Dorsey's Law

[Koontz John E]

On Tue, 2 Apr 2002, Nancy E Hall wrote:
> 1. 	Are there any examples of loanwords that have undergone Dorsey's Law?
> (I'm looking for evidence that it's synchronically productive).

WRT Omaha, where there isn't any subsequent accentual shift, and where the
vowel introduced into obstruent-sonant clusters is schwa, essentially a
phonetic element of the cluster, I've heard "England" in context in
English and Omaha produced with a consistent schwa in gl.

In Winnebago all I can offer is that searching the English loan words
(so-marked) in Miner's unpublished Field Lexicon yielded c^ra'ak and
sn^a'ap (n^ = n-hacek, or r as pronounced after a nasal vowel).  Although
the use of n^ instead of n is perhaps suggestive, it certainly appears to
me that both these forms should exhibit Dorsey's Law, if it were
productive, but do not.

Incidentally, VrT forms appear with V(V) for Vr, cf.  poa'c^ 'porch' and
one that I've always loved since first I noticed it, ko'iNnsta'c^
'cornstarch'.  There's also kearapuNuNniya' 'California'.

> 	Does anyone know how native speakers view Dorsey's Law segments?
> Have any tests for intuitions about syllabicity been tried?

As nearly as I can recall, all OP forms in ##bdh or ##(S)n are initially
stressed - notably in verbs.  (I guess xdhabe' 'tree' is an exception of
sorts.)  So that these clusters appear to be two-mora syllables.  I
suspect that the same is true of Ioway-Otoe and Winnebago (historically).
On the other hand, Dakotan handles these as one mora syllables as far as I
know.

For Winnebago specifically, it might be worth consulting Josie White
Eagle's dissertation on the use of Winnebago language introspection in
science education.  (I hope this is a fair assessment of at least part of
it - since I know it only at second hand.)  In addition, her intuitions
presumably underlie any descriptive material in her article with Hale.  I
don't recall any mention anywhere of syllabicity tests.