The intrusive Indo-Aryans had a huge demographic impact on South Asia

At the bottom of this post, I have posted a reformatted version of a table from the supplemental of The Formation of Human Populations in South and Central Asia. It shows a model of three hypothetical ancestral groups which contribute to the variation of modern South Asians:

  • AHG_related, a group distantly related to modern Andamanese
  • Indus_Periphery_Pool_related, a group that is roughly equivalent to the IVC population variation
  • Central_Steppe_MLBA_related, which indicates affinity to populations such as the Sintashta and Andronovo pastoralists

One of the things that people are doing is looking at “Central_Steppe_MLBA_related” as proxy-for Indo-Aryans. This is not totally wrong…but it is misleading. This fraction to me is indicative of the floor of the contribution of Indo-Aryans into modern Indians. Let me quote from the paper:

We next characterized the 2000 BCE Steppe Cline, represented in our analysis by 117 individuals dating to 1400 BCE – 1700 CE from the Swat and Chitral districts of northernmost South Asia (Fig. 2, Fig. 4). We found that we could jointly model all individuals on the Steppe Cline as a mixture of two sources albeit different from the two sources in the earlier cline. One end is consistent with a point along the Indus Periphery Cline. The other end is consistent with a mixture of about 41% Central_Steppe_MLBA ancestry and 59% from a subgroup of the Indus Periphery Cline with relatively high Iranian farmer-related ancestry ((13), Fig S50).

It seems very likely that a substantial proportion of the ancestry of the Indo-Aryans when they entered Punjab was already mixed with “Iranian-related” ancestry from further north and west. In the table below 13% of the Patel ancestry is from Central_Steppe_MLBA. All of this is from “Indo-Aryans,” but I assume some of the 60% Indus_Periphery_Pool is probably from Indo-Aryans as well.

GroupRegion AHG_related Indus_Periphery_PoolCentral_Steppe_MLBA
Kalash Pak 0.0420.660.298
Pathan Pak 0.0670.6530.281
Lohana Gujarat 0.0950.6530.252
GujaratiA USA 0.1280.6230.249
Khatri Punjab 0.1380.5990.263
Pandit Jammu_and_Kashmir 0.1590.6160.225
Yadav_Rajasthan Rajasthan 0.1630.6110.226
Dogra Jammu_and_Kashmir 0.1780.6010.222
Brahmin_Haryana Haryana 0.1880.5780.234
Muslim_Kashmiri Jammu_and_Kashmir 0.1970.5990.204
Yadav_UP Uttar_Pradesh 0.1970.5850.217
Baniya Haryana 0.20.6050.195
Rajput Haryana 0.2050.5770.218
Bhumihar_Bihar Bihar 0.2080.5180.274
Sikh_Jatt Punjab 0.2120.5350.252
GujaratiB USA 0.2130.5660.221
Brahmin_Tiwari Chhattisgarh 0.2320.5050.263
Bhumihar_UP Uttar_Pradesh 0.2380.5230.239
Brahmin_Karnataka Karnataka 0.240.5660.195
Brahmin_UP Uttar_Pradesh 0.2430.5030.254
Shiya Uttar_Pradesh 0.2430.5630.194
Havik Karnataka 0.2460.5780.176
Kshatriya_Durgvanshi Uttar_Pradesh 0.2470.5260.227
Brahmin_Nepal Nepal 0.2490.5040.247
Brahmin_Vaidik Andhra_Pradesh 0.2570.580.163
GujaratiC USA 0.2580.5650.177
Coorghi Karnataka 0.2670.6240.109
Oswal_Jain Gujarat 0.2690.5740.157
Panta_Kapu Andhra_Pradesh 0.2740.6510.075
Backward_Caste Haryana 0.2770.5110.212
Patel Gujarat 0.2790.5950.127
Brahmin_Catholic_Goa Goa 0.2850.5290.186
GujaratiD USA 0.2860.5930.122
Brahmin_Catholic_MangaloreKarn.0.2890.5650.146
Brahmin_Catholic .. 0.2920.5690.139
Chamar_Haryana Haryana 0.2920.5160.192
Meena Rajasthan 0.2920.5530.155
Jain Rajasthan 0.2930.5590.148
Agarwal Delhi 0.2930.5590.148
Brahmin_Catholic_KumtaKarn.0.2940.570.136
Brahmin_Bhatt Uttar_Pradesh 0.2980.510.192
Kurmi_UP Uttar_Pradesh 0.3120.5260.162
Nai Uttar_Pradesh 0.3130.5070.18
Srivastava Uttar_Pradesh 0.3170.5070.175
Baniyas Uttar_Pradesh 0.3190.520.162
Kurmi_MP Madhya_Pradesh 0.3220.520.159
Gaud_Karnataka Karnataka 0.3280.5960.076
Chaurasia Madhya_Pradesh 0.3280.5030.169
Reddy_Telangana Telangana 0.3290.5810.091
Lohar Uttar_Pradesh 0.3290.5530.118
Punjabi Punjab 0.3320.5110.157
Silawat Madhya_Pradesh 0.3320.540.128
Maratha Karnataka 0.3340.5510.115
Jatav Uttar_Pradesh 0.3410.4940.166
Lingayath_Karnataka Karnataka 0.3420.5450.114
Jogi Uttar_Pradesh 0.3420.5040.154
Kalinga Andhra_Pradesh 0.3430.5140.143
Yadav_Pondicherry Pondicherry 0.3460.580.074
Malaikuarvar Tamil_Nadu 0.3480.5440.108
Kanjad Uttar_Pradesh 0.3490.4920.159
Sindhi_MP Madhya_Pradesh 0.3510.5040.144
Lambadi Andhra_Pradesh 0.3520.5220.126
Kallar Tamil_Nadu 0.3570.5860.058
Narikuruvar Tamil_Nadu 0.3580.5290.114
Vysya Andhra_Pradesh 0.3580.5920.05
Naidu Andhra_Pradesh 0.3580.570.072
Ansari Uttar_Pradesh 0.3590.4850.156
Dhobi Uttar_Pradesh 0.3590.4980.143
Kuruba Karnataka 0.3610.5220.117
Ediga Andhra_Pradesh 0.3630.5480.089
Dhokkali Andhra_Pradesh 0.3630.5540.084
Baiswar Uttar_Pradesh 0.3630.4950.143
Pal Uttar_Pradesh 0.3650.5270.108
Nadar Tamil_Nadu 0.3660.5780.056
Hakki_Pikki Karnataka 0.3660.5230.111
Chamada Andhra_Pradesh 0.3670.5710.062
Achary Andhra_Pradesh 0.3690.5580.073
Gaud_Telangana Telangana 0.370.5530.077
Arunthatiar2 Tamil_Nadu 0.370.5560.074
Korava Karnataka 0.3710.5430.085
Muslim_Bihar Bihar 0.3740.4830.143
Scheduled_Caste_Haryana Haryana 0.3750.4830.143
Sonkar Chhattisgarh 0.3760.5210.104
Lodhi Uttar_Pradesh 0.3770.5060.117
Muthuraja Tamil_Nadu 0.3780.5660.057
Bestha Andhra_Pradesh 0.3790.5570.065
Dudhekula Andhra_Pradesh 0.3840.5420.074
Dushadh Uttar_Pradesh 0.3850.4790.136
Pattapu_Kapu Andhra_Pradesh 0.3880.5470.064
Pasi Uttar_Pradesh 0.3940.4510.155
Yerukali Telangana 0.3950.540.065
Kshatriya_Aquikula Andhra_Pradesh 0.4010.5260.073
Sah_Obc Bihar 0.4020.4570.141
Budagajangam Andhra_Pradesh 0.4020.4930.105
Lingayath_TN Tamil_Nadu 0.4090.5320.059
Vadde Andhra_Pradesh 0.4140.490.096
Hallaki Karnataka 0.4150.4920.094
Manjhi_MP Madhya_Pradesh 0.4150.4690.116
Chamar_UP Uttar_Pradesh 0.4150.4210.164
Paravar Tamil_Nadu 0.4170.5070.076
Vishwabrahmin Uttar_Pradesh 0.4210.5110.068
Dharikhar Uttar_Pradesh 0.430.4750.095
Oddari Telangana 0.4310.4870.082
Indumalayali Tamil_Nadu 0.4340.5250.041
Meddari Andhra_Pradesh 0.4340.4820.084
Bhil Gujarat 0.4340.4710.095
Scheduled_Caste_Karnataka Karnataka 0.4360.4850.079
Rathwa Gujarat 0.4370.4410.121
Satnami Chhattisgarh 0.4430.4480.109
Chaudhary Gujarat 0.4470.470.083
Madiga Andhra_Pradesh 0.4490.4910.06
Bhilala Madhya_Pradesh 0.4510.4430.106
Gamit Gujarat 0.4530.4490.098
Mahadeo_Koli Maharashtra 0.4570.450.093
Tadvi Gujarat 0.4580.4450.097
Changpa Jammu_and_Kashmir 0.4620.4850.052
Garasia Gujarat 0.4650.4260.11
Kunabi Karnataka 0.4660.4550.079
Yanidi Andhra_Pradesh 0.4680.4890.043
Sugali Andhra_Pradesh 0.470.480.05
Kumhar Uttar_Pradesh 0.4750.4490.077
Barela Madhya_Pradesh 0.4760.440.084
Adi_Dravider Tamil_Nadu 0.4790.4760.045
Chakkiliyan Tamil_Nadu 0.4790.4560.066
Mala Andhra_Pradesh 0.480.4550.066
Gugavellalar Tamil_Nadu 0.4830.4620.055
Kotwalia Gujarat 0.4840.4280.087
Arunthatiar1 Tamil_Nadu 0.4880.4630.05
Kurumans Kerala 0.4890.4540.056
Koli Gujarat 0.4940.4110.095
Kathodi Gujarat 0.5050.4180.078
Kurchas Kerala 0.5150.4170.068
Warli Maharashtra 0.5270.4120.061
Kolcha Gujarat 0.5310.3890.08
Pulliyar Tamil_Nadu 0.5620.3950.042
Irula Tamil_Nadu 0.5670.4030.03
Malayan Kerala 0.5810.3780.041
Ulladan Kerala 0.6070.3660.027
Palliyar Tamil_Nadu 0.6270.3430.029
Adiyan Kerala 0.6340.3310.034
0

26 Replies to “The intrusive Indo-Aryans had a huge demographic impact on South Asia”

  1. is there a way you could show individuals from the S Asian Ancestry project with the same components as this data?

    wasn’t indus peripheral like 25% AHG, averaging the three individuals they used? Granted, one individual at 42% skewed it a good bit.

    1+
    1. “is there a way you could show individuals from the S Asian Ancestry project with the same components as this data?”

      @thewarlock I also would like to see what those project members will score with ancient components. Razib should add an east Asian component as well for the members of the eastern part of the subcontinent.

      0
  2. Is there a way to estimate the actual Indo-Aryan contribution?

    Could the Iranian related ancestry further north and west be separated from the newly discovered ‘native Indian’ iranian related ancestry?

    0
  3. Could the Iranian related ancestry further north and west be separated from the newly discovered ‘native Indian’ iranian related ancestry?

    i think it’s more like the ‘native indian’ iranian anyway.

    ANI is 40% steppe and was ‘formed’ 1000 to 2000 BCE.

    1+
    1. i think it’s more like the ‘native indian’ iranian anyway.

      You mean the Steppe guys never picked up Levantine/Anatolian farmer components on their way through Iran to India (I haven’t read anything seriously yet, so pardon me if I’m bullshitting)?

      0
  4. They only used one Jatt population in the comparison? It would be nice to see the Sikh Jatts vs say the eastern/UP Jatts. Oh well, lets see what this gives us:

    “sample”: “Punjabi_Jat:Average”,
    “fit”: 1.8144,
    “IRN_Shahr_I_Sokhta_BA1”: 43.33,
    “RUS_Krasnoyarsk_MLBA”: 26.67,
    “Paniya”: 18.33,
    “IRN_Shahr_I_Sokhta_BA3”: 11.67

    “sample”: “UP_Jatt:Average”,
    “fit”: 1.9888,
    “RUS_Krasnoyarsk_MLBA”: 35,
    “IRN_Shahr_I_Sokhta_BA1”: 30,
    “IRN_Shahr_I_Sokhta_BA3”: 18.33,
    “Paniya”: 16.67

    I know that this is an amateur look at things at best, but still its something. Used BA1 as an Iran HG-rich source, the BA3 as an AASI-heavy ancestral population and added in Paniya as well since there has been additional AASI input after the IVC period. Both of the Shahr sources should also account for the WSHG + extra anatolia ancestry which inflated the steppe component in the previous south Asia population studies.

    Here is another run, this time I took away Paniya (hoping that BA3 would be a sufficient source for AASI and assumed that Andronovo could have mixed with some scattered Afanasevo survivors).

    “sample”: “Punjabi_Jat:Average”,
    “fit”: 1.3993,
    “IRN_Shahr_I_Sokhta_BA3”: 40,
    “IRN_Shahr_I_Sokhta_BA1”: 30.83,
    “RUS_Krasnoyarsk_MLBA”: 26.67,
    “RUS_Afanasievo”: 2.5

    “sample”: “UP_Jatt:Average”,
    “fit”: 1.9971,
    “IRN_Shahr_I_Sokhta_BA3”: 43.33,
    “RUS_Krasnoyarsk_MLBA”: 30,
    “IRN_Shahr_I_Sokhta_BA1”: 19.17,
    “RUS_Afanasievo”: 7.5

    Taking away an AASI source works better for the Punjab Jatts and worse for the UP Jatts (the UP Jatts still have a higher steppe ancestry). Looks like UP Jatts have more of both the AASI and steppe ancestry than the Punjab Jatts while Punjab Jatts have a higher Iran HG-related ancestry than eastern Jatts. The AASI and Iran HG-derived result is not too surprising but eastern Jatts having more steppe than Punjab Jatts is an interesting phenomenon that I have no explanation for.

    0
    1. “Is it true to say that Indians, especially Bengalis are the most mixed people on earth? ”

      @Patel It depends on what you mean by mixed. Even Europeans are mixed as they had different western Eurasian ancestors, but they are 100% Western Eurasian as their ancestors were closely related. For South Asian or Bengalis, the ancestral components are very diverse. Bengalis are Western Eurasian(IVC-Iranian type + Steppe) + AASI/AHG+ Eastern Eurasian(NE Asian +SE Asian). You can model Bengalis with middle castes from Gangetic plains with the addition of 10-12% East Asian, for example, Kurmi_UP who is 31.2% AHG,52.6% InPe and16.2% Steppe. So for a Bengali, there must be room for 10-12% east Asian, which mean Bengalis would have 5-6% less AHG and 5-6% less InPe.The steppe %age is the same or more for Bengalis, I think. Razib knows better.

      0
  5. Re: SCIENCE Paper

    Another controversial genetics paper (coincidentally?) in only two days. Again the same boring and meaningless ‘steppe, Indo-European, baltoslavics, etc’. Let see the first sentence in Conclusion:

    “CONCLUSION: Earlier work recorded massive population movement from the Eurasian Steppe into Europe early in the third millennium BCE, likely spreading Indo-European languages. We reveal a parallel series of events leading to the spread of Steppe ancestry to South Asia, thereby, documenting movements of people that were likely conduits for the spread of Indo-European languages.”

    >>> This paper contradicts the CELL paper regarding the migration and spreading the ‘steppe’ ancestry to SA. It mentioned a CONDUIT for the spread of IE languages!!! Which languages, what does it mean, conduits? Is it something as Latino-Americans spreading the Spanish language (and genes) through the conduits under the Trump’s wall?

    What the diagram says? The migration from Yamnaya to Europe started in 3300BC carrying IE languages? What about Lepenski Vir (9500-7200BC) and Vinca civilisations (5700-4500BC)? If you use a microscope you can see that ‘steppe’ migration came just meters (from other side of Danube) from these places. Migrations brought IE languages (which?) while In Vinca already had the alphabet for almost 2000 years. Which language was spoken in Vinca, which is this alphabet, time counting (now the year 7528), first wheel and thousands of other things which can be seen on wiki?

    Reich knows this because he wrote a paper about Lepenski Vir last year. They say that ‘hunter-gatherers’ lived in Lepenski Vir? LV is called ‘the first city in the world’ with one main settlement and 10 satellite ‘suburbs’, already developed trades, solid and planned houses which survived for thousands of years, etc. How likely that was the place of ‘hunter-gatherers’? Maybe, HG subset – ‘fishermen-catchers’?

    Diagram also says:
    “Location of the initial formation of Yamnaya ancestry is uncertain.”
    “2000BC: Path by which this ancestry arrived in South Asia is uncertain.”

    >>>They don’t know when the Yamnaya was formed (!) and how (which conduit?) they came to SA (!). Wow!

    Finally, in the paper:

    “Using data from ancient individuals from the Swat Valley of northernmost South Asia, we show that Steppe ancestry then integrated further south in the first half of the second millennium BCE, contributing up to 30% of the ancestry of modern groups in South Asia. The Steppe ancestry in South Asia has the same profile as that in Bronze Age Eastern Europe, tracking a movement of people that affected both regions and that likely spread the unique features shared between Indo-Iranian and Balto-Slavic languages.”

    >>> Ok, this something – genes of Eastern Europe (which one, who are they?) are the same as in South Asia.

    This thing about Indo-Iranian and ‘balto-slavic’ languages we can ignore because it does not say and mean anything…… Conduits? (ha-ha-ha).

    0
  6. @Milan

    The Vinca and other sites are ENF, the Anatolian hypothesis is dead at this point as far as I know. ENF definitely contributes to European (and thus indirectly to south Asian) ancestry but the IE languages themselves most likely originated in the Pontic-Caspian steppe.

    1+
    1. Also good commenting DaThang. Sorry for not giving you any award, so strong competition, excellent commenting – rackam, sdutta, Italian bambinos (Francesco and Carbone). I agree for Anatolian thing but Reich not. I did not want to destruct this topic. However, I mentioned in the Open Thread his (+121) last year’s paper re Lepenski Vir (Iron Gates). I think, he was a bit lazy and complacent and because we have there political statements (I mentioned in OT) and humorous elements (urban hunter gatherers). I will write about this soon (I guess, Razib will not answer my question, that’s ok).

      I strongly disagree about languages. When I talk (and make jokes) about Serbs, I am talking about Serbian language which is maybe the oldest in the world (and sometimes about R1a). Logically, if Yamnaya guys came to EU in 2700BC, which language was spoken between 9500BC and 2700BC? This is 7000 years!!! More than btw Yamnaya and us! And which alphabet was there for 2000 years before their arrival (there are now 22/30 letters from this alphabet in the modern Serbian language).

      Anyway, I agree with those suggesting that geneticists do not talk to media after publishing the paper nor holding press conferences (Razib can comment on his own blog!). I would add – just to explain the methodology and results but not interpreting them. And not talking about languages. It seems they hit the ceilings, sometimes conduits while some individuals, as we’ve seen here – the fan. We need the linguists (and mythologists) to step up.

      0
  7. @Razib, “3% of the Patel ancestry is from Central_Steppe_MLBA. All of this is from “Indo-Aryans,” but I assume some of the 60% Indus_Periphery_Pool is probably from Indo-Aryans as well.”

    In my opinon, Aryans arrived in India with 70%+ Andronovo ancestry.

    Andronovo meet two groups in Asia before arriving in India. Central Asian hunter gatherers (ANE-rich, East Asian) and BMAC (mostly Iranian). Indians don’t have a signal of ancestry from either group. Hence, I think Aryans arrived in India with 70%+ Andronovo ancestry.

    Considering Indians have 13-15% Andronovo ancestry that means the demographic impact of Aryan invasion was significant but not massive.

    0
  8. Andronovo meet two groups in Asia before arriving in India. Central Asian hunter gatherers (ANE-rich, East Asian) and BMAC (mostly Iranian). Indians don’t have a signal of ancestry from either group. Hence, I think Aryans arrived in India with 70%+ Andronovo ancestry.

    this is a geographical question, but i think they pretty clearly mixed a lot with ‘quasi-iranian farmer’ reservoir btwn bmac and when they hit the punjab.

    but 70% is not unreasonable

    0
    1. maxes at 30%. with most of the NW heavy groups at like 25%.basically still no majority aryans in S Asia. Hell I would come out at like 20-22%.

      2+
  9. While I broadly agree with the conclusions here, what most commentators fail to mention/ show is the origins of “Central_Steppe_MLBA” (MLBA – Middle-Late Bronze Age) (essentially the alleged/supposed “Aryan” ancestry).
    David Reich in his book a couple of years ago postulated (with some some good Ancient DNA evidence), that early Iranian Herders/Farmers went up the Caucuses, mixed with East European Hunter Gatherers to give rise to the Steppe population.
    My very broad hypothesis is that the “IQ” genes were contributed by this Iranian/Indus Periphery pool, while the Violence/Aggression/Physical-Strength genes were from the EEHG – giving rise to a “World Conquering” race…..

    0
  10. Well, let’s check after this SCIENCE paper where we are up to regarding my logic exercise. While “DTC (Doodlebug, Taki and Con-centric) con-genetics team” is sweating while going through this exercise, let me make their lives a bit easier. I will do the first 5 (easier) points and leave to them the most difficult, the 6th.

    1) Aryans existed.>>>>>TICK OFF (under the condition – ‘Conduit’ to be found)
    2) Aryans were R1A>>>>>>>TICK OFF (attested)
    3) Slavics were/are R1A>>>>>>>>TICK OFF (known)
    4) The term ‘Slavics’ was coined in the 7.c.AC.>>>>>>>TICK OFF (widely known)
    5) Previous term for “Slavics” (before 7.c.AC) was – Serbs>>>>>> TICK OFF (from other paper)
    6) Ergo – Aryans were Serbs. >>>>DTC – ????

    0
  11. Look at the map which Razib posted at the top of the OP – what he, the papers authors and the Euro-Centric Aryan Origins guys are ignoring is the “Back Story” of the Steppe_MLBA – that they themselves were from Iranian Herders/ Indus Periphey/Indus itself (not clear as yet) – and the original stock/cultural package took a rather large round trip via the caucuses, mixing with EEHG, maybe also mixing with other populations in Central Asia, and then into north India along the Hindu-Kush.
    Even David Reich hid this initial leg (from Iran/Indus to Steppe) in his European Chapter, with no mention of the initial leg in the South Asian Chapter!!!

    0
  12. The whole talk about IQ genes is purely speculatory. We don’t even have a full list for them yet let alone the ability to look for it in the damaged ancient DNA. Also, how would you explain people with higher EHG and lower CHG ancestry like Finns being smarter than other Europeans (on average) who have more CHG (in relation to the EHG ancestry among them)? Granted, Finns also seem to have a lower SD/variation when it comes to this characteristic (which may have something to do with lower stratification in the past and more significant effects of a bottleneck).

    As far as MLBA admixture is concerned, the Sikh Jat value in Narasimhan’s paper is roughly similar to what happens when you use Shahr BA1 + Shahr BA3 + Krasnoyarsk + extra Paniya to model them. The same combination results in the UP Jats being around 35% Krasnoyarsk. The number of UP Jats in the set was around 2 in total- so its not a very big set (which could be an issue in of itself), but eastern Jats have shown the tendency to have a higher steppe and the NE Euro admixture than their Sikh counterparts. I wouldn’t be surprised if they truly were between 30% to 35% Krasnoyarsk on average.

    0
  13. @founthead:
    A south of Caucasus origin is very dubious. Based on what I have read, it is distinctly easier to use CHG as an ancestral population to PIE instead of say Hotu cave HGs. We know for a fact that the Iran HGs who contributed to IVC separated away from the other Iran HGs 12,000 years ago. This combined with CHG being a better input than the Hotu cave HGs points to CHG being even further from the IVC Iran HGs than the other Iran HGs were. The split probably happened around the end of the Baradostian period I guess. The point is that the only migrations from the south of the Caucasus would have happened during the upper paleolithic itself and not recently from some Iran HG population source.

    0
  14. “It seems very likely that a substantial proportion of the ancestry of the Indo-Aryans when they entered Punjab was already mixed with “Iranian-related” ancestry from further north and west”

    Razib, few clarification question to understand this more :

    What was the Iranian related population they mixed with in North and west? Indus? What was the incoming steppe population like genetically, linguistically and culturally?

    Mixed population would be already “local” considering material and cultural continuity. Shouldn’t we look only at first mixer event for extent of steppe impact?

    0

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.