Subject: THEORY nouns and cases (was: Verbs derived from noun cases)

Philippe Caquant <herodote92@...> wrote:

<<<
I find the idea of a verb meaning for ex "to be a dog"
unsatisfactory. Anayway, saying so just means that the
corresponding word will behave, syntactically, like a
verb. Thus, we could have an English verb like "to
dog", for ex: I dog, he dogs, I dogged, I have dogged,
etc., meaning I am a dog, he is a dog, I was a dog, I
have been a dog (is this really useful ?)
<<<

"Satisfactory"? "Useful"? What do you mean?

I think you refer to my post suggesting to tell "verb" from "predicate". I
fail to see the problem about that. I did not say that zerocopula "A B" MUST
mean "A is B" (which is very different from B A), although a lot of natlangs
do so. And since here we go: Further, when reading on semantics I take it
for granted that the authors are clever enough--yes, some people having
spent years on a topic and made a book may understand the topic as much as
their readers do--to not categorize words, which may cover a great deal of
lexies, but the lexies themselves. As a result, when one of them talks of
"to hit" as a "verb of motion with a Path", then I take it for granted that
he excludes "to hit" as a verb describing an accident or incident. When he
talks of "to dance" as such, I immediately exclude "to dance" as a social
activity etc. Hence I don't have a problem with calling "GOAL" the argument
"tree" in "The car hit the tree", it being understood that we talk of "to
hit" as a verb of motion. Now, 'been there, 'done that too, and some people
suggest--as And bluntly stated it a few weeks ago in another thread--that
there are as many "semantic" cases as there are "verbs" in a language (which
is a good thing to believe anyway when some French linguists believe that
there are as many semantic cases as there are relations between all the
words of a language.)

µ.
http://conlang.free.fr