Order of verb affixes

[Koontz John E]

On Fri, 12 Nov 2004, Rankin, Robert L wrote:
> I notice in John's recap of the Comparative Dict. entry, that we
> recorded uuhaN in IO.

The form was from a source coded RR (Bob) and actually written u<raised
dot>h<a hook>.

> I wonder if this isn't a reflex of the *wo- prefix Ardis was discussing.
> It would come out [u-] or possibly [uu] in Omaha, of course and would
> ultimately be from *wa+o-.

I wondered about this issue myself.  Of course, I don't know how *wo- (I
guess *woo'-, really) comes out in IO.  I'm confident that it comes out
*oo'- in Dhegiha.  Winnebago has woo-, but I wonder if it doesn't also
have hoo- and if the woo- forms aren't secondary, with w restored.

> Would there be a phonological difference between reflexes of *o- and
> *wo- in Omaha? Other than probably initial syllable accent, that is?

There aren't any reflexes of *wo(o)- in Omaha-Ponca other than u(u)'- as
far as I know.  That is, there are no instances of wo- or wu-.  It is
true, as Ardis and I have pointed out, that some u- (*o-) verbs have wa in
the third person plural.  All nominalizations have u(u)'- with initial
stress.  The parallelism of these forms with the wa(a)'- and we(e)'-
nominalizations is the bulk of the evidence for PMV *woo- > PDH *oo- > OP
u(u)'-.  The rest of it is the alternation between accentuation and
non-accentuation of u in certain paradigms.  Note that paradigms that have
u'- in the third plural also have aN'gu- in the inclusive, i.e., wa-u-
appears as u'- and wa-aNg-u- appears as aN'gu-.  So I assume *wa-uN(k)-
develops in parallel with *wa-o-.

In essence, Omaha-Ponca speakers use a rule whereby accent shifting is an
allomorph of wa- with the u-locative forms.  Since wa- causes accent
shifting in other locative paradigms where it isn't lost, this is not so
weird as it seems.  Of course, I don't claim that speakers are necessarily
aware of this in conscious way.  They just know how to manipulate the
patterns.  Since some verbs do have a "secondary" or innovated wa-,
inserted after the u-, I would guess that the wa-u- > u'- pattern is
actually is not perfectly transparent.

> 'Pepper' is a noun and it seems to me ought to have had the initial
> *wa-.

That's true, though I don't think that all nominalizations have wa- and so
I'm not sure that wa- is a marker of nominalization per se.  I think,
however, that this is the kind of nominalization that would be expected to
have wa-, one in which a non-agent is the head.  Notice also that the IO
forms include the possibility of a wa-, as well as of an independent
"oblique of means" reference.  The structure is something like 'that/the
herb with which one cooks things'.

We haven't said so yet, but I assume all of us are taking the position
that these two forms, OP and IO, are more or less exactly parallel in
construction because speakers have calqued one from the other (or both
from a third form in another language).  The forms are then cognate in
their constituents because the two languages are so close, or because of
that plus some informal recognition of the equivalence or similarity of
morphemes.  This latter factor would apply if one was calqued from the
other and the resulting construction were not quite idiomatic in the
target languages.  Foe example, in English you can say for effect things
like "wrath incarnate" using what I take to be a French order of
constituents.  (At this point there doesn't need to be a French model for
the construction, of course.)  I don't think there's any evidence of
non-idiomaticity in 'pepper' of course.

I suppose one could also argue that the terms for 'pepper' in question
were existing, inherited, forms for 'flavoring agent' which have been
specialized or reassigned to refer to pepper.