Re: THEORY: phonemes and Optimality Theory tutorial

On Sun, 12 Nov 2000, Nik Taylor wrote:

> dirk elzinga wrote:
> > The significance lies in what the theory forces you to posit as the
> > phoneme. Ideally, there should be *one* possible phonemic solution.
>
> But isn't the phoneme simply the category that contains various phones,
> rather than the label given to said category?
Well, this is one view of the phoneme. It isn't the only one, though.
See an earlier posting from me on different definitions of 'phoneme.'

> > In OT, constraints only hold on the surface forms; constraints are not
> > allowed to operate on the underlying forms. If nothing may be excluded
> > from underlying forms, then it is possible to have underlying forms
> > which are fully specified, radically underspecified, or anything in
> > between. When I posit /b/ in underlying form (or /p/ or /B/ or /F/),
> > it is not "code" for the underspecified representation [+labial,
> > -nasal]; it really is /b/ (or /p/ or /B/ or /F/) in all of its fully
> > specified glory. This principle (the freedom of underlying
> > representation) is referred to in the OT literature as "Richness of
> > the Base". However, it has received surprisingly little attention,
> > partly because it is radical break from traditional phonemic analysis.
> > It *is* a tough pill to swallow, but if you take the OT idea seriously
> > that phonotactic constraints should only hold on the surface, Richness
> > of the Base is the inevitable outcome.
>
> I still fail to see how this is different from a phoneme.  It seems more
> like a slight modification of the phoneme concept than a radically new
> concept.
My analysis directly violates the Biuniqueness Condition. As I see it,
the phoneme--whatever its definition--had the following properties:

1)      it was segment-sized; that is, it was not decomposible into
features, prosodies, or elements (though it was characterized by
having certain properties such as labiality, voicelessness, etc);

2)      it was the unit of speech which enabled the expression of
opposition and contrast, and it was embedded in a system organized by
such oppositions and contrasts;

3)      it was part of representations which uniquely determined
phonetic forms; likewise, phonetic forms were analyzable into
sequences of phonemes (this is the Biuniqueness Condition).

The Prague School, Jakobson in particular, showed that it is possible
and profitable to decompose segments into features, where the true
contrasts of the phonological system were expressed. Firthian prosodic
analysis showed that these "features" (Firth never used that term--he
always spoke of prosodies) could span several segments (an idea which
Z. Harris later took up as "long components"). So property (1) does
not characterize the phoneme.

And already mentioned the distinction between different systems of
oppositions based on their syntagmatic properties (e.g., stressed and
unstressed syllables, which may have different sets of sounds in
opposition). If the phoneme is part of a system of oppositions and
contrasts, which system should be taken as definitional? So property
(2) does not characterize the phoneme, insofar as it is not possible
to find *the* system where opposition and contrast should be
expressed.

I've already talked about how Generative Phonology in general, and
Optimality Theory in particular has kicked the legs out from under the
Biuniqueness Condition. So this principle also does not characterize
the phoneme.

That being said, if the phoneme does still exist, then there must be
some property which characterizes it which I have not mentioned. If
you're talking about underlying segments as phonemes (by which I'm
assuming that what is meant the level of representation where contrast
and opposition are reckoned) just remember that it was not the only
part of the definition; the 'phoneme' was always something more than
that.

BTW, I'm ready to move this discussion off list if people are getting
tired of it.

Dirk

--
Dirk Elzinga
dirk.elzinga@m.cc.utah.edu