search this blog

Thursday, July 20, 2017

The Out-of-India Theory (OIT) challenge: can we hear a viable argument for once?


Recent weeks have seen a rash of activity from OIT proponents defending their "truth", largely as a response to a news feature in The Hindu on new genetic evidence backing the Aryan Invasion or Migration Theory (AIT/AMT). A few examples:

Genetics Might Be Settling The Aryan Migration Debate, But Not How Left-Liberals Believe

Genetics and the Aryan invasion debate

Propagandizing the Aryan Invasion Debate: A Rebuttal to Tony Joseph

Here We Go Again: Why They Are Wrong About The Aryan Migration Debate This Time As Well

The problematics of genetics and the Aryan issue

Too early to settle the Aryan migration debate?

The people who wrote these articles are able to string sentences together in a reasonable way, but apart from that, their efforts are clumsy at best. Not only do they not appear to completely understand what they're attempting to debunk, but they also fail to offer an OIT that realistically incorporates new findings from ancient and modern-day DNA.

AIT/AMT is now firmly backed by ancient DNA from Eastern Europe and high resolution modern-day DNA from South Asia. To quote myself from a week ago:

During the past couple of years ancient DNA has revealed the presence of Y-chromosome haplogroup R1a in Eastern European remains dated to the Mesolithic, Neolithic, Eneolithic and Bronze Age. Moreover, the Bronze Age remains, packed in ancestry derived from Eastern European hunter-gatherers (or EHG) and totally lacking any sort of South Asian admixture, belong to R1a-Z645, which is the ancestral clade of by far the most common types of R1a in Europe and South Asia today: R1a-Z282 and R1a-Z93, respectively. And on top of that, South Asians, especially those speaking Indo-European languages, show significant admixture derived from EHG.

The conclusion from this data is self-evident: during the Bronze Age R1a-Z645 became a very important Y-chromosome lineage in Europe and quickly moved to South Asia, in all likelihood on the back of the Indo-European expansion.

Pre-Indo-European Eastern Europe and South Asia were not the same world; they were world's apart. Thus, you will never read anything like this, no matter how much ancient DNA from South Asia is sequenced:

During the past couple of years ancient DNA has revealed the presence of Y-chromosome haplogroup R1a in South Asian remains dated to the Mesolithic, Neolithic, Eneolithic and Bronze Age. Moreover, the Bronze Age remains, packed in ancestry derived from South Asian hunter-gatherers, and totally lacking any sort of European admixture, belong to R1a-Z645, which is the ancestral clade of by far the most common types of R1a in Europe and South Asia today: R1a-Z282 and R1a-Z93, respectively. And on top of that, Europeans, especially those speaking Indo-European languages, show significant admixture derived from South Asian hunter-gatherers.

So, OIT proponents, what counter-arguments can you offer? And can you come up with a new vision for OIT that coherently takes into account ancient DNA from Eastern Europe?

However, to ensure that the debate is a fruitful one not derailed regularly by anti-AIT/pro-OIT red herrings, let's take care of the most obvious of these red herrings now. I reserve the right to delete any comments that attempt to go down these tired, irrelevant avenues without a very good excuse for doing so.

You: So and so found Y-haplogroup P* and other basal clades upstream of R1a in Papuans, therefore R1a and Indo-Europeans are from South Asia. Me: Nonsense. R1 and R1a are found in the remains of Eastern European Mesolithic foragers. Were these individuals recently arrived Indo-European-speakers from South Asia? Try harder.

You: It doesn't matter that Eastern European Mesolithic foragers belonged to R1a, because the most common form of R1a in the world is R1a-M417, and if it originated in India then OIT is a reality. Me: But what are the chances realistically that R1a-M417 is from India or South Asia, considering that prehistoric European samples, with absolutely no signals of ancestry from South Asia, belong to both M417+ and M417- lineages? In fact, Europe is the most likely homeland of R1a-M417.

You: India has incredible diversity in R1a, therefore it's the R1a and Indo-European homeland. Me: No it doesn't. India, and indeed, South Asia as a whole are dominated by one fairly young subclade: R1a-Z93. Europe is home to three different subclades that show up at perceptible frequencies: R1a-Z282, found throughout much of the continent; R1a-L664, mostly confined to Northwestern Europe; and R1a-Z93, mostly confined to far Eastern Europe.

You: Many unique Indian ethnic groups are yet to be tested genetically. They may show surprising results, including new subclades of R1a. Me: If you dig hard enough, you'll always find some exceptions to the rule. But how do you know where the ancestral lineages of such exceptions in South Asia were during, say, the Neolithic? What makes you think they were in South Asia? To prove that South Asia is the homeland of its by far most dominant R1a subclade, R1a-Z93, then at the very least you need to show that other, closely and distantly related subclades, are also found at perceptible frequencies in whole regions of South Asia, and therefore that they have some sort of history there. Otherwise we can safely assume that R1a-Z93 and the few exceptions to the R1a-Z93 rule in South Asia are relative latecomers from somewhere else.

You: But we have no ancient DNA from South Asia yet, and it may produce a huge shock. Me: For you yes, but not for me. What are the chances realistically that R1a was present among both European and South Asian foragers? I'd say practically zero. Feel free to raise it to a few per cent to make yourself feel better, but we both know the hard reality.

You: Ancient DNA from South Asia might show that Northern India was home to a population very similar to Yamnaya, and if so, then the Yamnaya-related ancestry in modern-day Indians is native to India. Me: There's no logic behind this. Yamnaya and other closely related Bronze Age groups were very specific mixtures of Mesolithic foragers and Neolithic farmers living in Eastern Europe and surrounds. There's absolutely no reason to assume that such unique mixtures would also form independently in South Asia, or even outside of Europe's generally accepted borders.

You: Bronze Age Europeans who belonged to R1a also carried southern admixture from Iran, or maybe even India. Me: In prehistoric samples, R1a is always highly correlated with Eastern European Hunter-Gatherer (EHG) ancestry, so positing that it also arrived in Europe with a southern population makes no sense. And why would this southern ancestry be from Iran or India? Why not the Caucasus? We know from ancient DNA that the type of southern ancestry that these ancient Europeans carried has been sitting in the Caucasus since the Upper Paleolithic. Moreover, they lack South Caspian- and South Asian-specific markers such as mtDNA haplogroup U7. How were such markers purged from their gene pool if they or their recent ancestors arrived in Europe from Iran or India?

You: Chickens and mice came from South Asia, therefore Indo-Europeans came from South Asia too. Me: Bullshit. Do better or go away.

Does anyone want to claim that I don't know what I'm talking about? Or perhaps that I'm just putting out Eurocentric propaganda? If you don't understand my arguments, and that they're indeed very solid arguments, then there's no hope for you. Go and find a new hobby or profession, because you're not cut out for this.

OK, now what we have the formalities out of the way, who wants to have a go at salvaging OIT in the comments? Don't be shy.

See also...

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

R1a-M417 from Eneolithic Ukraine!!!11

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

143 comments:

Tesmos said...

''So and so found Y-haplogroup P* and other basal clades upstream of R1a in Papuans, therefore R1a and Indo-Europeans are from South Asia''

People like to forget that modern haplogroup frequencies and diversity can be very misleading.

Ryan said...

I think you're getting yourself too

"So and so found Y-haplogroup P* and other basal clades upstream of R1a in Papuans, therefore R1a and Indo-Europeans are from South Asia"

"People like to forget that modern haplogroup frequencies and diversity can be very misleading."

Doesn't seem very misleading in this case. It's just that the location of Indo Europeans' paternal ancestors 35,000++ years ago tells us jack squat about where they were 6,000 years ago.

I'd stop banging your head against the wall David. Evidence won't change ideology.

Coldmountains said...

Even if we had a time machine and would bring Proto-Indo-iranians into the present, so that they themselves could say that they are not native to India, these people would still not change their views and ideology. It is better to ignore them

Lukasz M said...

Good job David! I propose to make such articles for few other strongly debated topics. To cut all useless discussions.

John Smith said...

It's not even worth giving the OIT people any attention. It's like trying to argue with the "moon landing was a hoax" people. Conspiracy theorists can't be reasoned with because no evidence is ever good enough.

Jingus Jendal said...

Excellent post.

Matt said...

Semi-off topic, semi-on:

Neighbour joining nets for South Asia based on Fst:

http://imgur.com/a/dWpGV

They reflect how Fst distances are not always high across South Asia, e.g. Brahmin Uttar Pradesh to Gond - 0.007.

0.007 is not actually that large on the European scale, e.g. about 0.006 Polish-Greek distance or 0.007 Scottish-Italian_South.

Even though qpGraph shows Brahmin around 36% Onge-like and Gond around 78% Onge like. Upper Caste to tribal, high ANI to low ANI not actually so distant, in itself.

OTOH, Paniya, who fit as 82% Onge-like in qpGraph, have Fst of 0.032 from Brahmin Uttar Pradesh.

High Fst in South Asia panels doesn't seem to be driven so much by ASI-ANI cline, but may be due to very high drift in tribal populations like Paniya, and in highly endogamous caste populations as well.

In the nets you see this, with lots of tribal populations separated on far "twigs", distant from all other populations.

(Also, seem like differences between same tribe/group in different sampling arrays / sets: North_Kannadi-Brahmin_UP in Chaubey has Fst of 0.016, while in the matrix Davidski calculated, only 0.005).

One thing that I can think of that hasn't been done in South Asia is an extensive analysis of f3 sharing. There was a study of Irish Travellers earlier this year (https://www.nature.com/articles/srep42187) which found quite a high Fst between them and Settled Irish, of around 0.0036 - 0.0086, depending on subgroup. But this went away when looking at f3 variant sharing.

Would be good to see if some of the groups who have these very high Fsts from each other have the same in f3. Like Pakistani_Pathan-Gond distance is quite different from Kalash-Chenchu, but I suspect it might become much more similar when comparing in f3.

I think that could also contribute to an counter-argument to claims that South Asia is necessarily a source because of high Fst populations indicating high long term diversity (e.g. "India was a second Africa" and all that kind of stuff).

Roy King said...

@All
I've been musing over some of the flagrant logical errors many of us make as consumers of or producers of archaeogenetic research. A common trope/figure of speech is termed synecdoche, which is substituting a part for a whole or a whole for a part. We tend to be more flagrant with this trope with uniparental markers, like claiming R1a-Z93 conveys the Indo-Aryans to India when, thus far we see that R1a-Z2124 is in aDNA in Europe/North Eurasia and in modern South Asia, or that Haplogroup G people are the first farmers to Europe, when several separate G-lineages dominate the European Neolithic. I've certainly done that myself arguing that J1-P58 conveys the Semitic languages, when subsets of J1-P58 are convected Arabic (L222) to North Africa and other J1-P58 are found in the Bronze Age Levant. Now, as Davidski is well versed at with a journalist background, rhetoric amply uses synecdoche to persuade others in argument, but the devil is in the details, and I would suggest that we be as specific and as limited as we can using factual terms.

Davidski said...

@Roy

What are the chances that Z93 isn't from Eastern Europe, now that we have its immediate ancestral lineages and recent offshoots in closely related EHG-rich prehistoric European samples?

Chances are small, and even though in theory these questions will always remain open, in practice they're no longer open.

Roy King said...

@Davidski
I agree with you that the bulk of the evidence so far is that Z93 is from Eastern Europe and not from India. That said, some Z93 lineages, like L657, may have expanded into Eurasia slightly before they adopted an Indo-European language, presumably from Yamnaya. Remember the old proposed Uralic/Dravidian genealogical relationship based on lexical and typological correspondences. IMO, there is no question that Z2124 and Z282 each carried a portion of individuals speaking Indo-Iranian and Balto-Slavic, respectively. The synecdoche fallacy would tend to conflate these into all Z93/all M417 without supporting evidence yet for such a conflation.

Davidski said...

@Roy

There is supporting evidence for such a conflation: Corded Ware.

Note that the R1a subclades in Corded Ware and the almost identical Poltavka_outlier and Andronovo already practically cover all of the current diversity under M417.

There's no reason to believe that Corded Ware and derived Trzciniec belonging to Z282 were Indo-Europeans, but their genetic twins belonging to M417(xZ282) and Z93 were not.

So with that said, why is L657 special? Why wouldn't it derive from the same, in all likelhood Indo-European, cluster?

Davidski said...

And by the way, Corded Ware doesn't have Yamnaya influence. Early Corded Ware is basically 100% Yamnaya living in Northern Europe.

In other words, Corded Ware is practically Yamnaya with M417, and obviously from the steppe.

That's where Balto-Slavs derive from, and so do Indo-Iranians. Our languages are even posited to be on twin branches on the Indo-European tree, and behold, we carry twin R1a subclades.

There's no problem here. It's all cut and dried.

akhil misra said...

Davidski wrote:-
“During the past couple of years ancient DNA has revealed the presence of Y-chromosome haplogroup R1a in Eastern European remains dated to the Mesolithic, Neolithic, Eneolithic and Bronze Age. Moreover, the Bronze Age remains, packed in ancestry derived from Eastern European hunter-gatherers (or EHG) and totally lacking any sort of South Asian admixture, belong to R1a-Z645, which is the ancestral clade of by far the most common types of R1a in Europe and South Asia today: R1a-Z282 and R1a-Z93, respectively. And on top of that, South Asians, especially those speaking Indo-European languages, show significant admixture derived from EHG.”
1. First and foremost not all people who criticize AIT/AMT support OIT. To equate critique of AIT to peddling OIT is infantile. It is a non-sequitur.
2. Most Indian geneticists, archaeologists who criticise AIT on its methodology say that present day samples of South Asians with which they are compared with ancient European DNA is compared with is wrong. You can’t compare apple with oranges. The assumptions underlying the above para are unscientific and flawed. Why?
a) Because present day South Asian samples is not representative of all the population of South Asia , especially India.
b) South Asian genetic samples (this adverts to all samples used by all geneticists in their studies till date including their critics) hardly capture the complex reality of present day India/South Asia(sample is too small and selectively collected). See Chaubey, Thangraj, Lalji Singh,etc.
c) This is what Chaubey/Thangraj say about Silva et al study:-
“However, India is a nation of close to 4,700 ethnic populations, including socially stratified communities, many of which have maintained endogamy (marrying within the community) for thousands of years, and these have been hardly sampled in the Y chromosome analysis led by Silva et al., and so do not provide an accurate characterisation of the R1a frequencies in India (several tribal populations carry substantial frequency of haplogroup R1a).”

d) Therefore we don’t know how much R1a or its various derivative subclades are present in present Indian population. It could be 1% or 0.1% or something different.
e) Therefore to claim R1a-Z645 is the ancestral clade of most common type of R1a-Z93 prevalent in South Asia is somewhat premature.
f) The design/models used for various genetic studies are flawed.Genetic studies confirm AIT because they assume AIT to start with. Genetic studies claiming AIT reconstruct a picture on the basis of today’s castes/tribes in South Asia are liable to err when they assume that those castes were identical three to four millennia back.’
g) The AIT conclusions drawn from genetic studies cannot prevail upon Archaeological studies which don’t find any archaeological evidence. To claim otherwise is to trash the entire Archaeology as a science which is manifestations of great intellectual arrogance.
3. Most of Indian geneticists who do not accept AIT today agree that there has been some migration in and out of India but the direction of R1a migration is inconclusive based of present genetic data. You need more detailed and representative genetic studies.
4. In view of above one cannot claim “that, South Asians, especially those speaking Indo-European languages, show significant admixture derived from EHG.”
5. We have to wait for ancient DNA studies of samples from India to come to some firm conclusions and they are likely to be available before long. This may or may not change views about AIT but at least we’ll have science to back our claims unlike the speculations running the roost here.

Davidski said...

@akhil misra

First and foremost not all people who criticize AIT/AMT support OIT. To equate critique of AIT to peddling OIT is infantile. It is a non-sequitur.

In that case, point me to some rebuttals to the Silva et al. paper and/or Joseph Hindu article which don't consider OIT as the better alternative to AIT. Thanks in advance.

The rest of your post is full of fact dodging and obfuscation.

You have no real arguments. All you have is hope that something will turn up in South Asia that might throw a spanner in the works once practically all of South Asia is sampled.

You should be more honest with yourself. That way you won't spend 10 years hoping for a miracle. Instead you can already start trying to figure out what really happened, because a lot of rich data is already available.

akhil misra said...

Read very very slowly Chaubey/Thangaraj, particularly this"With the information currently available, it is difficult to deduce the direction of haplogroup R1a migration either into India or out of India, although the genetic data certainly show that there was migration between the regions. Currently, CSIR-CCMB and Harvard Medical School are investigating a larger number of samples, which will hopefully throw more light on this debate." Where does this consider OIT as better alternative to AIT.
As things stand today, OIT is fairy tale and AIT is trash.
Geneticists-AIT is 'The theory',all Archaeology is trash.
Archaeologists-AIT is not even a hypothesis worth considering, Genetic studies peddling AIT is a trash.

Davidski said...

@akhil misra

I did read very slowly, or rather carefully. Chaubey/Thangaraj don't like AIT because they prefer OIT. That is clear enough. And keep in mind that I've seen other comments by them, especially Chaubey, on the issue, so I know where they stand.

And guess what, they're wrong. It's not difficult to deduce which way the migration went; it obviously went from Europe to India.

I covered this in detail in my post, so I don't need to repeat myself here. AIT is not trash; the idea that the migration went from India to Europe is trash, completely devoid of any logic.

Seinundzeit said...

At the end of the day, I think it's undeniable that the Indo-Iranian languages were intrusive to the Iranian plateau, southern Central Asia, and northern/central South Asia.

I mean, no one can seriously argue against this; long before aDNA, we had the linguistic data.

Also, it's fairly obvious that certain populations from Tajikistan all the way down to Uttar Pradesh have rather substantial amounts of Steppe_EMBA-related admixture, often at levels that equal Eastern/Northern Europeans.

So, I'd say that arguing against OIT is somewhat pointless, as the ship for that theory has sailed very far away.

Instead, I think that a discussion on how these languages spread across the Iranian plateau and the greater Punjab, and an attempt at understanding the underlying social dynamics, will prove to be far more constructive/useful. This is where we need archaeological data and analyses (especially with regard to population-levels in the Indus Valley, climatic conditions, cultural synthesis between BMAC and Andronovo, etc).

Regardless, certain questions concerning early Indo-Aryan culture, and their putative expansion into the greater Punjab from the contemporary Indo-Iranian borderlands (eastern Afghanistan/northwestern Pakistan), still linger in my mind.

For example, the "Aryans" of the Rig Veda seem to have been patriarchally-organized nomadic people who tended towards conflict/truculence/violence (which I find interesting on a personal level, since many Pashtuns are still, in this day and age, patriarchally-organized nomadic people who tend towards conflict/truculence/violence).

Yet, the ancient people of the Swat valley, who have long been identified as vectors for the spread of Indo-Aryan languages into South Asia proper (and I'm willing to bet that aDNA analyses will substantiate those theories/linguistic identifications) seem to have had aspects of matriarchy in their culture.

Of course, being matriarchal doesn't exclude widespread intra-personal violence + military subjugation of neighbors (the Iroquois nation were a good example of a matriarchally organized people who spent much of their time making war against their neighbors).

But, if I'm not mistaken, there are no hints of this sort of societal structure in the Rig Veda.

Then again, I'm no Rig Vedic scholar (not by a long shot). So, if anyone can correct me on this, they are more than welcome to do so.

Anyway, many aspects of material culture seen in the ancient Swat Valley cultural complex survive today among the Kalasha people, which is pretty cool.

Coldmountains said...

@Roy King

Prior to Aryans, Dravidians did not even exist in Northeast India (Nepal,Bengal.,,) and R1a-L657 is very high there. Actually in Nepal all R1a-z93 is L657. Saying that L657 is from Dravidians is the same like thing that R1b is from Basques. but actually R1a-L657 is not very high in South India and somehow there is even very much Z2124 among Dravidians. Brahmins got their steppe ancestry from L657 people that is now very obvious

Balaji said...

Davidski,

Could you run qpGraph with the following file?

https://drive.google.com/open?id=0B-aM8HL5oc9WT193ZWR4bDFNVG8

I hope it is in the correct format. Thanks.

Davidski said...

@Balaji

https://drive.google.com/file/d/0B8XSV9HEoqpFaGZoMDdCclJ5bkk/view?usp=sharing

So is this supposed to be an Out-of-India model? Because it looks like an Into-India model, using a more basal Yamnaya-like population, clearly from West Eurasia.

Balaji, you need to realize that India is nowhere near Europe and the Caucasus.

Peter Klevius said...

@Davidski: "Please note, I agree that the suggestion of a potentially violent invasion of South Asia by Indo-Europeans, and, indeed, Aryans, sounds provocative, and will always be politically controversial no matter how much evidence is gathered in its favor. But what if it really happened?"

Why is an Indo-European violent invasion "always politically controversial" when the violent invasion of islam by Arabs obviously isn't? According to historians (e.g. Hugh Kennedy) the main objective of islamic invasions was to get booty by the sword and taxation under sharia - plus sex and other slaves.

Palacista said...

I come to this debate from the historical linguistics viewpoint. There is absolutely no doubt that the IA languages came from an area contiguous with proto Slavic and more distantly with Greek.

mmghosh said...

Peter Klevius here is a clue why the Aryan Invasion Theory is opposed by Hindu fundamentalists.

https://www.pressreader.com/india/business-standard/20170708/281998967489596

Gioiello said...

@ Roy King

Very likely the first in the world who spoke of "synecdoche fallacy" (above all against the best of phallaciers: Ted Kandell) was I. Thus, if you know that someone used that before me, please give me the link.

July 21, 2017 at 6:20 AM

P.S. I deleted my post after having googled and found what I didn't know when I spoke of this fallacy, that Kenneth Burke spoke about that in The Phylosophy of Literary Form, but he spoke of "Synecdochic fallacy" and only I of "synecdoche fallacy" (probably I wrote "phallacy"), thus that Roy King has read my posts with Ted Kandell on the YFull page of FB is very likely and that from there he derived that.
I spent my life around the Textkritik...

Roy King said...

@Gioiello,

Sorry to disappoint you, but I learned of the synecdoche fallacy forty years ago from Kenneth Burke's book. I used to teach about the fallacy in my psychology/psychiatry classes on feminist/post-modern models of personality disorders and the arts. I don't read the posts on the YFull pages on FB.

Gioiello said...

@ Roy King
"@Gioiello,
Sorry to disappoint you, but I learned of the synecdoche fallacy forty years ago from Kenneth Burke's book. I used to teach about the fallacy in my psychology/psychiatry classes on feminist/post-modern models of personality disorders and the arts. I don't read the posts on the YFull pages on FB".

You learned synecdoche fallacy forty years ago: possibly true (but it seems that Burke spoke of "synecdotic fallacy" and not "synecdoche fallacy" as I wrote (not knowing then Burke's book, but of course I should verify).

You taught about that: I have no reason not believing you.

You "don't read the posts on the YFull pages on FB": clearly false, because you read and post there. Anyway I spoke of the "synecdoche fallacy" also here on Davidski's.
"Roy King con Ted Kandell.
16 luglio alle ore 3:32
I wanted to propose that Z2124-R1a might be the carrier for the Indo-Aryan expansion and not R1a-L657. Rise 386 from Sintasha is derived for Z2124 as well as the Poltavka outlier whom Davidski argues is autosomally the likely ancestor of the Indo-Aryans. However, Ted Kandell, finds that the Poltavka outlier is ancestral for L657, the major R1a clade in India".
Do you want to say that it is all from Ted's mind?

Roy King said...

@Gioiello,
Reading is different from posting. I find FB YFull group to be too tendentious to read. Davidski's blog at least has a large variety of expressed opinions, some of which are pretty sophomoric, but I'm used to that as a professor.

Ryan said...

@Peter - "Why is an Indo-European violent invasion "always politically controversial" when the violent invasion of islam by Arabs obviously isn't?"

Because Hindu nationalist politicians don't want to view themselves as similar to the Muslims they vilify.

Open Genomes said...

@David

What is this about, I thought personal irrelevant comments were supposed to be banned? What about enforcing this? Need I remind everyone of Marny?
The censored responses to the discussion here and the personal racial and ethnic attacks have "polluted" various Facebook posts as well (as many of you know) and I personally know that this has driven away many academics from engaging with people here or even reading the discussions. All of you suffer directly because of this, whether you know it or not.

Roy King explained this to me, the relationship to this Classical era poetic and rhetorical term to the present discussion of the AIT and the "Genetic Genealogy" in general.

This is more commonly known as a "faulty generalization".

In this case, I think we have a case of sampling bias.

I first will urge everyone to pay close attention to the YFull tree, which is the only Y (or any genetic) phylogeny to include accurate tMRCAs. The accuracy of these tMRCAs has been proven repeatedly through ancient DNA, and the dating methodology used is completely independent of the results.

I am going to show how the tMRCAs of the various South Asian Y-DNA clades are relevant to the discusson of the origins and ethnolinguistic composition of the Indus Valley Civilization and the AIT.

Open Genomes said...

We need to bases this discussion on our available evidence, not on ethno-nationalistic and religious-based speculation.
(As we all should know, the Rig Veda itself supports the AIT, regardless of current "caste politics" in India today. What constitutes a "religious bias" is not quite as "fundamentalist" as it seems.)

First, we need to determine the possible "aboriginal" (ASI) Y-DNA clades in South Asia, before trying to identify the Y clades of the IVC:

Here we see the YFull C-K281 ("C1b1") phylogeny, the South Asian clade of haplogroup C:
https://yfull.com/tree/C-K281/

As we know, Kostenki K14 from Ukraine, 38,000 years ago, was C1b-F1370*.
No other C1b1-K281s have been found in ancient DNA to date. A single family of C1b-F1370* has been found today which originates in Samarra, Iraq.

Given that C1b2-B477/Z31885 is the Austraian Aboriginal C clade, we can say with confidence that this is a rrepresentative of the "Eurasian ASI-Austroasiatic-Austronesian autosomal ancestry stream" which we can see in the following aDNA and modern samples 3-D PC plot:

http://www.open-genomes.org/analysis/PCA/Eurogenes_Broushaki_(2016)_and_Lazaridis_(2016)_Iran_and_Near_East_aDNA_PC_plot_1-2-3.html

Open Genomes said...

At this point, can only speculate as to the tMRCA of South Asian C1b1-K281.
Notice carefully that there is not a single high coverage C1b1-K281 Y-DNA sequence on the YFull 5.04 tree.

This is direct evidence of sampling bias in this debate.

I myself have paid for the YFull analysis of Y-DNA sequences members of unique Brahmin Y clades, as you will see below. (FYI, I've also personally contributed to the Eurogenes fundraising drive, because I value David Wesolowski's work.) I personally see "haplogroup racial bias" among Brahmins, who happen to be in what is perceived by a certain kind of "nationalist" as "non-Aryan" Y-DNA haplogroups, which they think of as being "not truly Brahmin". That means that I've spent real money when it comes to this debate, in order to obtain objective scientific data about the genetic ancestry of people from the Indian Subcontinent. It's not "just talk" on my part.

If any of you "Hindutva Nationalists" want to contribute to this debate, I challenge you directly to obtain high coverage Y-DNA sequences (preferably, from whole genome sequences) from South Asian Y-DNA haplogroup C1b1--K281, and submit them to YFull for analysis to obtain tMRCAs. Two at a minimum is what we need here.

I assure you that these "Scheduled Tribes and Castes" are indeed "Children of God" (as the Mahatmata taught us) and you will not be "polluted" from sampling people like the Irula who in all the MDS plots show up as nearly "pure ASI", in spite of them traditionally being rat catchers.

Put your money where your mouth is.

If you don't do something like that, you've lost your right to participate in this debate. Everything you say just gives more evidence of your bias and hypocrisy.
If you do this, you will have proven to the world that you don't harbor a scientific bias. (You might even get to confirm a certain aspect of the historicity of the Rig Veda.) Will you do it, or not?

Open Genomes said...

Now we can turn to Y-DNA haplogroup H:
https://yfull.com/tree/H/

Haplogroup H2-P96 (H2-M282) is one of the "Neolithic Y haplogroups" par-excellence.
The tMRCA of H is 46,800 years, which is just after the "H3" abrupt climate change event which allowed the (final) AMH settlement of Eurasia outside of the Near East.

Therefore, haplogroup H-M2826, nearly unique to South Asia, with a tMRCA of 45,000 years, is a primary candidate for a unique "ASI" Y clade, and also is a "terminus ante quem" for the initial AMH settlement of India.
https://yfull.com/tree/H-M2826/

Jijnasu said...

@opengenomes
"I assure you that these "Scheduled Tribes and Castes" are indeed "Children of God" (as the Mahatmata taught us) and you will not be "polluted" from sampling people like the Irula who in all the MDS plots show up as nearly "pure ASI", in spite of them traditionally being rat catchers." One of the most ridiculous comments on this post. The poster would like to know that the term 'harijan' is considered patronizing by the members of the scheduled castes today. Also many opposed to AIT aren't brahmin.

Jijnasu said...

@opengenomes
"Rig Veda itself supports the AIT" No it doesn't and neither does it support OIT. In anycase its a collection of hymns, not a history book all references to historical events are secondary

AJ said...

You constantly bringing up "EHG this or EHG that" does nothing considering they are heavily admixed with mesolithic forgers coming from EAST who brought R clads to Europe

1. EHG are admixed, as Lazridis et al. put it, with population on "Onge->Han" cline.

2. R and it's upstream clads K2b, P are all from deep in Asia. R1/2 diversified in Central Asia and migrated in all directions.

3. R1a/b are also only clads in Europe that have upstream clads deep inside Oceania/Asia, unlike Y-DNA IJG in Europe.

Deal with it.

AJ said...

There is also no point in discussing this without aDNA from South Asia and Central Asia Paleolithic, Mesolithic and early Neolithic.

OIT is similar to European "Anti-Invasionst theory" and "Paleolithic continuity theory", it has even less to do with recently constructed PIE language by linguists.

Davidski said...

Same tired, old, irrelevant nonsense.

Obviously aDNA from Paleolithic, Mesolithic and early Neolithic Asia won't be a game changer.

R1a-M417, Z645 and Z93 have already been found in Bronze Age Eastern Europeans who could not have come from South Asia: they and their ancestors from way back were native to Eastern Europe.

What this of course means, and what you're not getting for some unknown reason, is that South Asians have ancestry from Eastern Europe.

AJ said...

"Obviously aDNA from Paleolithic, Mesolithic and early Neolithic Asia won't be a game changer."

Not for you but it will further add to how admixed your EHG is and better proxies for it through aDNA. Along with better autosomal DNA model for South Asia through ancient DNA from the region and surrounding regions.

When it comes to steppe admixture, non-basal admixture in steppe will further break down into something you wont like as more aDNA from Asia comes in, that's all.

Davidski said...

This is completely irrelevant to AIT.

EHG could even be part South Asian (which it's not), and this still wouldn't make any difference to the fact that there were large scale population movements into South Asia during the Bronze Age originating in Eastern Europe.

Balaji said...

@Davidski

Thank you very much for running qpGraph with my suggested model. This model seems to be fairly good with a z value of 2.03.This is indeed supposed to be an Out-of-India model. The idea is that populations represented by nodes A, B, C and D were all resident in different parts of the Indian Subcontinent. Population C was from the Eastern part and Populations B and D in the western part (what is now Pakistan). Out-of-India migrations at different times in the Mesolithic gave rise to EHG, CHG and Iran_Neolithic. Bronze-age migration of a population related to ANI gave rise to Yamanya_Samara and the spread of Indo-European languages out of India and into Europe.

I expect that when the Swat Valley results are published, they will look like the ANI in this model and it will be possible to model Yamnaya very well as a mixture of ANI and EHG.

Davidski said...

@Balaji

The Swat Valley samples won't be like the ANI population that you describe. Instead, they'll be a mixture of EHG, CHG and, in various degrees, Neolithic South Asians, who themselves will look like a mixture of Iran_N and ASI.

In any case, how do you reconcile your model with the reality that in prehistoric samples R1a is highly correlated with EHG ancestry native to Eastern Europe?

Indeed, it's also highly correlated in South Asia with EHG ancestry. Bit of a problem, no?

Matt said...

@Balaji - quite an interesting model.

I think it really underscores that, with this limited set of outgroups, the model is mostly simplified to a continuum of basically EHG like and Basal ancestry in this region (Iran-CHG-Steppe); Yamnaya, CHG and Iran_N are mostly just points on this continuum.

Your model allows many more degrees of freedom in the precise EHG-like+Basal mix going into Brahmins, so can perform quite well against models which use India as Iran_N+Yamnaya+Onge.

That said, your graph still has the worst fitting stat as relating to a lower then reality signal between Yamnaya and Brahmin (relative to their relatedness to EHG and Onge).

Of course, there are also a limited number of outgroups here, which is a limit with the qpGraph method relative to qpAdm. qpGraph allows us to have these diverging "ghost" nodes, but the drawback is that it can only really consider all the outgroups you can fit on the graph, which is limited here.

The model doesn't really allow much independent drift between nodes, or test nodes in competition with each other.

A modified model that does have independent drifts between a CHG and ANI clade and then has competition for contribution of these to Yamnaya is: https://pastebin.com/raHRNh1E

I think the idea of your an A node and its descendant C and D nodes ever living in India is speculative in the extreme! Does not fit with Upper Paleolithic European evidence either.

Chad Rohlfsen said...

I have a feeling some will disappear from here, like OM. Can't handle or admit when they're wrong. Tears will be shed over Neolithic India.

nizam uddin said...

It's fairly disheartening to read people's post ridiculing the South Asian community; especially, when most scientific institutions favor the justification on how ethnically pure Europe is through statistics. Plus, you have to consider hundreds of years of propaganda, while the European man claimed himself to be a protagonist, and still mutilate millions for the sake of exploitation: you have to wonder, is it really worth it? To destroy the ancestral background of a nation to simply assert ones dominance.

Then again, the pious Western hemisphere has achieved more in the basics of social reality and not natural reality.
Anywho, what's with the 4th or 5th post about India. I thought this blog was about European genes not fortifying the evidence of the Aryan Migration theory. Especially, when the majority of the people who read this blog knows it did occur. It's definitely time to move on from this subject.

Davidski said...

I'm building to a massive crescendo.

Tesmos said...

''Doesn't seem very misleading in this case. It's just that the location of Indo Europeans' paternal ancestors 35,000++ years ago tells us jack squat about where they were 6,000 years ago.''

We need aDNA from South/South-East/East Asia to confirm that the haplogroups like K2/P has it's origins in one of these certain regions.

Davidski said...

@Matt

https://drive.google.com/file/d/0B8XSV9HEoqpFSjd4SzEybER2dTg/view?usp=sharing

Matt said...

@ Davidski, ah that lowers the Z score, *but* also like the modification of Balaji's model model still prefers the flexible "ANI" node to the CHG node - kind of looks like for whatever reason a specific signal linked to CHG doesn't show up here!

@ Davidski and others, off topic, and a bit peripheral for historical population genetics, but still interesting from a population structure perspective:

Huge UK Biobank paper with 500,000 samples. 1% of UK population, and as target at the 40-69 age cohort which represents 37% of British population, roughly 3% of that cohort in the whole population.
http://www.biorxiv.org/content/early/2017/07/20/166298

From population structure POV, good to see the PCA in the main paper and supplement (page 18 onwards) where they have been able to use 407,599 samples(!) via FastPCA.

In the dimensions that they visualise:

See this album: http://imgur.com/a/ViR94

PC1: European vs African, PC2: East Asian vs African, PC3: South Asian vs East Asian, PC4: Middle East vs South Asia (and Europe).

PC5: Not clear population affiliation, but more diverse in Europeans than others, with a weak differentiation between Irish at one end and Northern and Eastern Europe at the other.

PC6: Native America vs East Asia. The sample sizes appear to be massive enough that they are capturing some quite differentiated samples with high level of Native American, though the Latin American country of birth averages are much moderate on this than the most extreme samples (e.g. some Latin Americans have little Native American ancestry, some have much).

Funnily enough Eastern Europe has a weakly enriched signal on this dimension. Actually, includes both Southeast Europe - Bulgaria, Serbia - and Northeast Europe - Lithuanian, Russian (but less strongly on Swedish / Norwegian samples, so not just North Europe related). I do wonder if this reflects something to do with ANE or WHG.

PC7 and PC8 are dimensions that sit on Eastern Europe vs Middle Eastern/East African populations, and against Eastern Europe vs Native Americans.

Prominence of some of these dimensions is probably affected by the sheer weight of European and British samples in the panel.

Anyway, they also use clustering and projection on the first 1-6 dimensions to whittle down their 88% (431,059!) of "White British" samples to an 84% (409,728) sample size. So something like 95% of the self described "White British" samples fit easily within a cluster of the same ancestry.

Part of the interesting thing about looking at it is the sheer size and diversity of the set, which fills in a lot of PCA space, but notably, even with a sample set of that size, some areas are still empty, and the vast majority of the samples really fit into a small part of the PCA space in a way that's quite hard to intuit from the PCA plots.

JohnP said...

@Davidski
Is it possible, or even probable that the "European" admixture in the Chinchorro mummy is actually from a proto-ANE, pre-ANE population? It only signals WHG, Farmers, ASI and a little bit of Northeast Asian, no CHG, no EHG, no Siberian, no other Amerindian types, no Onge/Papuan types either, etc.
Or is it just bullshit?

Chad Rohlfsen said...

^^ It's bullshit. There's no movement from Europe to America.

epoch2013 said...

@nizam uddin

"To destroy the ancestral background of a nation to simply assert ones dominance."

Is Hinduism dead in India? Has India's religion been wiped away by conversion to the Christianity?

No

You rule yourself now. You have your own unique culture and religion. Stop victimizing yourself and start to become proud of what you are: The only surviving living Indo-European religion.

nizam uddin said...

@epoch2013
I see, there is no attempt to victimized the Indian culture but to acknowledge we are all imperfect creatures. Yet, the rest of your comment doesn't actually apply to me because I'm irreligious and an American. It's great that you mentioned it, but for labeling me as a Hindu is short-sighted. So, this helps me in lieu to acknowledge my ethnical background.
I'm actually mixed so in retrospect the idea of racial purity doesn't matter that much to me. Anyways, roughly to give you an idea of my background, father is Indian Muslim from Pakistan and mother is Polish Catholic American. I purposely visit the website to find out more about population genetics and anthropological studies simply because I've always been intrigued by it. And if, you find my comments are to persecute others on this blog, well simply don't take responsibility for it. I'm sure you're a great dude, thus not accusing you of anything because the Steppe-esque people who conquered India have yet to be discovered. Although, recent findings, claim that the Potapovka and the Kura-araxes have been excavated for the z93 subclade. I mainly interpret this information as the lowest extent of understanding ancient civilizations and people because there is much more knowledge to acquire in this world that we live in.

Unknown said...


>High Fst in South Asia panels doesn't seem to be driven so much by ASI-ANI cline, but may be due to very high drift in tribal populations like Paniya, and in highly endogamous caste populations as well.

Do you know any free program that can calculate TVD and ASD? These may be better for testing real genetic distance than FST influenced by bottlenecks.
Also what program do you use for Fst?

Unknown said...


>Funnily enough Eastern Europe has a weakly enriched signal on this dimension. Actually, includes both Southeast Europe - Bulgaria, Serbia - and Northeast Europe - Lithuanian, Russian (but less strongly on Swedish / Norwegian samples, so not just North Europe related). I do wonder if this reflects something to do with ANE or WHG.

It's from EHG which was mostly ANE and to lesser extent from CHG which also has ANE or EHG.

Vara said...

@nizam uddin

"Kura-araxes have been excavated for the z93 subclade."

Where did you get this?

nizam uddin said...

Vara, I read the news on wiki. Actually, it's based on the intercommunications between the steppe and the Kura-axes people. So, this scenario is only possibly if the m269 and z93 diversified within the Kura-axes culture. In other words, I completely forgot about lazaridis (2016) article stating that only the m269 has been found. Sorry about that...

Davidski said...

@JohnP

The Chinchorro mummy doesn't have any European admixture.

@nizam uddin

There's no Z93 in any of the Kura-Araxes samples.

Chad Rohlfsen said...

There's no m269 in Kura Araxes either. It was L389. A dead end.

nizam uddin said...

L389 was found in Iran and not Kura-araxes. Lazaridis (2016) claims that they found m269 in the excavations. If you know more than he or Reich. Please provide reliable sources, instead of the information that is provided from eupedia or other blogs.

Davidski said...

Here...

Armenia_EBA

I1635: R1b1-M415(xM269)

http://eurogenes.blogspot.com.au/2016/06/the-genetic-structure-of-worlds-first.html

The x means that it's not M269. And the R from Iran are both probably R2.

There's one R1b from ancient Iran, from an Iron Age sample that probably has steppe admixture. The subclade is Z2103, same as in (eastern) Yamnaya.

Chad Rohlfsen said...

I know what I'm talking about. You're confused.

nizam uddin said...

Davidski, the m269 was not found in the kura-araxes, then there is no specifics to the haplogroups they belonged to? Okay, Chad you right and I was wrong.

74d9d9b8-55ef-11e7-bd4d-43c094282792 said...

@Kjontendor001
If you live in a colonized land, then obviously you are a weak snowflake. Even the "degenerates" are stronger than than you are. Now go whine about it on infowars or wherever.

@Davidski It has already been pointed out that not supporting AIT does not mean supporting OIT. There is already evidence of Tazabagyab culture settling down after interacting with BMAC without actually "invading". Even the "cult of fire" is said to have began in that culture, which is one very strong Indo-Iranian cultural similarity. Before jumping the gun and investing in a hypothesis where people on horses and driving sheep invaded and toppled the IVC, wait for archeological evidence from central/south Asia. Instead of putting words into the mouth of people like Chaubey, why not just ask him directly?



Davidski said...

It has already been pointed out that not supporting AIT does not mean supporting OIT.

It usually does. It certainly does for most, if not all, of the authors of the articles I linked to above.

It's not crucial that Chaubey supports OIT. The reason that I linked to his article is because it's essentially based on the same logical fallacies as the rest of the articles.

Certainly, his view is that there was no large scale population movement to India from the Eastern European steppe during the Bronze Age, or even generally from West Eurasia at least as far back as the Neolithic. I don't have to ask him to know this; he's argued it himself repeatedly, and it's a view that is almost as silly as OIT, and 3/4 of the way there.

So what are you discussing with me? Degrees of nonsense? I'm not interested. And can't you at least get a decent nick for posting here?

Ritesh said...

It's either AIT or OIT because, hello!!!, Language did move. One way or the other. So opposing one means supporting other. Automatic.

Slumbery said...

Ritesh
Theoretically it is possible to imagine scenarios that are neither AIT nor OIT. For example a dialect continuum in Northern South Asia - South Central Asia - Iran, then an expansion of non-Indian branches from outside India. Or an assumption that IE spread as a trade language / lingua franka in most of its range, instead of any big scale migration/invasion. Or that Indo Aryan spread in India by the peaceful missionaries of a religion. And so on. People who don't like AIT and OIT can probably come up with even more narratives in their cushy armchairs.

Davidski said...

For example a dialect continuum in Northern South Asia - South Central Asia - Iran, then an expansion of non-Indian branches from outside India. Or an assumption that IE spread as a trade language / lingua franka in most of its range, instead of any big scale migration/invasion. Or that Indo Aryan spread in India by the peaceful missionaries of a religion.

But none of these scenarios are realistic, because they don't explain the massive shift in the genetic ancestry of South Asians during the Bronze Age, which will be clearly seen in the aDNA record.

Ritesh said...

"Or an assumption that IE spread as a trade language / lingua franka in most of its range, instead of any big scale migration/invasion. Or that Indo Aryan spread in India by the peaceful missionaries of a religion."

Any peaceful migrant/s can't impose a language in such a manner that even names of rivers of an inhibited land are changed. For eg, Sindhu, Saraswati etc have IE etymology.

batman said...

What were these rivers called before the IE? "Something else"?!

Jijnasu said...

Hard to say that there any evidence for the OIT. However some of the archaeology in question does create doubts about any large scale movements into India from the steppe. (Some archeaologists have also proposed a south-eastern origin for Andronovo linking it with S-C Asian cultures but as we now know that's not true).

There are problems with calling it an invasion though since it seems to suggest a motivated conquest under some central authority when in reality it involved several indo-aryan tribal factions, with clashes amongst themselves and with the indigenous chieftains who controlled what remained of the collapsing IVC

batman said...

Davidski,

"But none of these scenarios are realistic, because they don't explain the massive shift in the genetic ancestry of South Asians during the Bronze Age, which will be clearly seen in the aDNA record."

Proving OIT wrong don't prove AIT right. Please drop that playback-machine.

The "massive shift" you describe in BA-India don't need a language-shift to be explained.

The introduction of cattlefarming and horsebreeding may explain the observed GENETIC shift just as well.

There were massive shifts in the DNA of Mesolithic India and Iran too - introducing y-dna H and J, who became the predominant dynasties in ME/NE/BA India.

Which may explain the intro of the IE languages to India far better than any BA "invasion".

Creating ancient "invasions" as one see fit is nothing but armchair-intrusions to a desktop-reality.

Rob said...

Can the aDNA from India and Greece come out already, so I can formulate my new, correct model of PIE dispersal !

Ric Hern said...

We already see a adoption of Cultural Materials within the Tazabagyab Culture and a spread over most of the former BMAC territory after +-1900 bC. when Climate chance kicked in.

This dilution of Steppe Material Culture could have happened to the East of BMAC as well and by the time these migrations reached India there could have been little, if any traces left of the Steppe Material Cultures.

Matt said...

Unknown: It's from EHG which was mostly ANE and to lesser extent from CHG which also has ANE or EHG.

On the one hand, likely, but the patterns on the PC6 are:

* Strong+ = South America
* Strong-Intermediate+ = Northeast Europe
* Intermediate+ = Southeast Europe (Slavic speakers), South Korea, Japan, Pakistan
* Weak-Intermediate+ = Scandinavia, Iran, Turkey
* Neutral Signal= Western Europe (Britain, Spain, France), Africa (North, East, West, South)
* - = China, South India, Southeast Asia (most negative signal Philippines)

That's a little different from the normal rank ordering of Yamnaya rich or CHG / Iran_N rich populations (which is just latitude based in Europe, not longitude based)...

Also what program do you use for Fst?

I'm just using matrices that have been published by Davidski or in supplements to papers.

Do you know any free program that can calculate TVD and ASD? These may be better for testing real genetic distance than FST influenced by bottlenecks.

Well, I would quibble that genetic distance influenced by bottlenecks is *real* genetic distance. But I put that point aside, I don't know of software how calculate TVD or ASD. I would imagine TVD is quite intensive on processor power. The papers which use TVD measures should describe the software they use in the methods. ASD doesn't work in phylogenetic networks though btw - only matrices with a zero diagonal distance work, and in ASD measures the allele sharing distance within population is not a zero distance.

Ric Hern said...

Was there not some Indian Specific remains found among a minority of the Tarim Basin Mummies ? So surely the Tarim people also knew how to reach India.....?

Balaji said...

@Davidski

You asked, “In any case, how do you reconcile your model with the reality that in prehistoric samples R1a is highly correlated with EHG ancestry native to Eastern Europe?”. My answer is that EHG in Eastern Europe itself is a result of a mesolithic Out-of-India migration and therefore it is not surprising to find R1a in aDNA from them.

@Matt

Thank you for your comments and for thinking of an additional model. You wrote, “I think the idea of your an A node and its descendant C and D nodes ever living in India is speculative in the extreme! Does not fit with Upper Paleolithic European evidence either.” Let me suggest why this may be plausible. A single Out-of-Africa model is now widely accepted and many favor the Southern Coastal route. If this is the case, the mouth if the Indus river would have been the ideal first place for the Out-of-Africa migrants to settle since the Arabian and Iranian coast line is all fierce desert. The rest of the world would then have been populated by Out-of-India migrants. The Indian Subcontinent is large enough and ecologically diverse enough to have supported populations represented by nodes B, C and D.

@Open Genomes

Regarding your interest in Y haplogroup C, I looked up my DNA relatives in 23andme and found one individual who is listed as C-P92. Is it worth contacting him? I belong to a Brahmin caste in which everyone except a few stray individuals practiced strict endogamy. Nevertheless, the Y-haplogroups among my 23andme DNA relatives is quite diverse. They include C-P92, F-M89, G-L30, H-M52, H-M69, L-M27, Q-M346, R-L266, R-M173, R-M512 and R-P25_1.

epoch2013 said...

@nizam uddin

I think my point remains despite your etnicity: The ancestral background of India has not been destroyed, as you suggested. An most of the resistance against AIT theory comes from Hindu nationalists, or at least Hindu nationalists are among the most vocal.

Matt said...

@Balaji, I understand your reasoning for it, though I don't agree that there is any particular reason to think the "mouth if the Indus river" would be a particularly rich or likely environment for early humans out of Africa, and this may be a bit of a cultural -centricism, or a back projection of the "India as a second Africa" ideas of the 2000s (and general tendencies to bias towards South->North migrations) that have no real weight any more.

Certainly, even if a split of A to D and C happened within India (and I think there's no evidence at all that this ever happened), then we know that we have separation and structure within the D node outside India quite early on, through adna of Upper Paleolithic Europe, and that South Asia today suggests differential relatedness to particular D node descendants associated to North Eurasia, which is hard to reconcile with structured populations purely within South Asia.

If we ever do get adna evidence, then sure, but until then "speculative in the extreme!".

EastPole said...

@Rob
“ I can formulate my new, correct model of PIE dispersal !”

We all have our own correct models of PIE dispersals.

Could you start with defining PIE and the methodology which led you to such definition and the discovery of PIE homeland.

In my opinion we should look at what is common to Europe and India in terms of language, religion and culture and correlate it with what is common to Europe and India in terms of genetics.

Rob said...

Yes the palaeolithic record for India isn't exactly bountiful . This might relate to poor research to date, but it's not betokening positive evidence that the Indus should have been the chosen niche for out-of-Africans

Slumbery said...

@Balaji
EHG might have some ancestry from South Asia at some point, with Central Asian/Siberian populations as mediators, but Mesolithic is way too late for that. Don't you think that we would see a post-Mal'ta South Asian migration in ancient DNA, when Mal'ta is used as reference? The more it is back in time, the less it is relevant to the question at hand (spread of IE), and Mal'ta pretty much puts a limit on the time of the migration of P out of the South: it had to be earlier than Mal'ta, when none of the current language families could have possibly existed even in pre-proto form.

Jaydeep said...

I am amused as to why David is giving so much coverage to the issue of Aryan Invasion/Migration into India. Perhaps he is surprised by the fact that so many seemingly intelligent Indians are so vociferously opposed to the AIT/AMT.
If he bothered to find how the whole theory came about in the 1st place, he will not find the opposition of so many Indians surprising. A theory that was pulled out of thin air and which so far has no archaeological evidence to back it up and nor does it a strong linguistic argument in its support.

I am a bit late to join the debate but I think its necessary to give a reply to David's queries.

R1 and R1a are found in the remains of Eastern European Mesolithic foragers. Were these individuals recently arrived Indo-European-speakers from South Asia?

Y-dna R is also found in the 24,000 year old Malta boy which forms a significant part of EHG ancestry (the other being WHG). It is this ANE part of EHG's admixture that can be associated with Y-dna R1a found in EHG.

ANE is clearly related to South Asian ancestry and this relation is certainly not recent. If we were to compare the affinities of the ANE, EHG and WHG respectively with South Asians, ANE will be closest followed by EHG with WHG showing the lowest affinity to South Asian populations among the three. This clearly indicates a deep ancestral link of South Asian populations to ANE type ancestry.

This argument is also supported by the presence of ydna R2 in South Asia as well as the basal clades of R1b. Moreover, in a more recent paper on y-dna Q3 by Balanovsky et al, it was found that

" the deepest Upper Paleolithic split (Q3e vs the rest of Q3) to West and South Asia: most Q3e samples (green) were found in Pakistan and India, although some were also found in Europe and North Asia (in Kadom Tatars to the west of the Volga river and in a single Evenk sample from South Siberia); these latter samples fall into the narrower Q3e-YP4500 sub-branch and might represent a later spread of Q3e, rather than its place of origin."

Clearly the presence of Ydna R & Q is old in South Asia and therefore the ANE affinity in South Asians is also quite old. This therefore strongly supports the idea that y-dna R1a could have had a presence in South Asia atleast since the Palaeolithic. Either R1a travelled from Siberia to South Asia or the other way around, but the migration is likely to have happened before the Neolithic.

Therefore, both EHG as well as South Asian HGs could have had y-dna R1a. There is nothing strange in it.

Gaspar said...

You all need to read the Karafet 2014 paper on Marker P and R in South-east asia.
And also the follow-up 2015 paper by Karafet

Jaydeep said...

what are the chances realistically that R1a-M417 is from India or South Asia, considering that prehistoric European samples, with absolutely no signals of ancestry from South Asia, belong to both M417+ and M417- lineages?

How do you calculate what is South Asian admixture ? Does it require the presence of ASI ?

For starters, ANI and ASI are modern constructs. There is no evidence that two such distinct groups as ANI & ASI actually existed. Until we have aDNA from South Asia, this remains a hypothetical construct. This is one good reason why aDNA from South Asia is so important.

Secondly, as many studies have now shown, the West Eurasian type ancestry in South Asia dates to the Palaeolithic era and cannot be solely explained by Bronze Age or Neolithic migrations into South Asia. The deep split between Indian and West Asian y-dna J2, the equally deep split between West Asian & Indian y-dna Q3, mtDNA U7, etc also prove this. The Silva et al study that David has championed so much, also argues that the earliest uniquely Indian branch of west Eurasian mtDNA is as old as 21 kya.

This proves that proto-ANI ancestry in South Asia dates to the Palaeolithic.

On the other hand, the ANI-ASI admixture, if they indeed were real populations, is only dated to between 4200 - 1800 YBP. If correct, this means that before 4200 YBP, a ANI like populaton existed in South Asia since the Palaeolithic without any admixture with ASI. Can such an ANI-like population with high ANE affinity not be considered as a population source for admixture in Europeans ? Obviously this will be cleared out once we have aDNA from South Asia. But until then, on what evidence are you ruling out this very genuine possibility ?

India, and indeed, South Asia as a whole are dominated by one fairly young subclade: R1a-Z93. Europe is home to three different subclades

Clearly Z93 is most dominant in South Asia because it has spread during a Bronze Age expansion. But a Bronze Age expansion does not need to be from a source in the steppe. It can also be from one specific region in North India such as Haryana.

Also, a Bronze Age expansion of a young lineage does not rule out an older presence of ancestral lineages in a geographically more restricted setting. Older studies on Indian y-dna R1 indicated the presence of some basal lineages of R1a in South Asia. These have so far not been verified with the newer techniques of higher resolution. The Underhill et al study was hopelessly under sampled for South Asia and cannot be used to prove anything as far as South Asian R1a is concerned.

Plus, if ancient samples from Eastern Europe show higher R1a diversity than in modern Eastern Europe, why can this not be the case for South Asia too ? Why is it assumed that modern diversity in places like South or West Asia is a recent phenomenon and that unlike Europe, ancient genomes from South or West Asia will not show greater diversity ? This inference is based on what ?

how do you know where the ancestral lineages of such exceptions in South Asia were during, say, the Neolithic? What makes you think they were in South Asia?

What makes you think they were not in South Asia ? You have some secret access to aDNA from South Asia ? If Y-dna R1b could be in Italy by 14000 YBP while ydna R is present in Siberia by 24000 YBP, why cannot R1a or R1b not be present in more favourable climates of South Asia ?

Seinundzeit said...

For those among us who think that the notion of Indo-Iranians migrating into the Iranian plateau and the greater Punjab via the Af-Pak highlands (a migration which may have potentially involved substantial violence/conflict, in certain cases) has "no evidence" supporting it (besides the genetic data), please read this (very closely, and in full).

http://www.people.fas.harvard.edu/~witzel/EJVS-7-3.pdf

A beautifully clear and scholarly/sober analysis of these questions.

Every objection raised against the idea of intrusive Indo-Aryans is dealt with here (this is from long before the aDNA revolution, so keep that in mind though).

After having read this, I think it behooves me to further acquaint myself with Michael Witzel's scholarship. In the piece I linked to, he seems to be very erudite.

nizam uddin said...

epoch2013@
.
"You rule yourself now. You have your own unique culture and religion. Stop victimizing yourself and start to become proud of what you are: The only surviving living Indo-European religion."
The partition of India literally transpired less than a hundred years ago. Frankly, this is a dead end conversation. The fact is you'll never understand because white people believe there above racism and many forms of injustice. Oh sorry, I digress, concerning your sentence, in the context of your words (in quotations: It's above the first paragraph) you assumed that I was Hindu because obviously in my picture there are limited ethnicities I may belong to, but you failed to identify my name correctly. Furthermore, in your words, "the only surviving living Indo-European religion." This is redundant since Indo-European is referred to as language family and not attached to religion. Unless,the attempt to generalize culture based on premature theories like proto-indo-european which there are no archaelogical facts. And if you do, please read carefully most of which are identified such as: toponyms, typology, etymologies, etc., by which all are native to Asia and not Europe. To conclude, your summarized thoughts "ancestry wasn't destroyed," well, the British re-possessed everything that belonged to Indians. The British imperials persecuted the Hindu's destroyed there temples and replaced it with christian churches. They converted every single reluctant Indian to there ideologies. The avarice of the British was unudated and quite frankly, I don't know of all of the injustices that the British committed towards the Indians, but it doesn't look good.

Davidski said...

@Jaydeep

Your claim that an OIT expansion from North India to the steppe would not leave an ASI signal because ASI is a modern construct is very difficult for me to take seriously.

But OK, if we assume that this may have been the case, then you still have a big problem, because modern-day South Asian populations show West Eurasian markers typical of the Iranian plateau that aren't found in Bronze Age Eastern Europeans, like Y-hg R2 or mtDNA U7.

There are lots of examples of such markers, and the upshot is that even though the West Eurasian or ANI ancestry carried by Indians is, in variable degrees, part Yamnaya-related, and clearly from the EBA steppe, Yamnaya did not have ancestry from India.

So now you have to get even more creative, and claim that since the OIT expansion by the ancestors of one half of Yamnaya, North India has seen an influx of not only ASI admixture, but also Iran Neolithic and EHG admixture.

In other words, you have no argument. You're just repeating the same, tired, irrelevant, stupid shit that you've been posting here for years.

nizam uddin said...

@epoch2013
Honestly, I'm not here to argue. The point is you need to consider the pro's and con's with imperialism. Obviously, something that is blatant within imperialism and colonialization is the implication of propagandha. As I'm simply writing this for fun. FYI, I don't oppose to all of the ideals on this blog, but it's very uncanny how much of it supports antiquated forms of scientific racism. So restrospectively, people of that particular ethnicity can alleviate there guilt, because morally, imperialism and the enslavement is inhumane but that is beside the point.
I live in a country that was colonized by many different Europeans and has it's own history. So, merely to reiterate there are multiple factors and culture was in fact destroyed. Either way, It doesn't make much of a difference. America suffered many hardships especially the natives. All in all, I was born and loved being raised in the U.S. and in my opinion, is the best country in the world; and I stand by it.

nizam uddin said...

*Afterthought: to clarify, modern science supports the idea of scientific racism not only genetic and anthropological blogs. So, it's not something you can escape when reading about these subjects.

nizam uddin said...

@Rob
There is simply no comparison between the Balkans and India.

" because they wanted Russia out of the ports, It's all about money and self interest."
I don't disagree.

aniasi said...

On the invasion vs migration side, (FYI, I find evidence for OIT to be tortured and in the same category as creationism) it seems people are still making the same assertions without evidence, in this thread.

The insistence on calling this an invasion results in the rejection of an Indo-European migration/intrusion in to SA on the basis that this is another colonial theory of 'caucasoids' forming the civilising class over the natives.

Davidski never responded to my points on this in the last thread, so I am going to be annoying and repost:

aniasi said...

I wanted to get these on the last comment thread, so let me take them on right now.

'- the population in the northern part of the Indian subcontinent during the Bronze Age, even after the collapse of the Indus Civilization, was likely to have been very large for its time, and yet there was a massive pulse of admixture across South Asia from the steppe and a turnover in Y-chromosomes, especially amongst the ruling classes, suggesting that something very dramatic took place that had a major impact on the social and political fabric of the region'

Stilicho & Aetius. Collapsing civilisations show an increase in violence, but that does not always mean an invasion. One group of barbarians, like Stilicho, Aetius & Ricimer can come to power because the local men all die in war, and they are needed to fight other barbarians.

"- early Indo-Europeans in the Near East, from the Hittites to the Scythians, are often recorded as warlike and expansionist, with a habit of invading and subjugating other peoples, like the Hattians, Hurrians and Mitanni (who apparently ended up with an Aryan elite)"

I do not mean this to be personal, but this is some terrible historic analysis. Having studied History at a University that is excellent in these matters, and worked with those whose theses were published on Near Eastern History during this period, this doesn't really fit with the historical method as it applies to ancient societies.

1) The Indo-Europeans were no more violent or expansionist than any other society at this time. The difference is that they were remarkably successful in imposing their language and certain cultural attributes, though not always.

2) The early Hittites are first attested as vassals of the Assyrians. This is around 2000 BC, while their conquest period only peaks around 1300 BC. This isn't some Indo-European trait, since Near Eastern polities were constantly at one another's throats. We do not know how they came to power.

3) The theory of a violent takeover of the Hattians has been largely disproven. It was based on bad archaeology that postulated an apparent 'migration of peoples fleeing the Hittites.'

4) We do not know how the Hurrians wound up with a Mitanni Indo-Aryan elite. In fact, we only have a few words used by Kikkuli, references to gods in a treaty, and ruler's names. As a whole, their society was basically Hurrian, and outside the treaty of Qadesh, they seem to be worshippers of the local deities. Surprisingly, it seems that Hurrian society was largely the same, and that their Indo-Aryan language and culture was not very successfully imposed.

"if early Indo-Europeans outside of South Asia had a penchant for invasions, then there's no reason to believe that the M.O. of the early Indo-Europeans in South Asia would have been any different, unless some sort of direct empirical evidence says so, but what kind of direct empirical evidence?"

Good reason - we don't know how the Indo-Iranians arrived. They show up at the Medes, on the Iranian plateau, but we don't see a conquest of the Elamites. The Iranian arrivals are actually vassals of the Assyrians, which betrays the idea of a powerful conquering group.

Also, the steppe component in India does not peak with the warrior castes, but instead with the priests. Brahmins have more of the steppe component than Rajputs.

Further, the steppe component correlates with the Iranian Neolithic components, which indicates some continuity with the local ruling elites. If this was the invasion you propose, the numbers would line up differently.

Vara said...

@Seinundzeit

This model isn't as perfect as you think. If I have time later this week I'll comment on his dating of the Avesta and the Rigveda.

In short:
1. The Younger Avestas are probably a post Artaxerxes II text and most of it was written during the Sassanid Era, and were most likely propaganda tools by the Sassanids. They make no sense geographically all the way up to Rhaga and Ranga but not Kerman?

2. He makes the assumption that the Rigveda was written in one place during one period, by ignoring the different Sarasvaitis. Helmand, which is described similarly by both Indo-Aryans and Iranians, which flows into the lake and Ghaggar-Hakra which goes into the desert.

3. Assumption that fire worship = Iranians or Indo-Iranians. Evidence of fire worship in IVC is older than Sintashta.


https://en.wikipedia.org/wiki/Indigenous_Aryans#/media/File:Indo-Iranian_migrations_according_to_Kazanas.jpg

^ The route here makes more sense if we want to follow the texts, but it can't explain it DNA wise.

So, yes it's not clear as everyone thinks.

Rob said...

@ Aniasi

All very good points.
Perhaps the Indo-Europeanization process of Eurasia was more protracted and less of a 'manifest destiny' than is generally posited.

Vara said...

@aniasi

"Good reason - we don't know how the Indo-Iranians arrived. They show up at the Medes, on the Iranian plateau, but we don't see a conquest of the Elamites. The Iranian arrivals are actually vassals of the Assyrians, which betrays the idea of a powerful conquering group."

The Medes were described as mighty and were called Umman Manda by the Assyrians. The Medes, Cimmerians and Scythians had many victories and defeats against the Assyrians who built the largest empire of their time, so yes they were a powerful group. Also, the Indo-Iranian interaction with Mesopotamia is older than the Medes since there was a city named Baghadu during Hammurabi's time.


But you are right people forget that even after conquests Indo-Europeans didn't always impose their languages. For example, Persians ruled Mesopotamia for hundreds of years and Syriac was the main language of Mesopotamia when the Muslims arrived.

Davidski said...

@aniasi

Stilicho & Aetius. Collapsing civilisations show an increase in violence, but that does not always mean an invasion. One group of barbarians, like Stilicho, Aetius & Ricimer can come to power because the local men all die in war, and they are needed to fight other barbarians.

I also don't mean to get personal, but if this is any indication of what they taught you at that prestigious University, then I think you wasted your money.

The idea that you can draw useful parallels between Rome and the Indus Valley Civilization is at best very hypothetical and far fetched.

Keep in mind, for instance, that the Germanic barbarians did not make much of a demographic impact on Rome and post-Roman Italy, while the invaders from the steppe significantly re-shaped the genetic character of almost all of South Asia. So whatever happened in South Asia was huge, and potentially bigger than what happened in Rome.

This isn't some Indo-European trait, since Near Eastern polities were constantly at one another's throats. We do not know how they came to power.

So you admit that invasions and conquests at this time were common place, and yet you fail to consider the possibility that the IVC was invaded. Why specifically?

Also, the steppe component in India does not peak with the warrior castes, but instead with the priests. Brahmins have more of the steppe component than Rajputs.

Really bad reasoning here, because obviously once the Indo-Aryans seized power in South Asia, there was little incentive for many to remain serving as warriors, when they could choose safer and more lucrative and prestigious positions.

aniasi said...

@Vara

I am talking about a different time period. The Hamadan area was conquered by the Assyrians, who note the Medes as vassals. It isn't until the decline of the Assyrians, post 700 BC, that they assert their conquering power. When is the dating of your Umman Manda statement? Does it actually refer to their conquering power, or the bravery of their soldiers under the Assyrians?

On the Cimmerians, they don't show up until well after 800 BC, which puts them several hundred years after the original IE expansion into the area. We are also not 100% certain they were Iranian either. Further, they are originally defeated by the Assyrians, and they never formed a cohesive or imposing organisation of any sort. They vanish within 100 years of their first certain mention in Mesopotamia.

Also, the Scythians only show up as mighty conquerors well after the date range of the Indo-Iranian expansions, which again ties back into my original point that these do not provide any standard of evidence for what we can extrapolate about IE and I-I intrusions around Euraia.

Samuel Andrews said...

@Everyone,

After looking over Indian mtDNA, I'm pretty confident there's hardly any Steppe mtDNA in India. Their West Eurasian mtDNA is overwhelmingly derived from Iran Neo groups.

But mHGs that have basically been proven to be Steppe and EEF lineages with ancient DNA from Europe do exist in India. Keep in mind Steppe lineages also occasionally pop up almost everywhere in Asia. There's no exceptional mtDNA link between India and Steppe.

The occasional occurrence of EEF mtDNA could mean proto-Aryans did infact have EEF admixture like Andronovo.

Vara said...

@aniasi

"I am talking about a different time period. The Hamadan area was conquered by the Assyrians, who note the Medes as vassals. It isn't until the decline of the Assyrians, post 700 BC, that they assert their conquering power. When is the dating of your Umman Manda statement? Does it actually refer to their conquering power, or the bravery of their soldiers under the Assyrians?"

The Assyrian decline was after the death of Ashurbanipal, their greatest king, 627 BCE and it was thanks to the Iranians.

The statement was during Shamsi Adad's conquests of the Medes the word "dannu", meaning strong and fierce, was used to describe them, and these were the divided Median tribes. At the hieght of their power they were called the untouchable Medes.

"Further, they are originally defeated by the Assyrians, and they never formed a cohesive or imposing organisation of any sort."

It's because the Cimmerians were the same as the Scythians to the Mesopotamians. Their ultimate defeat wasn't by the Assyrians, in fact the Assyrians were afraid of them.

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


"which again ties back into my original point that these do not provide any standard of evidence for what we can extrapolate about IE and I-I intrusions around Euraia."

I agree with this actually. Maybe I didn't show that in the previous post? Apologies.


Samuel Andrews said...

Forget to mention, I think U2 in India is all of Iran Neo origin.

Samuel Andrews said...

And the ANE in India that can't be explained by Iran Neo or Steppe is probably due to Iran Neo ancestors who had lots of ANE? It looks like the Iran Neo ancestors of Indians were really rich in mtDNA U.

aniasi said...

@Davidski

I also don't mean to get personal, but if this is any indication of what they taught you at that prestigious University, then I think you wasted your money.

Bryan Ward-Perkins and James Howard Johnston. They worked with facts, and the point they made is that even though the Roman Empire fell to a Barbarian Invasion, by the end its own elites were of the same ethnicity as those of its conquerors. That point should not be lost. They were keen to emphasise the same point when it came to the Barbarian invasions of other societies.

The idea that you can draw useful parallels between Rome and the Indus Valley Civilization is at best very hypothetical and far fetched.

Of course! However, the fall of the Western Roman Empire is perhaps one of the best chronicled histories of civilisation's collapse in the face of invading peoples of lesser advancements. There are ample chronicles that show how the Goths first arrived as refugees, only later to become conquerors, or how Aetius (paternally Barbarian) was allied to other Barbarians to stop the Huns. That should be the point.

Keep in mind, for instance, that the Germanic barbarians did not make much of a demographic impact on Rome and post-Roman Italy, while the invaders from the steppe significantly re-shaped the genetic character of almost all of South Asia. So whatever happened in South Asia was huge, and potentially bigger than what happened in Rome.

Don't disagree at all there, but that may have to do with the fact that they in turn were being forced out by other peoples. The Visigoths lose southern Gaul to the Franks, who themselves dislodge other Barbarian invaders like the Lombards and Avars.

That aside, my point is more of a how, not a what


So you admit that invasions and conquests at this time were common place, and yet you fail to consider the possibility that the IVC was invaded. Why specifically?

Simply put, an invasion was possible, but not probable.

Also, this is quite a change from what you mentioned earlier, where you tried to paint a picture of textual evidence of Indo-European invasions in the region from 2000 bc. I definitely think it was a possibility, and I don't doubt that their arrival came with a period of violence.

However, that does not mean that the evidence on balance points to an invasion. You still did not address my point that Steppe ancestry correlates strongly with Iranian Neolithic ancestry. That points to continuity with the local elites, and arrivals moving into roles instead of throwing them out entirely.
.

aniasi said...

Really bad reasoning here, because obviously once the Indo-Aryans seized power in South Asia, there was little incentive for many to remain serving as warriors, when they could choose safer and more lucrative and prestigious positions.

This makes no sense for several reasons:

1) It entails a bunch of invading warriors deciding that they are going to spend the rest of their lives engaged with liturgical memorization, ritual minutae, and ascetic practices, while leaving those with more native admixture to become the new warrior class. This does not look lucrative, and certainly not prestigious to a hyper-masculine group of warriors without women.


2) Kings are still taken from the warrior class in Rg-vedic society, though they are subordinated to the priests. This means that a King with the most steppe heavy ancestry decides that instead of elevating his own position, he will become a priest and elevate that position instead. This goes against the norms of Indo-European societies, that tend to Sacral Kingship.

3) Both of these support my point. They indicate the strong influence of a local culture that places priests at the top, and delegates warriors and rulers to the second tier. That very unique structure, and its persistence points to continuity. The Indo-Aryans adopt and impose themselves on an existing heirarchy, instead of invading and throwing the whole thing out.


I would like to say that even though I do disagree on this point, I do appreciate the civil discussion.

Open Genomes said...

@Baladeep

Thank you very much for looking at your 23andMe matches for these Y haplogroups. To make it clear, the "bias" that I mentioned of course does not extend to everyone, it's just more "systemic" in the sense of academic and personal sample collection and focus, and also certainly a bias against non-R1a-M417 Brahmins, which I know causes people to hide their results and not want to investigate further.

I think it would be very worthwhile to contact your C-P92 match. This is really an excellent opportunity.

Can you politely / discreetly ask if they are Brahmins?
Sengupta (2006) has a list of Y haplotypes for specific caste and tribal groups, including Brahmins:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1380230/table/TB3/
As we can see, the Iyers and Iyengars are mostly L1-M76, G2a-Z30522, R2a, J2a, and J2a-M158 as well as a few H1a's (here, "H"), with only one C1b1-K281 (here, "C5"), and no H1b's (here, "F").

First of all, I know that the G-L30s are in fact Bhargava Iyer/Iyengar Brahmins in G-Z30522, which seems to be exclusive to the two Bhargava gotras, but the others may or may not be Brahmins, in spite of the endogamy.

We do have sequences for what 23andMe describes as F-M89*, which in fact is an early form of H, H1b-Z5857. This is yet another extremely early unique Upper Paleolithic South Asian lineage:
https://yfull.com/tree/H-Z5857/

The evidence from Sengupta (2006) indicates that at least South Indian Brahmins come from a variety of origins, although almost all of these would appear to be Neolithic and later, not of Upper Paleolithic South Asian origin.

Samuel Andrews said...

R1a Z93 in Sycthians and Indians, R1a M417* in Corded Ware. May I say more. This debate is over guys.

Seinundzeit said...

@Vara

"This model isn't as perfect as you think."

For what it's worth, he isn't really putting forth a fully-fleshed out model.

Rather, he exhaustively sketches out the parameters, and gives a rather sensible account of the impossibilities at play (and also examines how those impossibilities affect the possibilities).

@Nizam Uddin

"... to clarify, modern science supports the idea of scientific racism not only genetic and anthropological blogs."

No disrespect intended dude, but WTF are you talking about? Your statement above is so inaccurate/false, on so many levels.

Modern science does not validate "scientific racism".

FYI, since you seem somewhat unfamiliar with this subject, the people who support "scientific racism" are often just lonely, bitter, insecure computer nerds who can't openly vent in polite company; so naturally, they're drawn to genetics/biological anthropology blogs like flies to feces.

Regardless, just because you see a few Far-Right loonies leaving comments at blogs like GNXP or Eurogenes, that doesn't mean that the scientific data itself justifies conceptual racism.

If I had the inclination/interest, I could absolutely eviscerate "scientific racism", with arguments drawn purely from objective scientific fact/studies, and using only a couple of paragraphs.

But I have not the inclination/interest, nor are these topics the primary focus here at Eurogenes, so we'll leave it at that.

aniasi,

We only have Kshatriya samples from Uttar Pradesh; we don't know if all Indian Kshatriya have less Steppe_EMBA-ancestry relative to their local Brahmins.

Anyway, wouldn't it make more sense for "warriors" to have taken many more local wives/concubines relative to priests, and thus for their descendants to have more diluted Steppe_EMBA-ancestry relative to Brahmins?

Davidski said...

@aniasi

You still did not address my point that Steppe ancestry correlates strongly with Iranian Neolithic ancestry. That points to continuity with the local elites, and arrivals moving into roles instead of throwing them out entirely.

I'm not sure exactly what you mean by saying that Steppe ancestry correlates strongly with Iranian Neolithic ancestry? What is it supposed to correlate with in South Asia, considering that the Steppe peoples moved into a South Asia in which mixed Iran_N/ASI populations flourished for millennia and were probably still found at very high population densities even as the IVC was collapsing?

Actually, there are a couple of patterns worth noting in this regard. In Pakistan around the Indus Valley, in pops like the Balochi and Brahui, and in Dravidian speakers in India, the ratio of Iran_N ancestry to Steppe ancestry is often very high. On the other hand, in Indo-European speakers in North India, including in Brahmins, and in Pakistan some distance away from the Indus Valley, the ratio of Steppe ancestry to Iran_N ancestry is clearly higher.

I guess what this suggests is that the signal of pre-Indo-European South Asia survived better in more heavily populated former Harappan regions, and in areas relatively far from the main Indo-Aryan and Indo-Iranian settlement points.

But I don't know what this says about the nature of the Indo-European expansion into South Asia, apart from perhaps the obvious that in more densely populated zones, the locals were likely to be incorporated more often into the new Indo-European social and political structures, rather than by and large marginalized or killed, probably because such antagonistic practices were less practical than in areas where the locals were a minority?

nizam uddin said...

@Seinundzeit
This is a pointless discussion.
Personally, if anybody has a problem then simply ban me.
You respond only in the sake to criticize me. I mentioned scientific racism as an application for medical researchers to implement surveys to acknowledge genetic disposition; one example is common diseases such as, heart disease, and diabetes, which is suppossedly higher among non-Europeans. I don't believe this in true in the first place, but is still present in academic textbooks and that is ridiculous. If you don't find me adequate for any discussion on the blog then Davidski can delete all of my posts, given that, it's not my problem. As I'm fairly honest on this blog and that might be a curse. I don't post so I can glout how much I might know, but instead, to learn more about population genetics and anthropology.

epoch2013 said...

@nizzam uddin

"You respond only in the sake to criticize me."

If I were to guess he responds because you're uttering poppycock.

nizam uddin said...

there is no respect from people whom pretend to hide identity with computer screens. Attempting to know what they barely can grasp. Only to reevaluate themselves is to be negative towards another. Epoch, your comment was to vague culture is broad topic. Religion has also been controversial subject. From your words, you attempt was to label me a hinducentrist. As I told you no. I believe in the steppe theory to some degree but consider a bit exaggerated due to nationalism. If you don't think then tough luck for you. Furthermore, I'm not the only one who post halfwitted comments on this public blog.

Seinundzeit said...

@Nizam Uddin

"If you don't find me adequate for any discussion on the blog then Davidski can delete all of my posts, given that, it's not my problem. As I'm fairly honest on this blog and that might be a curse. I don't post so I can glout how much I might know, but instead, to learn more about population genetics and anthropology."

Lol dude, I don't want you to be banned, nor do I want your posts deleted. Lighten up.

Anyway, I applaud your interest, and your desire to explore something new (in terms of intellectual curiosity).

My problem was with the idea that contemporary science supports "scientific racism", but apparently you have a very unique definition of that phrase, so we probably have totally different things in mind when we say "scientific racism".

Anyway, forget about it.

Ric Hern said...

@ aniasi

How precisely does it go against Indo-European society ? We see the Senate in Rome and Druids among the Celts and Judges of today etc. whom basically decided who should be king and when to replace him....I can not imagine a nicer job for a retiree...especially a battle bruised King.

Shaikorth said...

@Matt

Gond having low fst to Pakistani populations may be a function of gene flow, not just drift. They don't just have an Austroasiatic connection that other Dravidian tribals lack, but the rest of their ancestry resembles non-tribals (which would bring down the Gond-Pathan fst). This is visible in Falush et al. 2016 fig 4. I agree that drift plays an effect in high intra-South Asian fst, but the genetic structure of Gond appears to differ considerably from other Dravidian tribes, more than Kalash do from Pathans.

Jaydeep said...

Sein,

Just because someone is erudite it does not mean that whatever he or she is arguing for is necessarily correct, especially when one is trying to fit in the data to support a particular narrative rather than allowing the data to speak for itself.

Let us have some critical evaluation of Witzel's arguments in the paper :-

1. Witzel argues on the basis of no mention of Iron in Rigveda that it must be a pre-Iron age text. He argues for 1000 BC as the start of Iron age in South Asia and hence argues that Rigveda is older to it. But now we know that the Iron Age in South Asia starts latest by 1800 BC. So what does this show regarding Rigveda's antiquity ?

2. Witzel also argues that Rigveda's horizon is limited to Punjab, when infact Rigvedic horizon encompasses a wide region from Uttar Pradesh to Afghanistan, and its core region is also not Punjab but Haryana and Western UP, the traditional homeland of the Purus, the main ethnic group that represented the Indo-Aryans and are the most important group in Rigveda.

3. He talks to Rigveda mentioning ruins, as proof of it being a post Indus text. This is a rather useless argument. A ruin can exist at any period of time. We do not know what ruins the Rigveda is talking about. How do we know it is talking specifically about the post-Indus ruins ? This is just a hyperactive imagination of Witzel at play here.

4. Witzel argues that since there were many invasions/migration into the Indian subcontinent during the historic period, why should one be so totally opposed to any sort of migration in the early period representing Indo-Aryans ? Well, to put it simply, all historic era migrations/invasions are recorded, none of them are hypothesized like the so-called AMT. Nor did any of these historic migrations leave any impact that can even marginally compare to the overwhelming impact an Indo-Aryan migration is suppossed to have played across the length and breadth of the Indian subcontinent. Yet there is no tangible evidence for such a watershed migration event in South Asian history and infact tradtional Indian history as recorded in the Puranas does not talk of any such migration but an emigration of a large tribe from Punjab into Gandhara and from there into Central Asia.

(conti...)

Jaydeep said...

5. Witzel argues that the Rigvedic spiritual and material culture as well as their language links it to the Northwest of the subcontinent. However, this is not a proof of Rigvedic people coming into South Asia from the Northwest. One should be well aware of the fact, that in the standard Indo-European expansion theory, the last group that was left in the PIE homeland were the Indo-Iranians and they are said to be the last to leave (the Slavs supposedly came back much later). In an Indian homeland scenario, it is only from the Northwest that you can move out, is it not ? You cannot go from the North & NorthEast due to Himalayas. So if all the other IE groups are moving out through the northwest and the Indo-Aryans are the last ones left to the east, is it not natural for them to show affinities with these groups to the Northwest ? How does this prove AMT ?

Here I may also add that the Indus civilization's affinities are also all towards its NorthWest and very little to its East. Does this also prove that the Indus civilization is intrusive to South Asia ?

6. He point to the Indo-Iranian intrusion in the Middle East during the LBA and suggests that the same phenomenon may be ASSUMED for the region of Punjab. But why assume ? Why not prove it ?

7. Witzel then argues that there is evidence that an unknown 'ghost' language of Indo-Gangetic region shows subtratic influence in Rigvedic Sanskrit and modern Hindi. This is all extreme speculation. Without any proof of any such language ever existing, just cook up whatever crap that comes into your mind. The reality is, the earliest attested languages in North India and Pakistan are all Indo-European. There is no proof of any other language family existing in any earlier period. The reason for the invention of this ghost language is to support the idea of an Indo-Aryan migration.

To supprt his argument, Witzel takes the example of Nihali, a Central Indian language. It is well known that early Indo-Aryans are recorded in Haryana and Western UP and their expansion into Central India was much later. So what does the example of Nihali prove ? Nothing at all. Why can't Witzel give an example of a Nihali-like language from Haryana or UP or Punjab ?

8. To show that the Rigvedic Indo-Aryans indeed preserved memories of a non-Indian homeland, Witzel gives example of some river names and tribe names found in South Asia which also existed elsewhere such the river Kuban (Vedic Kubha) or the tribe Sindoi. But how do these examples prove the direction of the migration ? Why can this be not proof for an Out of India migration ?

(conti...)

Matt said...

@ Shaikorth, thanks for the comment, good to have your thoughts, but to be honest, I don't think that is that important in a strong sense.

I really do think the difference (for Fst to ANI populations between Paniya and Gond) is almost all to do with recent, high degrees of populations specific drifts (or effects confined to a limit number of subpopulations).

Partly we could test with stats of the form:

D(Mbuti,Pathan)(Gond,Paniyas) and D(Onge,Pathan)(Gond,Paniyas)
and
D(Kalash,Pathan)(Gond,Paniyas)

That would lose some of any recent drift that might be connecting Pakistani populations and Gond (just as D stats often lose fine scale differentiation), but would provide a good check if there is any deep structured differences in how these two populations relate to Pakistani populations, theoretically with Gond as a representative of high ASI, low ANI, low drift and Paniyas as high ASI, low ANI, high drift.

...

Re; citing Falush 2016, doesn't the text seems to say the opposite of what you suggest?:

"Once population specific drift is accounted for, there is good evidence for four ancestry components, as can be seen by eye inspecting the palettes. For example, the four ATB populations each have higher proportions of all four ATB population palettes than any of the non-ATB populations. It is also possible to make deductions about the relationships between the ancestral populations. The ASI and ANI populations are relatively closely related, with high sharing of palettes between them, while ATB and AAA are more distantly related to each other and to ANI and ASI."

Finally, we can ask whether the variation in ancestry proportions in the ADMIXTURE plots are likely to be indicative of a history of admixture between populations. For example, ADMIXTURE assigns nearly 100% ASI ancestry to Paniya (PNY), while KDR and IRL are inferred to be admixed (Figure 4A) but this result is suspicious because of the tendency documented above for ADMIXTURE to assign pure ancestry components to highly drifted populations like PNY. Once population specific drift is accounted for, the palette for PNY actually has slightly less overall ASI ancestry than KDR and IRL (Figure 4C).

ASI = Ancestral South India = Paniyas, Kadar, and Irula
AAA = Ancestral Austro-Asiatic = Birhor, Korwa, Ho, Santal, Gond
ANI = Ancestral North India = Kshatriya, MRT, PLN, IYR (Iyer?), GBR (Gujarati Brahmin?), WBR

If we take Falush literally, using Gond as a reference for the Fst a drift free version of Paniyas would show from Pathan would actually *overestimate* Fst.

Caveat - Gond is the least AAA-like AAA population in Falush's panel!

Looking at the their palettes for Paniyas, Gond and Gujarati Brahmin together, I actually don't see that the Gond palette is obviously more or less like the Gujarati Brahmin palette than the Paniyas palette.

I hope this doesn't come across rude, but you might be overscrutinising the medium-dark green ASI modal components in the palette? Yes, Paniyas has an enrichment of those, but that's ofset that Gond has an equal enrichment of the dark grey, lilac, purple, light green components that are modal in Austro-Asiatic populations. The level of the ANI modal components/elements in both populations looks roughly equal.

Here's a set of images / screencaps which conclude where I've put the Paniyas, Gond and Gujarati Brahmin palettes side by side: http://imgur.com/a/SSWbS

(Not totally sure why the Gond group was put in the AAA group, as linguistically doesn't seem to make a lot of sense, but must be due to the component patterns).

Shaikorth said...

@Matt

High caste populations like Gujarati Brahmins or even South Indian Brahmins weren't in my mind when I talked about Gond similarities to caste populations. Lower castes (probably best represented here by Pallan) would be closer. There doesn't have to be much gene flow to decrease fst. "ANI" and "ASI" are indeed the two most closely related groups based on averages, and as you've noted Gond is an outlier in the AAA group, more similar to "ANI" than others.

To further investigate the possibility of recent Indo-Aryan geneflow to Gond specifically I compared them to other southern populations using Broushaki's haplotype modeling, which is also Chromopainter-based, with Indian donor populations removed to prevent them from donating to each other and confusing the picture. In this case the usual donors to Southern and Central Indian populations are Sindhi, Pathan, Ust-Ishim and for some tribal populations (Austroasiatics and the most ASI-rich Dravidians) Cambodian.

Malayan are modeled as 25% Pathan. They're "generic" Southwest Indian tribals in that they don't show any excessive accumulation of ASI or ANI compared to their reference group or extra Austroasiatic etc.

Gond, while they have higher Cambodian element than the Malayan consistent with Austroasiatic connection, have 36% Pathan! This even exceeds the 32% of Tamil Nadu lower caste sample. Then we can look at the relative pattern of their Pathan and Sindhi contributions, Sindhi usually exceeds Pathan in southern Dravidians but the Gond are again an exception. The Nihali, a possible language isolate, also show the Malayan pattern. I'd say the distinctiveness of Gond relative to other tribals in this regard is real.

Matt said...

@Shaikorth, Hmmm. I think we may be at slight cross purposes, in that I guess I'm really most interested in the relationship between Fst between South Asian populations and the ANI-ASI ratio. Rather than the more fine scale differences in ancestry of South Asian tribal groups per se.

Like, if you have a donation of 25% Pathan then assuming 80% ANI in Pathan, that's about 20% ANI, or if you have a donation of 36% Pathan then 28% ANI.

I've not got any reason to see that Gond *couldn't* be 28% ANI, rather than the 22% from qpGraph, vs 20% ANI in Malayan, and that there couldn't be some additional geneflow into Gond, which could be well as you say. (Though I'm not as sold on using haplotype donation as a superior method to judge that!).

But the main point I'm interested in is that when we go from Brahmin_UP at say 36% ASI, 64% ANI to Gond at 72-78% ASI, 28-22% ANI, with a 36-42% difference, and the Fst is only around 0.007, then the huge comparative Fsts we sometimes see from groups like Paniyas to Brahmin_UP at 0.032, are obviously not due to a small increment of increases in ASI between these populations, whether that's a 2% difference, or an 8% difference.

They have to be much more dominated by group specific drift

Shaikorth said...

I'm not suggesting Pathans are the ANI source for the tribals or even that Gond have more or less ANI than other tribals, but that their ANI is different due to different sources (ultimately this boils down to more steppe and less Iran_N) and that this affects the fst. Even though most of Paniya's fst differentiation is drift and I completely agree that intra-South Asian fst's based on groups like Paniya and Kalash will result in huge exaggarations, Gond's fst to Pathans should be lower than even "un-drifted" Paniya's - not just because of extra ANI but because it has more directly Pathan-related ancestry.

Matt said...

Well you certainly could be right and the "ANI" in these populations could be as you say!

Still, assuming a "driftless" situation, I'm not totally sure how much of that difference could drive Fst differences though.

For a quick thought experiment, say both Gond and driftlessPaniya had the same "ANI" proportion, at say 25%.

Then you have, as an extreme, the Gond have a Steppe:Iran_N ratio like Pathan/Kalash/Brahmin seem to, at 56:44, and the driftlessPaniya has a ratio like Baloch seems to, at 35:65. Then you have Gond having Steppe 14%, IranN 11% and driftlessPaniya at Steppe 9%, IranN 16%. And I think that seems an extreme scenario based on the models.

How much extra Fst differentiation would that drive, either from each other, or away from Pathan/Kalash? Seems like I would expect less than exists between Baloch and Pathan, from each other, which seems measured at around 0.003... But this is all very back of envelope stuff.

Seinundzeit said...

Shaikorth & Matt,

For what it's worth, I think that haplotype-based results in relation to Central/South Asia show a couple of interesting patterns (this is somewhat tangential to the Fst discussion, but in a way still related to it).

In terms of West Eurasian ancestry, we see this in South Central Asia (from the WC1 paper):

Pathan

78.45% Iran_N (WC1)

12.85% WHG (Loschbour)

6.11% ENA (Han)

(rest is self-copy)


Tajik_Pomiri

66.68% Iran_N (WC1)

24.16% WHG (Loschbour)

6.7% ENA (Han)

(rest is self-copy)


And in South India:

Mala

47.9% Iran_N (WC1)

47.87% Ust-Ishim

(rest is self-copy)


Considering the lack of a high-quality Steppe_EMBA sample in the mix, very sensible results for the South Central Asians.

I mean, we are just seeing a very stylized/extremely simplistic construal of the genetic reality that exists for these populations, which is that they are primarily a mixture between ancient European-related (of the very eastern/northern variety) and ancient Near Eastern-related (of the very eastern/northern variety) ancestries, with some minor extra ENA admixture (which other methods put at somewhere between 5%-25%; although most methods put these Pashtun samples at 10% ASI, and these Pamiri samples at 5% ASI + 5% East Asian/Siberian).

And as expected, the Mala are only Iran_N admixed (around 50%), with no ancient European affinity.

The differentiation seen in the West Eurasian streams of ancestry, when going from Tajikistan all the way down to Tamil Nadu, fully agrees with patterns in ADMIXTURE (and in PCA-based modelling).

Regardless, what I find even more interesting is the Han/Ust-Ishim dichotomy between South Central Asians and South Asians.

To put it somewhat facetiously, once one crosses the Indus River (the ancient border of India), one starts seeing Ust-Ishim as the primary ASI representative!

But only somewhat, because one actually does see that happen.

For example, just compare Sindhi samples to the Pashtun samples above:

Sindhi

82.4% Iran_N (WC1)

14.06% Ust-Ishim

1.2% WHG (Loschbour)

(rest is self-copy)


Compared to high caste Gujaratis from nearby:

Gujarati_A

72% Iran_N (WC1)

18.95% Ust-Ishim

8.49% WHG (Loschbour)

(rest is self-copy)


I wouldn't have given this pattern much thought, if it only occurred with this sort of data (although, as has been noted previously, David Reich and company are transitioning from f-statistics to haplotype-based analyses). But, I've replicated this pattern multiple times, using PCA data.

To see the very same pattern in both haplotype and PCA-based analyses is quite intriguing.

Anyway, I wonder if there's a way to test this in qpGraph?

Like by having a South Indian population as a mix of something Iran_N-related (with additional ANE, like Iran_Hotu), Onge/Ami-related, and Ust-Ishim-related?

Shaikorth said...

@Sein, yeah, I wrote a couple of posts on that modeling and South Asians last year. Loschbour is the sole representative of Euro-HG's there so it could mean both steppe ancestry in general (EHG in excess of Iran-N) and increasedly WHG-like (Sintashta/Andronovo as opposed to Yamnaya) steppe in particular. But then there's the question of why no CHG in South-Central Asia when it appears everywhere in Northern Europe and Iran_N does not. A high coverage Steppe_EMBA genome would be a goldmine for this kind of analysis.

Ust-Ishim in those haplotype fits can and likely does represent more than one thing. The alternatives are at least:
-Actual unique Crown Eurasian branches
-ASI or a separate old ENA branch (Onge and Papuan are basically Ust-Ishim and self-copy, and the East Asian proxy Han is all Ust-Ishim when not allowed to self-donate and self-copy)
-ANE (the Native American results make this plain)

On your PCA models, did Ust-Ishim appear when you allowed separate ASI donors?

Seinundzeit said...

Shaikorth,

"A high coverage Steppe_EMBA genome would be a goldmine for this kind of analysis."

I completely agree.

"On your PCA models, did Ust-Ishim appear when you allowed separate ASI donors?"

Looking back, I probably should have tried that. That being said, you've given me an idea.

Assuming that Ust-Ishim is proxying for a separate old ENA branch unique to South Asia (and to a far lesser extent, Central and West Asia), I've just created an ASI simulation for the Global_10 PCA; it's 40% Ust-Ishim, 40% Ami, and 20% Andamanese_Jarawa.

Just to see what happens:

South India

Paniya

59.90% ASI

32.85% Iran_Meso/Neo + 7.25% ANE

distance=1.7428

Pulliyar

51.0% ASI

40.0% Iran_Meso/Neo + 8.9% ANE

distance=1.5899

South Central Asia

Burusho

49.10% Iran_Meso/Neo + 7.85% ANE
21.25% Steppe_EMBA

12.75% ASI
9.05% East-Asian_ENA

distance=0.4508

Kalash

51.80% Iran_Meso/Neo + 1.05% ANE
37.75% Steppe_EMBA

9.40% ASI

distance=0.4306

Karlani Pashtun, Central Highlands

26.70% Iran_Chal + 25.30% Iran_Meso/Neo
38.45% Steppe_EMBA

7.65% ASI
1.90% East-Asian_ENA

distance=0.246

Shugni people

52.05% Steppe_EMBA
32.60% Iran_Chal + 7.05% Iran_Meso/Neo

5.60% ASI
2.70% East-Asian_ENA

distance=0.1704

Yaghnobi people

48.7% Iran_Chal
43.2% Steppe_EMBA + 3.0% Barcin_Neolithic

4.7% East-Asian_ENA
0.4% ASI

distance=0.1385

The patterns here are exceedingly interesting.

The results for the South Indian populations are sensible, and fairly typical.

On the other hand, for South Central Asians, the high ANE seen with the Burusho is kinda exciting, considering that the Burusho are a linguistic isolate, and have tons of Y-DNA haplogroup R2 (coupled with the lowest levels of R1a seen in this region).

Additionally, very interesting to see the Steppe_EMBA + Barcin_N combo in Yaghnobi speakers.

Anyway, although that worked out pretty great, I would still like to see this tested in qpGraph, with a solitary South Indian test pop.

Jaydeep,

I appreciate the time you took to look into the paper; I will eventually produce a detailed response (I don't want you to think that I'm ignoring you).

Reza said...

A rather basic question, but what is the presumed incursion date for Austroasiatic speakers into the subcontinent?

And when comparing more eastern groups like Bengalis or perhaps even some Biharis, is their ethnogenesis considered similar to or different to southern Indian populations?

For example, is the noticeable east Asian component in Bengalis / Bangladeshis a relatively modern admixture (<1500 years) or something older as Indo-Aryan languages spread east and interacted with AA speakers? As it does also appear, albeit in much smaller proportion, in West Bengali Brahmins.

Sein, would you be so kind as to run Bengalis in your models?

Jaydeep said...

Sein,

Take your time. I understand that the Witzel text is quite long and its not easy to read and comment on it after. It is time-consuming.

Nevertheless, forgive me, as I am going to make your task a little harder. Here's more of my criticism:-

9. Witzel says the retention and adaptation by the Iranians of earlier pre-Rigvedic river names points to an earlier IA settlement in Afghanistan

This is again a strange argument. First of all, how did he come to the conclusion that the river names are of pre-Rigvedic origin ? Based on what ? He does not offer any explanation. And the retention of river names by Iranians in Afghanistan & Central Asia, could just as easily be explained by Iranian groups moving out from their original homeland in North India where they lived around the afore-mentioned rivers.

Let us also bear in mind, that right from the Early Harappan period, the Indus civilization starts showing strong influence on both the civilizations of Central Asia and that of Eastern Iran (Helmand & Hari Rud), we have evidence of bronze age movement of Indian mtDNA towards those regions and we also see the unmistakable evidence of South Asian origin Zebu cattle making its appearance in both these regions.

10. Witzel talks of Rigvedic Rsis and their clans and other tribes as being mentioned to have crossed the river Indus. But he does not give any reference, nor does he show which direction that crossing is said to have taken place. And he also denies that there is report of westward movement in the Rigveda. He completely obfuscates the fact that the Battle of Ten Kings, mentioned in the Early Books of Rigveda, takes place when a Bharata king comes from the East (from Haryana) and crosses the river Parusni (modern Ravi) to invade the kingdom of Anus (who are Iranians) in Punjab.

Ofcourse the Rigveda does not talk of any emigration out of India into Central Asia because its main patron group are the Purus and their subtribe the Bharatas, whose core region remains Haryana & Western UP. But as we move from the Early Rigvedic family books (2-7) and into the Late Books, we see that the Rigvedic geographic horizon in the West includes regions in Afghanistan, where previously it was restricted to Punjab in the West.

11. then Witzel says The early YV Samhitås (KS 26.2, MS 4.7.9), however, continue to report such movements into the subcontinent. They state that the Kurus move eastwards or southwards victoriously, and TB 1.8.4.1 adds information about raiding expeditions of the Kuru-Pañcålas into the east

Does this look like a good credible argument from a big scholar at Harvard ? He clearly says earlier that the Kuru Panchalas in the later Samhita period expanded into the Indo-Gangetic regions. But this expansion was from their homeland in Haryana & Western UP. Where does it show that they lived further west in any period of time ?

Jaydeep said...

12. Witzel then mistranslates a passage from a Brahmana text about the migration of Ayu and Amavasu tribes. While Ayu is mentioned as going east, Amavasu is not said to be staying but moving towards the West.

Read here for the whole controversy, https://en.wikipedia.org/wiki/Baudhayana_Shrauta_Sutra

See also the amusing longwinded footnote of Witzel on it in his paper. He concedes that the text can also be translated to mean that Amavasu went westward. Moreover, the father of Ayu and Amavasu is said to have wandered in Kurukshetra, which is in Haryana.

Witzel also lets us know what the Vedic people themselves thought of where their homeland was -

9 The center of the world was, even according to the later parts of the RV (3.53), on the Sarasvatī in Haryana. This attitude continued to be the norm in the Bråhmaa period, and it is vaguely remembered in the Påli canon; it clearly referred to even in the Manu-Sm ti (ch. 2). The northwest, denigrated by the AV (5.22, PS 12.1-2), and depicted in Nirukta 2.2, cf. 3.18 and in Patañjali's Mahåbhå ya (ed. Kielhorn, I p. 9) as occupied by Avestan speakers of the Kamboja land in S.E. Afghanistan (Witzel 1980: 92), is regarded as non-årya.

Here it maybe noted that Rigvedic book 3 is not a late book but one of the earliest, perhaps the 2nd earliest after book 6.

13. Witzel then perfunctorily argues that there are some non-Sanskritic words and names in Rigveda. However, most of these names such as Agastya, Kanva, Kavasa, Kasyapa, Balbutha, Brbu etc. are clearly of IE origins. Nevertheless, Witzel never bothers to explain how all these names are of non-Sanskritic origin. You have to go by his word.

14. He also argues Local influence is indeed what the non-IE part of RV vocabulary suggests, by Kuiper's count some 380 words or about 3.8% of the vocabulary of the RV (Kuiper 1991, 1995: 261). Such local substrate words can easily be identified because of their isolation within the IE derived IA vocabulary, i.e. they always do not have Iranian, Slavic, etc. counterparts.

Here it is argued that a merely 3.8 % of the vocabulary of Rigveda is of non-IE origin. This 3.8 % contains many many words whose non-IE origin is highly questionable. So the real no. is likely to be even lower than this. So a nomadic group enters a region which had only recently a very advanced and vast urban civilization and they manage to live there in such a splendid isolation that next to nothing of the vocabulary of these civilization's language manages to make a impact on these nomads literature. At the same time, they managed to put such a spell on the descendents of this great civilization, that they accepted the religious text of these nomadic invaders in their own language and made all efforts to preserve the sanctity of even its pronunciation upto this day. And all the while there is not a single word of the suppossed non-IE language of this civilization that we know of. Do you see how incredible this sounds ? And yet the likes of Witzel want us to accept this without even taking the trouble to prove such a migration.

(conti...)

Ebizur said...

Reza wrote,

"A rather basic question, but what is the presumed incursion date for Austroasiatic speakers into the subcontinent?"

The subclades of Y-DNA haplogroup O-M95 that are found in the Subcontinent (not only among present-day Austroasiatic speakers, but also among many other non-caste populations whose members currently speak dialects of the regionally prevailing Indic language or their own Dravidian language, such as the Gond, Konda Dora, etc., many of which are socioeconomically among the poorest peoples in India) suggest a terminus post quem of approximately 5,000 years before present for the movement of Austroasiatic speakers into India. Prior to that, they share common patrilineal ancestors with people who currently live in Cambodia, Vietnam, Malaysia, Indonesia, etc. There are even some Yi and Dai people in Southwest China who are probably patrilineally related to those people within the last 6,000 years.

O-B426 Y-chromosomes also have been found among Bengalis in Bangladesh (e.g. HG04173 and HG04140 on YFull), so the Bengali population clearly does contain some descendants of the same migration wave, but I cannot say precisely when these individuals' forefathers may have been assimilated into the Bengali population.

Spencer Wells et al. (2001) also found O-M95 Y-DNA in 1/46 Sourashtrans, 1/84 Kallars, and 1/129 Yadavas from peninsular India. I suppose most of these scattered O-M95 individuals in South Asia are also probably ultimately descended from the same Austroasiatic-speaking ancestors who have produced the present-day Munda group of peoples, but one cannot be sure without higher-resolution testing of Y-SNPs.

Matt said...

@ Sein, those donation patterns are a puzzle, and I really have no answer for it!

@ Sein and Shaikorth, in case you are thought it worth looking at (on the off chance) here are a couple more of those phylogenetic Fst networks I was running, but in this case, using all World modern populations:

World Fst Network Analysis - Moderns : http://imgur.com/a/gTuvr
World Fst Network Analysis - Moderns (alternate dataset): http://imgur.com/a/CF204

Some look strange, particularly in the second dataset, some patterns as expected though and which make sense.

Seinundzeit said...

Reza,

That's an interesting question.

I think that if Muslim Bengalis weren't 5%-20% East Asian-admixed (depending on the individual), they'd be rather similar to some South Indian populations, like Primalai Kallar or Vellalar, albeit with a much higher Steppe_EMBA-to-Iran_N ratio.

If I recall correctly, I was sent PCA coordinates for a individual of Bengali heritage; I'll see if I can find those.

Jaydeep,

Honestly, I think it's great that you've looked at the whole piece, in such an exhaustive manner.

Before posting my thoughts, I'll reread the whole piece, and brush up on some general scholarship concerning the Vedas (quite frankly, I don't have much familiarity with the issues of dating, and zero knowledge concerning the Sanskrit translations), so that my response proves to be worthy of all the effort you've put into your critique.

Matt,

Those phylogenetic Fst networks are very interesting; thanks for posting them.

If possible, do you think you might be able to produce one involving ancient populations, Africans, West Asians, Central Asians, South Asians, and East Asians (from Siberia to Southeastern Asia)?

Of course, only if you feel it worth your while. If not, it's absolutely fine.

khana said...

I don't understand why so many are vehement in their opposition that a Northern or Western people could move in large numbers via the NW passes, conquer, and impose their culture on the plains of South Asia. If history repeats itself, then is it so far-fetched to believe that a people before the Achaemenids, Macedonians, Greco-Bactrians, Indo-Scythians, Indo-Parthians, Kushans, Hepthalites, Ghaznavids, Ghurids, Mamluks, and Mughals did exactly what they did -- invaded from the North or West -- ruled, brought their language/culture, and intermarried/assimilated!?

It's called the Hindu Kush ("Hindu-Killer") mountains for a reason folks.

How many native Indian empires expanded beyond parts of Afghanistan? How many foreign invaders have taken parts or the whole of India and eventually assimilated? I mean, what is really more likely, a mass movement of people from a confluence of river valleys into mountains or vice versa?

The only thing OIT should refer to is the Tamil Chola Dynasty and their influence on Southeast Asia, the number zero, and parathas.

khana said...

@Sein: if it's the coordinates I sent you, they are Reza's.

Matt said...

@Sein, no problem. I'm not sure how easy they will be for you to get look at though, as there's a limit to the resolution I can export from the software:

World Fst Network Analysis - Moderns : http://imgur.com/a/kCbBR
World Fst Network Analysis - Moderns (alternate dataset): http://imgur.com/a/rRH31

I could give a quick explanation of the way I've used the software and method if you'd prefer, then you can just zoom in and out of the plots within the software.

These are essentially kind of like neighbour joining trees in that they are main capturing distances samples from closest neighbours and the whole structure and positioning is influenced by that.

batman said...

Ezibur:

"I suppose most of these scattered O-M95 individuals in South Asia are also probably ultimately descended from the same Austroasiatic-speaking ancestors who have produced the present-day Munda group of peoples, but one cannot be sure without higher-resolution testing of Y-SNPs."

If y-dna O and N share parternal ancrstry from K/NO they should both have a pale-arctic ancestor - in common with the 'caucasoid' descendants of makrogroup F.

Comparing the spread of y-dna N and O we fond them bpth to occur and sprrad - just like hgs G/H/I/J and K/RST to the west - AFTER the paleolithic ragnarok called Younger Dryas.

Like hg N is mixing with a northern population of survivors from tropical Asia - today known as Chineese or Sino-Tibetans. Meanwhile, presumeably, Y-dna O became a part of a southern population, who populated the shores of Indo-China and the east-asian ('austro-asian') archipelagos.

At that time these areas were already fully populated, as the paleolithic populations of the tropical part of Eurasia - just like their australian, african and meso-american cousines (y-dna ABC) - could enjoy life, while the populations north of the 45th paralell was decimated. Except the lucky few - from which the extant descendants of macrogroup C/F all seems to descend.

The wide distribution of y-dna O in SE Asia seems to indicate that they were marrying into an already existing (five-cast) dynasty, where the 'royal seed' will determine the y-dna of all men belonging to the tribe -
as it moves down to every cast during the next give generations - creating a "grand-family" that grew to become large "etnicities" - in balanse with itself and its surroundings. This principle is seem symbolized by a variety of pyramidical symbols expressed in ancient art and architecture. Often at the very centras of the old civilizatons.

Unlile the depopulayed north where y-dna N and the uralic language spread to evry nook and cornet by themselves, the tropical part of East Asia was fully populated, with an "austro-asian" language - well estavlished in a wide range of branches since the Toba eruption, 70.000 yrs ago.

Consequently we have to assume that any caucasian women or men marrying into these societies would have to adapt to the native toungue.

The Dravidian/Hindi
-divide may be the result of the same variation - as the northern, semi-arctic parts of India was duely depopulated during the LGM/UD events - while the proto-indians survived the climate-catastrophe in large numbers, down south.

Seinundzeit said...

Khana,

I have some questions regarding the ancestry of an individual for whom you've sent me coordinates; was hoping to talk to you (via email) about them.

Matt,

"I could give a quick explanation of the way I've used the software and method if you'd prefer, then you can just zoom in and out of the plots within the software..."

Thanks! I'd very much appreciate that.

Ebizur said...

YFull Haplogroup YTree v5.04
O-M268 > O-M176 (Japan, Korea, Han, Vietnam, Manchu, Sibe, Nanai, Udege, Manchurian Evenk, Daur, NE Mongolia, W Mongolia, Micronesia) vs. O-K18 28,500 ybp
O-K18 > O-CTS10887 vs. O-PK4 22,200 ybp
O-CTS10887 > O-PF4341 (Southern Han, Kinh) vs. O-F779 (Han) vs. O-F417 (Han, Japanese, Dai) 15,900 ybp
O-PK4 > O-F838 vs. O-M95 12,400 ybp
O-F417 > O-F417* (Japanese) vs. O-CTS250 (Han, Dai) 11,400 ybp
O-M95 > O-CTS10007 (Southern Han) vs. O-M1310 10,900 ybp
O-M1310 > O-M1283 (Kinh, Dai, Yi, Gond, Konda Dora, Ho, Bengali, Cambodian, Singapore Malay, Java, Borneo) vs. O-F1252 (Ryazan Oblast, Han, Japanese, Hani, Akha, Tujia, Miao, She, Yao, Dai, Buyi, Zhuang, Kinh, Cham, Laos, Thailand, Malaysia, Java, Borneo, Philippines, Taiwan Aborigines) 10,500 ybp
O-F1252 > O-F1252* (Ryazan Oblast) vs. O-F2924 9,900 ybp
O-F2924 > O-CTS5854 vs. O-M111 9,900 ybp
O-CTS5854 > O-F4212 (Chongqing, Dai) vs. O-Z23810 (Beijing Han, Fujian, Guangxi, Dai, Japanese) 9,900 ybp
O-Z23810 > O-CTS10484 (Beijing Han, Dai) vs. O-CTS7399 (Japanese, Fujian, Guangxi, Dai) 9,900 ybp
O-F838 > O-F838* (Han) vs. O-F714 (Han) 9,700 ybp
O-F4212 > O-F4212* (Chongqing) vs. O-F2517 (Dai) 6,700 ybp
O-M1283 > O-M1283* (Kinh) vs. O-Y9322 5,800 ybp
O-Y9322 > O-Y9322* (Dai) vs. O-Y9325 5,800 ybp
O-Y9325 > O-Y9325* (Java) vs. O-Z39485 (Yi, Dai) vs. O-Y9033 (Singapore Malay, Cambodian, Kinh, Ho, Bengali, Konda Dora) 5,800 ybp
O-M111 > O-F2890 (Kinh, Dai, Southern Han) vs. O-F2758 (Miao, Kinh, Dai, Tujia, Southern Han) 5,700 ybp
O-CTS7399 > O-CTS7399* (Japanese) vs. O-Y13994 (Fujian, Guangxi, Dai) 5,600 ybp
O-Y9033 > O-Y9033* (Bengali, Konda Dora) vs. O-M6661 (Kinh) vs. O-YP3930 (Cambodian) 5,400 ybp
O-Z39485 > O-Z39485* (Yi) vs. O-Y26463 (Dai) 5,300 ybp
O-F2758 > O-F2758* (Miao) vs. O-Z24083 (Kinh, Dai, Tujia, Southern Han) 5,000 ybp
O-CTS250 > O-CTS250* (Hubei) vs. O-F2760 (Han, Dai) 4,600 ybp
O-Z24083 > O-Z24083* (Kinh) vs. O-Z24091 (Kinh, Dai, Tujia) vs. O-F923 (Dai, Kinh, Southern Han) 4,400 ybp
O-YP3930 > O-YP3930* (Cambodian) vs. O-Y18422 (?) 4,100 ybp
O-Z24091 > O-Z24091* (Kinh, Dai) vs. O-Y26364 (Kinh, Tujia) 4,000 ybp
O-Y13994 > O-Y13994* (Fujian) vs. O-Z23849 (Dai, Guangxi) 3,500 ybp
O-Z23849 > O-Z23849* (Dai) vs. O-Y31276 (Dai, Guangxi) 2,800 ybp

Karmin et al. 2015
TMRCA O-M1284: 5,965 ybp (includes Singapore Malay, Lebbo, Dusun, Cambodian, Vietnamese, Gond, and Ho according to Karmin's data set; also includes many Kinh Vietnamese, Dai Chinese, Yi Chinese, a person in Java, a Konda Dora, and two Bengalis in Bangladesh according to YFull)
TMRCA of branch containing two Dusun and one Lebbo from Borneo plus one Gond from Madhya Pradesh: 5,747 ybp
TMRCA of branch containing two Dusun from Borneo: 5,588 ybp
TMRCA of O-B426/Y9033, a branch containing two "Vietnamese_south" and one "Ho, Jharkhand" (or "Ho, Bihar" in Figure S33): 5,349 ybp (This branch also includes a Konda Dora, two Bengalis in Bangladesh, and a Cambodian according to YFull.)
TMRCA of O-B419, a branch containing one Lebbo from Borneo and one Gond from Madhya Pradesh: 5,322 ybp
TMRCA of branch containing two "Vietnamese_south": 1,588 ybp

Ebizur said...

O-F1252 seems to have been distributed mainly in some part of China (probably toward the south of the country; however, note that there is a divergent lineage in an individual from Ryazan Oblast, Russia) until the expansion of its well-known subclade O-M111 through Southeast Asia as far as central Indonesia (Java, Borneo) beginning about 5,700 ybp at the earliest. Neither O-M111 nor O-F1252(xM111) has been found in core South Asians (Indo-Aryans, Dravidians, Kolarians), though some members of O-F1252(xM111) eventually might be found among Tibeto-Burmans in India and populations that have been influenced by them (I recall having seen a Y-STR MJN that suggests such a possibility some years ago). Note that O-CTS7399* has been found in a Japanese individual who shares a MRCA with some individuals in southern China about 5,600 ybp; this also points toward a homeland in China rather than India.

It appears that members of O-M1284 increased in number over the course of the fourth millennium BCE and then expanded throughout Southeast Asia and into South Asia beginning about 5,000 years ago at the earliest. The area of the source population's pre-dispersal population growth (i.e. their "homeland") appears to have been somewhere between southern Sichuan and southern Vietnam based on the sparse data I have at present; additional data are needed especially from Cambodia, Laos, and Thailand (some high-resolution SNP data regarding the O-M95 lineages of the Wa in Myanmar, the Khasi in Northeast India, and the inhabitants of the Nicobar Islands also should be helpful), but the present data do seem quite coincident with Paul Sidwell's hypothesis of a spreading of the Proto-Austroasiatic language from the watershed of the Mekong River. The source of the Neolithic population that earlier colonized the Mekong River valley before expanding as the Proto-Austroasiatics is quite murky, however, as the MRCA of Austroasiatic O-M1283/M1284 and O-F1252 appears to have lived about 10,500 years ago. The discovery of some early branches of O-M1283 that have a MRCA with the rest of O-M1283 between 10,500 and 5,800 ybp might help to reveal the route that the Pre-Proto-Austroasiatic ancestors of the Proto-Austroasiatics have taken from an even earlier homeland (probably in some part of what is now southern China) to the Mekong River watershed.

I am not sure what language members of O-F1252 might have spoken originally. Its O-M111 branch is found frequently among some modern Austroasiatic-speaking peoples, such as the Kinh Vietnamese, but they are quite unanimously regarded as a population that has derived from recent large-scale admixture, containing important Dai- and Han-related elements in addition to authentically Austroasiatic ones. Its O-CTS5854 branch has been found among Dai, Han, and Japanese, and the basal O-F1252* has been found in an individual from Ryazan Oblast of Russia. I would like to see some high-resolution SNP data regarding the O-M95 lineages of other Daic- and Hmong-Mien-speaking populations from southern China.

khana said...

@Sein, shoot me an e-mail.

Seinundzeit said...

Khana,

Absolutely bro; I'll send over my questions tomorrow evening.

Bronze said...

@khana,

well all those invadors in south asia ultimately failed in destroying the native culture, many invadors also lost battles and where pushed back, and even those who managed to conquer south asia including all of those you mentioned, they where ultimately raped and killed by the native indians slowly but surely, even if it took hundreds of years in some cases. The ancestors of these foreign rulers where assasinated, their troops killed in many small battles until they lost control and where wiped out. This is why Hinduism still is the main religion in India after all those invasions.

And South asians did have influence further out than afghanistan, ancient bharata included lots of central asia, and the indus valleys influence reached all the way to mesopotamia, then you have the mittani. So your argument holds no value.

Default User94 said...

The debate is over. There is no way that South Asian populations from 2000 BCE could have been responsible for the spread of Indo European languages all the way to Western Europe in a couple of hundred years (Greek is attested from 1600 BCE).

Whichever kind of evidence you look at, linguistic, archaeological, genetic or historical, the conclusion is that of the introduction of the Indo-Aryan languages into India from Central Asian steppes. The only place where the Out of India theory is taken seriously is within nationalistic circles in India

Chetan_Vit said...

@Davidiski

You said "And by the way, Corded Ware doesn't have Yamnaya influence. Early Corded Ware is basically 100% Yamnaya living in Northern Europe."

Wasn't the Yamnaya an R1b dominated culture? So what do you think - did the Yamnaya R1b population Indo-Europeanize the R1a people to their north? That could be the origin of the Corded Ware culture

Davidski said...

No, R1a-M417 expanded with an Yamnaya or Yamnaya-related population from the steppe, mainly into I2 territory. See here...

http://eurogenes.blogspot.com.au/2017/05/late-pie-ground-zero-now-obvious.html

Chetan_Vit said...

+aniasi

I just chanced upon your earlier comment and I think it would be a grave injustice if such nonsense given in the form of an argument is not immediately corrected

It entails a bunch of invading warriors deciding that they are going to spend the rest of their lives engaged with liturgical memorization, ritual minutae, and ascetic practices, while leaving those with more native admixture to become the new warrior class.

First of all, your assumption that steppe admixture is highest in the brahmans all over India is plainly wrong (a case of Brahman supremacy? I wouldn't be surprised). Many of the warrior castes or even commoner castes in north west India have the same if not greater steppe ancestry than the Brahmans of those regions. The Brahmans of southern India OTH show much less steppe admixture although still higher than the non Brahmans of the south.

Kings are still taken from the warrior class in Rg-vedic society, though they are subordinated to the priests.

Rigvedic society had no dominating priestly class like later Indian society if you examine the verses carefully. Some of the hymns which might have given you such an impression (such as the purushasukta of book 10) are known to be much later interpolations to the Rigveda and do not reflect Rigvedic society. In fact the poets of the Rigvedic hymns whom the Brahmins later claimed descend from were nothing more than sycophantic bards and lickspittles of the tribe leaders (rajanyas) who were patronized and supported by these tribe leaders. They were in no way above the rulers and depended on singing praises to the rulers as a means of livelihood