Made visible · convergence

The deep root of Dravidian may point west — and four lines of evidence have been read that way.

Where did the Dravidian languages come from? For a century the honest answer was "unknown." But the linguistics, the ancient DNA, the Y-chromosome record, and the archaeology have begun to converge on one story — and it does not begin in the south. It begins between the Iranian plateau and the Indus, in the world of ancient Elam.

The linguist's case: Dravidian is cousin to Elamite.

McAlpin · Elamo-Dravidian

In 1974, in the journal Language, David McAlpin demonstrated that the Dravidian family is cognate with Elamite — the ancient language of southwest Iran, spoken at Susa, the rival and neighbour of Sumer. His evidence was not loose resemblance. He paired Elamite words — mostly verb stems, the hardest part of a language to borrow — with Dravidian terms from the Burrow-Emeneau Dravidian Etymological Dictionary, and argued for regular phonological correspondences in the word-roots and in the verbal morphology. Shared grammar, not just shared vocabulary, is what would turn a resemblance into a relationship — and whether McAlpin established it is exactly what the field disputes.The field’s objectionThe specific criticism is not general wariness. Critics hold that the regular sound correspondences required to demonstrate a genetic relationship are absent, and Bhadriraju Krishnamurti — author of the standard comparative grammar of Dravidian — regarded McAlpin’s proposed morphological correspondences as ad hoc and lacking phonological motivation. Elamite is generally treated as a language isolate. This is the strongest counterevidence to the page.

McAlpin, D. W. (1974). Toward Proto-Elamo-Dravidian. Language, 50, 89–101; (1975; 1981) monograph Proto-Elamo-Dravidian. Southworth's Linguistic Archaeology of South Asia (2004) develops the "Proto-Zagrosian" framing. citation checked · not consensus

The geneticists arrived at the same place — twice, independently.

the Koraga component · the L1-M22 lineage

The striking thing is that the DNA evidence was gathered to answer a different question, and it landed on McAlpin's map.

Sequeira et al. (2025), European Journal of Human Genetics (Nature Portfolio) — Koraga Proto-Dravidian ancestry. Peer-reviewed. Co-authored with linguist George van Driem (Univ. Bern); Pathak et al. (2024), iScience — L1-M22 Y-chromosome. Both independently point to a West-Asian / Iranian-plateau origin feeding into the Indus zone. source listed, not accessed

Four disciplines, one origin.

the convergence, stated plainly

A 2025 review in Mother Tongue pulled the threads together and concluded that the recent archaeogenetics support the Proto-Elamo-Dravidian model of McAlpin and Southworth — though that journal sits in the contested long-range-comparison tradition, so the weight here rests on the primary sources (McAlpin in Language, Pathak in iScience), not the review. Set side by side:

Set side by side, the lines are suggestive. They are not independent, and the platform should not have counted them as if they were.The independence test, applied hereThe platform’s own Sources page asks of every citation: “Independence — whether papers cited alongside each other share authors, samples or a laboratory, and so count once rather than several times.” Applied to this table: the L1-M22 paper is titled “…supports the Elamite and Dravidian connection”, and the Koraga paper states its component fits the Elamo-Dravidian theory. Both were interpreted against the hypothesis they are cited as independently confirming. The trade-contact row is the borrowing-and-areal-contact scenario critics propose as the alternative to common descent, so it does not discriminate between the two. What survives is one contested linguistic hypothesis and two genetic datasets whose authors read them as consistent with it. That is worth a page. It is not four sciences agreeing.

What it reframes — the Indus.

the platform's reading · marked as reading

If Elamo-Dravidian holds, one consequence follows, and it is the platform's own reading of the evidence: the Indus-periphery population was plausibly Elamo-Dravidian-speaking — arriving from the Iranian-farmer direction, exactly the ancestry the ancient-DNA studies find in the Indus Periphery Cline. And Dravidian, on this reading, was not born in the south — it was pushed south, as Indo-Aryan spread across the north after the Indus decline, surviving in the peninsula where it had once been part of a far wider northwestern world.

The doubt, kept in view.

what would strengthen or break this

The platform does not hide the seams, because hiding them is what the traditions it studies did. So:

• McAlpin's hypothesis is respected but not universally accepted. Even sympathetic Dravidianists — Burrow, Emeneau, whose dictionary McAlpin used — were wary of any external-relationship theory for Dravidian.
• Elamite itself is imperfectly understood, and its own script (Linear Elamite) was only recently, partially read — with no Indus or Meluhha reference found in it.
• The genetic "Proto-Dravidian" component (the Koraga study) is now peer-reviewed (European Journal of Human Genetics, 2025), co-authored with the mainstream linguist George van Driem — but its funding source is not disclosed in what we could verify, and the platform flags that as an open item rather than assuming. The synthesis review (Narayan) sits in a contested long-range-comparison venue and is used lightly. The load-bearing sources are McAlpin (Language) and the L1-M22 study (iScience).
• What would confirm it: decipherment of the Indus script showing Dravidian structure. What would complicate it: a competing origin for the Iranian-farmer-related ancestry, or the collapse of the Elamite cognate set under scrutiny.

Count (1973), Current Anthropology — Dravidianists' wariness; Blažek (2003) — Elam as bridge; Elamite decipherment status per the platform's Elamite page. source listed, not accessed

What this establishes.

the ledger