Re: Semetic-style Syllabry (was Re: I'm back!)

Henrik Theiling wrote:
> Oh, that's far less than I had quietly estimated.  Very doable indeed.
> I had expected like 400~500 at least (are there so few consonant
> phonemes or are the combinations restricted?)  Anyway, it would be
> very interesting to see the script in action!  A triconsonantal
> abudiga -- very nice and unique AFAIK! :-)
Both.  The basic permissible syllable types are
(C)(w,j,l)V(fric,nasal,l,gem), with certain restrictions, and the
consonant phonemes are /p t k b d g (tS) (dZ) f s (C) v z m n (N) l w
j/.  The ancestral language had even more restricted syllable types, so
that some syllables in the modern language must be written with two
characters (much as in Japanese kana, one must write ki-(yo) for /kjo/)

/tS/, /dZ/, /C/, and /N/ have become marginal phonemes, due to sound
changes.  Historically, /tS/, /dZ/, and /C/ were the palatized versions
of, respectively, /t/, /d/, /k/.  Later sound changes permitted those
phones to be followed by /i/, so there is now a contrast between /ti/
(realized [tsi] in some dialects) and /tSi/.  However, there is no
distinction in the orthography.  /N/ is derived from /ng/.  /g/ was lost
before /i/, and as a result, sequences such as [aNgi] became [aNi].
However, these are still written as _angi_, slightly problematic as the
same sound changes that created [ti] have allowed for [gi], making
_angi_ ambiguous between /aNi/ and /aNgi/.

Fortunately, the occurrences of [ti~tsi], [di~dzi], [ki], and [gi] are
quite rare.