Talk:Genetics and archaeogenetics of South Asia

error deleted.2A02:8108:9640:1A68:7975:5573:34AD:ED9D (talk) 08:00, 30 June 2024 (UTC)

Lede too long and other redundancies
The lede currently contains two summaries, a very brief one in paragraph three, and a longer one in the fourth paragraph. Then is there is the section "Overview" which has more detail, and finally "Reconstructing South Asian population history" where many of the details in "Overview" are repeated.

This is too much redundancy. Usually, we summarize a detailed section once in the lede. A mid-level summary may appear in a dedicated "Overview" section, which should just cover key points that are further elaborated in specialized sections. I am aware that much of this has grown organically, but time has come for us to take an effort to trim this massive overgrowth. My Bolo is ready, but maybe we want to establish first what is essential for a brief summary in the lede and what isn't. Austronesier (talk) 17:57, 9 December 2023 (UTC)


 * If I remember correctly, the 'Overview' section was made (bloated) into a second "Autosomal DNA variation" section by IP socks of WorldCreaterFighter sometime in August-October 2021. Before that it was like this. - Fylindfotberserk (talk) 18:05, 9 December 2023 (UTC)
 * Ugh, not really much better back then 😂 –Austronesier (talk) 18:08, 9 December 2023 (UTC)


 * and Maybe we shorten the summaries in the lead into just one paragraph (as opposed to two); reduce the overview section to only inform about relevant events: eg. compact information on peopling and migration events, as well as short info about the respective ancestral components (name, relationship, etc.); and move the more detailed informations and study views into the "Reconstructing South Asian population history" (kind of rewritting it). Maybe we try to "dechronologify" the "Autosomal DNA variation" section a bit too. - I have just prepared a possible version here:. It would be nice if you can review it, if we may use that one. Thanks. Regards–Wikiuser1314 (talk) 14:06, 10 December 2023 (UTC)


 * The lead looks kinda OK (small changes can be discussed). The overview seems too short, as big as the lead, to the point that we can merge both (this one for example doesn't have an 'overview' section). If we are not merging them, the overview section can be a bit bigger, 50% of the original?, plus with info on the haplogroups restored. Skimming through the 'autosomal' section, I believe the Yang 2022 part needs to be trimmed and merged, especially since the paper is largely a summary of what Narasimhan, Shinde and Yelmen discovered as far as AASI is concerned (Do we need large chunks of quotes from the papers?). For a proposed lineage, the article already seems to have more coverage of that particular component (AASI), need to balance those. Let's see what Austronesier has to say. - Fylindfotberserk (talk) 18:57, 10 December 2023 (UTC)
 * I agree it's a good start. But it still needs more "dechronologifying", after all, we don't want to present a history of research, but rather the present consensus of our knowledge about the topic. Some old sources might have been important building blocks to reach this consensus, but we have to break "the A said this, then B said that, then C said this etc."-pattern. Narasimhan et al. (2018) and Narasimhan et al. (2019) have separate citations, that's also a relic of the original version. This is why anyone wanting to add preprints in Wikipedia should get their balls wired and receive an electric shock when the peer-review version comes out so they remember to update the citation ;)
 * Personally, I think the "short" summary was good enough for the lede, while current draft summary makes a good "Overview" section (if needed at all; personally, I think it's a nice thing to have). Can't say more at the moment, because I'm enthusiastically worknig on a somewhat related project in my sandbox. –Austronesier (talk) 21:23, 10 December 2023 (UTC)
 * Thank you for your comments, I will try to "dechronologify" as much as possible. Also trying to find a smoother solution for the lead/overview. Regards.–Wikiuser1314 (talk) 12:24, 11 December 2023 (UTC)


 * This version's structuring is better than the current one you are trying in your sandbox IMO. The lead is succinct. After that many versions over the years, an overview section seems important, needs a little more in my opinion and would be better if kept separate from the lede. This image should be avoided, since terms like AAI (AAA) and ATB, etc are not mainstream unlike ANI, ASI, and AASI. The former only show up papers by the Basu and those group of scientists. - Fylindfotberserk (talk) 11:37, 12 December 2023 (UTC)
 * Ok, thanks for your input, than I will use the previous version.–Wikiuser1314 (talk) 11:43, 12 December 2023 (UTC)

Regarding this edit: I can see that the statement "closest to Southern Indian tribal groups" as such is wrong, but shouldn't we at least mention somewhere that the highest levels of AASI-ancestry are found among Southern Indian tribal groups? –Austronesier (talk) 10:30, 17 December 2023 (UTC)
 * Maybe in the "autosomal" section. Now that you mention, I see that the WP:OR part - "Paniya and Irula as better proxies for indigenous South Asian (AASI)" - is present in both the "overview" and the "autosomal" section. Needs to be corrected. - Fylindfotberserk (talk) 12:50, 17 December 2023 (UTC)
 * Loads of other things need to be corrected too. I've just exemplarily changed a passage based on Shinde et al. (2019). Actually, they note that the same set of qpAdm-sources has also been used for the Indus Periphery Cline in Narasimhan et al. (2019). I also don't like the statement "but others (Yelmen et al. 2019) note that both [= AHG and AASI] are deeply diverged from each other"; "but" insinuates that others haven't been aware of this before, which gives a wrong picture since Narasimhan et al. literally say the same thing ("even though the two populations [= AHG and AASI] are deeply diverged from one another in time..."). Narasimhan et al. describe a near-trifucation of AASI, AHG and Papuans, with a minimal shared branch length of 3 for AHG and Papuans; Yelmen et al. 2019 also split out their "S" component first in a near-trifucation of AASI, AHG and East Asians (no shared branch length given for the latter; maybe its buried somewhere in the Supplementary Information?).
 * Yes . This line - "The Andamanese people are among the relatively most closely related modern populations to the AASI component and henceforth used as an (imperfect) proxy for it" is problematic especially when Reich 2009 is used as one of the sources. AASI wasn't coined then. - Fylindfotberserk (talk) 18:16, 17 December 2023 (UTC)

A relevant book
Someone should summarize this book: https://books.google.co.in/books/about/Genomic_Diversity_in_People_of_India.html?id=UbI2EAAAQBAJ&source=kp_book_description&redir_esc=y here. It seems to be a relevant book. @Wikiuser1314 @Austronesier Ionian9876 (talk) 03:09, 8 January 2024 (UTC)
 * Hi, the book is focused "on mtDNA and Y-Chromosome polymorphism", as such I would not consider it that relevant. It may be relevant for haplogroup diversity, but we all know now that haplogroups are very affected by founder effects and bottlenecks, and may tell us nothing about the genetic makeup of individuals/ethnic groups. Regards.–Wikiuser1314 (talk) 07:57, 8 January 2024 (UTC)

Male Yamnaya component not attested in prehistoric India.
" the incoming mostly male-mediated Yamnaya-Steppe component to form the Ancestral North Indians (ANI)," I wonder whence the writer got this intelligence? Core Yamnaya had no R1a, and it remains a riddle how the R1a-Z93-94 reached India and where and for when it has been attested there.2A02:8108:9640:1A68:6CBA:FF08:B1F3:E6ED (talk) 16:56, 11 February 2024 (UTC)

Pl. Join the discussion
@45.129.86.225 and @PadFoot2008 I suggest both of you to join the discussions at talk page with your respective points and not to edit war any further. &#32;Bookku   (talk) 06:28, 21 June 2024 (UTC)


 * I agree to join this discussion. IP, I request you to join this discussion too and not edit war any further. What you are doing here is promoting a fringe theory. The main article (see Dravidian peoples) itself mentions that Dravidians are Indigenous to South Asia, but "might" have an Iranic origin (also see the hypothetical Elamo-Dravidian language family). However it is unproven as of now and a fringe theory. I shall also advise you to be civil. Also as far as I am aware "Hindutva" (incorrectly) claims that Aryans are indigenous to India (see Indigenous Aryanism). I never made such a claim. PadFoot2008  10:01, 21 June 2024 (UTC)


 * Obviously Dravidian speakers have deep rooted ancient Iranian ancestry. It isn't fringe per say as far as genetics is concerned, just that it is not well discussed. Obviously the nativism associated with 'Dravidian' speakers is politically motivated (Periyarist movement for example). Since we are discussing genetics here explicitly, I don't see a reason not to mention the arguments of some researchers and attribute to them. Note that Dravidian speaking mid / agricultural castes are majority Iran related, even the dalit and tribals are quite heavy in that ancestry. The few years older - Narasimhan and Shinde papers suggested that Proto-Dravidian might have origins in the IVC-InPe (which is in accordance with the commonly accepted view of "IVC were Dravidian speakers"), and considering how heavy the Iranian part of the samples were (Rakhigarhi even), it isn't far fetched to assume the origin of the language family to be in Iran, especially among a group of people that were related to but formed a distinct line with the Belt Cave HGs and Ganj Dareh herders as demonstrated by Shinde 2019, and these two recent researches  seems to allude to that. - Fylindfotberserk (talk) 11:38, 21 June 2024 (UTC)


 * Terminology question (because a lot of arguments are caused by terminology) when you say "Iranian" do you literally mean "Iranian speaking" or do you mean something like "from the West Asian area which is now Iran"?--Andrew Lancaster (talk) 11:44, 21 June 2024 (UTC)


 * "Iranian Neolithic" to be precise and yes "from the West Asian area which is now Iran". - Fylindfotberserk (talk) 11:49, 21 June 2024 (UTC)
 * @Fylindfotberserk, I am fine with the status quo as it currently says "maybe". Also, the middle castes (Reddy, etc.) have heavy Iran-related, as well as Steppes ancestry, indicating that the these might have been descendents of Indo-Aryan-speaking Ancestral North Indians who might've mixed with the AASI Dravidians. On the other hand, other castes (which form a bulk of the population) and Dravidian tribes have about two-thirds or more AASI ancestry, though with significant Ancient Iranic ancestry, but negligible to no Steppes ancestry, the Iranic ancestry likely coming from Harappan migrations after the decline of IVC. PadFoot2008  12:07, 21 June 2024 (UTC)
 * I agree with Fyl here. There is no contradiction, although we should be careful to include highly speculative (or fringe) proposals such as Elamo-Dravidian in this context. I have repeatedly explained to PadFoot2008 that Dravidian languages are indigenous to South Asia only in relation to the arrival of Indo-Aryan speakers four millenia ago. That doesn't preclude that the predecessors of present-day Dravidian languages also migrated into South Asia from outside prior to the Indo-Aryan migration (nor does it preclude that certain IE-speaking tribal groups can well be considered "indigenous peoples" in a modern context; this is, however, off-topic here). Proto-Dravidian is most certainly native to South Asia as Krishnamurti has shown based on basic vocabulary that can be reconstructed for Proto-Dravidian and that is compatible with a South Asian physical, faunal and floral environment. But this only applies to the time right before Proto-Dravidian diversified into its daughter branches. Before that, the ancestors of the Proto-Dravidian speakers might well have arrived from outside; or more precisely speaking: outsiders may well have significantly (or dominantly?) contributed to the genepool of Proto-Dravidian speakers. There's nothing fringe about that.
 * Finally, keep in mind that linguistic and genetic diversity to not necessarily match. Entire speaker groups can undergo language shift through social interaction without or with only little geneflow involved. It is highly likely that next to the extant language families (and isolates), there were other language families in South Asia that were completely submerged by the major current language families. There is not reason to assume a priori that AASI ancestry (or any other ancestral component) must be a tracer-dye that can be associated with a specific language family, e.g. Dravidian. –Austronesier (talk) 12:24, 21 June 2024 (UTC)
 * @PadFoot2008, Mid/Agricultural-castes like Kamma, Kapu, Reddy, Vallar, Velama apparently form the bulk in the Dravidian states and they have negligible steppe. They are majority ancient Iran derived (both of Belt Cave/Ganj Dareh like and Namazga/BMAC like) and are very much what should be the late IVC types migrating out after its fall. It is possible that these heavy Iran IVC migrant imparted their language into the existing high AASI tribal groups instead. Just did a model on Mala tribe and got around 34 to 39 percent Iran Neolithic related.
 * The IVC were (Proto)Dravidian-speakers per majority / general consensus and they were mostly Iran N derived. It is far more likely that the Iran N part of their ancestry is associated with the language. Not sure how a 'proposed' group of hunter-gatherers (AASI), who didn't have much in the name of society (as far as we know) and having minimal impact on the DNA of Early IVC would be instrumental in forming one of the oldest civilizations/languages of the world. Considering the existence of far older language isolates like Nihali, it is likely that the various streams of AASI spoke these languages/isolates. - Fylindfotberserk (talk) 12:42, 21 June 2024 (UTC)
 * @Fylindfotberserk, Firstly, Middle castes don't form a bulk of the population of South India. Take Tamil Nadu for instance, Backward Castes (45.5%), Most Backward Castes (23.6%) and Scheduled Castes (23.7%) together form 92.8% of Tamil Nadu's population. In Andhra Pradesh, Scheduled Castes, Scheduled Tribes and Other Backward Classes together account for 74.6% of the state's population.
 * Secondly, both Dravidian peoples and Indus Valley Civilisation articles mention that Proto-Dravidian might have been spoken there. Mostly, only scholars that support the Elamo-Dravidian hypothesis, support that the language might have been spoken in the IVC. However a reconstruction of Proto-Dravidian shows that it's vocabulary is characteristic of the dry deciduous forests of central and peninsular India, not Iran or the Indus Valley. It is certainly not the majority view. PadFoot2008  16:23, 21 June 2024 (UTC)
 * By 'mid-caste', I meant the non-high caste non-dalit non SC and ST population based on this. The Nadar caste which I believe is considered BC, are very similar to the Vellalar who are Iran N heavy. Even with the Madiga, an SC group, I get ~43% Iran related. The hypothesis that support a peninsular origin of Dravidian languages are based on re-construction of a Proto-language based on a few root words, but doesn't align with archaeology, genetics and material culture. There was an obvious large scale migration from IVC towards the south. I believe there are more proponents of IVC origin of Dravidian languages, many of whom are not necessarily supporters of the Elamo-Davidian hypothesis, bolstered by newer researches like these-this, that go hand in hand with linguistics and genetics. Note that this comes from pre-Vedic IVC, which might be a relict of the ANE people, and Iran N has copious amounts of ANE ancestry. Anyway, this is becoming too forumy, I'd suggest addition of these two sources, without changing any text. - Fylindfotberserk (talk) 17:38, 21 June 2024 (UTC)
 * Why is even the IVC discussed here? The proposed IP edit doesn't even mention it. But for the record, the statement "only scholars that support the Elamo-Dravidian hypothesis, support that the language might have been spoken in the IVC" (once again) betrays an unfamiliarity with the relevant literature. There are scholars like Asko Parpola who argue for a Dravidian affiliation the language of the IVC without ever supporting the Elamo-Dravidian hypothesis. And again, you're conflating genetic with linguistic evidence. I won't repeat what I have written above, but your claim that linking the introduction of Iranian hunter-gatherer-related ancestry with the introduction of the language that evolved into Proto-Dravidian on South Asian soil is fringe needs a source that calls it so. Sure, this link is sometimes embedded in an Elamo-Dravidian framework, mostly by Indian geneticists who uncritically accept the Elamo-Dravidian hypothesis as an established fact, but sure not by all scholars (cf. Mukhopadhyay's paper).
 * In case the IP will ever join this discussion: of course, Nature does publish fringe articles. To get published in Nature, it "only" needs a paper that appears to the editors worthy of further discussion by the scientific community. For articles in the core expertise of Nature, the threshold for inclusion is quite high, but for articles outside of their expertise, it can be astonishingly low, as in the case of said article by Mukhopadhyay. Likewise, a sensationalist title like "Human Y chromosome haplogroup L1-M22 traces Neolithic expansion in West Asia and supports the Elamite and Dravidian connection" (what does "Elamite and Dravidian connection" even mean?) is also not indicative of a differentiated discussion. Indeed, their conclusion is all iffy: "if Elamo-Dravidan is valid, then our findings support it". LOL. –Austronesier (talk) 17:46, 21 June 2024 (UTC)
 * Instead of citing lesser works (albeit published in good journals), why don't we use Narasimhan et al. (2019)? On page 12 (first column, second paragraph), they propose a link between the Dravidian languages and the Iranian-related portion of ASI ancestry. –Austronesier (talk) 17:55, 21 June 2024 (UTC)
 * I am not proposing a change in the article text, I think the status quo is good. Also, I initially claimed that it was a fringe theory (sorry, my bad), but in my previous comment, I revised it to say that it is not a majority view (as far as I am aware). I see that there is a significant section of scholars who propose the link between Dravidians and Ancient Iranians. Also to @Fylindfotberserk, I never refuted the large-scale migration of IVC people(s) to the peninsula after the decline of IVC, that is, in fact, what likely contributed Iranian ancestry in Dravidians. What I meant to say is that it doesn't necessarily support an Iranic origin of the Dravidians (but it is indeed possible). PadFoot2008  18:06, 21 June 2024 (UTC)
 * Perhaps, why the text is there in the first place. Narasimhan is referenced at the end of the paragraph, alongwith a bunch of other references. You know who jumbled up all those . If necessary we can attach the specific quote. - Fylindfotberserk (talk) 18:16, 21 June 2024 (UTC)