Re: NATLANG: Vowel harmony rules?

On 19 Jun 2004 Racsko Tamas <tracsko@FR...> wrote:

>   Meanwhile I was thinking it over and I found a possible test of my
> theory: the negative infix shows a similar dimorphism <-mA> ~ <-mI>.
  Since I did not study Osmanli Turkish morphology in detail, it
lasted me for a while to realize that my sudden idea is wrong. The
variant <-mI> of the negative infix <-ma> is a result of a simple
phonotactic rule before suffix <-Iyor> of (continuous) present:

  V -> 0 / _Iyor

  (Where "V" means any vowel, "0" is a zero phoneme [i.e. phoneme
loss], "I" stands for a close vowel affected by both back-front and
labial-illabial vowel harmony.)

  Therefore, aorist marker <-5r> = { <-Ar>, <-Ir>} seems to be
still unique.

On 18 Jun 2004 David Peterson <ThatBlueCat@AO...> wrote:

> Now the form of the non-lowering verb stems becomes an accident
> (which doesn't seem so bad, since there are exceptions)
  I would still insist on my analysis. We can study the morphology
according to a phonological descriptive methodology or according to
more conventional approaches.

  Phonological methodology is a bit like mathematics: there are not
"accidents" as 2 times 2 is always 4. If you choose these
principles you have to cover also "exceptions" (when 2 times 2
_seems_ to be 5) by rules.

  Of course, when you have too much trouble with these "accidental"
rules, you may consider that the methodology you used is
inadequate. But I think that it is not the case yet: it is still
worth rather introducing a new vowel harmony cathegory than to
reject phonological approach.

Reply