Kalini Sapak bits (or How to buse the letter "X" ...)

For reasons better not goten into, I've finally decided on Kalini Sapak's
phonemic inventory:

a i u

p t k
b d g
f s x h
v z G
m n N

w j
l r

Nothing fancy here (but see on assimilations below ...), but romanization
invloves a trio of problems: /x G N/.

Now, I don't want to use any digraphs (esp as that's only 23 phonemes and
the Latin alphabet 's got 26 letters). My current idea is "q" for /x/
(Somali does this), "c" for /G/ (after-all, "c" is related to gamma ...)
and, take a deep breath, "x" for /N/. I know this last is pretty
counter-intuitive, but "x" is used for a bit of everything around the world,
so why not a velar nasal? Also, pretty much everything else is already used.
Anyone got any comments/ideas?

The rest are going to be the same as IPA, except /j/ which'll be written "y"
for aesthetical reasons.

---

Since assimilation in Tairezazh is pretty much confined to draconian sandhi,
I wanted Kalini Sapak to do something more. First, the semivowels /y/ and
/w/ (for simplicity's sake I'll use romanization within slashes for phonemic
annotation, since there's an absolute one-to-one correspondence between
romanized letters and phonemes) affect the quality of preceding vowels.
Examples:

/yawal/ "star (ACC)" is ['jQwal]
/yiwal/ "starrier" is ['jywal]

/qayak/ "horse" is ['xEjak]
/quyaku/ "(he) rides" is ['xyjaku]

When the semivowel is followed by another vowel, /y/ don't affect /i/ and
/w/ don't affect /u/. When the semivowel is followed by a consonant, it's
lost and vowel lengthing occurs:

/aywal/ "starry sky" is ['e:wal] (also fem name)
/qiyk/ "horse-like" is [xi:k]
/uyC/ should yield [y:C] (no example yet, C=any consonant)

/yawl-/ "star (compound form)" is [jo:l]
/yiwl/ "starry" is [jy:l]
/uwC/ should yield [u:C] (no example yet)

/s/ and /z/ becomes are realized as [S] and [Z] before /i/:

/kazal/ "woman (ACC)" is ['kazal], but
/kazil/ "woman's" is ['kaZil]

/sapak/ "language (ACC)" is [sapak], but
/sipk/ "of language, linguistical" is [Sip@k] (final [-pk] is "illegal",
therefore an epenthic schwa is inserted)

In all likelihood, I'll come up with more, but what ya think this far?

                                                  Andreas


_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp

Replies