Re: Ditransitivity (again!)

I had a look at the definition for ditransitivity at
the address mentioned. As I suspected, it is purely
syntactic (notion of "two objects"; an "object" is
itself a syntactic notion).

I think that we definitely should think semantically,
and only use natlang syntaxes as indications (clues).
What is acceptable or not in a natlang is for a good
deal a contingent point: if it's not acceptable in
English, it might be in some - even close - other
natlang.

Somebody asked about Chomsky, how he was considered by
now. As far as I know about his theories, he just
tried to get rid of semantics, holding to pure syntax,
and that's why he came to a "cul-de-sac" (I'm not
denying his importance in a specific period).

Ditransitivity refers to the notion of valence of the
verb, and valence can only be based on semantics. For
example "to sell"  suggests that:
1. there is a human agent (the one who sells)
2. there is another human implied, whatever his role
will be called (the one to whom sthg is sold)
3. there is a "something" that is sold (presumably
non-human, but this is not mandatory, if you use to
sell slaves; doesn't have to be physical neither, you
can sell an access right for ex)
4. there is a price, expressed in some abstract unit
(otherwise it's only an exchange, not a sale)
(The very meaning of "to sell" itself being something
like: a change in property, which is a social notion).

The roles that have to be mentioned are normally the
agent and the "something" that is sold: John sold a
book, I won't tell to whom neither for what price; but
I can't say "sold a book" (except in the passive: the
book was sold), neither "John sold".

There can be other concepts more or less implied by
the act of selling. For ex: is the price paid at the
same time that the thing is sold, or is it prepaid /
delayed ? (think of stockmarket). If you think long
enough about "selling", you might come to the
conclusion that every term of a commercial contract
could be included in some sort of super-"to sell"
verb, a kind of a monster requiring about fifty roles
orbitating around it. The whole contract might be
lexicalized as a single verb. This would be maybe not
very easy to handle for a human being, but a computer
could easily do it.

--- Costentin Cornomorus <elemtilas@...> wrote:
<http://www.sil.org/linguistics/GlossaryOfLinguisticTerms/WhatIsDitransitivity.htm>
> bears this out. That some indirect objects can
> not be bare datives (i.e., they require a
> preoposition) doesn't seem to be enough to say
> that the verb isn't capable of taking two
> objects. Perhaps there is something about
> dativity that requires a person to be involved.
>
> Padraic.

=====
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/