More case alignment.

[Koontz John E]

On Mon, 15 Feb 1999, Robert L. Rankin wrote:
> > I should admit, that, unlike Bob, I'm inclined to see the -re suffixes
> > in Winnebago as part of the Winnebago version of the 'suddenly' or
> > punctual/inceptive auxiliaries that I've mentioned a while ago on this
> > list.  
> 
> The -re John's referring to here is the one that is etymologically a
> grammaticalized particle from *re: 'to go'.  It, along with some other
> motion verbs can appear post-verbally with aspectual meaning.  John is the
> expert on these and has a paper on them in Omaha.  We do something similar
> in English when we say "Sam went and ate the whole pizza." where Sam
> hasn't moved an inch.  'Went' is just a 'punctative' or 'sudden action'
> marker here.  

Nice English analog, Bob! A good point to make on the suddenly forms is
that they are comprised of several parts, none of which mean 'suddenly'.
In fact, I think that 'suddenly' is not a very good name for these items.
It arises from a common translation used in Dorsey's texts.  They also are
rendered various 'begin', 'start', and 'repeatedly'.  The components are
the a subset of the usual Dhegiha/Chiwere positional stems and motion
stems, sometimes causativized (when the sense of the main verb is
transitive).  The noncausativized forms can have the positional
reduplicated.  The formula is something like this:

motion-stem + positional-stem
motion-stem + reduplicated(positional-stem)
motion-stem + positional-stem + causative

The motion-stem or positional-stem can be omitted, and either or both can
take the vertitive/orientative *k-prefix.  These two stems, if included,
are in concord with some aspect of the action or result of the action. 
The personal inflection of the causative concurs with the agent of a
transitive main verb.  While the motion-stem and positional-stem do match
the action in some way, I'd have to admit that I'd probably do better
figuring out what is matched from an example than predicting what forms
would occur.  

The forms that occur are a lot like the Dhegiha positional verbs, which
are essentially i + positional stem + causative.  E.g.,

ihe=...dhe 'to lay'
ithe=...dhe 'to stand'
idhaN=...dhe 'to set'

Typical auxiliary sequences are dhe=...dhe or thi(g)dhe.  There are,
however, above 100 possible forms, with about 60 attested between OP and
IO.  In Chiwere and Winnebago the former sequence becomes something like
Winnebago re=...hii, by substitution of cognates.  The thi(g)dhe form
becomes Winnebago ji(ke)re.  In Dakotan the latter is hiNgla (Teton).  The
nasalization is a bit unexpected, but OP th : Wi j : Te h is the normal
set of correspondences for PS *th.

There are a genuine few oddities in the correspondences.  For example, Wi
and IO have (h)i, not (h)u, corresponding to OP i as a motion stem.  OP i
'to come' is from *(h)u, and one expects the u vowel in these branches. 
I'd be inclined to speculate on a new motion verb stem, except Dakotan
seems to have uN.  The nasal is a surprise, but at least it's a u!  Also
surprising is that Chiwere has we as the horizontal positional,
coresponding to OP he. 
 
> There is also a -re 'causative' and a 're' demonstrative and maybe even a
> 're' 'be'.  We have to sort these all out.  ...

I might add to these the pervasive Dakotan adverbial suffix -ya < ?*rE,
though that could be related to one or the other of these, too.