Re: New Pseudo-Dravidian Conlang

I've also thought some more about the internal history of the language,
particularly the assimilations that it went through.  In my first post, I
noted that the earlier syllable structure was CV(R)(S); that is, words like
*karkka and *pamppu could exist at that stage.  However, the 'modern'
language has a syllable structure CV(R/F,S), which is more complex (in
terms of rules) and simpler (in terms of permissible clusters).  The
question is, how to get from the first structure to the second?  Here's
what I have in mind:

1. Nasal + stop => nasal vowel + stop; e.g. *pamppu => päppu (where <ä>
denotes /a~/).  This is where (most of) the nasal vowels come from.
2. Rhotic + stop => lengthened vowel + stop; e.g. *karkka => kaakka.  If
there's already a long vowel, the rhotic becomes /u/; e.g. *ceer.t.t.i =>
ceeut.t.i.
3. Lateral + stop => /u/ + stop; e.g. *pelkku => peukku.  This change
occurs even when the vowel is already long, so *ceer.t.t.i and *ceel.t.t.i
would have the same outcome, ceeut.t.i.

Finally, I have a few thoughts about sandhi in the language.  The idea is
that word-final nasals will be pronounced in isolation, but are often
elided, with compensatory nasalization of the vowel, when preceding a
consonant-initial word, e.g.:

sampusan koyur. = ['sam.pu.sa~ 'ko.jur`]

In some dialects, the nasalized back vowels are fronted: /a~/ =
[{(~)], /o~/ = [2(~)], /u~/ = [y(~)].  (This opens up the possibility of
later vowel harmony... *evil grin*)

- Rob

Reply