Syntactic arguments always seem questionnable to me.
When we say for ex that the verb “to love” requires
Agent and (Thing acted upon / Patient / Object) :

- if the point of view is purely syntactic, then this
is not true. The French sentence “J’aime.” looks all
right, grammatically, to me. It means “I am in love”.
It may be old-fashioned by now, but I think than
earlier it was more common. Nowadays, it can be used
in another meaning, like in : “Comment tu trouves ma
nouvelle cravate ? – J’aime.” (How do you like my new
tie ? I like [it]). Also: “J’aime bien.”

- if the point of view is semantic, than the beloved
thing seems to be neither a “thing-acted-upon”, nor a
“patient” ; and “object” is syntactic and not
semantic. It might rather be called a focus.

We can always find a situation where a verb is
syntactically used with less arguments than it should.
Imagine you get a friend on the phone, you know he
wanted to sell his car, and you ask excitedly :
- Well, and ?
- Sold !
where “sold” is used without any agent, object, or
whatever (ellipse, meaning: I sold it, or It’s been
sold). You can even imagine that those two short
sentences could be the beginning of a best-seller (the
reader supposing that everything will be explained
later). If you come to questions of style, then just
everything is permitted, even “A rose is a rose is a
rose”. It is nearly impossible to say that “such
combination is impossible in that language”. It’s just
more or less probable, or usual. And what about poetry
? It is language too.

But that doesn’t mean that the definition of “to
sell”, like I can read it in my dictionnary, isn’t
something like “to yield at an agreed price” (this
definition omitting “something” and “to somebody”, but
il seems very hard to do without ; ellipse again).
This is semantic. What makes us confused is that we
are conditionned by the grammar we learned at school.
We think in terms of subject, and direct object, and
indirect object, and transitivity, and passive, and so
on. When I say “I follow him” (accusative)  in
English, vs “Ich folge ihm” (dative) in German, what
sense does it make to say that “him” is a direct
object and “ihm” is indirect ? Both sentences mean
exactly the same. The interesting question is: What is
the semantic role of “him / ihm” ?

--- "H. S. Teoh" <[log in to unmask]> wrote:
> Excuse me for barging in here, but I thought this
> was an interesting look
> into the semantic structure of languages.
> On Mon, Feb 02, 2004 at 02:03:07AM -0500, Roger
> Mills wrote:
> [snip]
> > I can open a conversation with:  "John loves
> Mary". After that, I can refer
> > to Mary as she/her, John as he/him.  I could not
> start off with "John loves
> > her" or "He loves her"and definitely not with
> *"John/he loves" nor *"loves
> > Mary/her".  Therefore, the verb "love" requires
> two roles/actants/"cases",
> > namely an actor/agent and a
> thing-acted-upon/patient/object.  Fillmore, in
> > particular, called these "cases, and schematized
> it as: LOVE [A, O].  (This
> > use of "case" is not to be confused with
> grammatical cases like Nominative,
> > Accusative etc.  How A and O are marked in their
> surface realizations is up
> > to each language.)
> >
> >  "Give" requires at minimum 3 actants, A/giver,
> O/thing given,
> > D(ative)/recipient. Only if the context is clear
> can O or D be omitted--
> [snip]
> IMHO, it's questionable whether the O here is the
> same as the O in the
> previous verb (love). It can easily be argued that
> the actants of "give"
> are A/giver, O/recipient, D/thing given, where D
> designates some third
> actant (probably not the dative as understood in IE
> langs).
> > "Sell" seems to require only an A and an O, the D
> is optional and in some
> > cases may be unknown. The price is helpful
> information but is optional, and
> > while it is part of the meaning of "sell" it isn't
> crucial to the
> > grammaticalness of a sentence (and may also be
> unknown).
> You could also say that D or O may be omitted from
> "give", too. You could
> give something to John, without specifying what was
> given, or you could
> give away a present, without specifying to whom you
> gave it. (E.g., "I
> gave away my old cupboard.")
> [snip]
> > It's in this sense, I think, that A O and D can be
> considered "core cases"
> > in the semantic framework of verbs
> [snip]
> If you assume that the O of a trivalent verb such as
> "give" is the same as
> the O of a transitive verb such as "love".
> You could also say A is optional, in cases such as
> "John was given the
> book". Of course, in English this is realized as a
> passive construct, and
> John has the A role; but it could be argued that the
> verb "give",
> semantically speaking, need only take the O and D
> actants in this
> particular instance.

Philippe Caquant

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

Do you Yahoo!?
Yahoo! SiteBuilder - Free web site building tool. Try it!