Re: Phonology Question (Coda vs. Word Final)

On Friday, October 17, 2003, at 01:29  AM, Nik Taylor wrote:

> David Peterson wrote:
>> Are there natural languages that do this--make restrictions on what
>> can end a syllable that are harsher than on those that can end a word?
>>  Are such restrictions realistic?  Any info would be much appreciated.
>
> Yes. Word-initial and word-final syllables can often be more complex
> than word-medial syllables.  Often this is due to earlier phonetic
> changes such as the loss of final schwas.  For example, English permits
> clusters such as /kt/ to end words (as in "fact" or "asked", which as
> /skt/), but, so far as I can tell, /kt/ never ends medial syllables
> except in compound words such as "fact-finding" (where it is still
> *morpheme*-final)
English is a good example of differing requirements in word-final
clusters compared to syllable final clusters. But it's different from
what David is asking. Since he wants syllables that are maximally CVC,
the issue isn't complexity in syllable codas, but rather varying
restrictions on place and manner of articulation in codas, such that
word-final codas are more permissive than word-medial codas. An example
of a language of this kind is Lardil. In Lardil syllables may end with
a non-coronal only if the following onset shares that place of
articulation; this effectively bans non-coronals from word-final
position. Coronals on the other hand may occur freely in medial and
word-final codas. However, only word-final coronal codas may be stops;
word-medial coronal codas must be sonorant. Also, the distinction
between apical and laminal coronals is available in codas in word-final
position, but not medial codas. I think these are the kinds of
distinctions David is after.

Dirk
--
Dirk Elzinga
Dirk_Elzinga@byu.edu

"I believe that phonology is superior to music. It is more variable and
its pecuniary possibilities are far greater." - Erik Satie