On Wed, Feb 29, 2012 at 5:56 AM, Paul Bennett <[log in to unmask]> wrote:
> On Tue, 28 Feb 2012 16:37:40 -0500, Garth Wallace <[log in to unmask]> wrote:
>> Personally, I'd try to keep it as similar to standard Ge'ez as
>> possible, so keeping it syllabic instead of turning it into an
>> alphabet. A virama of some kind could be introduced for consonants not
>> followed by a vowel; a second virama-like diacritic would be used for
>> unreleased final consonants. Geminates would be spelled out (either
>> prepending the same consonant with the plain or unreleased virama,
>> choose your poison).
> Since the /i\/ series is already used in Amharic for what is essentially a
> virama, and there is no appropriate /i\/ phoneme in Hamer, I think that we
> might have found a match.

Yeah, that makes things easy.

> As to the "unreleased" diacritic, I'm thinking "dot above" might be enough.
>> I'm not quite sure I understand how the vowels work from your previous
>> post. Can all of those vowels follow any consonant?
> Yes.
> There appear to be 19 vowels, of which 14 can appear in any given word.
> i i: e e: a a: o o: u u: E O a; i; - can appear in any word
> i_? i_?: e_? e_?: a_? a_?: o_? o_?: u_? u_?: - Category I words only
> i_h i_h: e_h e_h: a_h a_h: o_h o_h: u_h u_h: - Category II words only

Okay, so there's glottalization/laryngealization harmony, and a set of
vowels that don't take part in the harmony? How predictable is the

>> Are
>> long vowels distinguished from sequences of two short vowels of the
>> same quality?
> I don't think so. However, clusters of at least three consecutive vowels are
> documented.
> The Category IV vowels and diphthongs appear to arise under these
> circumstances, too. I need to get back and re-read that section.
>> You mention diphthongs but don't list them; are they
>> also distinguished from sequences of their components?
> This is not documented, but the answer appears to be "no".
>> If the answer
>> to both of those are "no", then long vowels could be spelled by
>> appending a vowel-only character of the same quality, while and
>> diphthongs would append vowel-only characters of a different quality.
> A vowel-only character? I suppose we'd need to choose to use either the
> Aleph or Ayin row and stick with one of them. I suggest the Aleph row, but
> I'm not tied to it.

That's what I meant. Whichever row(s) you decided would be used for
onsetless vowels.

>> If Category IV vowels are reduced from certain sequences, they could
>> just be spelled as if they were unreduced (like diphthongs).
> This might be a winner, depending on the native intuition. If we're sticking
> with Aleph for Categories I, II, and III, we could fall back to Ayin for IV
> if the native intuition is that they are un-analyzable.

Even if they're considered unanalyzable, digraphs may not be a bad
thing, particularly if diphthongs and IV vowels behave as
phonologically long, as long as they're predictable.

Depending on how harmony works, it may be possible to only mark
glottalization & laryngealization on onsetless vowels. Then you could
use choice of aleph/ayin/h row to mark it.