South Asian human geography as a post-Aryan synthesis


One of the things that is evident in the most recent work on Indian genetics is that some groups, often Brahmin, are enriched for “steppe” ancestry when looking at overall contributions of proximal ancestral components. But, there are other groups that are enriched for “Indus Periphery” ancestry. The plot above takes Indus Periphery on the x-axis, and steppe on the y-axis. You can see that Brahmins are above the main trend, but groups like “Panta Kapu” are below (click the image).

These trends can be hard to spot because of the complexity of the Indian genomic landscape, where geography is not entirely predictive. What explains them?

I outlined my general model in a blog post, The Aryan Integration Theory (AIT). In short, unlike Northern Europe, and like Southern Europe, pre-Indo-European cultural matrices have maintained some robustness in the face of agro-pastoralist intrusion. The persistence of linguistic isolates in the far northwest in the form of Burusho is indicative of this. But also the persistence of the Dravidian language family, which has pre-Aryan roots. The enrichment of “Indus Periphery” ancestry in groups in the west and south, in particular, as well as a Dravidian substrate in toponyms in Gujarat and Maharashtra, and the relative lack of such features in the Gangetic plain, point to the reality that Dravidian speaking peoples are not primal, but their current range is partially reflective of the human geography in the wake of the Indo-Aryan shock on the decaying IVC.

22 Replies to “South Asian human geography as a post-Aryan synthesis”

  1. ” pre-Indo-European cultural matrices have maintained some robustness in the face of agro-pastoralist intrusion.”

    I had a question. Sticking strictly to N-India (where the actual action really lies) , how are we presuming this robustness? For all we know everything IVC could be wiped off and what we see in N-India is just Aryan culture. How much part do you feel of today Indian culture is IVC and what part Aryan. (3/4th aryan , 1/4th IVC? 50-50 ?).

    This to mean could mean only 2 extreme things

    A) That the intrusion was so staggered that it wasn’t seen as an intrusion and the culture just slowly changed, so slow that it didn’t even seem to change.That even IVC groups (like the Dravidians) didn;t seem to have any popular recollection of how these intruding people were different.

    B) The Aryan victory was so complete that they just picked the remnants of IVC and force fitted in their culture and their culture/religion also became the “natural” religion of even the IVC groups.

    1. lohana and Brahmin. Lohanas are hindu migranta from Sindh

      B is a lot of rajputs (solankis and vanias)

      C (some patels and vanias)

      D (patels)

      Patels average less steppe but are more heterogeneous than other groups.

        1. yes. They form the plurality of Gujarat. Gujarat is diverse but they are used as the symbolic people of Gujarat. Parsis, Vanias, Rajputs, Brahmins, Lohanas, Memons, Bohras, and Bhils are just some other groups to name a few

  2. Razib, (I am trying to search for your relevant comments but finding it hard to search past blogs/comments.. so quoting you from memory).
    If I am not mistaken, you hypothesized that Indo-Aryans entered India (say around 1500 BC) and bypassed IVC area to settle (take over, replacing locals, etc.) in gangetic plains, at least initially.
    But then.. how did they replace IVC language/s in IVC areas so thoroughly in subsequent centuries?

    1. What is a IVC language ?
      Language doesn’t leave behind any evidence. So I doubt if we know what the IVC language is

      If you mean Dravidian languages, then Marathi & Gujarati speakers speak with a Dravidian accent

      & Tribal areas in MP, Chattisgarh also speak Dravidian languages

      Orissa / Mahanadi basin has the influence of Munda languages

      So Dravidian seems to have been confined to south of Vindyas

      Not sure, if Biharis ever spoke Dravidian

      1. “What is a IVC language ?”.. That is the million dollar question, isn’t it? OIT (at least some variants of it) says it was vedic Sanskrit. AIT variants have to postulate what it was (e.g. Dravidian, or some other “extinct” language), and explain how it disappeared so completely w/o even lending toponyms (names of rivers, at least) to the subsequent languages.

  3. If we could have geography as a third dimension in that graph it would help to understand better

    Also what are the sample sizes for each dot?
    Is it a sample of 1 or averaged across multiple individuals of the same demographic?

  4. These Gangetic plain communities(Srivastava, Nai, Baniya, Kurmi) stand in a balanced position for both X and Y-axis. I guess before migrating toward Ganges Delta, Bengalis were similar to these groups.

  5. Many Tamilians consider themselves and Tamil to be Arya.

    Agastya in particular (who is praised by Buddha, and to whom many Vedic Samhitas are attributed) is considered one of the founders of Sanskrit Vedagamic Samkhya civilization and Tamilian civilization.

    18 Siddha tradition (of which Agastya is most important) is formally Uttara Mimaamsa darshana–yet to my eyes is a fusion of Samkhya darshana, Yoga darshana and Uttara Mimaamsa darshana.

    Agastya has been deeply important in Java, Sumatra, Thailand, Cambodia, Loas, Vietnam and the Philippines for thousands of years.

    Could Agastya be the nexus that combines all the Darshanas, south east Asia, South Asia, central Asia, Iran, Xinjiang, Tibet, Yavana (Europe/Serbia), Tushara (east of Xinjiang and Tibet)?

    Might Agastya be tied to the civilization of Gunung Padang (a giant Javanese pyramid for which a samples have carbon dated to between 12 K and 28 K years ago)

    Is Agastya the founder of Arya or Aryanism?

    1. Agastya is probably a composite of many different figures with the components depending on the region and the importance that the various people give to different inputs. If an input into this mosaic-like figure comes from Gunug Padang, then those people would be more like Negritos/Hoabinhians than anyone else. Incidentally, it turns out that Jarawa people are super close to Hoabinhians on a PCA. Razib would know much more about this since he could probably use D-stats to check for relatedness.

        1. @Razib:
          Have you heard of the simulated AASI populations made by DMXX on anthrogenica? Probably not super accurate, but its the best I have come across in terms of a proxy for AASI (at least better than Onge). Apparently, you can model southern AASI using NW (northern) AASI + Onge/Hoabinhian. What do you think about this?

          “sample”: “Simulated_AASI_South_by_DMXX:Average”,
          “fit”: 5.6886,
          “Simulated_AASI_NW_by_DMXX”: 58.33,
          “LAO_Hoabinhian”: 41.67,
          “closestDistances”: [
          “Simulated_AASI_NW_by_DMXX:NW_SiSBA3: 15.55152”,
          “LAO_Hoabinhian:La368: 21.63812”,
          “Simulated_AASI_NW_by_DMXX:NW_SGPT: 27.94810”,
          “Simulated_AASI_NW_by_DMXX:NW_SiSBA2: 39.29861”,
          “Simulated_AASI_NW_by_DMXX:NW_GonurBA2: 41.78350”

          Doesn’t look too bad.
          And doing the opposite is not possible.

          “sample”: “Simulated_AASI_NW_by_DMXX:Average”,
          “fit”: 23.0301,
          “Simulated_AASI_South_by_DMXX”: 100,
          “LAO_Hoabinhian”: 0,
          “closestDistances”: [
          “Simulated_AASI_South_by_DMXX:Hakkipikki: 23.45924”,
          “Simulated_AASI_South_by_DMXX:Mala: 23.90492”,
          “Simulated_AASI_South_by_DMXX:Irula: 28.51050”,
          “Simulated_AASI_South_by_DMXX:Pulliyar: 29.16559”,
          “LAO_Hoabinhian:La368: 44.10563”

          It kinda makes sense. The folks at anthrogenica have theorized that perhaps Onge are a remnant of a migrating Hoabinhian population, and considering that the Andaman islands can only host small populations, it would make sense if most of the Hoabinhian migrants ended up in south India instead and mixed with the local AASI. As such I would assume that simulated NW AASI would be a simulation of AASI proper while simulation of southern AASI is a simulation of a mixed AASI proper + Hoabinhian population. So yeah, what do you think about this?

  6. Please tell me the about the African looking people living in the Andaman islands.
    What are their haplogroups?
    Are they the aboriginal people of south asia?

    1. They are an relic population closely related to Hoabhinians of southeast Asia and only distantly related to AASI people of South Asia. They belongs to Ydna D and MtDna M. They have pseudo african look probably due to convergent evolution as genetic wise Europeans are more closer to an African than Andamanese are to Africans.

      1. They are distant in the sense that any divergent east Eurasian population is also distant from AASI. You should have noted that they might have contributed to the southern AASI ancestry, though that hasn’t been proven yet.

  7. Does the “Pandit” in the Image refer to Kashmiri Pandits?

    If yes, the results are interesting in many ways.

    1. Very interesting that the admixture from rest of India is slighly more in Kashmiri Muslims than in Kashmiri Pandits (they are closer to Center of Gravity of Indian Cline, through AASI component?). Politically, that would be interesting to share with Kashmiri Separatists :-). BTW, we know that historically there have been migrations from rest of India to Kashmir over time e.g even Abhinavagupta’s (950 AD) ancestors were from Madhyadesha.

    2. Kashmiris are usually expected (anecdotally) in North India to be more lightly complexioned compared to other Indians. There is some truth in this, though I know for sure that the expectations are exaggerated and Kashmiris possess a wide range of “looks” and complexions. But the graph suggests they have much lesser “Steppe” component than UP/Bihar Brahmins (e.g Tiwaris/Kanykubjas), and there is usually no similar expectation of differential looks from UP/Bihar Brahmins.

    Seem to imply that either looks/complexion etc. is more a function of different climate over millennia and the Steppe/IVC is not the main determinant in outward appearance.

Comments are closed.