Re: PHONO: feature theory (was: vowel harmony)

Jonathan Knibb sikyal:

> When the vowel harmony discussion got on to comparing languages in
> terms of the features applied to their phoneme inventories, it
> reminded me of something that's been bugging me for a while.
>
> Question 1: in standard feature theory, is a phoneme uniquely
> associated with a particular set of features *no matter which language
> it's in*?
No. Phonetically identical consonants can be assigned to different feature
sets depending on the overall phonology of the language. However, the
feature values themselves do have *some* phonetic content--one can't go
around saying that [labial] distinguishes /t/ from /k/ without damn good
reason. One should think of the feature matrix as defining a set of broad
articulatory properties for a phoneme, the exact implementation of which
is left to the phonetic component and can vary from language to language.

> Question 1a: is it meaningful to speak of 'the same phoneme' cross-
> linguistically anyway?
Sort of. In the case of segments with the same phonetic value and the same
place in the overall phonological system, there is no reason not to call
them 'the same phoneme'. However, there are lots of cases of segments with
the same phonetic value but different places in the phonological system,
or segments with different phonetic values but the same place in the
phonological system. In either of these cases, it's a judgement call
whether we're referring to 'the same phoneme.'

> Question 2: how do you decide which value of a feature is unmarked?
Very carefully, and hopefully with lots of cross-linguistic evidence.

NB that the whole concept of classical binary features has been superceded
by feature geometry using privative features, which has a whole different
mechanism and set of assumptions regarding unmarked features.

> ObC - Telona's consonant inventory is ruthlessly feature-symmetrical,
> and I've had the devil's own job trying to get it to fit my personal
> aesthetic given this constraint. If I could make up my own features,
> it would make things a *lot* easier. :))
Feh. Don't constrain yourself too much with theory. Natlangs don't.


Jesse S. Bangs jaspax@u.washington.edu
http://students.washington.edu/jaspax/
http://students.washington.edu/jaspax/blog

Jesus asked them, "Who do you say that I am?"

And they answered, "You are the eschatological manifestation of the ground
of our being, the kerygma in which we find the ultimate meaning of our
interpersonal relationship."

And Jesus said, "What?"