A test for the valence of Lakhota verbs?

[Pamela Munro]

While 'put' requires a locative, it's an interesting question whether
everyone will call that locative an argument, since it is oblique.

I like your notion of "core valence" vs. some-other-kind-of valence. But
in fact I'd say that the patient of Lakhota 'give' is part of core
valence, even though it does not agree, while the locative with 'put',
though required, seems less argument-like (to moi).

The wa- test is very interesting! I'll be interested to hear how this
comes out.

Pam

ROOD DAVID S wrote:

>Yes!!  Thanks, Bob and Pam, for clarifying that; I should have thought of
>that term myself, since I also teach about "valence".  A verb's valence
>refers to the number of arguments it can take, but to reiterate what I was
>babbling about this morning, I think there are two kinds of valence: core
>and something else (I don't even have a name for the others). Thus I would
>hold that English "put" has a valence of two (I put something) in the
>core, but nevertheless requires a third participant in the form of a
>locative.  Would the rest of you use valence this way, or say that "put"
>has a valence of three, one of which must be locative?
>	And now I've thought of a Lakhota test for core valence: what does
>the wa- 'indefinite' prefix do when added to the verb?  I will repeat an
>example I've used many times -- sorry if this is old news to anyone --
>that the verb iyuNga 'to ask someone something' has a valence of three.
>Besides this, there is the form wiyuNga, with the wa- prefix, which means
>'to inquire about'.  The "wa" has replaced the "someone" argument,
>reducing the verb's valence from three to two.  There is also a third
>form, wawiyuNga, meaning something like 'go around asking lots of
>questions; be nosy'.  The second wa- has replaced the "something" argument
>of the original verb, and now we're down to an intransitive.
>	So the test would be: what (besides 'I gave it to him/her') does
>"wak'u" mean?  Does "wa-" delete the recipient and leave the other object,
>perhaps something like 'donate (something)'?  Or does it delete the
>equivalent of the English direct object and mean 'I gifted him/her'? And
>is there a "wawak'u" meaning something like 'be generous' or 'give stuff
>to people'?  If so, then I will concede defeat and admit that +AF8-k'u+AF8- has
>three arguments; if not, I think I may have found an objective argument
>for the position I've been trying to justify.
>	Sigh.  I'd like the question about "?wawak'u" 'give stuff to
>people' asked.  But it's not going to be so simple.  I just looked in
>Buechel, and discovered that the verb for 'make donations' is wawicak'u (I
>assume that's his spelling of wawichak'u).  If that's what I think it, is,
>then "wa" is replacing the thing I'm calling the adjunct, or the semantic
>patient, since "wicha", because it's animate, has to be standing in for
>the recipient(s).  Now, in a sense, we do have all 3 participants indexed
>on the verb.
>	I hope there is a speaker on the list with enough patience to read
>through this and help us.
>
>
>
>
--
Pamela Munro,
Professor, Linguistics, UCLA
UCLA Box 951543
Los Angeles, CA 90095-1543
http://www.linguistics.ucla.edu/people/munro/munro.htm