Re: Ditransitivity (again!)

I understand that the core arguments are those which
are logically requested by the meaning of the verb. It
is not a question of syntactic cases, but of semantic
actants. Every verb has its own scheme.

Of course, you may produce a sentence like "I sold my
car", this is grammatically correct, but the
information is incomplete. The interlocutor would be
right to ask: "to whom ?", and also "for what price
?", because these are parts of the concept "to sell".
Some arguments seem to be closer or further from the
"core". (Other questions like "When ? Where ? Why ?
etc. are not relied to actance, but to circumstance.
They are more peripheric than actants).

You may even say "This car was sold", but hardly "I
sold", except in case you just explained what you are
talking about in your last sentence (meaning: I sold
it).

The syntactic cases are not relevant. They change from
one language to another. In English and French, you
follow "somebody" (accusative), but in German, you
follow "to somebody" (jemandEM folgen, dative).
Instrumental has various meanings in Russian, in fact
it is a mix of different more primitive concepts, put
together (ex: ja rabotaju injenerom, I work as an
engineer - nothing to do with the concept of
"instrument").

So to me the notion of "oblique cases" has no meaning.
Only the semantic roles make sense.

The idea of a maximal number of arguments hardwired is
interessant, regarding the human mind, but it may not
if we consider a computer program. Nothing can prevent
to build a "to sell-function" using a dozen of
arguments, ten of them being facultative for instance.

Let's do it better than the human mind !


--- Nik Taylor <yonjuuni@...> wrote:
> Christophe Grandsire wrote:
> >
> > En réponse à Andreas Johansson :
> >
> > >It was argued that a verb like "to sell"
> logically should have four core
> > >arguments - the seller, the buyer, the thing sold
> and the payment - but that,
> > >for some reason, no known language has verbs that
> take more than three core
> > >arguments, so one of those arguments can only be
> introduced as an oblique (in
> > >the case of English "to sell", the payment; "he
> sold me the book for ten
> > >euros"). The conclusion drawn, IIRC, was that
> this limitation is hardwired
> > >into the language-handling parts of the central
> nervous system.
> >
> > Maybe it has something to do with another
> hardwired limitation, which seems
> > to be that the human mind cannot handle more than
> around 7 entities
> > simultaneously (i.e. our immediate memory is
> limited in size to around that
> > amount).
>
> Hmm ... just what exactly counts as a "core
> argument"?  I've always been
> confused about that term.  I understand that "core
> arguments" are
> generally taken to be nominative, accusative, and
> dative (or ergative,
> absolutive, dative), but what makes those core, and
> cases like genitive,
> instrumental, etc., oblique?

=====
Philippe Caquant

"Le langage est source de malentendus."
(Antoine de Saint-Exupery)

__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free web site building tool. Try it!
http://webhosting.yahoo.com/ps/sb/

Reply