Re: Optimum number of symbols

John Cowan wrote:


>Thomas R. Wier scripsit:
>
>> (A) alternating root-final plosive:
>>     [kanat]    'wing'      [kanadM]  'wing-ACC'
>>     [kanatlar] 'wing-PL'   [kanadMm] 'wing-1Sg'
>> (B) nonalternating voiceless plosive:
>>     [sanat]    'art'       [sanatM]  'art-ACC'
>>     [sanatlar] 'art-PL'    [sanatMm] 'art-1Sg'
>> (C) nonalternating voiced plosive:
>>     [etyd]     'etude'     [etydy]   'etude-ACC'
>>     [etydler]  'etude-PL'  [etydym]  'etude-1Sg'
>
>My analysis is that (A) is /d/, (B) is /t/, and (C) is a funky borrowing
>from French that breaks the rules.  Nobody says that Turkish doesn't
>have vowel harmony, e.g., just because there are Fremdworter with
>non-harmonic stems.
That would be my guess too-- though "kanatlar" ?//kanad-lar// implies a rule
somewhere that voiced (stops?) > unvoiced before /l/, which may or may not
be the case.

(Offhand, I'd suspect "sanat" of being a loan, too-- waiting to hear from
the Arabists/Persianists).

Years ago, a friend of mine encountered the Turkish name Mucahit, and
pronounced it a l'americaine, which was hilarious.  We now know it's just
their version of Arab. mujahid '(holy) warrior'. (I wonder if its Turk.
accusative is mucahit-V or mucahid-V)

Final devoicing is so ingrained in Indonesian that even the occasional loan
with a final voiced stop (written) has generalized voiceless even in
derivatives:  main ex. |jawab| [dZawap] '(vb.) answer' , |jawaban|
[dZawapan] '(n.) answer' (only those who want to show off their knowledge of
Arabic pronounce final [b])
>
>Also, is it possible that some people say [etyt] after all?  I remember
>reading that although the official pron. is [klyb], many people say
>[kylyb] instead, thus preserving the Altaic rule against clusters.
>
>--
>John Cowan <jcowan@...>     http://www.reutershealth.com
>I amar prestar aen, han mathon ne nen,    http://www.ccil.org/~cowan
>han mathon ne chae, a han noston ne 'wilith.  --Galadriel, _LOTR:FOTR_