argument structure of k'u 'give'

[Koontz John E]

What I've noticed is the examples Regina and Neva have assembled
is that what works for a two object-inflection verb is wic^ha- for a less
salient object 'them' and a "speech act participant" inflection for a more
salient argument, often a benefactive, but but not in the 'give in
marriage' example.  I think we've observed similar things in the past in
other contexts.  One that I recall (for OP, anyway) is that the verb
embedded under the causative can take wa-, but otherwise inflection is on
the causative.

Regina suggests that with k?u a wic^ha- here is essentially derivational
Could this maybe be argued in the other cases, too?  Is there a test for
derivational status?  (I could use some broad hints on this in some OP
contexts!  So far I only know the arguments in terms of (lack of)
preductivity and (lack of) predictability of the sense of the whole from
the sense of the parts and statistics, which are hard to apply easily.)

Actually, I think that Regina doesn't need to concede that wic^ha- is
derivational in any absolute sense in any of the cases as long as it's
possible to characterize it in some way as being on the boundary between
derivational and inflectional, or at least as being less inflectional than
mi- and ni-, etc.  The synchronic argument might be in terms of ranking,
along the lines that "a less salient pronominal category can only encode a
less salient object."  Or maybe it's "a more salient object must be
encoded and can only be encoded by a more salient pronominal category."
I'm trying to think of a way to preclude ni-ma- or wic^ha-wic^ha-, but
allow wic^ha-ni- and wic^ha-ma-.  (And where does uN(k)- fit into this?)

Linda's example and its rejection by Lakota speakers confirm what David
said earlier about this differing from place to place and speaker to
speaker.  Assiniboine and Lakota might well be entirely different in this
respect.  I'm always amazed when I see actual Osage, etc.  The
similarities to OP only make the differences more astounding.  I go along
thinking "Same.  Same.  Same.  Whoa, what was that?"