Indian onager

[Venkatraman Iyer]

>I would appreciate guidance on predicting the ancient form, say
>of ca. 3000 BC in Rajasthan, of the related lexemes. CDIAL 4054 has
> gardabha (Skt.). DEDR 1364 has Parji gade, garad; Gondi ga_r.di;
> Kuwi ga_r.de; Te. ga_d.ida; Ta. kar..utai.
> Is the consonantal change k>g or g>k?

The change is from 'k' to 'g'. The other way as in the Telugu voicing
is secondary and much later. Bh. Krishnamurthy in his Telugu verbal
 bases, p. 24,
"Caldwell's assumption that there were no initial voiced plosives in
PDr.as reflected in Ta. and Ma. was contradicted by Jules Bloch on the
 basis of the history of a few words like ghOTaka horse and drAviDa
name of a people, which, according to him, were originally native to
Dravidian. This view rceived support from some quarters and
contradiction from others. Burrow argued for Caldwell and analytically
put forth evidence to show that voicing of initial stops appearing in
some of the Dravidian languages, was only secondary and did
not represent PDr."

Jules Bloch wrote about drAvida as the original word predating the
word "tamizh". Zvelebil has analyzed this problem and shows
that drAvida is ultimately derived from the root word, "tamizh".

Hence, the consonantal change is k > g.

Sincerely,
V. Iyer


Get Your Private, Free Email at http://www.hotmail.com