Re: Underspecified verbs?

Jesse writes

>Essentially, my question is this: would it be plausible for a language to
>have verbs be vastly underspecified syntactically, but to have nouns be
>massively overspecified?  The system I have in mind would have a very
>small verbal vocabulary supplemented by a large nominal case system which
>would do the work for the verbs.  No verbs would have specified valency,
>and would be fairly fluid in meaning depending on the nouns associated
>with them.  Is this plausible?
This is certainly possible in principle...but I'm a little skeptical that
the 8 cases you've outlined would really be able to greatly reduce the
number of verbs, unless some other aspects of the language assist in this.
You cases seem to allow for more subtlety of meaning than is possible in
English, but (at least in the example you give) don't seem to extend the
basic meaning of the verb far enough to make other verbs superfluous.

Whenever I have these "plausibility" questions, I work on them by doing
some translations into the conlang. After a few translations, you can get a
feel for how much stretching is needed to get the proposed mechanism to
cope with a range of meanings.

>The system I've designed has six different core noun cases, operating
>across three different semantic variables.  The variables are Intention,
>Causation, and Participation, and the various roles that I have are:
>
>           Intent.    Cause.    Part.
>Actor       +          +        +
>Agent       -          +        +
>Participant -          -        +
>Origin      -          +        -
>Causant     +          +        -
>Object      -          -        -
This is great--I love the logic at work here. (I'd agree with Matt, though,
that you should swap the words "agent" and "actor" around.)

Cheers, Tom


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Tom Tadfor Little            tom@telp.com
Santa Fe, New Mexico (USA)
Telperion Productions        www.telp.com
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~