Re: Moraic codas [was Re: 'Yemls Morphology]

Thomas R. Wier wrote:
>"SMITH,MARCUS ANTHONY" wrote:
> > Syllables are also theoretical constructs, yet people can break a word
> > into syllables without any training. My Pima consultant started doing this
> > spontaneously, for example. He even deleted very salient epenthetic
> > segments. As an illustration, he broke up [Tok@Dot] 'spider' as [Tok
> > .Dot], and [komkIdZ1t] 'tortoise' as [komk. dZ1t]. (This epenthesis is
> > subject to a regular phonological rule, it is not simply phonetics.)
>
>But see, I'm not sure that that's really a certain guide, since even
>educated English speakers do not always agree on the syllabification
>of English words they use everyday.
It doesn't matter whether or not they always agree. Educated speakers don't
always even agree about vowel quality or stress placement in derived words;
and these are as concrete as phonology gets.

>   Furthermore, even assuming that they
>are like your informant, could most untrained educated English speakers
>*predict* how underlying forms work themselves into syllables?
Yes. But not with 100% accuracy. Ambisyllabic segments will throw them off,
as will some syllabic liquids. The problems with the ambisyllabicity is
understandable, they are trying to divide a word up, but these darn
segments don't seem to fit right.

Prediction does not necessarily mean that you can get everything right
every time; as long as they are right most the time, that's significant.

>  Can they
>*predict*, even in informal terms, based on some kind of generalization,
>what kinds of onsets and onset clusters and what kinds of codas and coda
>clusters are allowed in the speech that they use everyday (that is, without
>being asked about each possible onset or coda cluster individually)?
Of course not, for this one. That requires too much conscious knowledge
that speakers just don't have. Speakers don't consciously generalize very
well. Linguists mess that kind of thing up all the time.

>   That's
>the problem:  that even when you grant that they can parse words into
>syllables, can they say *why* those syllbles end up the way they do?
No, but it doesn't matter if they can say why. They can't say why they have
subject-aux inversion, but they sure can make a solid prediction about
that. The "why" is the domain of linguistic theory, not the intuitions of
speakers.




Marcus Smith

Unfortunately, or luckily,
no language is tyrannically consistent.
All grammars leak.
                -- Edward Sapir