Brahmi Ttha 𑀞

The aspirated retroflex is not especially common in Sanskrit and is entirely ordinary in the Prakrit of the Ashokan edicts, and the difference is a regular sound change rather than a matter of style. Clusters of a dental stop with a following sibilant, and several related groups besides, resolved into a doubled aspirated retroflex as the language moved from its Old Indo-Aryan stage to its Middle one. Words that a Sanskrit text writes with a cluster of two or three signs the edicts write with this single letter, doubled, and the correspondence is regular enough that an editor can work back from the Prakrit form to the Sanskrit one with some confidence.

That is one of the reasons a reader cannot approach a Brahmi inscription as though it were Sanskrit in an unfamiliar script. The language is different, the sound system is different, and a transliteration that is accurate letter by letter can still be unreadable to someone expecting classical forms. Editions of the edicts therefore carry a grammatical apparatus alongside the text, and the standard handbooks of Indian epigraphy devote as much space to Middle Indo-Aryan phonology as to the shapes of the letters themselves.

The letter also illustrates how much of the earliest Brahmi record is not in the language most readers associate with the script. Sanskrit inscriptions are a later development; the Mauryan material and much of what follows it for several centuries is Middle Indo-Aryan, and the sound systems differ enough that a reader trained only on the classical language will misread ordinary words. The standard handbooks treat this as the first thing to establish about any inscription, before the letters are even discussed.

Whatever the ultimate source of the Brahmi letter shapes, the shape of the inventory is Indian. The retroflex class is the cleanest demonstration. Five signs are provided for a series of sounds that no script of western Asia had any letters for, because no language written with those scripts had the sounds, and the five sit in the table exactly where the analysis puts them, third of the five classes, between the palatals and the dentals in a sequence running from the back of the mouth to the front.

An inventory taken over wholesale from a Semitic model would have had to invent these letters from nothing and find somewhere to put them, and the obvious place would have been the end of the list, which is where such scripts normally file their additions and where Brahmi itself files the retroflex lateral and where Tamil-Brahmi files its three Dravidian consonants. That the retroflex class instead sits in the middle of the order, integrated into the pattern, indicates that the arrangement was worked out as a whole rather than assembled by additions. The question of where the shapes came from remains open and epigraphists continue to argue it; the question of where the structure came from does not.