Kolami

During 1935-38 he visited India and did extensive fieldwork on the language and culture of several nonliterary Dravidian languages of South and Central India, mostly Toda and Kota and for a shorter period Badaga in the Nilgiri hills, Kodagu in Karnataka, and Kolami in Central India.

Krishnamurti suggests the conditional clitic =te in Kolami (337) comes directly from the Telugu conditional suffix -te, suggesting Kolami speakers somehow parsed a bound verb suffix and reanalyzed it as a postclitic particle; Steever (1993: 125ff.) provides an alternative etymology in which the Kolami clitic results from contraction of a pre-Kolami form *ate 'if X becomes'.

Using Steever's (1993: 4) names, three distinct and coherent groups are universally recognized: 1) South: Tamil, Malayalam, Irula, Kodagu, Toda, Kota, Kannada, and Tulu; 2) South-Central: Telugu, Gondi, Konda, Kui, Kuvi, Pengo, and Manda; 3) Central: Kolami, Naiki, Parji, Ollari, and Gadaba.

6, 1.4: for DEDR 4240 'smoke', Tamil pukai, Kannada poge, Telugu poga, it is doubtful whether South Dravidian (plus Telugu) can be reconstructed beyond the problematic *pulokay (placed in DEDR as if with *u), but o and o found in Kolami pog, Gondi pog-, poy-, and Gadaba pog-, Kui pok-, Kuwi boy- are evidence for Proto-Dravidian *pok- and *pok-.