Irregular "to eat" in Dakotan

[Koontz John E]

On Sat, 15 Sep 2001, Hu Matthews wrote:
>        sg           pl
> 3rd  duush¡k      duus£uk
> 1st  buush¡k      buus£uk
> 2nd  dil£shik     dil£suuk
>
> 'eat' is irregular also in Crow.
> The final k in these forms is the declarative marker.
> (In case the accents don't make it through, - in the 1st and 3rd
> person, the accent goes on the first vowel of the last syllable; in
> the 2nd person it is on the middle syllable.

For the record, I believe that this is the regular pattern for the lu-
instrumental in Crow, and reflects the Crow outcome of th *r-stem
syncopating conjugation.  Thus it would follow the alternate pattern of
inflection for 'eat' attested in Assiniboine and Winnebago (or Dhegiha,
etc.).

It's particularly neat to find such a nice exhibition of this pattern in
Crow, of course, and it's on the strenght of this that I think that the
*r-stem pattern can be read into Proto-Siouan, not just Proto-Mississippi
Valley Siouan,

I've wondered if the -u- vowel of plurals in Crow might not reflect the *p
of =*pi as a plural in MV, though this may be stretching things!  I don't
know of any parallel development of *p to u in the enclitic sequences of
Crow (or Hidatsa).