Ho-Chunk Wa

[Koontz John E]

On Tue, 11 Mar 2003, Henning Garvin wrote:
> One function of this form seems to be 3PL object marker on verbs.  It has
> been identified as such in most of the literature.  However there seems to
> be a slight twist.  In eliciting forms, I have been told that this form can
> only refer to human beings.  For example:

I definitely don't know the literature on Hochank as well as I'd like to,
but I think this is a new observation.  I'd have to review Lipkind,
Marten, etc., not to mention Helmbrecht's recent paper, to be positive
what they say.  I know Johannes has been very interested in the functions
of wa-.

> waNk  naaNka waacanaN.
> men   those  3PL-1S to see- DEC.
>
> caa  naaNka  hacanaN.
> deer those   1S-to see-DEC
>
> When I started to ask, speakers informed me that this was the case.
> ...  I noticed in the Boas and Deloria Grammar of Dakota that a
> similar form exists in that langauge.  Namely ’wic’a’ which they state
> is the plural object for human and animate beings.  They do not go
> further into depth so I have no idea what they mean by ‘animate
> being’.

This is always a sort of fuzzy term, and I don't know of a discussion or
definition of the category animacy in Dakotan, but linguistically I would
usually interpret it to mean people and (other) animals, but not plants or
objects or natural phenomena or abstracts generally, subject to the
proviso that particular things in these latter classes might be
arbitrarily animate in a particular language, and particular animals (in a
Linnaean sense)  might be inanimate.  For example, bows or fire, etc.,
might be animate, and insects or snakes or fish, etc., might be inanimate.

> I haven’t run across any other examples of this through the course of my
> studies.  I also have no idea what to call this phenomenon other than trying
> to describe it.  Is this happening in other Siouan langauges?

Yes.  Within Mississippi Valley Siouan some third person patient plurals
are indicated with wa- (Hochank, Ioway-Otoe, Dhegiha) or wic^ha-
(Dakotan).  I have noticed that there are animacy restrictions on wa- in
Omaha-Ponca. To my shame, I don't recall the details off the top of my
head.  Either that or I never worked them out!  I do recall that wa + u is
usually u' (accented), presumably because *wa + o > *wo(o)'- > *o(o)'- >
u(u)'-.  However, some u- verbs reintroduce wa, leading to u-wa-.  These
seem to be (some of?) the verbs that take human objects.

Another issue to look into is whether wa- in this sense (third person
patient) in used only as an object of transitive verbs or also occurs with
stative verbs.  Also, what, if any pluralization formula is used when wa-
is not, e.g., none, a plural enclitic or reduplication, etc.  I'm pretty
vague on both these issues.

> Could it be some type of remnant of a noun class system that has ,over
> time, dissappeared?  I would appreciate any comments you all might
> like to make.

Dakotan wic^ha- looks like a worn down (or perhaps just unextended)
variant of the longer independent forms wic^has^a or wic^hasta (also
wic^has^ta?) 'person' (varies with dialect).  In essence, wic^ha- is an
incorporated noun.  It might be argued it was a classifier, but I'd feel
better with this characterization if there were other opposed categories
besides "everything else."

In the other languages my suspicion, not original with me, is that this is
a special alternative use of the wa- that serves to detransitivize
transitive or experiencer stems, or at least to render their object or
patient vague and unspecified.  The idea is that third person plurals are
tantamount to third person idefinites, both as subjects and objects, in a
variety of languages around the world, and that the reverse - indefinite
to plural - might also be a reasonable path of development.  Of course, I
think it is usually plural to indefinite, but I'm not sure.

The indefinite wa- is found throughout Siouan languages.  It might well be
an old incorporated noun or a classifier of some sort.  I'll leave Bob
Rankin the job of describing the set of classifier prefixes he found
relicts of with various nouns in various Siouan languages, though several
of these seem to have the form *wV-.

It is a moot point how many wa- morphemes there are, historically.
Presumably first person and indefinite are different, at least in the
attested languages (where *wa- first person often develops as *(h)a- or
has a different length or accentuation) but historically even that's not
absolutely guaranteed.  An indefinite makes a good source of first
persons, too.

Thanks, Henning!  I really enjoyed hearing this!