Document ID: L_2_06
Section: L_Genetics_Origins
Keywords: South Asian genetics, Indian subcontinent, ANI, ASI, Ancestral North Indian, Ancestral South Indian, Indo-European migration, Steppe pastoralists, Harappan, Indus Valley Civilization, caste genetics, endogamy, founder events, mtDNA, Y chromosome, R1a, Dravidian, tribal populations, Rakhigarhi, AASI
Category Tags: genetics, human-origins, civilization
Cross-References: L_1_06 — Human Migration Synthesis · L_1_09 — Ghost Populations · L_2_07 — European Genetics · F_4_11 — Indo-European Migrations · W_1_03 — Harappan / Indus Valley Civilization
Reliability Tier: Tier 1 (well-supported by aDNA, modern genomics, archaeology, and linguistics)
Last Updated: Mar 9, 2026 | Source Count: 14 | Weighted Score: 38 | Source Confidence: [4/5] | Confidence: High

QUICK SUMMARY

South Asia harbors one of the most genetically diverse and internally structured population histories of any world region, reflecting deep settlement, repeated admixture, and long periods of extreme endogamy. The best-supported broad model uses at least three major ancestry streams: (1) AASI-related ancestry — representing the deepest layer of South Asian ancestry and inferred from population-genetic models rather than directly sampled prehistoric South Asian hunter-gatherer genomes; (2) Iranian-related ancestry — present in populations associated with the Indus/Harappan world and closely related to, but not identical with, early populations from the Iranian plateau; and (3) Steppe pastoralist-related ancestry — arriving after the decline of the urban Indus civilization and strongly associated with later Indo-Iranian dispersals.

The landmark study by Reich et al. (2009) established the ANI/ASI framework, but more recent ancient-DNA work has shown that those labels are endpoints of later mixtures, not pristine ancient populations. Narasimhan et al. (2019) and Shinde et al. (2019) support a model in which populations related to the Indus Periphery and Harappan sphere lacked detectable Steppe ancestry during the Mature Harappan phase, while Steppe-related ancestry entered South Asia later and mixed differentially across regions and status groups. At the same time, caste endogamy and founder effects intensified genetic drift over the last ~2,000 years, meaning that many present-day differences between groups reflect social isolation layered on top of older admixture rather than separate biological origins.

1. VERIFIED CLAIMS (Tier 1 — Peer-Reviewed / Established)

1.1 Three ancestral components of South Asian ancestry

• AASI (Ancient Ancestral South Indian): The deepest layer — inferred from models using Andamanese hunter-gatherers as a proxy, not from direct prehistoric South Asian hunter-gatherer genomes; represents a deeply diverged eastern non-African lineage likely present in South Asia since ~50,000+ BP.
• Iranian-related ancestry: Associated with populations linked to the Indus sphere and earlier South Asian population formation; shares ancestry with Neolithic groups from the Iranian plateau but is not identical to a simple migration of Zagros farmers into India.
• Steppe pastoralist (Yamnaya/Sintashta-related): Arrived ~2000–1500 BCE; highest proportions in upper-caste groups of northern India (20–30%); strongly correlated with Y-chromosome R1a-Z93 (Indo-Iranian branch); brought Indo-European languages.
• ANI/ASI model (Reich et al., 2009): Still useful as a shorthand for the main modern Indian cline, but later work shows both ANI and ASI were themselves mixed populations formed after the decline of the Indus civilization rather than untouched ancient source groups.

1.2 Ancient DNA from the Indus Valley Periphery

• Rakhigarhi (~2500 BCE): Ancient genome from a Harappan burial (Shinde et al., 2019) — showed Iranian farmer-related + AASI ancestry with no detectable Steppe ancestry; consistent with an in-situ development of the Harappan civilization without Central Asian nomadic input.
• Narasimhan et al. (2019 — Science, 523 ancient genomes): Comprehensive study showing that "Indus Periphery" individuals cluster with a mix of Iranian-related + AASI-related ancestry; Steppe ancestry appears in South Asian contexts only after ~2000 BCE — well after the urban phase of the Indus Valley Civilization.
• Implications: The Harappan/Indus world did not require Steppe ancestry for its formation. At the same time, because ancient DNA from the civilization itself remains sparse, the genetic profile of the IVC is inferred from a small number of direct and indirect samples rather than a dense time series across multiple Harappan cities.

1.3 Steppe migration and Indo-European language spread

• Y-chromosome R1a-Z93: Found at high frequency (30–70%) in upper-caste males and in Indo-European-speaking populations of South Asia; R1a-Z93 is the Indo-Iranian branch (distinct from R1a-Z282 in eastern Europe); highest ancient frequencies in Sintashta (~2100–1800 BCE, southern Urals) — the likely proximate source.
• Autosomal Steppe ancestry: Correlates with caste rank — highest in Brahmins (20–30%), lower in Dalits and tribal groups (~5–10%); also correlates with Indo-European language affiliation vs. Dravidian language affiliation.
• Timing: Steppe ancestry arrival in South Asia dated to ~2000–1500 BCE by Narasimhan et al. (2019), coinciding with the late/post-urban phase of the Harappan civilization and consistent with early Indo-Iranian dispersals.
• Best-supported interpretation: Genetics strongly supports an external Steppe-related contribution to later northern South Asian populations, and that signal is a major line of evidence in favor of Steppe-linked Indo-Iranian language spread. It is stronger evidence for population movement than for a simple one-to-one mapping of genes onto any single text, culture, or caste identity.

1.4 Endogamy and caste-driven founder effects

• Moorjani et al. (2013): Used linkage disequilibrium decay to date the onset of endogamy in dozens of South Asian groups → many groups began practicing strict endogamy ~1,500–2,000 years ago (roughly corresponding to codification in Dharmaśāstra texts); some groups even earlier.
• Founder effects: ~25% of South Asian groups show founder effects as severe as those in Finns or Ashkenazi Jews; effective population sizes as low as a few hundred individuals despite census populations in the millions (Nakatsuka et al., 2017).
• Medical implications: Extreme founder effects concentrate specific disease alleles — high frequencies of alleles for sickle cell disease (tribal populations), certain cardiomyopathies (MYBPC3 25-bp deletion — ~4% of South Asians, ~8% in some endogamous groups; Dhandapany et al., 2009), and other recessive conditions.

1.5 South Asia includes additional ancestry layers beyond the main Indian cline

• Austroasiatic-speaking groups: Some populations do not fit cleanly on the simplified ANI-ASI cline and preserve signals of ancestry profiles with more AASI-related ancestry than reconstructed ASI, showing that peninsular and eastern South Asia remained heterogeneous after the Harappan decline.
• Tibeto-Burman and northeastern groups: Many northeastern South Asian populations carry substantial East/Southeast Asian-related ancestry not captured by a simple three-part summary focused on the Indo-Gangetic and peninsular core.
• South India remained connected, not isolated: More recent work on ancient and historical South Indian ancestry continues to recover West Eurasian-related genetic inputs in regionally variable proportions, reinforcing that southern South Asia was shaped by admixture rather than by complete isolation from wider Bronze and Iron Age population movements.
• Implication: The standard three-component model is a useful scaffold for much of South Asia, but it does not erase later and regionally important gene flow from the northeast or the survival of populations that fall partly outside the "Modern Indian Cline."

2. CREDIBLE BUT DEBATED CLAIMS (Tier 2 — Academic / Debated)

2.1 Timing and route of initial AASI settlement

• The earliest modern human occupation of South Asia is poorly dated archaeologically — few sites with secure dates >40,000 BP exist; some Middle Stone Age-like assemblages at Attirampakkam, India, have been dated to ~385,000 BP (Acheulean-to-Middle Paleolithic), but whether these represent modern humans is uncertain.
• Genetic divergence dates between AASI-related and other Out-of-Africa populations suggest arrival ~50,000–65,000 BP, but direct ancient DNA from early South Asian sites is largely absent due to poor tropical preservation; as a result, the deepest layer of South Asian ancestry is still reconstructed indirectly.

2.2 Nature of the "Steppe migration" — invasion, migration, or elite diffusion?

• The strong sex-biased admixture (Steppe ancestry much higher on the Y chromosome than mtDNA) suggests a predominantly male-mediated migration — consistent with pastoralist elite groups integrating into existing populations.
• Whether this represents large-scale population replacement, gradual diffusion, or small-group elite dominance remains debated; autosomal Steppe proportions of 10–30% suggest significant but not overwhelming gene flow.
• Political sensitivity: In India, the question of Indo-Aryan migration vs. indigenous origin remains culturally and politically charged; genetic evidence strongly supports external entry of Steppe ancestry.

2.3 Dravidian language origins

• Dravidian languages (Tamil, Telugu, Kannada, Malayalam, etc.) are often linked to populations rich in Indus Periphery-related + AASI-related ancestry, and Narasimhan et al. (2019) regarded an Indus-linked spread of proto-Dravidian as plausible.
• However, no ancient inscription securely ties the Indus script to Dravidian, and genetics alone cannot demonstrate that Harappans spoke proto-Dravidian.
• The Brahui language (Dravidian isolate in Pakistan/Balochistan) is suggestive but not decisive; its speakers are not a frozen Harappan relic population and show substantial later admixture.

2.4 Iranian-related ancestry does not equal direct descent from western Iranian farmers

• Later work on West Asia, including Lazaridis et al. (2022), shows that ancient West Asian populations were themselves structured and mobile; therefore, labels like "Iranian farmer-related" should be read as ancestry affinity, not as proof of a single, clean migration of a known farming people from the Zagros into South Asia.
• This matters because popular retellings often compress a complex gradient into a misleading story of one incoming "Iranian farmer" population replacing local groups.

3. SPECULATIVE CLAIMS (Tier 3 — Possible but Unverified)

3.1 Indus Valley script and language

The undeciphered Indus script — if the language was Dravidian (one leading hypothesis), it would support the association of the Iranian farmer + AASI component with Dravidian linguistic ancestry; other proposals include an unknown language family; cannot be resolved without decipherment.

3.2 Ancient connection to Southeast Asian Hoabinhian

Some genetic analyses suggest AASI-related ancestry may extend to Hoabinhian hunter-gatherers of Southeast Asia (McColl et al., 2018), pointing to a broadly distributed pre-Neolithic population across southern Asia; however, the relationship is distant and poorly resolved.

4. DUBIOUS OR FRINGE CLAIMS (Tier 4 — No Credible Source / Contradicted by Evidence)

4.1 "Out of India" theory for Indo-European languages

The claim that the Indo-European language family originated in India and spread outward — contradicted by linguistic phylogenetics (Anatolian and Tocharian outgroups), archaeological evidence (Yamnaya/Sintashta expansion), and genetic evidence (Steppe ancestry absent in Indus Valley Periphery individuals pre-2000 BCE; appears in South Asia only after ~2000 BCE).

4.2 Genetic basis of caste as natural hierarchy

Claims that caste reflects innate biological differences — not supported by genetics; caste correlates with varying proportions of the same three ancestral components, and differences in allele frequencies result from genetic drift under endogamy, not from functional adaptation to caste roles.

COUNTER-ARGUMENTS / LIMITATIONS

• AASI is still an inferred ancestry, not a sampled ancient population: Much of the deepest South Asian model depends on proxies such as Andamanese-related ancestry because tropical preservation has left a severe ancient-DNA gap for prehistoric South Asia.
• The Harappan genetic profile is still thinly sampled: Rakhigarhi is important, but one direct Harappan genome and a limited number of Indus-periphery individuals cannot capture the full diversity of a civilization that covered a vast area for centuries.
• Language cannot be read straight from ancestry: Steppe ancestry is highly relevant to Indo-Iranian dispersal, but genetics alone cannot prove exactly which language every Bronze Age population spoke, and Dravidian/Indus language claims remain inferential.
• Caste patterns are historically layered: Present-day caste differentiation reflects older admixture, later social closure, and region-specific drift. It should not be mistaken for evidence of timeless or biologically fixed social categories.
• Geographic focus can distort the model: The classic ANI-ASI discussion works best for much of the Indian cline, but northeastern South Asia and some Austroasiatic-speaking populations carry ancestry components that do not fit neatly inside that simplified framework.

IMAGES

BIBLIOGRAPHY

1. Narasimhan, Vagheesh M., et al. eaat7487 | 2019 | "The Formation of Human Populations in South and Central Asia" | Science | ∅ | 365:: | ∅ | ∅ | doi:10.1126/science.aat7487 | ∅ | ∅ | ∅
2. Reich, David, et al | 2009 | "Reconstructing Indian Population History" | Nature | ∅ | 461::489–494 | ∅ | ∅ | doi:10.1038/nature08365 | ∅ | ∅ | ∅
3. Moorjani, Priya, et al | 2013 | "Genetic Evidence for Recent Population Mixture in India" | American Journal of Human Genetics | ∅ | 93::422–438 | ∅ | ∅ | doi:10.1016/j.ajhg.2013.07.006 | ∅ | ∅ | ∅
4. Shinde, Vasant, et al | 2019 | "An Ancient Harappan Genome Lacks Ancestry from Steppe Pastoralists or Iranian Farmers" | Cell | ∅ | 179::729–735 | ∅ | ∅ | doi:10.1016/j.cell.2019.08.048 | ∅ | ∅ | ∅
5. Nakatsuka, Nathan, et al | 2017 | "The Promise of Discovering Population-Specific Disease-Associated Genes in South Asia" | Nature Genetics | ∅ | 49::1403–1407 | ∅ | ∅ | doi:10.1038/ng.3917 | ∅ | ∅ | ∅
6. Underhill, Peter A., et al | 2015 | "The Phylogenetic and Geographic Structure of Y-Chromosome Haplogroup R1a" | European Journal of Human Genetics | ∅ | 23::124–131 | ∅ | ∅ | doi:10.1038/ejhg.2014.50 | ∅ | ∅ | ∅
7. Lazaridis, Iosif, et al. eabm4247 | 2022 | "The Genetic History of the Southern Arc: A Bridge between West Asia and Europe" | Science | ∅ | 377:: | ∅ | ∅ | doi:10.1126/science.abm4247 | ∅ | ∅ | ∅
8. McColl, Hugh, et al | 2018 | "The Prehistoric Peopling of Southeast Asia" | Science | ∅ | 361::88–92 | ∅ | ∅ | doi:10.1126/science.aat3628 | ∅ | ∅ | ∅
9. Dhandapany, Perundurai S., et al | 2009 | "A Common MYBPC3 (Cardiac Myosin Binding Protein C) Variant Associated with Cardiomyopathies in South Asia" | Nature Genetics | ∅ | 41::187–191 | ∅ | ∅ | doi:10.1038/ng.309 | ∅ | ∅ | ∅
10. Majumder, Partha P | 2010 | "The Human Genetic History of South Asia" | Current Biology | ∅ | 20::R184–R187 | ∅ | ∅ | doi:10.1016/j.cub.2009.11.053 | ∅ | ∅ | ∅
11. Metspalu, Mait, Mayukh Mondal; Gyaneshwer Chaubey | 2018 | "The Genetic Makings of South Asia" | Current Opinion in Genetics & Development | ∅ | 53::128–133 | ∅ | ∅ | doi:10.1016/j.gde.2018.09.003 | ∅ | ∅ | ∅
12. Ahlawat, Bhavna, et al | 2023 | "Deciphering the West Eurasian Genetic Footprints in Ancient South India" | Genes | ∅ | 14::963 | ∅ | ∅ | doi:10.3390/genes14050963 | ∅ | ∅ | ∅
13. Silva, Marina, et al | 2019 | "Untangling Neolithic and Bronze Age Mitochondrial Lineages in South Asia" | Annals of Human Biology | ∅ | 46::140–144 | ∅ | ∅ | doi:10.1080/03014460.2019.1623319 | ∅ | ∅ | ∅
14. Chakraborty, Subhendu; Analabha Basu | 2019 | "Reconstruction of Ancestral Footfalls in South Asia Using Genomic Data" | Journal of Biosciences | ∅ | 44::74 | ∅ | ∅ | ∅ | ∅ | ∅ | ∅

CROSS-REFERENCE INDEX

• L_1_06 — Human Migration Synthesis: Global dispersal background and steppe-linked migrations
• L_1_09 — Ghost Populations: AASI and proxy-based deep ancestry inference
• L_2_07 — European Genetics: Shared Steppe ancestry, Indo-European expansion
• F_4_11 — Indo-European Migrations: Steppe ancestry and Indo-Iranian dispersals
• W_1_03 — Harappan / Indus Valley Civilization: Archaeological context for Rakhigarhi and the IVC decline

Last verified: Mar 09, 2026 — Broader internal cross-checking and external literature review added; all listed sources are peer-reviewed or established academic literature