Lakota wa- 'variety object'

[Koontz John E]

On Thu, 11 Dec 2003, R. Rankin wrote:

> It looks as though maybe some experimenting would be possible with Omaha
> and Ponca with the support of the 1890 Dorsey text collection.  So, how
> many different WA- morphemes would you posit?  And what are their
> different meanings/functions?

Well, I still posit just two or three, depending on whether we distinguish
wa as an animate third person marker from wa as a "valence reduction
marker."  I definitely distinguish wa in wa ~ wa...a ~ awa P12, although I
concede that it might be historically connected, too.  And, of course, in
Dhegiha, the regular first person agent is a, not wa.

My logic in associating the first two wa's has been that, apart from an
occasional difference in accentuation, the two forms seem to be
morphosyntactically identical.  In other words, it's as if the plural
formation was a specialization of the detransitivized form, which is
certainly historically plausible.  Except maybe for these wawe- cases, and
maybe the wawa- cases I mentioned, too, the two wa's don't co-occur.
Naturally, since they're usually both object forms.  In the wawe- cases
I'm not positive that one of the wa's is a plural, because I think
wa-plural is generally animate.  In the wawa- cases the (first?) wa is
part of the stem, and is always present.  These are usually "experiencer"
subject verbs, though I don't think wa...s^i 'hire' is.  On the other
hand, given (1) the difference in accentuation, which might suggest that
there is sometimes a difference in length, and (2) the complex
interactions of wa with the locative u (which I've only hinted at), we
have at least some morphosyntactic differences.

So, let's say we have three wa's:  detransivizer, animate plural, and P12.
The one in which it is possible to do any subdivision is the
detransitivizer.  I think in a position-class approach you'd have to
distinguish the wa's in wawe and wawa, and you might want to distinguish
the the wa in deverbatives from statives, e.g., was^a(a)'be 'hunt chief's
standard' (dark thing).  These are all in some sense positionally or
functionally difference.

However, I think that instead of looking at Siouan morphology as a series
of ordered slots filled with slot-specific fillers, we have to look at it
as a series of ordered rules, with several paths through the rules, and
some multiple application of rules.  In that case these wa's all reduce to
"attach wa detransitivizer" even though the role that wa is filling
depends on the time at which the rule is applied.  For example, the
initial wa in wa...khega 'be sick', a non-dative experiencer verb (so it
looks stative in inflection) is a blocked reference to the body or the
part of it experiencing the sickness. (Though I am not positive you can't
include a noun referring to this body part in the sentence - will have to
check.)

JEK