Re: Consonant allophones in Minza
From: | Paul Bennett <paul.w.bennett@...> |
Date: | Friday, September 28, 2007, 16:19 |
On 9/28/07, Philip Newton <philip.newton@...> wrote:
> On 9/28/07, Henrik Theiling <theiling@...> wrote:
> > In a conlang sketch, I once used
> >
> > |a e i o u| for /a e i o u/ (/a/ as more like [A]),
> > and |ä ë ï ö ü| for /& V M 2 y/.
> >
> > Maybe it would be an option for you? This way, you have the diaeresis
> > as a 'swap front-back' diacritic and still retain the standard values
> > for the unmarked vowels. Furthermore, only |ë| and |ï| are lightly
> > off-standard.
>
> And since /M/ is the value of Turkish dotless-i (AFAIK), you could
> think of the second dot in |ï| "cancelling out" the first one,
> resulting in a dotless-i, for which /M/ is an accepted pronunciation.
> IYSWIM.
Using trema for /& 2 y/ is more than cromulent, and using dotless-i
for /M/ vs i for /i/ would also seem sane. Using e-trema for /V/ is
both loosely analogous to /& 2 y/ and somewhat precedented in
Albanian's use of e-trema for /@/. Another option for /@/ might be
a-breve, a-caron, or something along those lines.
The full grid according to the above might be:
i ü ı u
e ö ë o
ä a
or:
i ü ı u
e ö ă o
ä a
Paul
Reply