Re: Consonant allophones in Minza
From: | John Vertical <johnvertical@...> |
Date: | Sunday, September 30, 2007, 12:46 |
On Thu, 27 Sep 2007 22:36:41 -0500, Herman Miller wrote:
(...)
>The other thing is a writing issue: how to spell a sound that alternates
>between [G] and [j\], similar to what happens in Modern Greek. I've been
>writing this phoneme as "j", which works fine in words like "jönžix" (to
>rain), where it represents [j\]. But I don't like how it looks in words
>like "jovi" (excessive, too much). Another possibility is a "gh"
>digraph: "ghönžix", "ghovi", or some kind of "g" with a mark over it:
>"ġ" or "ğ". I could use just plain "g" if I didn't already have a /g/
>phoneme. Certainly there are precedents for languages using "g" to
>represent /j/, so that might be the best alternative. Or just pick some
>unused letter like "q", but I can't find anything that's even as
>suitable as "j".
How about <gj>? Does nicely for /j\/ in Swedish, and extending it to /G/
might not be too much of a stretch, esp. since unless I've overlooked
something, this would be the only place where <j> appears, and so it would
essentially be a diacritic for <g>.
>For the vowels, it's tempting to use a diacritic for all front vowels:
>
>Front: ï ë ä ö ü
>Back: ı e a o u
>
>But then one of the most common vowels would be ï and there'd be dots
>everywhere. Well, more dots than i's normally have...
Wouldn't the thinness of <i ~ ı> justify having <i ë ä ö ü> for the front
series, if <ï> is too much dots? You could consider it "dotted line over",
not "two dots over". (Hmm, and now I'm imagining <w> with three dots over
(quadrema?) for /H/. Too bad none of my langs have /H/.)
Consider also other Latin-1 diacritics, maybe the acute:
front <í é á>
non-front <i e a o u> (or <ı> if you prefer)
<é e> for /e @/, at least, looks a LOT better to my eyes than <ë e> in
whichever order. If you added front rounded vowels, <ú ó> for /y 2/ might
not have a precedent, but IMHO it looks OK; acute for length is not rare,
and didn't some northern English dialects have an alternate GVS where /u:
o:/ went to /y 2/?
John Vertical
Reply