On Sat, 4 Sep 2010 08:29:26 -0700, Leila Kalomi wrote:
>So, what do y'all think of my phonology? Are there any sounds it needs/doesn't
(updated version)
>Stops: bilabial, alveolar, velar (romanized as p, t, k)
>Affricate: alveolar (ts)
>Flaps: lateral alveolar, alveolar (rh, r)
>Fricatives: bilabial, alveolar, lateral alveolar, velar (f, s, lh, h)
>Nasals: bilabial, alveolar, velar (m, n, ng)
>Approximants: bilabial, lateral alveolar, palatal (w, l, y)
>Vowels: open back unrounded, open-mid front unrounded, near-close near-front
>unrounded, close-mid back rounded, near-close near-back (a, e, i, o, u)

An original inventory without being overly contrived. I don't think I've
seen a contrast of central/lateral flaps before (while surely that's no
worse, tho, than the contrast between central/lateral approximants found in
eg. English).

>Phonotactics: All combinations of vowels are allowed, and diphthongized or not
>depending on how fast one is talking. Two-consonant clusters are allowed
>anywhere but the beginning and end of a word, and consonants other than
>fricatives and nasals may not end a word. In compounding words, nasals
>assimilate to following stops/affricates/fricatives, which are then voiced
>(though voicing is not phonemic.)
>*pants* And I think that's it. Thoughts?

Beyond banning consonant+x as you mention'd, have you thought about
constraints on allowed consonant clusters, or is anything else from -yng- to
-rhl- to geminates fine?

BTW, I would say <rh> as lateral but <lh> as voiceless isn't consistent
either. By contrast, if <h> is /x/, <nh> makes equally much sense for /N/ as
<ng> does, and would allow you to do away with <g>.

There's always also the "just fill the slots with leftover letters"
approach, which would here probably be something like <z x q g> for /ts K l\
N/ (your current <ts lh rh ng>, if you're not familiar with X-SAMPA).

As for phonetic details, does the assimilation of NP clusters involve just
place assimilation (nk > [Ng]) or total assimilation (nk > [gg])?

John Vertical