Where the Retroflex Stops Come From

Retroflex consonants are pervasive in Sanskrit and they are not straightforwardly inherited. Most of the retroflex stops in the classical language can be traced to specific sources: to clusters that assimilated, to the effect of a neighbouring retroflex sibilant, and to a substantial body of vocabulary that Indo-Aryan took up from the other language families of the subcontinent. A great many Sanskrit words containing a retroflex have no convincing Indo-European etymology, and the concentration of the sounds in that part of the vocabulary is one of the central pieces of evidence in a long argument about how much Indo-Aryan owes to its neighbours.

That argument is genuinely unsettled and the atlas records it as such elsewhere. What is not in dispute is that the sounds are there, that they are shared across the Indo-Aryan and Dravidian families alike, and that they are one of the features that make South Asia a linguistic area rather than merely a place where several families happen to be spoken. Brahmi gives the whole class five letters of its own, which is a provision no inventory borrowed unchanged from western Asia could have made, since nothing written with those scripts needed it. The class is written and read exactly as the other four are, with the same four stops and a nasal.