Indian Antiquary Vol 59

Author(s): Richard Carnac Temple, Charles E A W Oldham, S Krishnaswami Aiyangar, Devadatta Ramkrishna Bhandarka
Publisher: Swati Publications

Page 217

OCTOBER, 1930 ]
THE VELAR ASPIRATE IN DRAVIDIAN
190
The following facts, however, militate against this opinion :
(a) If it had been a borrowing from Sanskrit, it is strange that it should have been borrow. ed only in the very limited contexts in which it occurs in Tamil.
(6) The use of the visarga immediately before a plosive generally brings about in certain contexts phonetic changes in Sanskrit, which convert the aspirate into a different sound altogether, e.g., nih+kama = nişkama; nih+cinta = niścinta, etc.
It is not explained why Tamil should have failed to take note of this change.
(c) There are a number of contexts in the rude uncultivated dialects of Central India. where the presence of the plosive has given rise to an aspirate immediately before the plosive, As in Gôndi plural ending - hk and Gôndi causative ending -ht. The phonetic features of the change in Gôndi are, as will be seen below, exactly on a par with the features characterising the production of the Tamil dydam.
(d) The dydam of Tamil is produced also by Tamil sandhi rules: The groups 1+t and I+t immediately after short vowels in word.combinations change into h and hf in old Tamil, e.g., kal+tidu= kandidu (the stone is a bad thing); mul+tidu=muhdidu (the thorn is & bad thing), etc. The change here is entirely independent of any Sanskrit influence; the initial plosive surd of the second component, in the process of its assimilative change into the alveolar or the cerebral (as the case may be) passes through the fricative stage which has given rise to the slight aspirate.
(e) There are a few derivative old Tamil forms which in contexts similar to the above show the slight aspirate, e.g., ahal (mark), eham (wheel), etc. These are, unlike shdu, etc., without any alternative forms, and could certainly not have been “invented for prosodio lengthening."
It is quite possible, therefore, that the slight aspirate h known as dydam in Tamil was not a borrowing" from Sanskrit. It would probably be more proper to consider it as more or less a native development in Tamil, which was recognized and fixed in the literary dialect, probably by those who were acquainted with the Sanskrit visarga.
As the slight aspirate does not occur in similar contexts in any other dialecta of Dravidian except Gondi, we may presume it to have been a secondary development in Tamil.
(C) Intervocal g in common speech changes into the fricative and often into the aspi. rate in Tamil: pahu (to divide); padahu (boat) ; vehu (to burn); pôhu (to go), etc.
This aspirate, however, is stable only in syllables which carry some amount of emphasis with them ; in unemphasized syllables, the aspirate disappears, sometimes lengthening the adjoining vowel, if it is short, e.g., pagalu (division, etc.) > pahalu > pdlu.
2. Malayalam :(a) The Tamil dydam is not found in Malayalam.
(6) The influertce of Sanskrit on Malayalam has been so great that the intervocal aspirato shows, both in Sanskrit borrowings and in native developments (from intervocal plosives), greater stability than in Tamil; but the general tendency of Dravidian to eliminate the aspi. rate is, nevertheless, observable in changes like the following -vahiyd (not bearable, not permissible) > vaiyd > vayya.
3. Telugu
(a) Telugu does not show the aspirate in native roots or forms; the Tamil dydam is not met with either.
(6) Intervocal aspirates are not developed from the intervocal plosives; in intervooalie positions the plosive never even changes into the fricative,
4. Kannada :(a) The remarks made above in regard to Telugu apply to Kannada also.
(6) A very interesting secondary change in Kannada is the development in middle and modern Kannada of a velar aspirate from p, very commonly in initial positions and less