Re: 1. YAESR

On Wednesday, April 20, 2005, at 02:45 , Muke Tever wrote:

> Joe <joe@...> wrote:
>> Tim May wrote:
>>> That was what I was thinking.  I don't see any better solution
>>> (although you clearly know more about Welsh than me).  You could cut
>>> down on <ss>s a bit by writing 's and plural -s as <s> even when
>>> they're phonetically [s]; e.g. <cats> rather than <catss>.  It's
>>> underlyingly /z/ anyway, IIRC.
>>
>> Or possibly underlyingly /s/.  I can't imagine an [s] occuring after a
>> voiced consonant.
>
> But there *is* [s] after voiced consonants: e.g., |fence|
..and dance, prince, France, lance etc etc. Some people do insert a
non-phonemic [t] between the [n] and [s]; but very many people (including
myself) do not.

> (and vowels, |loss|).
Plenty of those: ace, mice, lice, fuss, ass, voice, etc etc

We also have [s[ after [l], e.g. pulse, false etc. Altho some do pronounce
a [t] between [l] and [s] of _walz_, that is to do with the spelling, not
ease of pronunciation; the most common pronunciation IME is /wQls/.

Anyway, there is certainly a contrast of /s/ ~ /z/ in _false_ ~ _falls_

> Given that:
>
> 1)  [s] can appear after voiced sounds (fence, loss)
> 2)  morpheme |s| is /z/ after voiced sounds (fens, laws)
> 3) [z] can't normally appear after unvoiced sounds (heat-zone et al.
> maybe)
> 4) morpheme |s| is /s/ after unvoiced sounds
>
> strongly suggests morpheme |s| to be /z/.
There is also the fact the allomorphs are {s}, {z} and {Iz}. IMO it is
easier to explain {s} as a devoiced variant of {z} after non-sibilant
voiceless obstruents, and {Iz} as a conditioned variant with epenthetic
vowel after sibilants.

In fact, as the three {z} morphemes (plural*, 3rd
sing.pres.indic.active, & the possessive clitic) behave exactly the same
way as the reduced forms of _is_ and _has_ (written as 's), a good case
can be made that the morph is {Iz}     :)

*Yes, I do know the plural morpheme has other allomorphs as well   :)

But to be fair to Jo, I think his cymricized YAESR was meant to be a bit
tongue in cheek and that he was, as we say, 'thinking on his feet' when he
made the remark about /s/.

This question is of importance only if:
(a) we are making a serious analysis of English morphology;
and/or
(b) we a seriously designing YAESR on a morphophonemic basis, rather than
a phonemic or phonetic basis.

Neither of these seem particularly on topic to me.

Ray
===============================================
http://home.freeuk.com/ray.brown
ray.brown@freeuk.com
===============================================
Anything is possible in the fabulous Celtic twilight,
which is not so much a twilight of the gods
as of the reason."      [JRRT, "English and Welsh" ]