Tony Joseph has an interesting piece up, Who built the Indus Valley civilisation?, which people are asking me about via email. First, I don’t have any inside information. Last I heard in September was that the Rakhigarhi results were “one or two months away,” like they have been for a year or so. So I put it out of mind.
In any case, here are the important points:
All this could now change thanks to the science of genetics and four ancient skeletons excavated from a village called Rakhigarhi in Haryana. The four people to whom these bones once belonged — a couple, a boy and a man — lived roughly 4,600 years ago when the Indus Valley civilisation was in full bloom.
…
In the three-and-a-half years since its excavation, Shinde has brought together scientists from Indian and international institutions like the Centre for Cellular and Molecular Biology, Hyderabad (CCMB), Harvard Medical School, Seoul National University, and the University of Cambridge to work on different parts of the project, including extracting and analysing DNA from these ancient people, reconstructing their faces, and studying the remains of their habitation to understand their daily habits and ways of life.
The DNA analysis will also help figure out their height, body features, and even the colour of their eyes….
Joseph also asserts that the publication will happen in a “leading international journal” in a month or so. If I had to bet, I’d say Nature.
Harvard Medical School suggests to me they finally got David Reich’s group involved. As for Cambridge University, Eske Willerslev now has an appointment there. He’s apparently assembling a paleogenetics group.
The piece specifically highlights Y and mtDNA. But if they are talking about height, body features, and color of eyes, they must have gotten genome-wide data. If Eske Willerslev is involved they may have sequenced the whole genome at some coverage of at least one of the samples.
If I had to bet I think the Rakhigarhi samples will be Y haplogroups J2 or the Indian branch of L, and the mtDNA will be an Indian branch of M. In terms of genome-wide patterns they will exhibit a mixture between West Eurasian ancestry, with strong affinities to Near Eastern farmers from the Zagros, and what we now term “Ancestral South Indians” (AS), who descend from the aboriginal peoples of the subcontinent, and are genetically somewhat closer to East Eurasians than West Eurasians (to be fair, I think it is not implausible that much of ASI heritage is the product of westward migration out of Southeast Asia during the Pleistocene and early Holocene).
Overall, genetically these samples may look the most like South Indian non-Brahmin middle-to-upper castes. Think the Reddy people of Andhra Pradesh. Additionally, going back to R1a1a-Z93, I do think it was intrusive with the Indo-Aryans. Its highest frequencies do tend to be among upper castes, and there is an increasing cline toward the northwest of the subcontinent.
ButR1a1a-Z93’s presence at appreciable frequencies in South India among non- Brahmins, including tribal populations, indicates a more complex ethnogenesis of Dravidian speaking groups than we might have realized. Priya Moorjani told me specifically that 4,000 years ago there were “unmixed ANI and ASI groups” in the subcontinent. I think for the former she’s picking up the signal of intrusive Indo-Aryans. But what about the latter? I doubt there were unmixed ASI in the Indus Valley. But they probably still persisted to the south and east when the Indus Valley people were in decline and the Indo-Aryans arrived. The South Indian Neolithic dates from 3000 to 1400 BC.
Here my moderate confidence sketch. The collapse of the Indus Valley civilization was probably ultimately due to the fact that these early antique societies were not very robust to exogenous shocks and endogenous decay of asabiya. Once these societies, which have accumulated some level of surplus wealth by squeezing it out of the Malthusian margin, start to totter social collapse and dissolution can happen fast, and barbarian groups outside of the gates with more social cohesion can engage in a takeover.
In the case of the collapse of the Sumerian-Akkadian civilization, the barbarian Amorites actually took over and maintained cultural continuity. In post-Roman Britain, the Roman civilization collapsed in totality, and “Roman Christianity” had to be reintroduced from the European continent and from the Celts into Anglo-Saxon England. The barbarian takeover resulted in the total cultural obliteration of the Britons. Finally, you have instances such as post-Roman Gaul, which transformed into Francia. Unlike the case of the transition from the rule of the Third Dynasty of Ur to that of the Amorites, the Frankish rulers oversaw a wholesale reimagining of the identity of the people of Gaul. Even as late as 800, a ruler such as Charlemagne still spoke a dialect of German as his first language. And yet the Franks of Neustria were ultimately transformed and became one with the “Romans” whom they ruled.
In the post-Harappan world of northwest India I suspect something close to the Anglo-Saxon precedent is likely. Though the majority of the ancestry of the Upper Gangetic plain is not Indo-Aryan, a substantial proportion is. And this ancestry is detectable at lower fractions even among non-Brahmin Bengalis. In Central and South India the situation was probably more like Mesopotamia around ~2000 BC or Gaul post-500 AD. There were various sorts of interactions between Indo-Aryans and local populations, as well as the final assimilation of aboriginal peoples into Indo-Aryan and Dravidian speaking peoples.*
* The Munda people clearly have some East Asian ancestry. And, they are mostly a mix of ANI and ASI. But whenever I look at their genome-wide results it strikes me they may not have any Indo-Aryan ancestry. This may ultimately be totally comprehensible in light of the chronology of migration and segregation.
Update: One of the researchers involved indicates Eske Willerslev is not involved.
If Moorjani asserts that around 4000 years BP, ANI and ASI were present unadmixed, the clear inference from it would be that the Indus Civilization people were the unadmixed ANI people.
In that case, they are also unlikely to have mtDNA M and more likely to have some “west Eurasian” mtDNA.
Dear Razib, I would like to inform you that Prof. Eske is not involved in this research.
“The Munda people clearly have some East Asian ancestry. And, they are mostly a mix of ANI and ASI. But whenever I look at their genome-wide results it strikes me they may not have any Indo-Aryan ancestry.”
The Munda/Santhal/Khasi people are really interesting, also they are an important contributor of Bengalis and East Indians genetic, maybe even to all North Indians.
They are very likely paleolithic inhabitants of East India/Bengal/Southeastern Nepal area that survived till now.
If they are a mix between ANI-ASI and SE Asian, ANI must be present from paleolithic times in that region. I found this paper interesting, though im not sure how reliable it is:
http://ispub.com/IJBA/4/2/5591
The ANI-ASI ratio remains unclear if we count the third component “ANE”.
Perhaps Ust-Ishim is closest proxy for ASI?
Ust-Ishim originally supposed to be closer to East Asians, but 8000 years old “La Braña” or 24000 years old “Mal’ta 1” shares more derived alleles with the Ust’-Ishim individual, that means they are in between ANE and East Asians?
Tony seems to be claiming that one can disprove a theory about the composition of population can be proved or disproved based on just four skeletons. Well, journalist.
That said, is there any other new info that we can possibly gain from studying these? Can it confirm or reject some existing theories?
Whatever is published will be highly politicized.
Davidski has said (in his comments section) the delay in publication is likely because the Indian scientists involved can’t believe the results which support Aryan migration, and keep asking for more data and delay.
On the other hand, if Moorjani says there is pure ANI in India 2000 BC, then she is implying Out of India, or a close variant (out of everywhere).
The only official leak so far, to the Times of India in August 2016, from Haryana archeological department, suggested cultural continuity from harrapan to haryana today. I think that tells you the outcome the hindu nationalist gvt would like.
Whatever is published will be highly politicized.
yes, but as the 2009 reich paper suggests in the long-run that’s irrelevant. the indian scientists and media explicitly inverted the most parsimonious conclusion IMO and that had no long-term impact.
On the other hand, if Moorjani says there is pure ANI in India 2000 BC, then she is implying Out of India, or a close variant (out of everywhere).
no, she thinks the likelihood is that the ANI probably came from outside of india. i’d interpret her:
ANI = indo-aryans newly arrived from central asia
ASI = relic populations of ASI through the subcontinent; we know from europe that unmixed mesolithic type populations persisted in areas for several thousand years after the arrival of farming. the neolithic begins in s india 3000 BCE
Tony seems to be claiming that one can disprove a theory about the composition of population can be proved or disproved based on just four skeletons. Well, journalist.
this is a fair point, but remember that if you get genome-wide data you are able to make an inference about a whole lineage.
if the individuals turn out to be ANI-ASI mix but the mix seems old, that tells us something. if they are very recently mixed that tells us something. if they are 100% ASI that tells us something.
remember, there are no people in all of india today who are 100% ASI or 100% ANI. unmixed people would be pretty major.
They are very likely paleolithic inhabitants of East India/Bengal/Southeastern Nepal area that survived till now.
culturally, no. they are intrusive. the Y chr lineages they have seem to be derived from SE asia not other way around.
here is my timelime
1) west asian farmers + ASI = ASI/ANI mix
2) east asian farmers + ASI/ANI mix = munda in parts of eastern india
3) central asia steppe ppl + ASI/ANI mix = indo-aryan speaking peoples
4) some of the eastern indo-aryan ppl, like bengalis may have munda ancestry. but their east asian ancestry is almost certainly more than munda. probably some of it is even tai, not austro-asiatic
If Moorjani asserts that around 4000 years BP, ANI and ASI were present unadmixed, the clear inference from it would be that the Indus Civilization people were the unadmixed ANI people.
no, the “ANI” is a composite for west eurasian. so new west eurasians without asia could arrive, and mix with ANI/ASI. and, one can imagine relic ASI populations that late (or later).
the methods she was using tend to detect the *last* admixture event the best. does not preclude earlier admixture events.
Davidski has said (in his comments section) the delay in publication is likely because the Indian scientists involved can’t believe the results which support Aryan migration, and keep asking for more data and delay.
I would give you a word of advice – do not take such statements from Davidski too seriously. He is a loose cannon and also a bigot. He is not at all credible. The Indian scientists surely have more integrity and sense than him.
Thank you for the explanation.
You raised the issue of Roman Britian and gaul. I am in the middle of listening to The Fall of Rome Podcast https://soundcloud.com/fallofromepodcast It is very good, and I recommend it to one and all.
He discusses the Saxon takeover of Britain, and talks about theories of elite replacement vs theories of mass migration. I am sure there is genetic evidence available on this issue. You may have even discussed it in the past, can you point me to some information on this subject?
patrick comments on this blog now and then.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4632200/
“here is my timelime
1) west asian farmers + ASI = ASI/ANI mix
2) east asian farmers + ASI/ANI mix = munda in parts of eastern india
3) central asia steppe ppl + ASI/ANI mix = indo-aryan speaking peoples
4) some of the eastern indo-aryan ppl, like bengalis may have munda ancestry. but their east asian ancestry is almost certainly more than munda. probably some of it is even tai, not austro-asiatic
—
the methods she was using tend to detect the *last* admixture event the best. does not preclude earlier admixture events.
”
Basically we Bengalis are probably :
ANI/ASI mix + Munda + Indo Aryan + Tai.
Or very likely :
Indo Aryan + Munda + Tai. Since Indo Aryan is “Central Asia steppe people + ANI/ASI mix” and founder Ethnic group of Bengalis possibly migrated from west, maybe from cow belt region.
Munda population was formed after Indus velly collapse? Or “ANI/ASI mix + east Asian farmers” occurred before IVC?
If Moorjani et al. detects the last admixture event the best, the ASI might be already admixed with west Asian farmers as you stated and That would change the real ANI/ASI ratio.
Perhaps west Asian farmers had multiple migration( including pre-neolithic before the spread of agriculture) toward the subcontinent?
Thank you Razib:
“patrick comments on this blog now and then.”
I assume you mean Patrick Wyman, author of the Fall of Rome Podcast.
He has a new podcast series titled Tides of History:
https://wondery.com/wondery/shows/tidesofhistory/
Which you also linked on today’s “Open Thread’
The 21 Sept 2017 episode is titled: “Interview: Dr. Jennifer Raff on DNA and History”
I have not yet gotten to that one. Have you? Do you know Dr. Raff?
I found two more links from the PubMed link you gave:
“Iron Age and Anglo-Saxon genomes from East England reveal British migration history” by Schiffels, Haak, et. al.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4735688/
“Genomic signals of migration and continuity in Britain before the Anglo-Saxons” by Martiniano, Caffell, et. al.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4735653/
Are either of those worthwhile?
Razib, I am a historian by training, not a biologist. I truly appreciate your interest in and knowledge of history. I look to you for information as to which biological research I should pay attention to.
Looking only at Bayesian graphs, it must be the case that the Late Harappan (not early, not middle, but just about the pre-Aryan ones, the ones which experienced their civilisational fall) were about 50% ASI and 50% Iran_N.
Another good proxy for the Late Harappans in my view are today’s Vishwabrahmin and Mala peoples, who are about ~80% ASI and ~20% Iran_N.
Now, how were the Early and Middle Harappans? Those that founded and flourished the civilisation? I think they would be very much closer to Iran_N, something like +90% Iran_N at Early periods and +70% Iran_N at Middle periods.
If they had contact with the Aryans, if they mixed or if the Aryans were the cause of the Harappan’s downfall is also uncertain (this could also be attributed to the ASI).
I have not yet gotten to that one. Have you? Do you know Dr. Raff?
we’re friendly. not sure she says anything you wouldn’t find here though. but she knows the americas really well.
the papers you talk about are of interest, but don’t add much to the general picture.
Looking only at Bayesian graphs, it must be the case that the Late Harappan (not early, not middle, but just about the pre-Aryan ones, the ones which experienced their civilisational fall) were about 50% ASI and 50% Iran_N.
i think this is likely.
i think punjabi dalits are likely to share lot of drift with them. there is a huge range within punjab samples, even excluding jatts, who seem to have even post-aryan admixture from the NW.
Perhaps west Asian farmers had multiple migration( including pre-neolithic before the spread of agriculture) toward the subcontinent?
likely.
Indo Aryan + Munda + Tai. Since Indo Aryan is “Central Asia steppe people + ANI/ASI mix” and founder Ethnic group of Bengalis possibly migrated from west, maybe from cow belt region.
when india comes back into history it looks like magadha was a marcher state/society, and what became bengal was not part of aryavarta. so i think yes it got ‘indo-aryanized’ by ppl from magadha, but the magadha people themselves were highly admixed (ashoka’s lineage was not of the conventional aryan kshatriya class than siddartha came from from what i recall).
the indo-aryan signal, the steppe affinity, does exist in non-brahmin bengalis. but it’s attenuated as the signal of admixture kept getting weaker from the punjab.
the people who remained culturally munda seem to be relatively insulated from this indo-aryan signal, though they are surrounded by indo-aryan speaking peoples. i suspect a lot of munda speaking ppls were absorbed by indo-aryans in NE 1/4th of india in the lowlands. though i’m not totally sure since i don’t see any derived EDAR or haplogroup O in biharis…but cow belt samples outside of brahmins and kshatriyas are few and far between.
@Jaydeepsinh Rathod “Davidski is a loose cannon and also a bigot. He is not at all credible.”
Internet speak for “he disagrees with my cherished doctrines and nationalistic origin myth”.
Can I ask you to please remove comments calling others bigots, etc.
In any event, the idea of Indians calling someone else bigot is somewhat strange. We are the world champions in being intolerant towards people who hold differing opinions.