Friend, town words.

[Koontz John E]

This is taken from an ongoing exchange among Bob Rankin, David Costa, and
myself. 

On Sat, 21 Nov 1998, Robert L. Rankin wrote:
Per Koontz on differences between Winnebago and Chiwere:
> > One interesting phonological difference is that *R ends up unaffricated
> > d in Winnebago (written t), while Chiwere seems to have merged *R and
> > *r/*y (except for the few *y forms that show as y/z^ in an exceptional
> > way). 
> 
> So WI has the archaism there -- and we'd reconstruct something like /d/ if
> not /R/ to proto-Chiwere-Winnebago.  WI probably has a few more archaisms
> where CH has more recently innovated.

It's interesting that Chiwere should have neutralized *R (merged with *r),
though this also occurs in Osage and Kansas, though there *R merges with
*t. 

Nothing else occurs to me as a Winnebago archaisim in the realm of
phonology, but in morphology there are some obvious modifications to the
syncopating paradigms in Chiwere that I've mentioned, leaving Winnebago
more conservative.  For the most part these are signalled by double
(regular + syncopating) inflection in Chiwere. Other than that I think
honors are about even. 

Oh yeah, Winnebago has that waNaNg-...-a-... inclusive form for the
patient, where Chiwere has wa-...-wa-..., which I think is innovated (and
looks close to Dhegiha wa-...-a-..., for what that's worth).  

I've mentioned that I think that Winnebago allowing the augment/plural =wi
with the first person strikes me as an archaism. 

In keeping with this, I think Winnebago does not allow the inclusive to
combine with either first (standard Siouan) OR SECOND (unusual) persons in
paradigms, at least as far as I can tell. It's always by its lonesome. I
realized/rediscovered this when I started to look at why the aN vowels
might go to iN in the pronominals.  Jean Charney had included it without
commenting on it in her little sketch, and it must be in the other
grammars, but I think no Siouanists have ever noticed it is an anomaly. 

> > shift of a to e after k in final position (though Winnebago loses the
> > e's which obscure this and may make it a figment of my analysis), 
> 
> The fact that WI keeps the final -e in such cases after consonant clusters
> ending in K firms up your analysis, I think.  (But I'm not thinking of
> examples off the top of my head right now.)

A good point, and one that I've considered in the past.  I wasn't sure it
would convince anyone, since some consider the final vowels to be
"variable and unpredictable" due to the absence of a simple a = a = a =
... = a correspondence and presence of morphological considerations.  They
are a bit of a shock after typical Siouan vowel correspondences.  An
example would be 'heart' naNaNc^ ~ naNaNc^ge < *yaNt-e ~ *yaNt-ka vs. Da
c^haNte' < *yaNt-e vs. OP naN'de < *i-yaNt-e vs. Ma naNaNte ~ naNaNtka <
*yaNt-e ~ *yaNt-ka (should the Ma have a wa- prefix? - this is all from
memory).  There are some nice sets with -Vke from *-h-ka, but I don't
remember them.  Heron?  Badger?  I'm sorry - I don't have any CSD files on
this computer. 

> Chiwere has frozen relics of the -re portion of the causative.  Sag-re and
> the like. 

Yeah.  I guess these are like the -ya "adverbial" suffix in Dakotan,
though you have to consider as well whether these -re might be from 'to
go'.