The Elamo-Dravidian languages are a hypothesised language family which includes the living Dravidian languages of India, Sri Lanka and Pakistan, in addition to the extinct Elamite language of ancient Elam, in what is now southwestern Iran. Linguist David McAlpin has been a chief proponent of the Elamo-Dravidian Hypothesis. In the light of the Elamo-Dravidian hypothesis, it has been suggested that the extinct Harappan language (the language or languages of the Indus Valley Civilization) may be part of the same family.

Evidence

McAlpin (1975) identified several similarities between Elamite and Dravidian. According to McAlpin, 20% of Dravidian and Elamite vocabulary are cognates; a further 12% are probable cognates. Elamite and Dravidian possess similar second-person pronouns and parallel case endings. They have identical derivatives, abstract nouns, and the same verb stem+tense marker+personal ending structure. Both have two positive tenses, a "past" and a "non-past".
The Elamo-Dravidian Hypothesis is based on several other pieces of evidence. It appears that agriculture developed in the Near East and later spread to the Indus Valley region, suggesting that Elamo-Dravidian agriculturalists may have brought farming from the Near East to the Indus Valley. Later evidence of extensive trade between Elam and the Indus Valley Civilization suggests ongoing links between the two regions. Proponents of the hypothesis noted similarities between the early Harappan script, which has not been definitively deciphered, and early (Proto-)Elamite script that, however, is hardly understood so far.

The disjunct distribution of living Dravidian languages, concentrated mostly in southern India but with isolated pockets in South Eastern Iran and Southern Afghanistan and Pakistan and northeast India, suggests a wider past distribution of the Dravidian languages, and that the Indo-European languages of modern India and Pakistan were later arrivals in the Indo-Gangetic plain, leaving isolated islands of the older Dravidian languages in the surrounding mountains. A variety of Dravidian loan words (i.e., phalam- ripe fruit, khala- threshing floor) in Vedic Sanskrit suggests that the two languages existed for a time in proximity. Retroflex consonants, which exist in Vedic Sanskrit and Dravidian but do not exist in Iranian or European languages could suggest a Dravidian substratum or adstratum in Vedic Sanskrit.

Some who claim to have deciphered the Harappan script, including Asko Parpola and Walter A. Fairservis Jr., suggest that the Harappans spoke a Dravidian language, while others, for instance Shikaripura Ranganatha Rao, suggest that the Harappan script represents an Indo-European language, similar to Sanskrit.

Criticism

The theory has been criticized on linguistic grounds.

In addition, the Dravidian Brahui language of Baluchistan, per McAlpin the supposed link between Elamite and the Central Indian Dravidian languages (), has been suggested by J.H. Elfenbein to be a late, c. 1000 year old immigrant from Central India. As such, it cannot reflect a remnant of a Dravidian language speaking Indus population.
Georgiy Starostin criticized McAlpin's proposed morphological correspondences between Elamite and Dravidian as no closer than correspondences with other nearby language families. Proto-Nostratic was already hypothesized to be an ancestor of Dravidian, and Václav Blažek had proposed that Elamite was related to Afroasiatic, so Starostin performed a lexicostatistical comparison using the Swadesh list between Elamite, Proto-Afroasiatic, Proto-Nostratic (a version of Nostratic not including Afroasiatic, similar to Joseph Greenberg's Eurasiatic), and Proto-Sino-Caucasian. He concluded that Elamite is related to Afroasiatic and Nostratic but not a member of either, with Sino-Caucasian being more distant from those three.