Hallo conlangers!

On Wednesday 16 November 2011 18:08:40 R A Brown wrote:

> On 16/11/2011 12:22, BPJ wrote:
> > On 2011-11-15 20:09, R A Brown wrote:
> >> One snag I noticed yesterday as I was trying to write
> >> up a page on Brx grammar: it's darn difficult to
> >> differentiate pronunciation between, say, _xa_ /'çi.a/
> >> and _xj_ /'cija/, or between _da_ /'tu.a/ and _do_
> >> /'tuwa/.
> >> 
> >> This afternoon I re-assigned _a_ as /ɰa/ - but /'çiɰa/
> >> and /'tuɰa/ ??

If you want your phonemes to be easily mastered and kept apart
by as many people in the world as possible, stay away from /ɰ/
- most languages don't have it, and most people won't be able
to tell what is going on here.  (You should also replace /ç/
by /ʃ/ if that is your goal.)

> > Why not simply _a_ == /ha/ where /h/ = any
> > fricative/approximant further back than palatal? And why
> > not also allow _i_ = /hi/ and _u_ = /hu/, or _xi_ and
> > _du_ will be as difficult.
> 
> I had actually thought of _a_ = /χa/ (with allophones raging
> from /xa/ to /ha/).   But allowing _i_ = /χi/ or /hi/
> actually does _not_ make _x_ = /çi/ easier.  If you recall,
>   both */ti/ and */ki/ fall together /ci/; likewise, if I
> reinstate /χ/ ~ /h/ as a phoneme, then we should expect both
> */si/ and */hi/ to fall together as /çi/
> 
> But you right that I would need /hu/ (or /χu/).  In fact I
> played around with such a system yesterday, using _f_ =
> /hu/; but I finished up having to import Greek π for /pi/ -
> I really don't want to import Greek letters if I can avoid
> it (tho I suppose π is well enough known).

It is well known - but *not a letter of the Latin alphabet*,
and you should avoid such bags on the side of an elegant idea.

--
... brought to you by the Weeping Elf
http://www.joerg-rhiemeier.de/Conlang/index.html
"Bêsel asa Êm, a Êm atha cvanthal a cvanth atha Êmel." - SiM 1:1