14.2554, Qs: Trivalent Verbs (fwd)

[Koontz John E]

Please excuse this repost from Linguist, which I thought of general
interest to at least Mississippi Valley Siouanists.  I know not all of you
follow Linguist.

John E. Koontz
http://spot.colorado.edu/~koontz

---------- Forwarded message ----------
Date: Thu, 25 Sep 2003 02:34:56 -0000
From: LINGUIST List <linguist at linguistlist.org>
To: LINGUIST at LISTSERV.LINGUISTLIST.ORG
Subject: 14.2554, Qs: Trivalent Verbs

...

-------------------------------- Message 1 -------------------------------

Date:  Sun, 21 Sep 2003 14:05:32 +0000
From:  Florian Zellmayer <zellmayer at chello.at>
Subject:  Trivalent verbs with 1/2 person theme or patient

Divalent verbs in head-marking languages with two-slot agreement
reference agent and patient. With trivalent verbs, in many of those
languages the agent and the goal rather than the agent and the patient
are agreed with. The same is the case in beneficatives and other
applicatives where the agent and the benefactee or applied object
rather than the agent and the patient is agreed with. In trivalent
verbs, beneficatives, and applicatives the patient is often restricted
to 3rd person, then, because it is not indicated in the agreement
system.

[JEK:  This seems generally reasonable as a description of how things work
in Siouan verbs, e.g., basic verbs like 'give' (rare, I think) and dative
verbs fall under the heading of "trivalent verbs ... and beneficatives
...," collectively or separately.  Such verbs generally can have a third
person patient that is unmarked in the verb, except as the dative prefix
itself, or its traces, might be considered in this role.

We tend, of course, to refer to the person markers - concords, if you
prefer - as the agent and patient series, using agent and patient as
surface category terms, which conflicts with using agent, patient, and
benefactee as terms for the underlying or semantic relationship of the NP
to verb.

Verbs with locative prefixes that govern the patient agreement fall under
the heading of "other applicatives."  Some locatives do this; others
don't, more or less on a case by case basis.

Resume Zellmayer:]

Now, many of these head-marking languages (without case) do have
possibilities of expressing 1st or 2nd person patients in trivalent
verbs, benefactive verbs, or applicative construction. Some of them
encode e.g. ''I killed you for him'' as ''I killed your body for him''
or the like, thereby providing an ''escape hatch construction'', so to
say, for the 1st or 2nd person patient that cannot be expressed by
agreement.

Information on how 1st or 2nd patients in trivalent verbs or
beneficatives or beneficatives or applicatives is expressed or
circumlocuted in head-marking languages with two-slot agreement is
rarely contained in the relevant grammars.

So, if you work on such languages, or if you have materials or
references on this topic, please let me know.

[JEK:  I'm not really aware of escape hatch constructions like this in
Siouan languages, though I think causatives are perhaps sometimes used in
this way - "I made him have you." - and then there are transitivized
serial verb patterns like "having him, he came here" which in Omaha-Ponca
could be combined with the s^u 'near you' locative to produce forms like
"having me, he came here near you."  I haven't checked to see if there are
any examples in the texts with "having you/me/us."]