--- In qalam@yahoogroups.com, "Peter T. Daniels" <grammatim@...> wrote:

> If you will consult e.g. Masica, *The Indo-Aryan Languages*, you

will learn that Prakrits basically didn't have clusters, hence the
rarity of Kharaosthi conjuncts in general, and the nonexistence of CVC
syllables.

Well, I suppose you don't count the Gandhari Prakrit or Pali as
prakrits. I won't dispute the rarity of CVC (other than via gemimates
and homorganic nasals, whose ignoring can be justified), though,
possibly under Sanskrit influence, Pali certainly has them, e.g.
_tasma:_ therefore and initial vy-/by-. I lack data for Gandhari,
though I am aware of 'Gandhari metathesis' (CVCrV... > CrVCV...).

> Similarly, there was no need to mark words as final-vowelless, as

Prakrit words did not end with a consonant.

But Kharoshthi was also used for Sanskrit, and, though confirmation is
hard to come by, Tocharian B. Tocharian B had non-trivial CVC and
final consonants, though that may not be relevant.

There a collection of conjuncts in
http://www.andrewglass.org/downloads/Glass_2000.pdf for anyone who
wants to go beyond the Unicode-related sources. The section header
'virama' actually refers to syllable-final consonants, though that
merely adds that the examples come from Niya, so presumably not
Tocharian. There's also a section arguing that anusvara is not
subscript <m>.

Richard.