Proposed Sound change

Uatakassí has a bias against /u/.  In grammatical suffixes, it's not too
noticeable, but in lexical items it's very noticeable.  I decided to
explain it by borrowings.  Early Common Kassí borrowed a lot of words,
and one of the main sources had very few back vowels.  In that lang, the
reason was that at an earlier stage, short back vowels were fronted, and
then derounded, so /u/ -> /y/ -> /i/ and /o/ -> /2/ -> /e/.

Anyways, I want to balance it out a bit in descendant langs.  One
thought I had was the following:

/i:/ -> /ji/ -> /ju/ (dissimilation) -> /u/

Does this seem reasonable?  I got the idea for the third phase from
Spanish uo -> ue

Early on (in fact, probably from the very beginning of the koine), long
vowels could not exist in closed syllables.  So, /i/ would have three
reflexes:
/i:/ ->> /u/
/i/ (open) -> /i/
/i/ (closed) -> /e/

This would make /u/ a little more common, but do nothing for /o/

Thus, the same word could have an interesting alternation between /u/
/i/ and /e/, for example:

ualíi -> valú
ualíkaf -> valíka [Class I long vowels shorten the vowel and add -k
                   before plural and before genetive and dative
                   suffixes]
ualíiv -> ualív -> valé

Codas were lost, but some trace survived, I haven't figured out the
details.  Probably nasalization in the case of -n, and some kind of
tonal distinctions for other codas, perhaps rising for voiced codas and
falling for voiceless ones, with -l having been vocalized.

--
Dievas dave dantis; Dievas duos duonos
God gave teeth; God will give bread - Lithuanian proverb
ICQ: 18656696
AIM Screen-Name: NikTailor