Re: An aesthetic question

Ok, thank y'all for the opinions.
I've given it some thought, and I'll try to address all your points best I
can:

On vowel lowering/nasality transference:
I rather like these ideas.  I think that finally, the system will actually
be a laxifying system, where /i~/ is rendered as [I~] in all cases, and /u~/
as [U~].  (/i/ is already [I] in unstressed positions, so this isn't a wild
stretch, and the laxed vowels are more comfortable for me, and don't lose
distinctness so much as E,O would.  Joy.)
I think I'll also adopt this nasal mutation thing, in two forms.
So, finally, rules:
If a nasalized high vowel precedes a voiced stop, then it denasalizes and
the stop becomes a stop+nasal cluster.
Otherwise, it is laxed.
An internasal voiced stop, affricative, or fricative becomes nasal.  Stops
and affricatives become normal nasals, fricatives simply assume
nasalization.  Voiced h ignores this rule.

In stems, this is not reflected orthographically.  In fact, stems can
violate the last rule freely.  (There are actually a great number of rules
that stems can ignore; it makes me feel guilty, but makes for a greater
variety of words.)
However, when a morphological process triggers a rule, then spelling does
change, in accord with all the other phonological rules.  Thus, if one
wanted to say "hot water", one could say àànnà kulèdekk, water
hot-Vkk(3p-nom-present tense), or strip the inflection -Vkk from kulèd and
affix it to àànnà, yielding kulènàànnà (grave marks nasalization).  Dhan
astthììkk compounds to astthinhan, creating the interesting phenomenon of an
aspirated nasal, something that's non-phonemic, but it seems it will be
appearing, as all processes retain aspiration if they can.  (<th> and <dh>,
while technically the fricatives /T/ and /D/, are treated by the phonology
as aspirates.)  Quite possibly, this might lead me to turning all fricatives
into aspirated nasals instead of making them into nasal fricatives, which
are mildly inelegant. This would cause confusion between astthì+zan and
astthì+dhan (though at the moment this doesn't matter as none of those three
words is more than a meaningless noise), but one would imagine that context
can disambiguate.

As for orthography, that's pretty well-behaved, though I'm forced to use <.>
as a marker for retroflexes in ascii.  Otherwise, long vowels are
circumflexed, nasal graved, and long nasal doubled and graved.

Again, thank you.  That discussion resulted in something unexpectedly
graceful, which is always a blessing.

---
Shreyas