Re: Phonemic vowel and consonant length.

At 11:03 AM -0500 2/2/03, Roger Mills wrote:
>Steven Williams wrote:
>
> >3. Quite a few languages hold phonemic consonant length contrasts--Italian,
> >Japanese, Finnish and so on. Is it at all common, or even possible, to have a
> >three-level distinction? In stops?
>
>Interesting idea.  Here's a possibility:
>
>/t/ realized as [d]
>
>/t:/ realized as [t]
>
>/t::/ realized as [t:] or maybe [?t] or some other "emphatic" pronunciation.
>
>Would probably work best in a system that lacked contrastive voiced stops. But how
>would voiced stops work, if they were also present?
>
>/d/ realized as [D]?
>
>/d:/ realized as [d] (overlap with /t/, not good)?
>
>/d::/ realized a [d:] or [?d] or other emphatic.
But this isn't a three-way length distinction anymore. For voiceless stops it's a
combination of voicing and length (or glottalization); for voiced stops it's a
combination of frication and length (or glottalization). It's seems plausible
that these distinctions would form a lenition/fortition series (i.e., [d] < [t]
< [t:]/[?t], etc), but I don't see how it can be reduced to a single phonemic
distinction.

And there's the rub with scalar distinctions; they seem always to be reduceable to
the intersection of two or more separate binary distinctions. (For the
theoretically minded out there, Robert Kirchner published an interesting paper
on this problem in the journal _Phonology_ -- 1997 I believe. His contention
was that contrastiveness was an epiphenomenon of Optimality Theoretic
constraint ranking and that there is no principled distinction between
phonetics and phonology. I don't buy it, but the arguments are clever.)

Dirk
--
Dirk Elzinga                                               Dirk_Elzinga@byu.edu

"It is important not to let one's aesthetics interfere with the appreciation of
fact." - Stephen Anderson