The Toda are different

A new paper on Southwest Indian genetics highlights the Toda sample from Genomes Asia. People in the comments of this weblog have asserted this small southern tribe may have the most “Indus Valley Civilization” ancestry in the subcontinent. This is perhaps an exaggeration, but, looking at the admixture plots the Toda clearly have hardly any steppe ancestry, but a lot less “ASI” ancestry than their tribal neighbors, with the balance being something like the IVC ancestry.

Genomes Asia doesn’t make it’s data public, and for ancestry purposes I don’t think they’ve done the best job.

0 0 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
1 year ago

“for ancestry purposes I don’t think they’ve done the best job.”

Except for the Kapla, the p-values of their qpAdm models for various populations didn’t pass the threshold. I believe this has to do with scarce sampling of the subcontinent regions across space and time. Btw, some people have been citing these failed model estimates on social media, do you think that the estimates of ,say IVCp ancestry, of these failed models are close to accurate ?

1 year ago
Reply to  td

He is talking about this

‘Genomes Asia doesn’t make it’s data public’

Any other problem in paper?

1 year ago

Razib, is it correct that Northwest Indian upper castes like Jat and Pakistani Sindhi/Pathan/Punjabis generally carry higher IranN ancestry in their IVC-like ancestry as compared to the IVC-like ancestry of the Southern Indian castes which generally tend to be AASI-shifted? If I remember correctly, certain South castes like Vysya and Kodava score highly on IVC (over 60%) with qpadm’s steppe-ivc-aasi grouping but have over 50% south-indian in harappaworld whereas Sindhis, Pakistani Gujjars/Arains/Jats have over 40% Baloch and under 30% South Indian.

If you get the time, you should really try to do an qpadm of Steppe-IVCp-AASI on a set of samples like this one:

1 year ago
Reply to  Lazimo

I am not Razib, but those components might be splitting things unevenly (some Iran-like goes in Baloch, some in S-Indian maybe). IMO it is worthwhile to wait till high quality IVC era and AASI samples are dug up and published, so we have an array to compare to.

1 year ago
Reply to  Lazimo

1. Those aren’t technically “upper castes.” You listed ethnic groups and one high steppe group that was typically avarna/considered to be shudra historically because they likely have a different ethnogenesis than the typical upper caste groups, despite more steppe. The ethnic groups have a mix of low and high castes. There is a conflation of steppe proportion and caste. That proportion is more applicable to Gangetic Plains. In the NW, many peasant groups are still high steppe. But caste like structural endogamy just exists in the deep genomic structures of the populations. Look at Razib’s posts for more information on that.

2. I agree with Dathang. The data isn’t disaggregated enough yet to get a “pure” version of AASI vs. the iranic like ancestry. The end result is that modeling assumptions will then change with how percentages are assigned.

3. I see Pak nationalists use this area of ambiguity to push an agenda that maximizes the Indian AASI and Indian nationalists do the opposite. Pak nationalists are driven by the this new “Indus Gang” theory for two reasons.

A. They always wanted to show that they are different than Indians to justify
partition. This fell apart on religious grounds with the Bengal split. The new
theory then came on emphasizing “Middle Eastern” roots. Modern genetics
put a dagger in that, when it showed that component was minimal among S
Asian Muslims. This “Indus” nationalism is the final piece to do it on racial
grounds. This probably has the most credence since there are clusters for
many groups. Though many lower caste Pak groups, like 30% of Pak
Punjab do cluster with many Indian groups, albeit some upper caste ones.
But Pak Christians, who are no less Pakistani cluster with dalits. And many
Ashraf Mohajirs with upper castes of Gangetic plains. The minority
Pasmanda with the Pasmanda. These tend to get disowned. Also, the AASI
heavy aspect of the cline aka (50% AASI, 50% iranic) also tends to get
disowned, which is what Gujarat, a state that was part of IVC- tends not to
get included in the discussion. Including it would include a ton more groups t
to the South and East who would fit that cline. Also, Gujarat is not as
sympathetic to the Pak cause. A minority of Punjab is. And Haryana fits the
racial narrative.

B. Indian nationalists either do OIT, which is losing credence year by year.
Or they use the aforementioned ambiguity to pick models in the opposite
direction, so as to show a minimum amount of difference. This then allows
them to claim maximal similarity, thus trying to prove that partition occurred
on purely religious grounds and we are all in fact more or less “the same

4. I see none of this resolving in the future. I am more sympathetic to the view that all of S Asia can lay historical heritage to the IVC. The population proportion components are similar but most importantly the cultural aspects are there. The dancing goddess type imagery, the yogic poses, the possible references to some Hindu Gods. It is hard to prove because the script is not deciphered. But there seems to be a clear cultural legacy. I notice that there are ulterior political motives for the two different moods of the nationalists. But I think the end result of pushing the most accurate narrative is the best. But if one has to pick one, I am more in favor of a unifying one. Indians tend to view Pakistanis and Bangladeshis as brethren lost to an unfortunate ideology. Bangladeshis seem less neurotic about this. They focus less on how to “get at India” and more on economic growth. On the other the other hand, I view the modern “Indus Gang” narrative pushed by Pak Nationalists as a renewed, more sophisticated attempt, exaggerate racial differences with Indians to not only justify partition, but also continue their more colloquial claims of racial superiority. Unfortunately, Indians are not immune to doing this among themselves. But we are talking about things on a nation-state level. Indian has massive internal problems with this. The historic dominance hierarchies of more West people on top in India has really made AASI ancestry something people want to run from, the proxies for it are skin color and tropical features. We see manifestations of this over and over again. It just gets more sophisticated with genetics.

Brown Pundits