears to be V88 or pre-V88, IIRC
The Samara region has M73 and an xP297 branch in Khvalynsk which is also found in Kura-Araxes
This makes some of the Latvia HG P297s the most proximal 'cousins' to date. And a Z2103 Ukraine Eneolithic outlier that is suspiciously WHG/EEF in outlook.
Let's hypothesize that Yamnaya is in large part descended from Repin, which expanded from the Dnieper-Don steppe, east (Volga) & to west. Repin pottery has been seen to have Combed Ware inspiration, suggestion a northern link. So L23 could have emerged from a late Dnieper-Donets group.

Shaikorth said...

@Alberto
Was Dali_EBA's population one of the mediators on the road or more widespread? I'm asking because there might have been a different population in easternmost Kazakhstan/Inner Asian mountain corridor already, the Kanai MBA sample is like Okunevo and has Y-DNA C2b. This is probably responsible for increasing Okunevo-like ancestry in Zevakinskiy MLBA and LBA.

Alberto said...

@Shaikorth

I don't know about the Kanai site, but it's further north and related to Okunevo. Dali, on the other hand, is part of the IAMC, in SE Kazakhstan. The Dali_EBA sample is also some 700 years older, and has admixture from (pre-)BMAC. The site is part of the Tasbas/Begash complex, so I think this is the sample that matches the nomadic Central Asian population that has evidence for commerce between China and SC Asia. Kanai seems unrelated, though I could be wrong (no information about it on the paper. Seems to be spelled as Kenes too).

Shaikorth said...

Kanai's from 50 N 84 E but the authors think it has something to do with IAMC.

"However, we can show that Kanai_MBA from further south is a genetic clade with Okunevo.SG relative to our outgroup set using qpAdm (p-value = 0.15). This suggests that the range of the ancestry type that was represented in the Okunevo culture extended beyond the Minusinsk Basin (where the Okunevo material culture is documented) and into the Inner Asian Mountain Corridor. It also provides further evidence that the Afanasievo made little genetic impact on the populations here and indeed further south during this time."

Matt said...

@Alberto: No, Fig.2C (unless I'm reading wrong the number) is the qpAdm models using the 6 main ancient populations involved, and represented in coloured bars.

Ah yeah that's the one then.

I'd guess your thinking here is that the fits seem inconsistent with any EHG ancestry in Saidu_Sharif_IA and Butkara_IA, and have large excesses of Anatolian ancestry, which is immediately inconsistent with modelling then as having Steppe_MLBA ancestry (no EHG!), and rather shockingly in contradiction to the main conclusions of the paper!

For your interest, I created a West Siberian ghost (20Ulchi30EHG50ANE) and then used that with the AASI ghost and the real Iranian and Anatolians to create some simulations for the bottom half of the figure: https://pastebin.com/wWTxsV0k

Graphics: https://imgur.com/a/HkgIi

These simulations without any steppe ancestry and modelling Indus diaspora related ancestry purely through

Iran_N+Anatolia_N+"West_Siberia"+AASI look a bit more off for high dimensions where the SA cline is very specifically at the peak. (This is more clear when I resample the G25 through PCA again than in the above graphics though).

Compare and contrast the Shahr-i-Sokhta sims from the Indus_Diaspora sim.
These are supposed to be the same things, but have different positions here which imply different things. Depending on which the real samples resemble, there will be different very conclusions (very, very different conclusions for history and IE).

Potentially it looks the BMAC related samples, in the flesh, may be more relevant to Makrani/Balochi, and not so much so to the SA cline.

@Sein, cheers, it was a very simple thing to do.

Also, your simulations do seem to demonstrate that populations like the Kalasha and the HGDP Pashtuns are somewhere between 10% and 15% AASI.

I'm gonna be stubborn on this ;).

Looking at Alberto's first mass run, for Kalash, assuming 23% AASI in Indus_diaspora_sim at 35% and 4.1% Onge, plus 1.3% Han, plus 0% extra AASI, then AASI+Onge (4+14) 18% / ENA 19% (above from other ancestry streams).

Pashtun in the same run is 8.4% Onge, 1.6% AASI, 2.7% Han and finally 23% of AASI in Indus_diaspora_sim at 27.3%. Then AASI is (8+10+2) 20% / ENA 23%.

Or looking at the one with the speculative West Eurasian proto-IVC (that may or may not have ever existed), then Kalash simply 14.6% Onge, 1.3% Han and Pashtun 8.1% Onge, 10.1% AASI, 2.7% Han.

So I'd still bet on anywhere between 15-20%. But it's not like it's a large range we have uncertainty over.

Anthro Survey: Do you still have the PAST3 PCAs for Iran_Chl by chance? :-)

Lol, I think I just overdrew Lazaridis PCA from memory (which was pretty amateur tbh).

Eric said...

@Davidski

"And no, it's not very plausible that R1a was already present in South or even Central Asia during the Neolithic"

Maybe not Iran(although it's quite possible unless we want to say ANE related people have been invading Iran since the paleolithic), but given the two Gonur1_BA outliers and the closer similarity of Kelteminar culture to European HG cultures rather than the mesolithic/neolithic Levant, looks like EHGs or West Siberian HGs(the paper said they were EHGs with some East Asian related admixture but it could also be Afontova Gora shift)are going to be native to atleast as far south as southern Turkmenistan. The authors of the paper seem to think that. There's really no geographic barriers between the Volga region and central Asia, the mountains start around Iran/Afghanistan.

Honestly, I wouldn't be surprised if something like Mal'ta boy pops up as far south as Baradostian culture(related to Aurignacian) before the more Natufian like Zarzian culture came in to form N Iran.

Alberto said...

@Matt

Depending on which the real samples resemble, there will be different very conclusions (very, very different conclusions for history and IE).

Yes, and that sounds a bit scary, given how sensitive this subject can be for some.

(Un?)fortunately, I placed the SPGT_Sim as the target with the same setup as the previous runs and it didn't take any Indus_ Diaspora-related or Indus_ Diaspora_related_-23%ASI_sim. Not even Iran_N.

So for some reason it seems it's not working as good as the previous simulations. And honestly, I'm a bit relieved. Davidski mentioned that we might get the real samples within a day, so I it might be better to wait for them instead of risking to post some results that could create false expectations.

Davidski said...

@Eric

The Gonur1_BA_o2 outliers have steppe pastoralist ancestry from Eastern Europe (Steppe_MLBA), so the only thing they prove is that steppe people moved into BMAC settlements. They say nothing about the Mesolithic connections between Eastern Europe and Central Asia.

The Gonur1_BA_o outliers are related to West Siberian HGs, not Eastern European HGs. That's probably because they have ancestry from either Botai or Kelteminar.

And considering that there's a clear difference between Eastern European HGs and West Siberian HGs, and a clear difference between steppe groups west and east of the Urals prior to the Afanasievo expansion, I wouldn't get my hopes up that anyone resembling Eastern European HGs was living east of the Caspian.

Central Asia is terra incognita no longer; the data from this paper show that it was not simply an extension of Eastern Europe at any time.

Matt said...

Alberto: (Un?)fortunately, I placed the SPGT_Sim as the target with the same setup as the previous runs and it didn't take any Indus_ Diaspora-related or Indus_ Diaspora_related_-23%ASI_sim. Not even Iran_N.

Hmm... The SPGT_Sim should take Iran_N ancestry when put into a model with Iran_N, Barcin_N, AASI_Sim and West_Siberia_Sim, because this is what I've made it from, assuming all those are at play.

But it won't take any Indus_ Diaspora-related_Sim, because the regression equation that has been used to estimate it places that simulation has been placed in a far position beyond the end of the Anatolia_N->Iran_N cline and other related dimensions (in a way that should make it a superior choice to model the present day SA cline, but an inferior point to model linear combinations of Iran_N, Barcin_N, AASI_Sim and West_Siberia_Sim, which the other simulations are). The method for creating the Indus_ Diaspora-related_Sim using regression from present day populations and proportions from the hierarchical model is totally different from simple additions and subtractions of other population averages (used to create the other sims in my post today just above, because they're the only thing that's possible). I'll be very interested to see which hits closer ground truth (I don't see any way it can't be the regression method that does though).

Carlos Aramayo said...

@ Nirjhar and Davidski

After the first good observation that there is no Steppe ancestry in Swat`s samples by Nirjhar and the reply by Davidski commenting briefly on a c. 500 BC sample there, I found the paper says that in "isolated" Swat region samples between 1200 and 800 BCE "we find no evidence of Steppe-pastoralists-related ancestry"(see p.9). But after using a qpAdm modelling they conclude they found "Steppe pastoralist ancestry" in "the Swat Iron Age individuals" dating to c. 1000 BCE (see p. 12). A very confused way to explain things by Narasimhan et al.

Davidski said...

@Carlos

You have to understand that when it comes to this topic at least, Nirjhar never has any good observations. His observations are always horribly bad.

Now that we that out of the way, you obviously didn't understand the text on page 9. Here is an outtake. I bolded some of it to emphasize that the authors are talking about the Indus_Periphery trio, not any Swat samples.

Third, between 3100-2200 BCE we observe an outlier at the BMAC site of Gonur, as well as two outliers from the eastern Iranian site of Shahr-i-Sokhta, all with an ancestry profile similar to 41 ancient individuals from northern Pakistan who lived approximately a millennium later in the isolated Swat region of the northern Indus Valley (1200-800 BCE). These individuals had between 14-42% of their ancestry related to the AASI and the rest related to early Iranian agriculturalists and West_Siberian_HG. Like contemporary and earlier samples from Iran/Turan we find no evidence of Steppe-pastoralist-related ancestry in these samples. In contrast to all other Iran/Turan samples, we find that these individuals also had negligible Anatolian agriculturalist-related admixture, suggesting that they might be migrants from a population further east along the cline of decreasing Anatolian agriculturalist ancestry. While we do not have access to any DNA directly sampled from the Indus Valley Civilization (IVC), based on (a) archaeological evidence of material culture exchange between the IVC and both BMAC to its north and Shahr-i-Sokhta to its east (27), (b) the similarity of these outlier individuals to post-IVC Swat Valley individuals described in the next section (27), (c) the presence of substantial AASI admixture in these samples suggesting that they are migrants from South Asia, and (d) the fact that these individuals fit as ancestral populations for present-day Indian groups in qpAdm modeling, we hypothesize that these outliers were recent migrants from the IVC.

Carlos Aramayo said...


@Davidski,

OK, so in the statement:"...between 3100-2200 BCE we observe an outlier at the BMAC site of Gonur, as well as two outliers from the eastern Iranian site of Shahr-i-Sokhta, all with an ancestry profile similar to 41 ancient individuals from northern Pakistan who lived approximately a millennium later in the isolated Swat region of the northern Indus Valley (1200-800 BCE)" the word similar (for them) means not absolute coincidence. Of course through latter qpAdm modelling in Swat and other modern South Asian samples they found Steppe ancestry.

Davidski said...

@Carlos

The Swat samples are native South Asians for the most part, so it's not surprising that they resemble the earlier Indus_Periphery trio, who are likely Harappan migrants to Central Asia.

But there is obviously steppe admixture in many of the Swat samples that is missing in the Indus_Periphery samples, so everything is in line with the paper's conclusions.

Lenny Dykstra said...

No IVC samples....

Reich definitely got fed up with the Indians that had custodianship over those IVC samples, trying to contort the data to fit with an "Mesolithic North Indian Continuity" model or some such BS.

Reich releasing this paper without their input forces the hand of the would-be saboteurs. Now if they try to release an IVC paper without mentioning BA steppe input into modern Indians, or even more absurd they leave the door open for an Out of India IE model etc and then refuse to publicly release the sample data, then we'll all know the jig is up.

Nick Patterson (Broad) said...

@Lenny Dykstra:

I am extremely grateful to my South Asian collaborators
(and this goes for the whole Reich lab.); they are honest,
excellent scientists. There's a simple reason we have not
published on DNA from the IVC proper -- we have no useful DNA, and as you might guess this is not from lack of trying.

Nick

Kurti said...

West Iran Hajji Firuz 5500 BCE R1b M269. No Steppe ancestry. Indo European ancestry into South Asia most likely from Steppes it seems though.

Jijnasu said...

@Nick Patterson
I'm relieved to hear that, but a little disappointed that we have no DNA from the IVC yet. I'm thankful to the Reich Lab for the work done in uncovering the Indian past. Ancient DNA has revealed to us that the past was far more complex than we imagined bringing down dominant accounts historical narratives stained by the ethnic or racial prejudice of its authors

Kurti said...

Also interesting J2a in MLBA Steppes.

Chetan said...

@Kurti We know where it comes from though. The paper is pretty unambiguous on that.

Kurti said...

But what is also interesting though. No R1a in IA South Asia?

Chetan said...

@Kurti Yes indeed. After all these years of thinking Swat culture represented an early Indo-Aryan settlement, it now seems they were descendants of the IVC. So not surprising

Jaydeep said...

The current paper is really a major collaboration and it is wonderful to see so many samples from so many different and important sites from Iran & Central Asia. However, when it comes to South Asia, the interpretations of migration are very questionable.

The Harvard Lab has been fumbling with interpretation of Indian Genetic history for quite some time. They first interpreted the Indian population as an admixed from two groups - ANI - related to West Eurasian & ASI - distantly related to ENA.

Next they came up with dating the admixture between the 2 groups as having happened between 4200 ybp to 1900 YBP.

Then they realised that the ANI itself could not be derived from any one single ancient West Eurasian population outside of South Asia. So they argued that ANI was formed by mixing of two West Eurasian populations - Iran_N & Steppe_EMBA.

Now, with this paper, we are being told that ANI was formed by the admixture between steppe_MLBA & Indus_periphery. Indus_periphery is a combination of a measly set of 3 ancient samples from Eastern Iran & BMAC spread over a period of 900 years and having differential ratios of ancestry components. But it has Iran_N + Old ASI + ANE related ancestry. In other words, the Indus_Periphery already had all the 3 ancestry components which were said to be necessary to form a modern Indian population - Iran_N + Steppe (Western Siberian HG) + Old ASI.

So now the only real way to argue for a steppe migration into South Asia is to look at the extra steppe related ancestry most Indian populations have in relation to Indus_periphery and argue that that has come from outside.

Ofcourse this line of data interpretation is aided by the fact that we really do not have any ancient Bronze Age or earlier samples from South Asia. Let those samples come. We will then see, if the extra steppe component is not found among some of the IVC samples.

As far as the Indus_Periphery samples are concerned, I am curious. The archaeologists working on Eastern Iran & Central Asian sites have often remarked that these sites have some features common with the chalcolithic cultures of Baluchistan. So could these Indus_Periphery samples' ultimate origin be around Baluchistan ? That would perhaps explain their low Western_HG (aka steppe) related ancestry. As for the AASI high samples from Shahr - Sokhta, maybe he was just an anamoly for his time. Who knows ?

Jijnasu said...

I don't think we're going to find any Indo-aryans in south asia looking like they just got out of yamnaya. There was constant interaction between the steppe pastoralists and local farmers and hunter gatherer groups all along the way. Also the lack of r1a in swat may have to do with the increased preference for cremation amongst the indo-aryan religious orthodoxy.

Jaydeep said...

One more point I should make is that Razib & Ryukendo were making the same argument as is being made in this paper for quite sometime now.

Razib & Ryukendo were arguing that the Iran_N like ancestry was spread across South Asia while a unique steppe related ancestry was more concentrated in the Northern regions and in higher castes. The present paper is making the exact same argument as a base for the steppe migration.

My answer is the same as it was to Ryukendo and Razib. Let us have Bronze Age or earlier samples from Haryana. Let us find out where this extra steppe is really coming from. Unfortunately we will have to wait for those samples a little more.

---------------

capra internetensis said...

@Nick Patterson

Good to hear that, though too bad about the state of the aDNA. :)

Jomon said...

Based on Mr. Patterson's comments, this is the famous and so expected "South Asia paper". There will be no other coming soon. But this is not disappointing, since important findings have been reached.

1) Existence of an anatolian/iran neolithic cline through iran and turan. This explains the presence of "EEF" in modern central asian populations like tajiks.

2) The ASI component was widespread throughout the Indian subcontinent since the neolithic and par of the IVC population.

3) The excess of ANE component in the Indian subcontinent is now explained by mixing with Siberin HG.

4) Finally, we have a date of arrival of the Steppe mix in South Asia.

Eric said...

@Davidski

I did say or. They are different, yeah, but at the same time, still really closely related(there's an admixture graph on page 145 of the supplementary mats with both EHGs and W. Siberian HGs just incase you haven't seen it already).

Point was, nothing Basal Eurasian related native to that part of Central Asia, instead a northern like ANE-rich population, which means R1a in pre-neolithic Central Asia is probably really likely, if not atleast R1. Not Z93 ofc.

That's all assuming that the heavy W. Siberian HG admixture wasn't from a recent immigrant(that seems less likely though).

Jijnasu said...

@santosh - I think there was an AASI population across the subcontinent. One of the Indus periphery samples was 42% aasi I think, pretty unlikely if they were limited to the deccan. Some lower caste groups from the North have relatively high amounts of steppe ancestry but much more aasi as well, so the AASI reach pops existed in the gangetic plains too. They should probably do away with the ASI - ANI terminology, it is getting far too confusing especially since they weren't even homogenous groups. Also the ANI in Indo-Aryan speakers is quite different from the ANI in dravidians. Two groups of Brahmins from Karnataka have about the same amount of aasi as the dravidian coorgi from the same state yet he has only about half the amount of steppe and ancestry and excess steppe.

Jijnasu said...

@santosh I'm actually more curious about the Iran neolithic like ancestry in the the ASI fraction. Could it atleast be partially pre-neolithic? there isn't a single agricultural tradition that spans south asia. Also groups like the paniya and irula have very different y lineages from neighbouring non-tribal populations.

Santosh said...

@ Jijnasu

Oh okay, thank you! Sorry for deleting my comment (in which I basically asked if the researchers had any strong reason to still retain the label "South Indian" in the name of AASI especially if they believed that the three East Iran and BMAC outliers were connected with the Indus valley). I really look forward to the day when they correctly and definitively characterise the mesolithic populations of various parts of the Indian subcontinent.

Santosh said...

@ Jijnasu

It's all such a mess frankly; they seem to suggest that the varied amounts of AASI ancestry in the three Indus_Periphery individuals implied that an Iranian agriculturalist plus AASI cline existed in the Indus valley (i.e. it was not homogeneous) and they considered some kind of average of this Indus_Periphery and obtained an ASI in the peninsula with the other component of the mixture being AASI and remark that this ASI could have "formed" only in the 2nd millennium BC; elsewhere, they also remark that this ASI itself was like a cline with varying levels of Indus_Periphery (which is already cline-like) and peninsular AASI, not a stabilised population, before it itself began to rapidly mix with the similar mess that is ANI, to result in the one final ANI-ASI cline that they originally posited.

Bogdan said...

“Finally, we examined our Swat Valley time transect from 1200 BCE to 1 CE. While the earliest group of samples (SPGT) is genetically very similar to the Indus_Periphery samples from the sites of Gonur and Shahr-i-Sokhta, they also differ significantly in harboring Steppe_MLBA ancestry (~22%). This provides direct evidence for Steppe_MLBA ancestry being integrated into South Asian groups in the 2 nd millennium BCE, and is also consistent with the evidence of southward expansions of Steppe_MLBA groups through Turan at this time via outliers from the main BMAC cluster from 2000-1500 BCE.”

If I’m not mistaken, ~20% is roughly the frequency of R1a haplogroup among modern day Kalash men...

Slumbery said...

@Santosh

About the nomenclature: I do not really see the point of renaming ASI to AASI as the authors did. Because - at least as it was used in here in the discussions - we called ASI what they call now AASI and they use ASI as a name of a late pre-mixture. I do not see the gain from this.

Also I agree, if their working hypothesis is that there was only one pre-Neolithic HG population up all the way to the Indus valley, they could as well name it Ancestral Indian or Indian HG or whatever.

Santosh said...

@ Jijnasu

Also, regarding Iran neolithic like thing being pre-neolithic in south India, not sure; it may be possible because the groups like Malayan, Ulladan, etc. which they deem to be pure ASI-like and are like genetic relics, actually all speak Malayalam languages- this might mean that, if their results are correct, then Indus_Periphery-like ancestry did not necessarily give them Dravidian languages or that it gave them such languages but which were later wiped out by South Dravidian-I Malayalam, assuming that the seemingly single ancient admixture event also brought some ancient language shift in those communities. But pre-neolithic in any case is doubtful because they say that ASI was "formed" in 2nd millennium BC only and by this time, south India (except perhaps the far south) is definitely in the neolithic stage; even full fledged archaeobotanic assemblages from various sites including wheat, barley from Indus-influenced regions and native and African millets in somewhat settled contexts are available from 2000 BC. I don't know about the precise locations of these deemed-to-be relic groups. It could mean that their ancestors were not in the neolithic in the 1700 BC time frame but only if they were in the far south, much southerly than the Western Ghats of Karnataka and northern Tamil Nadu. On the other hand, it might be that the 1700 BC event was really associated with a farmer-like Indus_Periphery or Iran_N but that these communities did not shift to farming or that they did that then and shifted back to foraging-like activity again, much later.

Alberto said...

@Matt

To be honest I just tried that fast run with SPGT_sim as target, and when I saw it was not getting the more obvious components I decided not to continue investigating (it took Iran_ChL instead, so maybe that explains why no Iran_N).

*If* the data comes in a day or so, I'll wait for the real samples. If not, and it will take months, then I'll be glad to experiment with simulations more while we wait.

About the thing about Fig.2C, it was not so much the EHG and West_Siberian_HG that I found strange (after all they're quite related, and qpAdm is using a max of 4 source pops, so I wouldn't expect a perfect match between components there). I'm more concerned that looking at the data, the verdict that anyone can have on whether there is evidence of some steppe migration or not is not clear at all. And that uncertainty is not translated to the main text, that seems completely off the mark, IMHO.

I guess that if the genotypes don't come out, we'll have time to go into all the details on the preprint. If they do come out, probably we can better move onto the samples directly and see what they show.

Santosh said...

@ Slumbery

"About the nomenclature: I do not really see the point of renaming ASI to AASI as the authors did. Because - at least as it was used in here in the discussions - we called ASI what they call now AASI and they use ASI as a name of a late pre-mixture. I do not see the gain from this.

Also I agree, if their working hypothesis is that there was only one pre-Neolithic HG population up all the way to the Indus valley, they could as well name it Ancestral Indian or Indian HG or whatever."

I remember reading somewhere that this putative Indian hunter gatherer component was most diverse in peninsular India perhaps pointing to its origins there and thus the choice of the term "ASI"- is it the same reason now for "AASI"? Perhaps the narrative is that something like only a minor off-shoot of this "AASI" was living in the northwest when the Iranian agriculturalist people came there? I really hope they modify their terminology at least somewhat if the Ganga mesolithics also turn out to have a bit of this "AASI" in them, even if it turns out they are an off-shoot too.

EastPole said...

@Jomon

“important findings have been reached.

1) Existence of an anatolian/iran neolithic cline through iran and turan. This explains the presence of "EEF" in modern central asian populations like tajiks.”

You are mistaken, EEF in central Asian populations like Tajiks comes from Tripolye/CWC/Andronovo, not from Iran and Turan. Just look at Fig. 2 A and B.

Santosh said...

"I'm actually more curious about the Iran neolithic like ancestry in the the ASI fraction. Could it atleast be partially pre-neolithic? there isn't a single agricultural tradition that spans south asia. Also groups like the paniya and irula have very different y lineages from neighbouring non-tribal populations."

Oh I actually seem to get your point only now. You think that all of the Iran neolithic-like thing in ASI may not be because of Indus_Periphery-related admixture which can be assumed to be related to first neolithisation or later neolithic-stengthening and thus pre-neolithic? Perhaps. I don't know anything about the likelihood of it. I earlier used to entertain the notion that Pre-Proto-Dravidian and earliest Proto-Dravidian was first spoken by Iran_N-like mesolithics belonging to India but I definitely can't evaluate that; even in this situation I can't evaluate it because the groups who forage and have Iran_N related admixture in them and who are deemed to be the southern counterparts to the seemingly-ancient-relics like Kalash who speak a divergent language in Indo-Iranian, actually all speak Malayalam languages.

Jijnasu said...

@santosh
I'm not saying that they are like the kalash at all. I'm saying that these tribes were genetically ASI pre-agricultural groups pre-dating dravidian expansion into the extreme south. Whatever the ultimate origins of dravidians, the people responsible for spreading south dravidian were initially probably ANI rich (biased towards Iran Neolithic) like Coorgs or Reddis, most local ASI hunter-gatherers were assimilated into their society while a few remote groups survived. Eventually these hunter-gatherer groups adopted the dravidian language of the dominant agriculturists surrounding them and received some gene flow in the process as well to form irulas, malayans paniyas etc. (the lower bound for the admixture dates seems to be quite low in some of these tribes, as low as 400 BCE)

epoch2013 said...

@Nick Patterson (Broad)

Does that mean that the Rakhigarhi samples didn't yield enough autosomical DNA? That would be a pity.

Santosh said...

@ Jijnasu

"I'm not saying that they are like the kalash at all. I'm saying that these tribes were genetically ASI pre-agricultural groups pre-dating dravidian expansion into the extreme south. Whatever the ultimate origins of dravidians, the people responsible for spreading south dravidian were initially probably ANI rich (biased towards Iran Neolithic) like Coorgs or Reddis, most local ASI hunter-gatherers were assimilated into their society while a few remote groups survived. Eventually these hunter-gatherer groups adopted the dravidian language of the dominant agriculturists surrounding them and received some gene flow in the process as well to form irulas, malayans paniyas etc. (the lower bound for the admixture dates seems to be quite low in some of these tribes, as low as 400 BCE)"

But they say in the paper that Malayan, Ulladan, etc. are likely pure-ASI relics from 1700 BC formed from one Indus_Periphery-like plus AASI admixture without any ANI admixture later! That 400 BC bound makes sense though, if that's correct as some kind of west coast dialect of Tamil may have been in existence at that point. But 1700 BC as a time period for some single language shift, to a Malayalam language is unimaginable. It is in this context that I talked about Kalash who they deem to be a purely-ANI relic, with Indus_Periphery-like plus Steppe_MLBA without excess AASI by ASI admixture later.

So you think the people bearing Dravidian languages first came to south India already having Steppe_MLBA (that is ANI) and mixed with the earlier Indus_Periphery-related plus AASI neolithic ASI groups who were living there? I was imagining a situation something like Dravidian languages associated with ASI (perhaps brought by the Indus_Periphery-related part of the combination) and mixture with ANI having Steppe_MLBA (and likely slowly in the process of shifting language to Indo-Aryan), happening later, without any language shift to Indo-Aryan/something else. Any of the above two scenarios is possible going by the time frames though- the latest Proto-Dravidian is placed by Krishnamurti in 1500 BC and his short chronology seems to be correct to me currently.

Anthro Survey said...

@Alberto and Indian commentators here:

Alberto said "I'm more concerned that looking at the data, the verdict that anyone can have on whether there is evidence of some steppe migration or not is not clear at all. And that uncertainty is not translated to the main text, that seems completely off the mark, IMHO."

You know, I strongly agree with this. Well, to a large extent.

Thanks to its geographical location and topo features, Swat area lends itself to the following influences:
1.migration of "Western" ANF-rich migrants from Turan/Iran.
2.higher than normal forager admixture thanks to the elevation and mountain surroundings which would have provided a better refuge for them than then flat oases of Turan.

Some may contest the first point because Indus Diaspora samples showed no ANF, but Swat isn't the IVC core! If this isn't satisfactory, then perhaps ANF-rich migrations swept the area a bit later than dates for those diaspora samples.

Steppe admixture is likely quite minimal here, indeed, but should this area
or even the Punjab be scoured for signs of the Aryan invasion? The truth is, Uttar Pradesh and Haryana were the Vedic heartland and continue to be the "heart of India" in many ways. THAT is where we need to be looking for unambiguously and profoundly steppe_MLBA admixed people bearing R1a-z93.

Gill said...

Well, I had been guessing all along that ancient Pakistan inhabitants would have had higher ANE-type ancestry than Neolithic Iranians. And so would Central Asians in general (like a western encroach of ANE). I'm guessing that's where the West Siberian HG comes in?

Has the story of when and where R1a1 appeared changed? Is it Ukraine/Eastern Europe? I remember some were saying it arose with CWC in Northern Europe and then was pushed into Central Asia and the Andronovo horizon. But clearly if Afanasievo is the ancestor to South Asian R1a-L657, then R1a1 had to have come there via Yamnaya directly? Is that even possible? Is the candidate then an eastern Andronovo pop?

We still haven't found L657.

I was hoping we'd find some J2b2-M241 here. Someone on Anthrogenica said they found some L283?

If you look at the main map depicting the dispersal of farmers from Anatolia and the Zagros, and the divergence of J2b2-M241 trees (continuously being developed at YFull and FTDNA), it's a pretty good fit:

https://yfull.com/tree/J-M241/

We've got enough to show that the Z2432 side had a large divergence from 8700ybp to 6400ybp with multiple ancient branches everywhere. Whereas L283 itself only seems to start at 6200ybp.

This poses a problem. They seem to have set their dates in stone, almost in parallel. After 7000 BCE, farmers spread in both directions. They hit Europe by 6100 and Afghanistan/Pakistan by 6000.

But according to the differentiation of J2b2-M241, odds are there was more pronounced activity in the Zagros and East of it. Odds are J2b arose somewhere further West (Levant?) or North (Caucasus?), and then perhaps J2b2 arose in Iran, and then one of its branches makes it out east to Anatolia and beyond by 4200 BCE while we have multiple strains spreading from a sister clade in Asia since 5700 BCE? Obviously dating them genetically isn't as sound as archaeologically/forensically, but the tree differentiation pattern is still there. One is older and more diverse and widespread than the other and dominates much of the Middle Eastern J2b2-M241 found today (so far).

I think the Saudi might be a post-Islam migration to Arabia from India, but still waiting to see where YF12601, the Kuwaiti, falls. If it's another ancient divergence underneath Z2433, that could indicate a really wide area for the differentiation of the entire Z2432 side of the family tree. Or who knows, maybe something happened later on to create a founder effect if we don't find enough ancient J2b2-M241 to correlate (if none is found in BMAC for example).

I guess considering the scarcity of genuine Neolithic Iranian ancestry in modern Iran, it makes sense that J2b2 was always found in this weird Europe/India dispersal. Now we're finding a few Middle Eastern samples to fill in the blanks, and they're so far only from Arabs or the Levant (I don't know if the Syrian YF08649 is an Arab or from another ethnic group like Druze).

Gill said...

Correction: Asian branch of J2b2-M241 was differentiating since 6700 BCE... and now we have a ton of lines in India. I really wonder if the IVC facilitated that.

mzp1 said...

I am quite critical of the authors statement that their findings show a migration of Indo Europeans into South Asia. As per my understanding contributors to the paper are not experts on Indo European, they are experts on genetics. The interpretation that this is an Indo European migration does not belong in the paper at all.

Indo European studies is 200 years old, archeogenetics barely a decade. Those people looking for simple solutions to the origin of Indo European betray their lack of competence in the field of Indo European studies. Co-opting the noble literature and great cultural achievements of IEs to further political claims is disrespectful to true progenators and inheritors of said culture.

I appreciate that IE studies would not exist without Western academics who first made the connections in India. However, it should not be forgotten that without the Brahmins in India none of this would have been possible. They were the original custodians of the literature that is the basis of IE studies, that which shed light on the origins of European civilization. Perhaps it was a mistake for those Brahmins to hand over the sacred literature, which their forefathers had passed to them, to those opportunistic foreigners who would use it for such lowly purposes. I imagine many a Vedic sage turning in his grave on hearing how their work is being appropriated, and of the character of those involved.

I think Davidski has a good blog here, though I dont understand much. I know he is interested in genetics, and I respect him for it. I do, however, seriously question his competence on IE studies, and that of contributors to this paper.

It should be remembered that IE is a defined by a combination of Linguistics, Culture and Religion, and not by Genetics.

Christianity spread throughout Europe but (as far as I know) there is no genetic trace.
On the other hand, the Migration Period has significant movements of people with limited linguistic impact.

English and Western European technology/culture/democracy are globally adopted yet there is 0, I mean literally 0, genetic impact.

I'm sure you will have a lot of fun looking for arab ancestry in countries that have adopted Islam i.e India, Pakistan, Iran, Turkey. If anything you will find Iranian, or Turkic genetics and will have to conclude Islam originated in either of those regions.

I am sure I can find many more examples.

If someone wants to claim they know about IE origins, I expect them to know the Rigveda and Shahnama inside out. They should also have good knowledge of other IE literature.

I am pretty confident most of the posters here dont know what they are talking about. Infact, I challenge just one person to answer this simple question.

Who is an Aryan (I mean in terms of pre 100BC) ?

Now guys, I know it'll only take a single paragraph to shame this entire post and show me up, so dont all line up at once!

Finally, I would like to end this post with a few very IE lines from a famous poet, likely a descendant of those illustrious vedic sages of the past...

Yūnān o-Miṣr o-Rūmā, sab miṭ ga'e jahāṉ se
Ab tak magar hai bāqī, nām o-nis̱ẖaṉ hamārā

Suyindik said...

See below quotes of some sources about the archaeology and anthropology of the Tepe Hissar culture. Could it be that the Ubaid, Uruk, Leyla-Tepe, Maykop, Early-Kura-Araxes(like Areni-1) and Sumerians will have a similar Y-haplogroup structure as seen within the Tepe Hissar results? Could it be that the carriers of the first "Metallurgy"(Varna, Uruk, Leyla-Tepe, Maykop, Early-Kura-Araxes and Sumerians) were the same people and originated within an area between Mesopotamia, Eastern Anatolia and South-Caucasus?

1.
The Chronology of the Third Cultural Period at Tepe Hissar:


There is yet another weapon typical of Hissar III ? and C, the adze-axe, This type of axe probably evolved, as Prof. Gordon Childe has always maintained, from the amalgamation of the Sumerian type tubular shafthole adze and axe, but the region where this took place would appear to have been that of Maikop where all three types are found, and from this centre it spread.

2.
Ancient Metallurgy in the USSR: The Early Metal Age:


Metal ornaments occur mainly in burials at Kuro-Araks cemeteries in Elar, Urbnisi, Khizanaant-gora, and others. Long pins with flat, broad, double-spiralled heads are found in the later levels of Kuro-araks settlements. They mark the beginning of a major series of similar ornaments which occur throughout the caucasian cultures of the MBA and LBA. They are also known from Iran, from the Tepe Hissar II period, and in eastern Anatolia, from the so-called Late Chalcolithic levels of the Tepecik settlement which must date to the second half of the fourth millenium BC.

The latest levels of Namazga III may be compared, in terms of their ceramic and terracotta assemblages, via Tepe Hissar IB and IC(in Iran), with Tepe Sialk III/5-7, and via these with Uruk XV-IV.

Only the knives from Kyul-tepe I and Tekhut provide an indication that these forms were later to occupy one of the central places in the production of Transcaucasian metallurgical centres, already linked with the Kuro-Araks culture. Relatively similar forms of tanged knives occur at sites of the Sialk II-III and Hissar I types.

3.
A COMPREHENSIVE STUDY OF ANTHROPOLOGICAL MATERIALS OF THE LATE ENEOLITHIC FROM THE ARENI 1 CAVE:


Results of the intergroup statistical analysis identified close morphological affinities between the Eneolithic skulls from Areni 1 with groups from Tepe Hissar II, Ginchi, Catal Huyuk, Alishar Huyuk and the bearers of the Kuro-Araxes from the Southern Caucasus. As for race and genetics, the Late Eneolithic inhabitants from Areni 1 cave, more probably, originate from the territory of the Middle East.

4.
The Indo-Europeans: Archeological Problems:


The royal tombs at Maikop and Tsarskaja in Kuban, with burials in house graves built of timber or of stone slabs equipped with a fantastic amount of gold, silver, copper, pottery and stone vases, gold figurines of bulls and lions sewn on garments, gold and silver bull figurines adorning canopies, gold beads and rings, gold, silver, turquoise and carnelian beads, as well as copper axes, daggers and spearheads show close relations with northern Iran (Tepe Hissar III) and with the royal tombs of Alaca Huyuk and Horoztepe in northern central Anatolia (Lloyd 1961).

5.
THE CHRONOLOGY OF THE CAUCASUS DURING THE EARLYMETAL AGE:


While the pottery found in the first group of kurgans is close to the Kura-Araxesculture, the pottery in the second, and later, group is characterized by the so-called ‘pearl-like’ ornaments. This decoration is typical of the Novosvobodnaya(Tsarskaya) stage of the north Caucasian Maikop culture and Early Bronze Age north-east Iranian sites (Tureng Tepe III C, Shah Tepe III, Tepe Hissar II B, Yarim Tepe);two such sherds were found in the ‘Late Chalcolithic’ levels of Alishar(central Anatolia)

6.
The New Chronology of the Bronze Age Settlement of Tepe Hissar, Iran:


We could not disregard the striking resemblances between ... Early Dynastic Sumer and Hissar IIIC.

Suyindik said...

7.
Studies in ancient technology. 5:


Hancar considers the Copper Age of Kuban, Egypt and Mesopotamia individual growths on a common base, but connects Kuban with Tureng Tepe and more closely with Tepe Hissar III.

8.
The Southern Caucasus in Prehistory:


Skulls from the time of formation of the Kura-Araxes culture are representative of the Mediterranean type of the Europoid race. In a number of traits they are close to the series of skulls from such sites as Tepe Hissar II-III, Ubaid, and Kish.

Seinundzeit said...

Matt,

"I'm gonna be stubborn on this ;)."

Haha; I wish I could agree with you on this, especially considering that it is such a minor detail.

But, for the sake of reality, we must continue to disagree, ;). (Anyway, it is fun to talk to you, so I don't mind the friendly debate. You've always been an intelligent, and reasonable, interlocutor)

It just doesn't make sense for these populations to have so much AASI. With them at 20%, you would need to have populations from Iran to be on a cline involving 5%-6% AASI in the northwest (near the Caucasus) to 15%-16% AASI in the southeast (near Bandar Abbas and Balochistan). Pamiri highlanders, as well as Farsi/Dari/Tajiki speaking Tajiks, would be between 10% and 15% AASI. Again, this just doesn't make sense. It's possible, but implausible.

"Or looking at the one with the speculative West Eurasian proto-IVC (that may or may not have ever existed), then Kalash simply 14.6% Onge, 1.3% Han and Pashtun 8.1% Onge, 10.1% AASI, 2.7% Han."

The Han percentages disappear, when using Ulchi and Mongola (so, probably a reflection of West_Siberian_HG, not to mention Turko-Mongolic admixture in the case of Pashtuns).

So, if we assume that only the AASI_SIM/Onge are reflective of actual AASI, your sims do lead to the Kalash having 14.6% Onge, which isn't exactly the 20% we've been discussing. Rather, that's within the range I've seen (10% to 15%).

I'd ignore the Pashtun average, as it isn't based on the usual samples. Rather, it's mostly composed of the most Pakistani/Northwestern Indian-like HGDP Pashtuns (at least 5 of these are Qasibghar, so as ethnically Pashtun as a Lithuanian Ashkenazi would be ethnically Lithuanian), plus 3 Afghan Pashtuns.

If you look at the only sample which resembles the usual HGDP Pashtun average, it gets 13% AASI, using your simulation. The Afghan Pashtuns get 10%.

In addition, Kho_Signali receives around 12%.

So, again, I'd say somewhere between 10%-15% for these populations, somewhere between 5%-0% for Pamiri highlanders/Yaghnobi, and somewhere between 0%-1% for northwestern Iranians.

Anyway, it'll be exciting to play with the actual samples!

Davidski said...

@Gill

Have a look at the Indus_Periphery samples in this paper. They don't show much West_Siberian_HG on top of their Iran_Neolithic-related ancestry, and that's especially true for the sample that has a high level of AASI.

So it's unlikely that there was a lot of extra ANE in Northern South Asia on top of what was brought there by Neolithic farmers from Iran. Probably just a few per cent across the board.

@mzp1

The paper includes co-authors who are experts in Indo-European studies. It's not just a bunch of lab and computer nerds trying to make sense of the data, as you seem to be implying.

And please note that my own views on the Indo-European homeland are fully in line with the consensus held by most historical linguists.

It's not like I'm pushing for a new consensus on this blog, which is what many people seem to be suggesting here and elsewhere. The only reason they're talking nonsense about me is because they're seriously confused, and don't know what the consensus is.

Gill said...

https://yfull.com/tree/R-Z93/

It's crazy how quickly they had to have gotten around most of the world.

R-YP294 has both Asians and Europeans splitting from 1500 BCE!

Gill said...

"So it's unlikely that there was a lot of extra ANE in Northern South Asia on top of what was brought there by Neolithic farmers from Iran. Probably just a few per cent across the board."

Yeah, that's about how much I expected. Nothing too crazy, but a little bit more than the Iranians up until you hit Western China and then quite a bit more on the eastern steppe.

@Davidski:

What do you say about when/where R1a1 and then R1a1a1b2-Z93 arose?

Davidski said...

@Gill

Ukraine_Eneolithic I6561 probably belongs to R1a-Z93, and might be on a branch leading to L657. So that might be a big clue.

Bogdan said...

@Gill:
“Well, I had been guessing all along that ancient Pakistan inhabitants would have had higher ANE-type ancestry than Neolithic Iranians. And so would Central Asians in general (like a western encroach of ANE). I'm guessing that's where the West Siberian HG comes in?

Has the story of when and where R1a1 appeared changed?”

I’m fascinated by this too. Kalash, Kho (and presumably Nuristani) apparently have high ANE-type ancestry and seem to have been in the area for a very long time. There is a lot of R1a among male population among these peoples, but I can’t seem to find reference to specific R1a subclades. Are there two major waves of R1a into the region first from Siberia HGs and later R1a subclades from Steppe through the Indo-Aryan Migrations?

Seinundzeit said...

Also, here's something you may find interesting Matt...

The most AASI-like sample we currently have is Bhumij1. It's in Global_25. No other contemporary population comes close to this much AASI.

Yet, even Bhumij1 has 25% West Eurasian ancestry. David tried a qpAdm model for it.

Bhumij1 turned out to be 75.9% Onge, 16.2% Iran_N, and 7.9% MA1. The chisq is 10.928, and the tail prob is 0.14177.

We now know that Iran_N + MA1 is correct, as West_Siberian_HG are primarily ANE (which is why they work so well for ancient Central Asians/Indus_Diaspora).

So, it might be profitable if you tried to create an AASI sim that is based on the Bhumij1, but which accounts for their 25% AASI (Iran_N, but with additional West_Siberian_HG/ANE).

old europe said...

mzp1

agree with you especially when, as I did some posts ago, you say that genetics is not exclusive factor and vector of cultural change. Nice that you mentioned the example of christianity, I forgot in the example I quoted ( shame on me!) Obviously it can be a proof of a cultural transformation but the approach to a deeper understanding must be multidisciplinary. Also genetics unlike mathematics seems not to be yet an "exact science" in the sense that his principles methods are still in formation.

Seinundzeit said...

^^

Lol, above, I meant 25% West Eurasian, not 25% AASI.

Arza said...

Can this preprint tell us anything new about "Tokharians" from the Tarim Basin?

Davidski said...

@Bogdan

Indian R1a is almost all Z645+, so it's more closely related to Eastern European/Norse R1a, than Eastern European/Norse R1a is to Northwest European R1a, which is L664+.

So Indian R1a comes from Eastern Europe (via Central Asia) for sure.

This is basically in line with the historical linguistics consensus, which posits that Balto-Slavic and Indo-Iranian languages share a closer relationship with each other than with most other Indo-European language groups.

old europe said...


suppose that in the year 5000 CE people have no possibilities to have books or video about the past. They are puzzled because they have been told that back in the early III millennium all the world communicated in English. They start to wonder where english comes from? They must rely only in genetics and archeology.

suppose we have a guy that gives importance to archeology. He will ask himself which kind of cultural traits there were, which kind of society there was in early III millennium. By digging he discovers that back at that time there was a society based on factories and information technology. Maybe he will start to think that english is connected with these cultural trait. If he is smart he will try to detect where this cultural traits pop up first. Keep on digging and he finds that the first factories and the first computers pop up first in a tiny island far off the cost of france ( factories) and in northern part of the american continent ( computers) . His mind set will push him to say that english comes either from ......Britannia or the north american continent.

But then we have another guy that believe only in genetics and in movement of people. He will say that since back at that time many people were on the move from subsaharian africa and from asia to europe and north america ( so a gene flows from africa and asia to north america). He will proclaim that english must have its root in subsaharian africa ( because the genes that are on the move nowadays are predominantly from africa).



Davidski said...

@old europe

You'll have to come up with more convincing arguments than that.

Indo-European studies is a field that is well over a century old, and there's a strong consensus based on linguistics data alone where the Proto-Indo-European homeland was located.

You might want to look into that first before assuming, wrongly, that this is all about ancient DNA.

Bogdan said...

@Davidski:

Thanks for this insight. Did more upstream R1a also came in to N.Pakistan Hindu-Kush area a lot earlier than Indo-Aryan Migrations?
(through Western Siberian HGs?)

See: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4570283/

old europe said...


I forgot...the first guy is not only smart but very very smart and he will soon realize that factories in the isle north of france are older than computers in the north american continent.....he will proclaim Britannia as the source of the english language.

Kurti said...

"Ha! Ha! His ego is as big as the sky. He won't admit it. He will spin every which way to delude himself."

Most of Indo European groups seem to have spred from the Steppes though.

Davidski said...

@Bogdan

Almost all of the R1a in South Asia is Z645 derived. Also, in regards to the Kalash, see here...

http://eurogenes.blogspot.com/2015/04/the-enigma-of-kalash.html

@Kurti

Many of the Y-haplogroup assignments in this paper are wrong and that supposed L23+ sample from Hajji Firuz doesn't have a C14 dating. You might want to take note of this before you start bragging.

Matt Thomas said...

I believe the below happened.

Steppe_MLBA_West - (Steppe_EMBA + 26% Europe MN). The proportion of Europe MN would be much smaller as stated in the paper due to geographical location for Steppe_MLBA_East.

Most South Asians - Indus Periphery + Steppe_MLBA_East / Steppe_EMBA
Northwest South Asians / Brahmin - Indus Periphery + Steppe_MLBA_East / Steppe_EMBA + Steppe_MLBA_West (Andronovo)

This would make complete sense and would account for the extra WHG ancestry in Northwest South Asians.

Kurti said...

"Many of the Y-haplogroup assignments in this paper are wrong and that supposed L23+ sample from Hajji Firuz doesn't have a C14 dating. You might want to take note of this before you start bragging."

Most are not wrong, just low coverage. That's why the A's and many other upstream Haplogroups. However that does not include the L23 sample in any way. Many bloggers have re-checked the sample. Seems the R2 argument is not working here.

Davidski said...

@Kurti

Many are plain wrong. Have a closer look before commenting on the issue again here.

Twasztar said...

If Hajji Firuz and Iron Gates M269 gets confirmed, will it mean that the Middle Eastern or Balkan homeland of R1b is more likely than the steppe? If so, anybody got an idea what language did those people originally speak?

Roy King said...

@Davidski
Unfortunately, the mtDNA e.g. U2e seem to amplify in Swat Valley and are from the Steppe. The R1a in Swat Valley is relatively rare. It fits with a female mediated steppe ancestry in Northern India.
Perhaps, the AIT is female mediated?

a said...


We will have to wait and see if Hajji Firuz turn out to be R1b-Z2103 or R1b-M269 since there is a closer relationship between R1b-M269 and R1b-M73 than R1b-Z2103. R1b-M269 branches R1b-M73. Sample I0124, EHG, Lebyazhinka IV (Russia)---R1b-M73, 5640-5555 calBCE is classed as Hunter Gatherer. There have been no ancient R1b-M73 found outside of Europe. Yamanya samples buried close to Russian Hunter Gatherer(within 7km) --Yamnaya, Luzhki I, I0438, 3021-2635 BC and Yamnaya, Lopatino II, I0443, 3300-2700 BC. So a good test to validate origin of Iranian farmer would be to compare/parse levels amongst the steppe samples {Samara-R1b-M73}--{KhvalysnkR1b-I0122, Khvalynsk II, Volga River, Samara (Russia), 5200-4000 BCE}---Yamnaya samples Yamnaya, Luzhki I, I0438, 3021-2635 BC and Yamnaya, Lopatino II, I0443, 3300-2700 BC.

Davidski said...

@Roy King

That's very funny.

But of course stochastic effects, like drift, often have a bigger impact on Y-haplogroup frequencies than mtDNA haplogroup frequencies, for one, because male lines usually derive from fewer founders.

And small, isolated populations are most susceptible to such stochastic effects; small, isolated populations like those living in massive mountain ranges, like where the Swat Valley is.

Davidski said...

Yeah, the 50-80% frequencies of R1a-Z93 across North India and Pakistan are also female mediated, especially in the upper castes.

Oh, no wait, that's a Paleolithic relict...or something.

Anthro Survey said...

@Roy and Rob

Guys, read my comment above addressed to Alberto and the Indian commentators. It's true--Swat samples have but a spit of steppe(probably not all, at that) and most of that orange is likely western BMAC-related influx, but we shouldn't put too much weight on them. It's a gatehouse, a checkpoint to South Asia proper, not exactly South Asia. Also, places like that are also bound to be exotic(for lack of a better word), so I don't know how much use it will have in modeling modern South and South Central Asians to be honest.

Not going to find conclusive evidence of a steppe_MLBA surge there. We need to be looking on the Ganges or at the very least in Punjab proper/Saraikistan.

Arza said...

Description from the supplement:
• FH8, K10 B1 (I2327): Archaeological Date of 5900-5500 BCE. Genetically male. This is a
bin ossuary burial in Phase F-G, which is skeletally a 25- to 35-year-old male.

doesn't match anything here:
Hajji Firuz Tepe, Iran: the Neolithic settlement (pdf, 651MB)
http://digital.library.stonybrook.edu/cdm/ref/collection/amar/id/101623

Square K10 wasn't excavated to the Phase F-G at all (neighbouring Operation IV was). Phase F-G is about 2 meters below A3 which was carbon dated to 5900-6000 BCE. And between K10 and Op IV there was a part that eroded and collapsed between excavations 1961 and 1968.

https://s6.postimg.org/5ve2z2ntt/HF1.jpg
https://s6.postimg.org/e0w4x84cx/HF2.jpg
https://s6.postimg.org/aha77etxd/HF3.jpg

Either there were later excavations not cited in the preprint or a major f*ck up happened here.

But if so... why the autosomal profile matches other samples from this site?

Salden said...

The babble by old europe and the hinduvata is just babble.

1. The spread of Christianity isn't remotely comparable to the situation of IE in India. Christianity from the start was proselytizing and more open (see the quote of all groups being one in Christ) than its rivals, to the point of accepting slaves and women into its fold. There's no evidence that the bringers of IE to South Asia had any plans of proselytizing or were remotely more open than their rivals in the region.

2. The Arab, Moorish, Turkic invasions with the Sub-Saharan Slave Trade did leave impacts genetically across the world. Whether it's Spain having significant amounts of Maghrebi admixture, North Africa to Mespotamia having Arabian admixture, or even the Far Eastern admixture in modern Turks.

Sorry, but People did carry Pots with them.

Mr. Kulkarni said...

R1a seems to have entered north india too late, post 1000bc.
Is it possible something like J2a is responsible for Ie expansion, originating in Georgia 7000bc and going north (Karelia) as well as east into iran.

Might explain similarity between balto Slavic and indo Iranian.

Seinundzeit said...

Anthro Survey,

You're right on all counts.

There is definitely Steppe_MLBA_East (or perhaps Steppe_MLBA_West + Steppe_MLBA_East; we'll find out when these samples are released) ancestry everywhere in modern Southern Central Asia, Northern South Asia, and the Iranian plateau. These Steppe_MLBA_East/Steppe_MLBA_West populations brought both R1a-Z93 and the Indo-Iranian languages into the interconnected regions of Iran, Turan, and Hind.

Quite a few people seem to be forgetting this, due to the fact that many ancient South Central Asians were, in a roundabout way, sorta like Yamnaya. I mean, Yamnaya was CHG/western Iranian + EHG, while BMAC was eastern Iranian + West_Siberian_HG/ANE. In terms of deep ancestry, these populations were similar. I used to think this wasn't possible, but I guess it really was the case. Still, the existence of Yamnaya-like Central Asians doesn't eliminate the need for Steppe_MLBA (probably Steppe_MLBA_East) in contemporary South Central Asia, northern South Asia, and Iran.

With this being duly noted (again, there is Steppe_MLBA in Turan, Iran, and Hind. These Steppe_MLBA populations brought Indo-Iranian languages), I think the paper is probably wrong about Steppe_MLBA admixture in the ancient Swat valley.

Rather, these ancient Swat valley samples seem to be mixtures between ancient South Central Asians and ancient South Asians, but with an excess of ANE/West_Siberian_HG.

Even today, the Swat valley is a transitional region, a meeting point for the Pashtun highlands (southern Central Asia) and Punjabi lowlands (northern South Asia), with a diverse population of Central Asian-affiliated Pashtuns/Dardic highlanders living alongside South Asian-affiliated Hindkowans, Dehqans, Qasibghar, and Doms. Furthermore, these various populations of Central Asian and South Asian affinity are thoroughly mixed with each other (they marry freely). For example, the Pashtuns of the Swat valley are much more genetically South Asian in comparison to the Pashtuns neighboring them on their west, and Swat valley "Indians" are more South Central Asian-shifted when compared to true Punjabis.

So, I guess these samples show that the same situation may have been at play at that stage in prehistory: basically, the samples have an affinity to both BMAC and IVC, but with the local distinction of having excess ANE.

As you say, we really should be looking in the Greater Punjab, if we want to find samples enriched in Steppe_MLBA_East (or again, perhaps Steppe_MLBA_West + Steppe_MLBA_East).

Samuel Andrews said...

@Everyone,

The only India/Pakistan samples being from BA/IA Swat valley is very disappointing. But the huge amount of data from Iran & Central Asia & Siberia confirms the Steppe signal in India is real. It can't be explained by anything but real Steppe ancestry.

That alone is convincing evidence for a Steppe origin of Indo Aryan languages and Indo European languages in general. Add to that, the fact the Steppe people who moved deep into Central Asia in the 2nd millennium BC belonged almost exclusively to R1a Z93.

Knowing all this, it's hard to argue the Aryan invasion didn't happen.

Samuel Andrews said...

Recall the post on Steppe mtDNA I wrote a few months.

http://mtdnawiki.com/2018/01/27/steppe-mtdna/

Look at the haplogroups I listed as Steppe mHGs then look at Afanasievo & Andronovo/Sintashta mtDNA results from this study. Here's the frequency of mHGs I identified as Steppe haplogroups in that post (Northern Europeans score ~20%).

Afanasievo. N=39.
Steppe mtDNA=54%

Andronovo, Sintashta. N=109
Steppe mtDNA=48%

Plus look at these unique mtDNA links with modern South Asians....

T1a1b: Tajik=4, India, Sycthian=2
H7b: India=2, Tajik=4, BA Russia, BA Kazakhstan, Sycthian.'
K2a5: Sindhi, Pathan, India, Corded Ware=2, Sintashta=1
J2b1a2: Tajik=3, Sintashta, Srubanya, BA Kazakhstan=2
N1a1a1a1: Turkmen, India=2. Sintashta, BA Kazakhstan=2, Sycthian, Sarmatian.
U2e1h: Kalash=3, Tajik=8. Sintashta=3, Potapovaka

Samuel Andrews said...

H7b & W1c & U2e1b hadn't been found in any ancient Steppe DNA before this study. I predicted those mHGs were of Steppe origin because I had found them in India, SC Asia, Iran, and several locations in Europe.

Mr. Kulkarni said...

@samuel
Great, you see steppe mtdna in turan, bmac adna but not y haplogroups. Maybe better to hold off on the 'invasion' proclamation.

Samuel Andrews said...

Afanasievo/Yamnaya R1b Z2103+ share mtDNA with Sintashta/Andronovo R1a Z93+. Northern Beaker R1b P312+ & Corded Ware R1a M417+ also shared much of the same mtDNA.

IMO, this means they all derive from the same PIE community who lived sometime around 5000 BC and carried R1a, R1b, I2a, and probably other yHGs. Eventually, a R1a M417 & a R1b L23 group formed. Only begging in 3rd millenium BC IE languages began spreading outside of the European Steppe.

Davidski said...

@Kulkarni

Those Andronovo groups from the Kazakh/Uzbek/Tajik borderlands are packed with R1a, almost all of which is probably Z93.

You don't see a connection yet with India?

Samuel Andrews said...

I know I'm ignorant of lingustics. But I'm thinking, this scienero can explain how PIE can date to 5000 BC but still have words for technology which didn't arrive till many years afterwards.

5000 BC: CHG, EHG mixed PIE community emerges.
5000-4000 BC: IE speakers spread across the European Steppe, replacing most of the previous inhabitants.
4000BC: Distinct IE communities form in the European Steppe. Some carry R1b L23, some R1a M417. For a good 500 years, they still speak the same language. After 1,000 years, they can barely can't converse but share much of the same vocab.
4000-3000 BC: New inventions, like the wheel & metal, enter IE communities. The multiple IE communities, because of contact with each other & similar languages, adopt the same words for each invention.

This process would create the impression their common ancestor language had words for wheel, metal, etc.

Salden said...

Anyway, on a different note:

http://archive.is/JlxcH

The paper that came out last month on the First Intermediate Period to Middle Kingdom Egyptian mummy head had its haplogroup discovery supported by the team behind last year's Egypt Ancient DNA paper.

Kristiina said...

@ Aram

A and B are probably Africa-rooted lineages, and A and B in the paper are unresolved lines. However, there are reasons to presume that there was one exit via the Horn of Africa to Arabia and Pakistan as the paper proposes. This exit could correspond to yDNAs CT and DE and mtDNAs L3'M'N. Then, there were multiple exits out of circum-Pakistan to North Africa, Near East, Central Asia, Southeast Asia and Indonesia and Australia. As DE has a different distribution from the rest, E may have back-migrated from Arabia to Africa earlier, i.e. c. 65 kya ago which is the age of E on y full. However, I do not know the structure of E so well, hence I won't argue which line would be specific to East Africa, which to North Africa, which to Arabia and if there is a line that could be old in southwest Pakistan. Let's see if these deep roots will be resolved one day.

Salden said...

http://www.icaane2018.vorderas-archaeologie.uni-muenchen.de/index.html

https://anthrogenica.com/showthread.php?8066-DISCUSSION-THREAD-FOR-quot-Genetic-Genealogy-and-Ancient-DNA-in-the-News-quot/page131

A bunch of abstracts of what'll be covered at a gathering are listed in both links. The second one focused on ancient DNA.

Mr. Kulkarni said...

@davidski
look at the dating of that kazakh/uzbek/tajik R1a. let me refresh you.

R1a moving south.
Kazakh r1a - 1700bc
southern Kazakh r1a - 1600-1500bc
Uzbek r1a - 1200-1000bc
Swat r1a - 500bc

too late brother.

postneo said...

@davidski
"The paper includes co-authors who are experts in Indo-European studies"

they called brahmins and bhumihars custodians of old Sanskrit "text"... there was no text. This is not even close to accurate and neither are the authors trying to be. Its a perfunctory sentence written in a hurry.

Swat should have been steppe chariot central but it seems it has little steppe dna , or steppe burials. It starts picking up some steppe by 400 bc.

Mr. Kulkarni said...

@postneo

I loled at at the bhumihar part. They were classified as shudras by early britishers, which the bhumihars protested. They wanted to be classified as Brahmins.

They were land owners and certainly not priestly, as was the argument put by them to the British "Priestly functions alone do not define Brahmins"

The authors were trying too hard.

Ric Hern said...

Let's not forget 362 individuals.....and we are going cookoo about a hand full only...lets wait for the big picture...

Alberto said...

@Anthro Survey

Yes, I agree with all that you wrote. Good to know it's not just my impression, but it's shared by others too.

@Arza

Can this preprint tell us anything new about "Tokharians" from the Tarim Basin?

Well, it can confirm something that was quite clear already:

However, this sample [Dali_EMBA] is an important data point in our observations relating to the geographic range of the Steppe_EMBA. We know from the geographic and genetic positions of Afanasievo that the Steppe_EMBA were a highly mobile community and were capable of large scale movement within the steppe zone. However, several arguments, including the lack of admixture in the BMAC sites or the late Copper Age sites in Turan (which are roughly contemporaneous with the Afanasievo culture), provide evidence against their movement and contact with agricultural settlements to the south. The Dali sample (which is contemporaneous with the majority of our Steppe_EMBA samples) is from due south of the Afanasievo on the Inner Asian Mountain Corridor, an ancient migratory route between Turan and Central Asia (84). It does not display admixture from the Steppe_EMBA but does have Iranian agriculturalist-related ancestry, suggesting that populations like those from the Neolithic forest zone remained in Kazakhstan and the southern Steppe, and that Steppe_EMBA may have been restricted to the northern Steppe zone at this time. This is also consistent with our observation of an outlier sample from the earliest BMAC city of Gonur which did not have any Steppe_EMBA-related ancestry.

Further:

The samples from Siberia can be modeled with virtually no ancestry from Anatolian-related agriculturalists, though they have additional ancestry from Iranian-related agriculturalists (a phenomonen we saw in the Dali_EBA sample from Eastern Kazakhstan), as well as a significant proportion of their ancestry from West Siberian hunter-gatherer related groups (similar to those from the Tomsk and Tyumen oblasts) that did not admix with European groups (Table S3.54-S3.55). We are unable to model Okunevo.SG with proximal models, presumably due to the fact that we do not have the right proximal source in our dataset. However, we can show that Kanai_MBA from further south is a genetic clade with Okunevo.SG relative to our outgroup set using qpAdm (p-value = 0.15). This suggests that the range of the ancestry type that was represented in the Okunevo culture extended beyond the Minusinsk Basin (where the Okunevo material culture is documented) and into the Inner Asian Mountain Corridor. It also provides fiurther evidence that the Afanasievo made little genetic impact on the populations here and indeed further south during this time.

So any connection between Afanasievo and the Tarim Basin is discarded, and therefor any connection between Afanasievo and Tokharian.

Chetan said...

@David Do you have the data to model Brahmin groups from different parts of India using this new info? There may be a lot of surprises now different from what was seen in the Harappa DNA project.

Aram said...

@All

I had a close look to other Y dna assignments. Not only adna but also modern.
Many of them completly wrong.
Numerous E1a in modern India.
M269 in Iron Gates, Kura Araxes and Mal'ta boy.
I2a2a in BA Turkmenistan
Kotias is J2a1h.. Plain wrong
Deep subclade of E1b1a (Niger Congo) in Iran

Etc etc

The software is wrong doing and it is possible that some of this E1a-s are real R1a.

Aram said...

Alberto

I am not ruling out the possibility that L1a and EHG came to Areni from Turan. But then again why they are ANF shifted and no trace of that WSiberia Hg. Also notice that y dna L1b is purely West Asian y dna. So hard to say.

Anthro

I think L23 level formed in Steppe. But it is possible that M269 split into L23 and Pf7562 occured in West Asia. This is a possibility that we can't rule out.
Pf7562 moved into C/BA Anatolia creating Hittites and others. While L23 expanded from Steppe.

M269 ultimately came to West Asia from EHG side. Either directly either via Central Asia.

Seinundzeit said...

Finally reading the supplementary materials (in detail)...

Lots of complexity, and many things that deserve comment. Will do so tomorrow.

Aram said...

I must add. By EHG side I mean that Volga region.

Btw Hajji Firuz mtdna has no singla match with Yamna and Afanasievo.

a said...

@Arza

"Can this preprint tell us anything new about "Tokharians" from the Tarim Basin?"

So far there is only one branch R1b-Z2103 that is known to have practiced wagon burials. Afansievo and Yamnaya match ydna and autosomal and location of known proto Indo Iranians.
https://erenow.com/ancient/the-horse-the-wheel-and-language/the-horse-the-wheel-and-language.files/image018.jpg

Anthro Survey said...

@Rob

The question isn't so much where, but where exactly. As for where: Punjab or western Ganges between 1500 and 500BC would be ideal, of course. That's where history happened. As for where exactly? Well, with cremation being commonplace during this period, where would we potentially find well preserved individuals (of higher caste?) carrying higher-than-average steppe-MLBA? Himalayas(think Indian Oetzi)? Thar flats?

Rob said...

@ AnthroS/ Alberto
Yes I understand. But isn’t Swat meant to be the centre piece of the AIT? Some predictions had it as 70% steppe
Anyhow, let’s wait for the data and re-do of haplogroups ...

Anthro Survey said...

@Seinundzeit

Yeah, pretty much what I'm thinking. The presence of ANF-related ancestry(BMAC era introduction) is ofc another factor playing a role in elevating steppe-MLBA ancestry for C and SC Asia(but not so much S Asia) across various methods. ANF+Iran_N+West_Siberian_HG make for a lot of deep-ancestry overlap with Steppe_MLBA, indeed. To make matters more uncertain, we don't know what Mesolithic North Indians were like: AASI or AASI-basalANE?

Re/what you said about modern Swat----I find it interesting how so many genetic deja vu scenarios played out across geographical zones over the millenia: either in terms of deme bifurcations or patterns of migration. For example, look at how south Kazakhstan's Taraz area historically represented a border zone between the hot, arid, urban Turan zone to the south and the less densely populated, forested/stepped greener, yurt-dweller zone to the north. Seems like this was the case even in the prehistory, but the genetic clines were somewhat diff, obviously(Iran_N-ANF and West_siberian-like HGs).
Another example is how the region of today's Pontic-caspian steppe seems to have been used as a conduit for HG migrants into Europe(e.g. Kostenki) back in Paleolithic times.

Rob said...

@ Anthro

“have been used as a conduit for HG migrants into Europe(e.g. Kostenki) back in Paleolithic times”

Where did you get the idea that the Aurignacian is from Russia ?

aniasi said...

@Chetan

"@David Do you have the data to model Brahmin groups from different parts of India using this new info? There may be a lot of surprises now different from what was seen in the Harappa DNA project."

Ditto. Would you know who could get an MLBA Steppe/Indus Periphery/AASI utility up on GedMatch?

Ariel said...

I don't understand some of the comments here, based on this paper it's pretty obvious that neolithic south asians lacked steppe admixture compared to the modern counterparts, even late in the bronze you have samples with little to no steppe, so that has to be a recent phenomenon. And the staggering lack of R1a in these samples clearly doesn't support the idea that R1a is there since the mesolithic. Yes, the picture is fairly complicated but these theories about a secret para-ANE signal hiding somewhere in India are clearly wrong, south asians today have signals from an EHG groups, so a WHG+ANE mix which was native of the European steppe, there are many glaring evidence of that. If that admixture is connected or not to IE it's a different topic altogether...

Davidski said...

The first thing I'll do is to update the Global25 datasheets with most of the new ancient samples from Central and South Asia.

I only have Brahmins from North India and Tamil Nadu in those datasheets, and I can't add any others because they're not freely available. Maybe one day.

But there are plenty of other Indians there, and you guys will be able to model them any which way you like using the Global25 coordinates.

Jijnasu said...

@seinundzeit While the swat valley is is peripheral to the Indosphere. I think it would be strange if it continued to be occupied by pre-Indo-European people until relatively late in its history. The Rig Veda itself mentions swat and names some of the western tributaries of the Indus. By the late Vedic Gandhara was a centre of language learning noted for its conservative indo-aryan speech. It does seem rather strange if the indo-aryan settlers along the Indus and the kabul rivers left swat uninfluenced

The Aviator said...

Hi, I've done a 23andme DNA test and (this might be noobishly unnecessary) but if any of that data is valuable to you guys for collation/comparison etc. I can provide it :)


Background - Punjabi Jat.

Paternal Haplogroup - H-M69
Maternal Hapologroup - M5a2a1

https://i.imgur.com/rxht4QK.png - results in the GEDMatch "GedrosiaDnA -K9 ASI" - suspect it's typical.

But if anyone needs any more data, I'll be happy to oblige!

Suyindik said...

I think that most of the A, BT, CT, A0-T, DE data are incorrect and should belong to other haplogroups, i hope they can fix this too. And what do you guys think about the prediction of Leyla-Tepe, Maikop and Early Kura-Araxes(before 3500 BCE) cultures? Could they have the same kind of haplogroups(J, J2a1h2, L2, T, T1a, H3) like in the "Tepe Hissar" which is also the basis of the Ubaid/Uruk/Sumerians? All these cultures are archaeologically and anthropologically proven to be close. Maybe the results will indeed show a migration of Mesopotamia into the South-North Caucasus + Eastern Anatolia?

Mr. Kulkarni said...

The current data sheet if correct shows:

J2a from Georgia moving west into Anatolia/west turkey, and to east Austria.
Moving north/north west to Karelia. Both around 6000bc. Moving east into Iran and later India.

A good contender for origin of Balto slavic, greek, indo iranian. Might explain indo iranian loanwords in Finno Ugric as well.

Bronze said...

@Ariel

Nope, we dont know what kind of R1a lieages might have existed in the indus valley before any theoretical migration from the steppe, others have mentioned cremation as an explanation why R1a has barely been found in these remains, and we didnt get dna from south asia proper like the punjab region. Also those siberian hunter gatherers could be the source and not steppe, we didnt get any Y-dna from them but they could be R1a, since Mal”ta was R1b and likely basal to european R1b, same could be true of R1a. Proto indo european likely came from Iran and was brought to the steppe along with paternal admixture.

Cpk said...

Hajji Firuz no C14 argument is already getting old. It has not c14 and also: no steppe in it. Either way Reich wouldn't put the homeland there in his book and on the whiteboard with only one sample. He has probably many of them unpublished.

Santosh said...

It is quite clear that the indus valley people where dravidian/ or proto dravidian. I would even call the indus valley the dravidian urheimat. There are several reasons.

1) http://www.thehindu.com/news/states/Symbols-akin-to-Indus-valley-culture-discovered-in-Kerala/article16883884.ece
2) Y Haplogroup L. The distribution is clearly indus valley, western coast of india up to Kerala.

And I am pretty sure the Rakhigarh sample have L. Precisely the reason they are not publishing it.

epoch2013 said...

@Cpk

This would be the paper to publish them in, wouldn't you say?

Cpk said...

@epoch2013

The Genomic Formation of ''South and Central Asia'' paper?

Lenny Dykstra said...

@David,

Doesn't look like Global 25 has been updated yet, can you notify when/which samples from this paper you'll be adding?

Also, will you be adding any samples to G25 from Lipson etal 2018 or McColl et al 2018??? The "Hoabinhian" Mesolithic foragers from Laos, Malaysia etc are probably better "pure" AASI proxies than anything reported in this paper, since Swat is the closest we got to the heart of the subcontinent.

I have a feeling modern South Asians from Afpak to Tamil Nadu (but not "easterners" like Bengalis etc) can be modeled in nMonte very well now as a 4-way mix of Steppe_MLBA, West Siberian_HG, Iran_N, and "Hoabinhian" AASI.

Arza said...

@ Santosh
Indeed it's strange as over one year ago everyone here was confident that a lot of L was already found in Rakhigarhi.

http://eurogenes.blogspot.com/2017/02/rumors-leaks-thread.html

@ epoch2013
Maybe Maykop?

Botai is also missing, despite that they have Botai samples for sure as Botai appears in the stats in the supplement (page 170, f3 stats involving Swat Valley):

https://s6.postimg.org/ljjvjn341/botai.jpg

epoch2013 said...

@Arza

Then why publish this sample in this study?

Chetan said...

@Santosh Like we discussed before, the lower Indus valley, the Sindh region of today may indeed have been the Dravidian Urheimat. In that case, haplogroup L may be linked to its expansion to the south. But alas we don't have genome data from IV that would confirm this.

But unlike what many believe before, it is clearer now that both Dravidian expansion in the south and IA in the south were roughly contemporaneous.

Mr. Kulkarni said...

Its clear that R2a entered NW India from Iran.
Its also clear that J2a entered NW india from Iran.

Current population has R2a max freq in SE India in Telugu people, roughly correlates with Dravidian speakers. Map here https://ipfs.io/ipfs/QmXoypizjW3WknFiJnKLwHCnL72vedxjQkDDP1mXWo6uco/wiki/Haplogroup_R2a.html

Current population has J2a peaking in South Pakistan, Gujarat, Rajasthan, roughly coinciding with IE speakers. Map is here https://commons.wikimedia.org/wiki/File:The_spatial_distribution_of_M410(J2a)_clade_in_worldwide_populations.png
It seems J2a dominant people pushed R2a people to the SE corner, and that J2a entered later. Male invasion, if any, seems to from J2a populations.

R1a entered much later.



Matt Thomas said...

@Kulkarni

J2 definitely entered South Asia much before R2a. The highest frequency of J2 is found amoung Vellalar in Tamil Nadu. The oldest R2 is found among Burusho People and is the major haplogroup among them along with R1a. The highest frequencey of R2 is South Asia is among Jaunpur Kshatriyas. It is also the second most important haplogroup among UP Brahmins with a frequency (20 - 35%) after R1a. Some of the R2 most likely moved into South Asia from Central Asia along with R1a after the arrival from the steppe.

R2 also has higher percentages in North Eastern and South Eastern Iran indicating that it moved there from Central Asia and decreasing to very small percentages in Western Iran. R2 was most likely present in populations in Central Asia before the arrival of R1a. It is present at about 8% in Tajiks and Pamiris.

It was most likely present in the initial movement of ANE westwards through Central Asia as it also found at higher percentages in (North Caucuses) Ossentians, Georgians, Chechens and Armenians but not present in South of Caucuses or Turkey.

Singh said...

@MattThomas

That is not accurate. I don't know when R2 entered South Asia but it did not come with the I.E steppes.

R2 in South Asia has high frequency in Eastern India, Karmalis tribes at almost 100% R2 (shaoo et al.), Lodha tribes show is at 43% (Vikrant et al.) and as we move West it decreases greatly, Burusho show it at 18% (firasat et al.)

None of these above group are Indo-European speakers, including Burusho (language isolate).

We have one known sample of Iran Neolithic who were R2, that should give us some clue, and Iran_Neo themselves are partially ANE. So, ultimately R2 could have arrived from common source in C.Asia.

Taymas said...

Sorry for the deletes. I don't fully understand why they ruled out a test of South Asian = [BMAC + IndusPer + AASI], and trying it to piece it together. Any input would be appreciated.

Matt Thomas said...

@Singh

I agree, R2 did not come from Steppes but rather from Central Asia. It probably moved along with some of the R1a Steppe Migrants into South Asia as it has a very high presence in Punjabi, UP and West Bengal Brahmins, also higher percentage among New Delhi Hindus.

It also has a higher presence in Eastern Central Asia (Kalmyk) among where it most like originated, whereas R1 seems to have a higher presence in Western Central Asia as well as the steppe.

Mr. Kulkarni said...

@matt thomas

what I do know is that there is both J2a and R2a in Swat samples of 900BC, and also BMAC samples 1700-1800bc, but no R1a in both..

Matt said...

@Taymas, BMAC and the Indus_Periphery are pretty close, so in the first instance you'd need a lot of admixture from BMAC to shift the population average.

In the second instance, judging by their reported results, BMAC wouldn't push later populations in the right kind of direction; BMAC is modestly shifted in the direction of Levant Neolithic and the Barcin Neolithic, while later South Asians need admixture to shift them towards the north of Eurasia, esp. NW Eurasia, relative to Indus_Periphery.

Those together are strong reasons not to try and model that way. (Though I'd guess they did anyway to be thorough and there will be a line in the paper to that effect, just did not explicitly report on it in detail).

The most they could do that I could see would maybe would work (judging by the Fig 2C models) is come up with some fantasy population which is like BMAC+more Anatolia+a really strong resurgence of West_Siberian_HG; but this probably wouldn't work anyway in the fine detail, and it's not very straightforward compared to the Steppe_MLBA populations who were already about anyway (most plausible way a West_Siberian enrichment could probably happen is from agro-pastoralist populations like Okunevo or Dali_EBA, which is in any case a migration from the steppe again, so hard for us to be motivated to prefer a ghost steppe migration and an extra movement from the southwest to just using people who are generally about).

Santosh said...

@Chetan

"@Santosh Like we discussed before, the lower Indus valley, the Sindh region of today may indeed have been the Dravidian Urheimat. In that case, haplogroup L may be linked to its expansion to the south. But alas we don't have genome data from IV that would confirm this.

But unlike what many believe before, it is clearer now that both Dravidian expansion in the south and IA in the south were roughly contemporaneous."

LOL! I am a different Santosh than the guy who wrote recently attaching Dravidian to IVC.

Seinundzeit said...

Anthro Survey,

"Yeah, pretty much what I'm thinking. The presence of ANF-related ancestry(BMAC era introduction) is ofc another factor playing a role in elevating steppe-MLBA ancestry for C and SC Asia(but not so much S Asia) across various methods. ANF+Iran_N+West_Siberian_HG make for a lot of deep-ancestry overlap with Steppe_MLBA, indeed. To make matters more uncertain, we don't know what Mesolithic North Indians were like: AASI or AASI-basalANE?"

Interesting possibilities.

"Re/what you said about modern Swat----I find it interesting how so many genetic deja vu scenarios played out across geographical zones over the millenia: either in terms of deme bifurcations or patterns of migration."

I often find myself thinking about this phenomenon.

All,

Just some things I found interesting in the supplementary materials...

1. Their method of determining the "Indian Cline" is the primary factor which goes into the conceptual scheme they've applied to India. At the end of the day, it also explains the unnecessarily complicated terminological system.

2. A major benefit of having these samples in Global_25 will be the possibility of running "distal" and "proximate" sources together (something which can't be done in qpAdm).

3. At one point, they say that the "Indus_Periphery" samples are the only ancient Central Asians which display Onge-related admixture.

Yet, they detect "AASI" for Turkmenistan_Gonur1_BA (4%-5%), Sappali_Tepe_BA (3%-5%), Dzharkutan1_BA (3%-6%), Bustan_BA (7%-8%), and Shahr_I_Sokhta_BA1 (7%).

So, the primary differentiating factor between these native Central Asians and those outliers isn't necessarily AASI. Rather, it's the total lack of Anatolian_N-related admixture in the outliers/IVC migrant samples.

Interestingly, the IVC migrant samples range from 13%-8% West_Siberian_N. Surprising that this West_Siberian_N (well, ANE) stream of ancestry has a noticeable presence in samples with roots in the Indus valley and beyond (I mean, the sample that seems to be 40% Onge-related is probably from beyond the Indus valley, yet still has West_Siberian_N).

4. Speaking of which, good that we finally have adequate ANE representatives; the West_Siberian_N samples are just what we needed.

I wonder though, is the West_Siberian_N (essentially ANE) ancestry in ancient Turan and the Swat Valley also admixed with East Asians? I think not, as this sort of ancestry in Central Asia isn't actually from the steppe/west Siberia.

Rather, it's just local HG ancestry, so perhaps missing the 10% ENA (as per supplementary materials, 11% Han) seen with the Siberian ANE samples. Or not (lol).

5. As Arza noted, Botai comes up in the strongest f3 stat for SPGT. I knew they had Botai samples (and I was quite disappointed that these were not presented), so this is kinda funny.

I wonder how the models will look with those samples?

6. I think they would benefit greatly from including CHG, Levant_N, and Iran_Chl in their analyses.

7. Looking at how BMAC is modeled, there is substantial BMAC-related ancestry in contemporary Tajikistan/Turkmenistan/Uzbekistan/Afghanistan/northwestern Pakistan, with some minor percentages in parts of the Greater Punjab. This was expected.

8. I think Steppe_MLBA_East are the most sensible candidates for the recent steppe influx into India.

Speaking of which, such incredible diversity among steppe populations. Some with southern Central Asian admixture, some with extra ANE (West_Siberian_N, which explains Srubnaya_outlier, among others), some with East Asian admixture, some who resemble Steppe_EMBA or Steppe_Eneolithic, etc.

I very much look forward to seeing these samples in Global_25.

Anthro Survey said...

@Rob

To some, perhaps, but there's no reason to think it an epicenter of the Vedic phenomenon any more than it is to treat a roadside inn as the focal point of 12 day concert.

Don't know why I mentioned that sample. Was thinking more about the later presence of basal R1b clades in Mesolithic and Early Neolithic Europe. I'm unconvinced they arrived with HGs from West Asia.

Bogdan said...

Good grief, there is a lot of serious “deep in the weeds” discussion on this one... Some of you would be best served to create your own blogs, or perhaps write your own novel...We are fortunate to have Davidski as a blog creator here, to give us a medium of sorts....

As a reminder, don’t think the pre-print is a final paper yet, nah...

Perhaps some more topical discussion would lighten things up:

Mankind, both in mentality, spirit, drive, needs, obsessions and lust (you name it), hasn’t changed much over many millennia...

If there was substantial shiney stuff being mined/smelted/forged/crafted/bartered or sold (Copper, TIN, Silver, Gold) in the Zeravsan Valley and general BMAC area circa 2000-1500BC, you can be damn sure Andronovo and Central Steppe people’s were part of that action, coming in ready to pounce, on horses and perhaps chariots... I am sure others from the South immigrated up too... It doesn’t mean there was a lot of admixture there specifically...Probably a lot of blood letting to be more precise... Remember, not much has changed. In all likelihood, he who dominated that Zeravsan area through force won, and from the looks of prelim genetic evidence, rolled right up into Iran, Hind and Swat and from there it’s all down slope...

Anthro Survey said...

@Seinundzeit

"Interestingly, the IVC migrant samples range from 13%-8% West_Siberian_N. Surprising that this West_Siberian_N (well, ANE) stream of ancestry has a noticeable presence in samples with roots in the Indus valley and beyond"

Yup. This strengthens my suspicion of AASI-ANE forager hybrids roaming across North India prior to the arrival of Iran_N people. Those Damdana specimens from Mesolithic Haryana may just have been that.

"I think they would benefit greatly from including CHG, Levant_N, and Iran_Chl in their analyses."

Hopefully, we'll be able to get at least a cursory idea in G25 whether all of that "blue" in BMAC era Central Asia was Iran_N or if there was a layer of CHG influence(via CHG-ANF migrants). I suspect mainly Iran_N for pre-BMAC stuff.
Would also be interesting to glimpse whether Hajji Firuz was CHG-Iran_N or just Iran_N like Iran_Chl samples. The former area traditionally had stronger ties to SE Anatolia/N.Mesopotamia, while the latter to Iraq proper.

Anthro Survey said...

@Kulkarni

Why should we have expected R1a in BMAC? We'd expect to see it in pastoralist groups contemporaneous with late BMAC and post-BMAC hybrid or unmixed pastoralist groups.

Sadly, BMAC-era steppe-rich Dasthy Kozy were all females, but those Kashkarchi samples from Ferghana are all R1a. So there's that.

Seinundzeit said...

Anthro Survey,

"Yup. This strengthens my suspicion of AASI-ANE forager hybrids roaming across North India prior to the arrival of Iran_N people."

I share the same view; it's a solid possibility.

Taymas said...

@Matt

Thanks for the insight. Yeah, they had a line about BMAC carrying less steppe-related ancestry than the South Asians. Figure 2C is part of what's throwing me off, I think their WSibHG component is itself a mixture. But I'm also not seeing them ever establish EHG in South Asia. Part of my confusion is that I didn't appreciate that EHG dropped significantly in the steppe pops early-to-late BA, from >50 to 34. [[Side note: whoa. Probably wrong and totally crazy, but makes we wonder if EMBA~IndoHittite and LMBA~IndoEuro.]]

I agree it's really hard to explain extra ANF and (EHG or WSibHG, unclear) without a steppe pop.

Their admixture graphs are also a little misleading (eg not differentiating between steppe/ANF nor between EHG/WSibHG). I don't trust admixture components too much, but they reference it a lot in the analysis.

@myself
I wrote this on 1/12/16 on the Poltavka Outlier post:
"I think anyone posting an alternative viewpoint should include their testable predictions for Harappa. I don't have an alternative viewpoint, but I'll record my boring, derivative take on any samples older than 2000BC:
-mix of (neolthic ancestry forming a clade with CHG) and (hunter-gatherer ancestry currently strongest in S Indian tribals)
-no R1a"

Not giving myself points for the prediction, I basically just copied Davidski, but I will give myself points for at least offering something I couldn't later wiggle out of.

Vara said...

Just because R1a wasn't found in Swat doesn't mean it wasn't Indo-Aryan or just a roadside inn. Maybe it's time to get rid of some preconceived notions like R1a+Steppe ancestry = Indo Aryans. It's not because of R1a Swat is considered Indo-Aryan but because of the culture itself and it's mention in the Rigveda. Besides it was an important Indo-Aryan center for centuries not far from the famous Taxila city.

As for Iranians, I think the paper ignores them for a reason. Hint: Early West Iranian Grey Ware culture that is also found in Marlik has it's roots in early 2nd Mil Hissar and Tureng Tepes.

PS. Despite what the paper says it looks like there was a contribution by the Scythians.

Samuel Andrews said...

@Vara.
"Maybe it's time to get rid of some preconceived notions like R1a+Steppe ancestry = Indo Aryans"

That's unlikely considering all supposed early Indo Europeans carried 90%+ of some subclade of R1b L23 or R1a M417.

Once the data is released we need to pay attention to Y DNA results for East Andronovo. If they come out R1a L657, then they were definitely the Indo Aryans who moved into India.

Vara said...

@Samuel Andrews

"That's unlikely considering all supposed early Indo Europeans carried 90%+ of some subclade of R1b L23 or R1a M417. "

Not according to Reich and Krause.

"then they were definitely the Indo Aryans who moved into India."

That's cool and all, but there is no R1a except for that one dude in 500BCE. Don't get me wrong , I'm still waiting for the revised version but they seem so sure of their data.

Davidski said...

@Vara

Not really sure what you're trying to communicate there, but if you're actually suggesting that Andronovo people weren't the early Indo-Aryans who moved into India, then you're not making any sense.

Obviously, there's a reason why Balto-Slavs and Indo-Iranians are close relatives linguistically and in terms of paternal origins, with both showing high frequencies of R1a subclades under R1a-Z645, and considering that Andronovo is packed with R1a-Z645(Z93+), then there's not much to argue about.

Bogdan said...

@Sein:

"Yup. This strengthens my suspicion of AASI-ANE forager hybrids roaming across North India prior to the arrival of Iran_N people."

Even a layman could see it after cursory look at the evidence thus far. I thought it mattered too, but it actually doesn’t mean squat.

You, your family, house, neighbors, town, country and region could be taken over, dominated, subdued / suppressed within days/weeks with superior technology and tactics in modern times.. In the past, months or years, or further in time years or few decades, or even tens of decades. Either way, not even a blip on the horizon....

Davidski said...

@All

Interesting paper here from earlier this year about potential ivory carvers from the Indus Valley being present at Gonur Tepe. So those Indus_Periphery individuals may have been such people or their relatives.

https://www.sciencedirect.com/science/article/pii/S2352226716301131

Free version...

https://www.researchgate.net/publication/319453538_Manufacturing_and_trade_of_Asian_elephant_ivory_in_Bronze_Age_Middle_Asia_Evidence_from_Gonur_Depe_Margiana_Turkmenistan

Jaydeep said...

I think we also need to talk about the South Asian admixture in BMAC.

Besides, the one outlier Gonur2_BA, who has around 14 % AASI, the main BMAC cluster has around 5 % AASI. Assuming, with a very high probability, that the IVC or South Asian migrants who contributed to the BMAC, were about 15 % AASI, this would mean that the South Asian admixture into the main BMAC cluster stands at roughly around 30 to 35 %.

Compare that to the questionable and limited evidence of admixture from the steppe among some BMAC outliers.

Very clearly, if BMAC had about 30 - 35 % admixture from people of the IVC, and if this happened only after 3000 BC, such a high admixture from IVC would have definitely led to the BMAC people being very strongly influenced linguistically, religiously and culturally.

So I believe the probability is very high that the BMAC people were infact speaking an IVC language rather than a steppe one.

Is this not strong argument for an Out of India movement albeit in the Bronze Age ?

Let us note further :-

There are 3 outlier steppe samples along the Inner Asian Mountain Corridor, which are said to have received admixture from Turan or BMAC, I think they are Dashly Kozi, Kyzlkum & one another. These samples also seem to have some level of AASI admixture. Although I would request someone to confirm this.

And we know, that BMAC definitely had contacts with Sintashta and David Anthony's theory seems to be that the Sintashta folks were supplying metal to the BMAC people. So, we also need to check if there is some sort of BMAC admixture into Sintashta.

------------------------------

Last but not the least, the IVC outliers in Shahr i Sokhta, also raises the possibility that the Iranian civilizations of Helmand and Jiroft were also heavily influenced by IVC migrants.

--------------------------------

So the 1st samples we have from South Asia, we already start seeing evidence for Out of India movements.

With more aDNA from South Asia, expect more evidence for Out of India movements.

Seinundzeit said...

Bogdan,

"Even a layman could see it after cursory look at the evidence thus far. I thought it mattered too, but it actually doesn’t mean squat."

You're quoting Anthro Survey, although I share his views.

"You, your family, house, neighbors, town, country and region could be taken over, dominated, subdued / suppressed within days/weeks with superior technology and tactics in modern times.. In the past, months or years, or further in time years or few decades, or even tens of decades. Either way, not even a blip on the horizon...."

Please go easy on the weed bro; too much, and you start getting paranoid. Bad trips can be a b*tch...

postneo said...

@davidski
"Not really sure what you're trying to communicate there, but if you're actually suggesting that Andronovo people weren't the early Indo-Aryans who moved into India, then you're not making any sense."

How did IE languages spread to Iran? Was it by Andronovo as well ? Or was Iran an IE speaking region from before?

The steppe MLBA seems to have south asian outliers as per the weird unexplained plot 1d. Its possible that what makes east MLBA differentiated from west MLBA is south asian components. and a bit more of Iran.

Davidski said...

@Jaydeep

There's no evidence of any Out-of-India population movement in these new samples. The minor South Asian admixture in BMAC is the result of economic and political links between the proto-cities of Central and South Asia, as hypothesized and described here...

https://www.sciencedirect.com/science/article/pii/S2352226716301131

At the same time, like I told you already, there's absolutely no evidence of Indus Valley or even BMAC ancestry in ancient or modern-day Eastern Europe.

And it's very amusing how quickly you've forgotten about those Andronovo groups from the Kazakh/Tajik/Uzbek borderlands, who just happen to be packed with R1a and appear to be perfect references for the ancient steppe ancestry in Indians.

Suffering from amnesia?

Davidski said...

@postneo

How did IE languages spread to Iran? Was it by Andronovo as well ? Or was Iran an IE speaking region from before?

No it wasn't. Proto-Iranians are also supposed to derive from Sintashta, via Srubnaya. I don't see a problem with that.

The steppe MLBA seems to have south asian outliers as per the weird unexplained plot 1d. Its possible that what makes east MLBA differentiated from west MLBA is south asian components. and a bit more of Iran.

There's no South Asian ancestry in Steppe_MLBA, east or west.

You get no points for creativity here. Quit acting like an idiot.

Bogdan said...

Sein,

“Please go easy on the weed bro”...

My apologies, given your board name, I thought you were a European, not a “bro-bagger”. Keep up the good posts though. I do enjoy content you bring, no matter how arcane and generally irrelevant...

Seinundzeit said...

Bogdan,

"My apologies, given your board name, I thought you were a European, not a “bro-bagger”."

Perhaps we are dealing with a clash between distinct "lebenswelten"?

To put things in perspective, I don't deal much with the intellectually retarded side of internet culture, which may partly account for why I don't understand what the f*ck you're saying. Quine is smiling.

"Keep up the good posts though. I do enjoy content you bring, no matter how arcane and generally irrelevant..."

Ah, now I get it; I'm talking to a fan here! You should have said so, right in the beginning.

If my thoughts constitute a well-spring of intellectual pleasure in your life, then doggone it, just another reason for me to soldier on! For you, "bro", I'll always keep the comments coming...

epoch2013 said...

@postneo

"How did IE languages spread to Iran? Was it by Andronovo as well ? Or was Iran an IE speaking region from before?"

If that latter case would be the case, wouldn't it either be very noticeable in the surrounding languages, or wouldn't those surrounding languages leave a large influence in that language? And would the current Iranian language show a clear substrate from an older PIE-like language? Does it?

Matt said...

@Jaydeep

Interesting suggestion; 5% AASI scaled up assuming 23% in the Indus_Periphery is representative of IVC (a fairly substantial assumption, but no more than any other modeling using the I_P in the paper I suppose) means about 21% admixture from IVC->BMAC.

Assuming some West Siberian like ancestry comes as well, could be a 25-30% level of total migration. This could well have been gradual accumulation over time (less likely to change language) or a mass migration.

Hence the IVC could have been a profound influence on BMAC via migration, not culture and trade alone. Whether this was a mass migration of gradual migration even less clear.

I would, however, think it's not parsimonious to assume language shift from a previous language. Judging from historically similar events 20-30% is probably safely *enough* of a mass migration for language shift in the right circumstances, but I'd say it's less probable than language continuity, if we don't actually have any direct evidence indicating a culture change. It's more of a leap to assume a ~20% migration leads to language change! Likely they were not speaking an IVC language (nor a steppe language).

This also means though, we must remember the other end of the spectrum, the other side of ancestry must have been more Anatolian like than BMAC, since Indus_Periphery by and large lacks this.

If one side is 20-30% I_P, then to get to BMAC probably something like 60-70% of the Copper Age Seh_Gabi sample, from Western Iran (https://bit.ly/2H7FDfA). This would mean in turn a pretty high level of influence from Western Iran / Armenia / Mesopotamia, and probably more arguably if any language shift at all this dominant proportion as a source of language.

Considering chronology, with broad strokes assumptions possibly some combination of minor influences from IVC with a greater degree of influence from the early city states of Mesopotamia / Elam?

Chetan said...

@postneo said How did IE languages spread to Iran? Was it by Andronovo as well ? Or was Iran an IE speaking region from before?

Iranian arrival on the plateau happened later maybe after 1000 BCE. There are no samples from this preprint in this time range for Iran/Middle East. But I think we already have Iron Age samples that may be indicating such a movement.

Chetan said...

@David I think Srubnaya better represents the origin of Thracians/ Cimmerians. Iranian groups like Scythians are supposed to have moved to Eastern Europe from Central Asia later, around 600 BCE.

Davidski said...

@Matt

IVC migrants to Central Asia may have been genetically heterogeneous, like the two Indus_Periphery individuals from Shahr-i-Sokhta, with many showing around 40% or even more indigenous South Asian ancestry.

If so, it'd be hard to estimate how much total admixture from South Asia there was at BMAC, but as you say, any figure around 30% is hard to justify without any mental gymnastics.

Chetan said...

correction: around 800 BCE it seems. But Greek writers say the Scythians replaced the Cimmerians on the Pontic-Caspian steppes.

Mr. Kulkarni said...

@anthro survey

theres no evidence in this paper of male mediated steppe migration.
"Sadly, those BMAC era steppe rich samples were all female". Exactly

"but those Kashkarchi samples from Ferghana are all R1a. "
There are 15 male samples from Uzbekistan earliest being 2000BC.
None have R1a except the latest 2 samples 1200-1000BC. Much too late for Rig Vedic people brother.

Suyindik said...

1. Group A
I think that the main location of the people of the Gobeklitepe, Ubaid, Uruk and Sumer cultures is located in Upper Mesopotamia(South-Eastern Turkey, North Syria and North Iraq). Their haplogroups were G, H, J, L, T.

2. Group B
The main location of the people of the Khvalynsk culture(and earlier hunter gatherers) was the Steppe and Northern Regions(North of the Black Sea). Their haplogroup was R. The haplogroup of hunter gatherer people of Paleolithic Scandinavia and Europe was I. In time these two hunter gatherers mixed.

3. Group C
The main location of the people of the Natufian culture was the Southern Levant Region. Their haplogroup was E.

4. Mixing and Migrations
-During the "Pre-Pottery Neolithic period", Group A mixed with Group C. During the 6th millennium BCE they migrated to Central and Western Europe.

-During the "Chalcolithic/Eneolithic period"(5th millennium BCE) "Group A" migrated into the South and North Caucasus and created(found) the Kurgan type of cultures(Leyla-Tepe, Maykop and Early-Kura-Araxes/Areni). In the same period(and the following periods), "Group A" also migrated to South-Eastern Europe(Bulgaria) and South Central Asia(Turkmenistan, Eastern Iran). In South-Eastern Europe they found the Varna culture(4600-4200 BCE), oldest metallurgy(gold). Group A moved this metallurgy system in later periods to the Leyla-Tepe and Maykop.

-Maybe before the 5th millennium BCE a small part of Group B(R2 and maybe some R1b) already migrated to some parts of the Iran/Caucasus regions(no Kurgans associated with them).

-During the 3th millennium BCE(and the end of the 4th millennium BCE) Group B migrated partly into South and North Caucasus, mixed with Group A, and went back to the Steppe to form the Yamna culture. Now, the people of the Late-Kura-Araxes period are a mix of Group A, B and C.

Martin Clifford Styan said...

Martin Clifford Styan said
For a long time I more or less accepted the idea that Indo-European originated from Neolithic Anatolia. I abandoned the idea several years ago, when it appeared to be clear that Anatolian Neolithic people came west into Europe, but did not go east to India. Now it appears that BMAC and many Steppe people had a considerable amount of Anatolian ancestry. I would like to ask:
Is Anatolian Neolithic ancestry found to any significant extent in India?
Is there a significant eastern Steppe group with little or no Anatolian ancestry, which could have gone to India?

Rob said...

@ Suyindik

Interesting synthesis. If i were to critique.

- the Majkop's kurgan tradition probably isn't from Uruk, but from NW Black Sea region.
On the other hand, Kura-Araxes isn't really a 'kurgan culture'.

- it seems counter to evidence to suggest that R2 migrated from Khvalynsk (4000 BC) to Iran, when R2 is already found in Neolithic Iran (7000 BC).

- groups A didn't found the Varna culture after a migration from northern Syria in 4600 BC. The genetic evidence suggests that it formed from preceding local (Karanovo) culture groups which engaged in contact with Pontic steppe already by 5000 BC.

- haplogroup I wasn;t the sole lineage of Mesolithic Europe. R was also found fairly widely - from southern France to Baltic, Balkans and western Russia (and do note all those places are in Europe, even by Cold War Era definitions). Lastly, there were no Paleolithic foragers in Scandinavia because it was under Ice.

Davidski said...

@Martin Clifford Styan

The Steppe_MLBA_East groups that moved into India, and probably spoke early Indo-Aryan languages, did carry some Anatolian farmer ancestry.

On the other hand, based on the Indus_Periphery samples and modern-day South Indians who lack steppe admixture, it appears that South Asians initially acquired admixture from a West Asian source with practically no Anatolian ancestry.

So you could, in theory, say that perhaps the arrival of Indo-Europeans in South Asia was correlated with the first spread of a fairly significant level of Anatolian farmer ancestry into the region, but the idea that early Anatolian farmers were the Proto-Indo-Europeans doesn't make any sense on so many levels, that I don't know why you would want to do that.

Anthro Survey said...

@Kulkarni

"theres no evidence in this paper of male mediated steppe migration."
The evidence is staring us in the face and the inference is not difficult to make.

Uzbekistan....look closely at:
-the spread of those dates: 2000-1800BC(morning of the Aryan invasion)
-the CULTURAL CONTEXT of those samples: nearly all from urban BMAC dwellers. These Aryans would have initially been separate communities living outside, but nearby, the metropoli. We simply have not extracted DNA from them----yet.

"Exactly"
So, you'd have us believe that those Neolithic sites for which we only have female genomes were....parthenogenetic communities a-la Wonder Woman's island? :-)

Limited AVAILABLE evidence is NOT the evidence of absence. You know that's not an argument, mate.

Suyindik said...

@ Rob

-The oldest kurgans are from Leyla-Tepe and Maykop cultures. The attributes that define the root of these two cultures are archaeologically proven to come from the Ubaid/Uruk cultures from Mesopotamia. So, the Ubaid/Uruk peoples are the ancestors of the oldest kurgans.

-I didnt say R2 migrated from Khvalynsk. I mentioned "Khvalynsk culture(and earlier hunter gatherers)". In some point(thousands of years ago) R1a, R1b and R2 have a common ancestor, so before the 5th millennium BCE the R haplogroup was ancestral in the Steppe. And from here they could have migrated to Iran regions.

-The Uruk is the continuation of Ubaid(7nd millennium BCE). And the Varna people genetically had Y-haplogroups from Neolithic farmers in Europe. If the Neolithic farmers in Europe came from the Near-East(=Mesopotamia), then the same people could have spread the metallurgy to Varna, Leyla-Tepe and Maykop. And in time mixing between steppe and mesopotamian people happened.

-I wrote within group-b section that the two hunter gatherer people mixed within time. And better to use Paleolithic/Mesolithic terms.

Anthro Survey said...

Re/South Asian ancestry in BMAC---

Can't discount it at all, ofc, but have you guys considered limited archaic AASI presence on the west side of the Khyber?

@Vara

I'm not saying Swat area wasn't IA CULTURALLY. I believe it got absorbed into the IA fold fairly early on, but it wasn't until many centuries later that its steppe affinity "caught up". Also, I have to concede something: we might be looking at the wrong samples. What if high-steppe specimens are much harder to come by due to cremation being more uniformly practiced among the higher proto-castes(presumably)?

Davidski said...

There is Steppe_MLBA_East ancestry in the Swat samples, and it's somewhat variable.

When I get the data we'll be able to identify which individuals show the highest levels of such admixture and then speculate why. Till then, I really don't know what all the fuss is about, considering we don't even know what the Y-haplogroups of these Swat guys really were.

Rob said...


@ Suyindik

“The oldest kurgans are from Leyla-Tepe and Maykop cultures. ”

No they’re not.
Kurgan monuments are seen in west Ukraine already in 4500 BC.
The M-L/T horizon is from 3800 BC onwards
Then it appears in east Anatolia (Aralantepe) 3000 BC.
(Although I’m not advocating any simple unidirectional movements).

Suyindik said...

@ Rob

At 4500 BCE in Ukraine there were no kurgans. Oldest kurgans are from Leyla-Tepe culture(second half of 5th millennium), and the ancestors(the people that founded) of the Leyla-Tepe people are the Ubaid people. And the Ubaid people are located in Mesopotamia. But eventually in time the people of the Steppe and people of the Mesopotamia mixed.

Davidski said...

@Suyindik

Where are you seeing the Mesopotamian ancestry on the Bronze Age steppe? Can you point out a clear example?

Jijnasu said...

@davidski
I don't think the y haplogroups matter too much. The samples are from several centuries after the Indo-Aryans entered India. Males from surrounding societies were likely assimilated into Indo-Aryan society. Given that both cremation and burial were practised the rareness of of r1a in the graves might not reflect the frequency of r1a in NW Indo-aryan society. Some have suggested that the 'steppe' in these samples is actually BMAC. Is this possible. It looked unlikely to me

Davidski said...

@Jijnasu

I wouldn't worry about what some (idiots) happen to blurt out online. BMAC ancestry can't be confused with Steppe_MLBA_East ancestry. Like I said, I'll lay it all out when the samples are released.

Suyindik said...

@ Davidski
I didnt say anything like "Mesopotamian ancestry on the Bronze Age steppe".

What i said is a cultural(Kurgan) ancestry of the Bronze Age steppe(Yamna culture) coming from the Mesopotamia(Ubaid/Uruk culture). Oldest kurgans are from Leyla-Tepe, their ancestors are the Ubaid people, and they come from the Mesopotamia, and they eventually passed over the kurgan culture to the Yamna culture.

As for the genetically mixing between the Steppe and the Mesopotamia, i think the Varna Necropolis culture can serve as a good example. The Varna culture(4600-4200BCE) has the oldest gold, so they are the oldest metallurgy users. This metallurgy moved to the Maikop when the Varna culture disappeared. And if we look at the genetic structure of the Varna culture(study: "The genomic history of southeastern Europe"), we see that they have Y-haplogroups G2a, R1b and T. We can see in here a mix of the Steppe with Early European farmers(which are in fact descendants of the West-Asia/Mesopotamia = Ubaid/Uruk).

And as i pointed in my previous post:

3.
A COMPREHENSIVE STUDY OF ANTHROPOLOGICAL MATERIALS OF THE LATE ENEOLITHIC FROM THE ARENI 1 CAVE:

Results of the intergroup statistical analysis identified close morphological affinities between the Eneolithic skulls from Areni 1 with groups from Tepe Hissar II, Ginchi, Catal Huyuk, Alishar Huyuk and the bearers of the Kuro-Araxes from the Southern Caucasus. Their connection with bearers of the Maikop, Khvalynian, Pit-Grave and Catacomb cultures was also revealed. As for race and genetics, the Late Eneolithic inhabitants from Areni 1 cave, more probably, originate from the territory of the Middle East.


The Areni-1 culture(3900 BCE) is the oldest form of the Kura-Araxes culture(a type of Kurgan culture). And just like the Leyla-Tepe and the Maikop the origin of Areni-1 is located at the Mesopotamia(=Ubaid/Uruk).

The Areni-1 culture results showed Y-haplogroup L. And the recent results of the Tepe Hissar culture(origin in Mesopotamia and very close to Maikop and Kura Araxes) showed Y-Haplogroups H, J, L, T.

As for the MT-DNA haplogroups found in the study "Six complete mitochondrial genomes from Early Bronze Age humans in the North Caucasus", this is completely different than is found from the Yamna MT-DNA. This could show the fact of mixing between the Steppe and the Mesopotamia within the Caucasus(South/North) regions. But the original ancestry of the Leyla-Tepe and Maykop cultures(oldest kurgan cultures) lies within Mesopotamia.

Arza said...

Does anyone know anything about this study?

https://www.ebi.ac.uk/ena/data/view/PRJEB20658 (description only, no samples available)

Submitting Centre
CENTRE FOR GEOGENETICS, NATURAL HISTORY MUSEUM OF DENMARK

A population genomic history of the Eurasian steppe

Description

The Eurasian steppe, stretching about 8000 kilometres from Hungary and Romania in the west to Mongolia and western China in the east, is culturally among the most dynamic areas in the world. In the past four millennia, it has been variously dominated by Iranian-, Turkic- and Mongolic-speaking groups, and its temperate grasslands have been a crossroad for extensive movements of peoples, goods, and ideas between Europe, Siberia, South and East Asia. In order to understand the genetic history of the Eurasian steppe populations, we have sequenced 137 ancient genomes (~1X average coverage) spanning a 4000 years time series. We also genotyped 502 individuals from 16 contemporary self-reported ethnicities. We find evidence of a highly dynamic population history; the Iranian-speaking Scythians that dominated the Eurasian steppe throughout the Iron Age (~1 millennium BCE to common era) emerged following admixture between Late Bronze Age herders of western Eurasian descent and East Asian hunter-gatherers. The steppe nomads later further admixed with Turkic-speaking groups of East Asian ancestry that spread westward across the steppe in multiple waves: firstly, the Xiongnu confederations that emerged in Mongolia around the 3nd/2nd century BC; secondly, the Huns (4-5th century CE), infected with plague basal to the Justinian Y. pestis strain that destabilized the eastern Roman Empire in the 6th century CE; and thirdly during various short term dynasties, including the Mongol Empire of Genghis Khan and his descendants. These recent historical events transformed the Eurasian steppe populations from being Indo-European speakers of largely western Eurasian ancestry to the present-day Turkic-speaking groups, primarily of East Asian ancestry.

mzp1 said...

Does anyone know what the Indo Aryan Migration theory looks like that people here are working off in terms of deriving Indo-Aryan and Iranian from IIr?

I ask because though I understand the theory has Sintashta/Andronovo as IIr I do not know how they propose the derivation of IA and Iranian from IIr.

All of this happened in Central Asia around in Andronovo/BMAC/South Asia. Now we have some good population genetics from that time period how does that fit in with the literary and linguistic data.

IVC and BMAC were clearly very closely related and were the biggest civilizations of their time by far (IVC alone was the biggest in the world). Funny they leave no trace of their language.

I never understood where the M migration model derives
IA from IIr
Iranian from IIr
given they place IIr in Sintashta/Andronovo

Davidski said...

@mzp1

BMAC and IVC weren't closely related. They clearly derived from two highly differentiated populations that mixed to a minor degree due to long distance economic ties. So it's extremely unlikely that they spoke the same, or even closely related, languages.

As for the Indo-Europeanization of Iran, it's generally accepted that it was a rather late process during the Iron Age. You'll find a brief discussion on that here, plus a bibliography.

https://www.researchgate.net/publication/319949929_A_palimpsest_grave_at_the_Iron_Age_cemetery_in_Estark-Joshaqan_Iran

Chetan said...

@mzp1 Indo-Aryan and Iranian have a common ancestor around 2000 BC. So they were not that diverged at all during the Andronovo culture. It's just the groups that migrated south earlier were linguistically closer to proto-IIR (the Indo-Aryans) and the groups that migrated a couple of centuries later were more diverged (Iranians)

postneo said...

@echo@chetan
I don’t care that much about linguistic reconstruction or language = genes association. That is your job.
We have IIr in 1400 bc Syria and Assyrian reference to Persians and medes around Lake Urmia in 900 bc

On a different note “steppe mlba” outlier in bmac is earlier than most such samples from the eastern steppe itself

Davidski said...

@postneo

On a different note “steppe mlba” outlier in bmac is earlier than most such samples from the eastern steppe itself.

You're such a massive straw clutcher, that it's not even funny anymore.

mzp1 said...

@Davidski,

We are not talking about Indo_Europeanization of (the modern country of) Iran, because Iranian culture (Median and Persian, as discussed in the link) are later Iranian developments. You are equating the country that is Iran today with historical Iranian culture. Historical Iran (Greater Iran) extends from BMAC all the way to the Caspian sea and has its roots in the East, specifically in and around the BMAC area.

Ofcourse BMAC and IVC (and Helmand Civilization) were closely linked.

"BMAC materials have been found in the Indus Valley Civilisation, on the Iranian Plateau, and in the Persian Gulf.[9] Finds within BMAC sites provide further evidence of trade and cultural contacts. They include an Elamite-type cylinder seal and a Harappan seal stamped with an elephant and Indus script found at Gonur-depe.[13] The relationship between Altyn-Depe and the Indus Valley seems to have been particularly strong. Among the finds there were two Harappan seals and ivory objects. The Harappan settlement of Shortugai in Northern Afghanistan on the banks of the Amu Darya probably served as a trading station.[5]
https://en.wikipedia.org/wiki/Bactria%E2%80%93Margiana_Archaeological_Complex

thorin23 said...

Swat only gets one r1a sample and that too in the late iron age. So instead of saying - yes this doesn't look like an invasion, more like long term slow migration or conversion maybe - more work needs to be done. That would have been credible, but instead they double down on massive migration/invasion with mediocre evidence. My argument is that this paper in fact proves there was no invasion/massive migration - but they don't draw that obvious conclusion and state it explicitly.

Obvious things like how the mighty war chariot would have been completely useless in India. They may be great in the steppe and deserts, but try going through Indian forests with them. Heck even in Rome, chariots were used mostly for races and show and not for actual combat.

Incredible claims need incredible evidence. And i don't see it in this latest paper. Show me a kurgan in Swat with weapons and chariots and tons of r1a - I'm sold. But instead they make excuses - Oh the kurgan thing - we can explain turns out after 3500 years of using kurgans to bury the dead, we completely changed our method right before launching the massive invasion campaign. By the way, we also burned all our chariots and cremated everyone with the r1a genes. Its things like this that make them lose credibility and give more ammo to people who want to claim out of india.

This paper isn't the smoking gun they're pretending it is.

Davidski said...

@mzp1

Fairly close economic ties =/= similar ancestry or languages

Why do you think those Indus_Periphery outliers at BMAC sites are such obvious outliers and so obviously South Asian?

Davidski said...

thorin23

I'll lay it out for you in more detail when I get the samples.

«Oldest ‹Older   201 – 400 of 877   Newer› Newest»