Questions on word-final vowels in Sanskrit

[Madhav Deshpande]

One possibility to consider is that Sanskrit words passed into Tamil
through their somewhat Prakrit-like pronunciation.  In many Prakrits, the
nominative singular of Sanskrit (masc.) words in -a ends in -o.  This o
vowel is often reduced to u.  One can see this already in Buddhist Hybrid
Sanskrit continuing all the way into Old Marathi of JnAnezvar in the 13th
century A.D.
                                        Madhav Deshpande

On Wed, 25 Nov 1998, Sudalaimuthu Palaniappan wrote:

> Some of the words borrowed from Sanskrit into Tamil give information about how
> Tamils perceived the pronunciation of some Sanskrit sounds. (If there was no
> difference between their representation in Tamil, their perception of the
> Sanskrit sound and the Sanskrit sound in reality, then it might say something
> about the pronunciation of Sanskrit itself in Tamilnadu more than 1100 years
> ago.) An analysis of pairs of words where most IA scholars agree on the
> direction of borrowing as from Dravidian into Sanskrit suggest that the
> converse may also be true.
>
> Consider the following.
>
> The word-final "a" in Sanskrit seems to have been perceived by Tamils in many
> cases as close to the extra-short "u" in Tamil. For instance, the word "soma"
> in Sanskrit has been rendered by periyAzvar as "cOmu" in the following line.
>
> tUya nAn2maRaiyALar cOmuc ceyya  (Periya tirumoLi 2.10.1.6 or nAl. 1138.6)
>
> (translation: with the pure ones of the four vedas performing soma)
>
> Another interesting case seems to be where Sanskrit "kapAla" is rendered in
> Tamil as "kapAl" as shown below.
>
> vaittavan2 kapAl micai                            (tiruccanta viruttam 42.4 or
> nAl. 793.4)
>
> (translation: having placed on his skull)
>
> This seems to be very unusual. Normally, one would expect "kapAlam". The
> absence of any sound after "l" in the Tamil word suggests that the Sanskrit
> final "a" was so imperceptible as to justify removing it altogether.
> Alternatively, one could interpret it as the author hearing it as "kapAlu" and
> removing the final "u" to make it sound literary. (Many literary Tamil words
> are pronounced colloquially with an enunciative vowel often represented as
> extra-short "u".)
>
> Interestingly, there are Sanskrit words accepted by Indo-European scholars as
> borrowing from Dravidian which show a final "a" while the equivalent Tamil
> forms have final extra-short "u" or expected to have an enunciative "u"
> colloquially as shown by the pair, Sanskrit karambha and Tamil kuzampu or
> Sanskrit kuvalaya and Tamil kuvaLai (equivalent in pronunciation to kuvaLay).
> (This does not mean Sanskrit word was borrowed from Tamil.)
>
> I would like more information on the pronunciation of word-final vowels in
> Sanskrit and what happened to them in MIA and NIA? Is there any research on
> the presence or absence of enunciative vowels and their conditioning in
> Sanskrit? Why is there such a mapping between Sanskrit -a and Tamil
> zero/extra-short u?
>
> Thanks in advance.
>
> Regards
> S. Palaniappan
>