Development from Old Ulion to Modern Ulion

[Omzinesý]

In this thread, I devide the development of Ulion into two stages: Old Ulion and Modern Ulion.

----------------------------------------------------------------------

Old Ulion phoneme inventory:

p t k
b d g
s
m n
l r
(w ?) j

i u
o
ɛ
ɑ

The allowed syllable structure is (C)(w/j)V(C).
There is no length distinction.

Stress is phonemic and appears on any syllable but usually on one of the last ones.

----------------------------------------------------------------------

Disappearance of voiced plosives in onset

/b/ => /v/
/d/ => /z/ => /r/
/g/ => /j/ before OU *e and *i
/g/ => /ʀ/ before OU *ɑ, *u, and *o

In a chain shift, *r is backed to /ʀ/

----------------------------------------------------------------------



First palatalization:
I decided to go the Indian way.

/t/ becomes an affricate t͡ʃ before front vowels /i/ and /ɛ/.
/s/ also becomes /ʃ/ in that environment.

/k/, on the other hand, is backed to /q/ before /ɑ/

Later, /ɛ/ and /ɑ/ later merge and become /ä/.
/u/ is also fronted to /y/ and later merges with /i/. (In the chain shift, /o/ becomes /u/.)

----------------------------------------------------------------------

Second palatalization:

/t/ or /k/ + /j/ becomes/t͡ɕ/.
/s/+/j/ becomes/ɕ/.

Other vowels preceding /j/ are usually considered codas and deleted like other coda consonants.

----------------------------------------------------------------------

Reduction of codas

Middle-Ulion only had open syllables.

Old Ulion open syllables go on as short vowels in Modern Ulion.
OU /i/ → MO /ɪ/
OU /u/ → /y/ → /i/ → MO /ɪ/
OU /ɛ/ → /ä/ → MU /ä:/ or /ə/ or /ʌ/
OU /ɑ/ → /ä/ → MU /ä:/ or /ə/ or /ʌ/

Any consonant could appear in the coda in OU.

Voiceless plosives /p/ /t/ /k/ became glottal stop /ʔ/ in coda. It later became part of the vowel. The vowel developed like those of open syllables above, though only /ʌ/ is the product of /ɑ/ and /ɛ/ in that context.
So, Modern Ulion has the glottalized vowels /ɪʔ/, /ʊʔ/, and /ʌʔ/.

/ts/ and /ks/, however, make an exception to the rule above and become /t͡s/.

Voiced plosives /b/ /d/ /g/ disappear in coda. They though compensatorily lengthen the predecing vowel.

Nasals nasalized and lengthened the preceding vowel.
I'm still considering how /ɛ̃/ and /ɔ̃/ came about.

Other coda consonants compensatorily lengthened the preceding vowel. Othen they also affected the vowel quality.
/äd/ => /ä:/
/id/ => /i:/
/ud/ => /yd/ => /y:/ => /i:/
/od/ => /ud/ => /u:/
/äb/ => /ɔ:/
/ib/ => /y:/ => /i:/
/ub/ => /yb/ => /y:/ => /i:/
/ob/ => /ub/ => /u:/
/äg/ => /ɛ:/
/ig/ => /i:/
/ub/ => /yg/ => /y:/ => /i:/
/ob/ => /ug/ => /u:/
/äj/ => /ɛ:/
/ij/ => /ej/ => /e:/ => /ɛ:/
/uj/ => /yj/ => /i:/
/oj/ => /u:/




----------------------------------------------------------------------
Reduction of word-final vowels

Unstressed word-final vowels disappear.
Stressed word-final vowels lengthen.

----------------------------------------------------------------------

Consonant clusters

----------------------------------------------------------------------


----------------------------------------------------------------------


To be continued!

[Omzinesý]

Etymologies of words existing in Ulion

pr̠ạlū [pʀʌ'l̰u:] 'language' <= *padatlú
jōri ['jɔ:ri] 'friend' <= *jábra-ja (good-DIM)
sịc [sɪʔt͡s] ' house' <= *síkitsa
kjũvi ['kjũ:ʋi] 'girl' <= *kujúmb(a)-ja
pjēcạ ['pjɛ:t͡sʌʔ] 'family' <= *pajájtsa-k (k 'Plural')
mēsuk [mɛ:'sʊk] 'ground' <= *majsúku
miprãš [mɪ'prɑ̃:ʃ] 'event' <= *mipámpadase
kạrā [kʌ'r̰ɑ:] 'boy' <= *kekadá
prēv [prɛ:v] 'house', 'home' <= *pedágba