Re: CHAT: THEORY/CHAT: Re: Jackendoff's "Semantic (?) Structures"

I can't reply to the question, but I can give some
further reflexions about J.'s book.

I was very pleased to discover, when reading
Jackendoff ch.5.2, that he said exactly what I thought
as I was reading ch.2.1.

Ch. 5.2. is entitled "Verbs of Manner of Motion and
Configuration". J. says:
"The sentences in (1) describe motion of an object but
do not imply traversal of a Path.
(1) a. Willy wigled
b. Debbie danced.
c. The top spun.
d. The flag waved.

Compare (1) to (2), where there is an implicit Path.
(2) a. Emma entered.
b. Aaron approched.
c. Leon left. (...)"

Later he says that "sentences in (1) describe only the
internal motion of the subject, with no implications
with respect to their location, change of location, or
configuration with respect to any other object."

Well, since I myself came to that conclusion BEFORE I
read this passage (and before reading anything from
Jackendoff), I think I have good reasons to take this
for grounded. I'm more dubious about the next
sentence: "This suggests that, unlike the previous
Event- and State-functions, the function in question
takes a single argument." But anyway, how comes that,
when reading an earlier chapter, I couldn't understand
why he didn't mention this at that moment?

So I found back the earlier chapter (2.1) and noticed
he was both talking about "into" and about "run".
"Run, which express a GO-function, requires two
arguments: the Thing in motion and the Path that
specifies the trajectory of motion. (...) If no PP is
syntactically present, the Path is simply unspecified:
'John ran' means in part "John traversed some
(unspecified) trajectory".

Haha. When I read this, I thought: this is not always
true, "to run" can be understood as a physical
exercice, as a motion of the body, with no reference
to any Path at all. So why does J. consider that this
would be true for "to dance" and not for "to run" ?
Because one is supposed to run in a straight line,
from point A to point B, while dancing is more on one
spot, or maybe as a brownian movement on the
dancefloor ? This is not always true. As I said, one
can run on a training outfit, and in that case (for
ex), the concept is very similar to "dancing": it
means, to move one's body in a special way, and NOT to
go from some place to some other place. Also think
about children running around when playing: they don't
run FROM or TO any special location, they just run,
that's all: this is exactly the same as for dancing.
And, from an opposite point of view, you also can
imagine actions like "dancing away" or "dancing to",
or "from", or "along", or whatever.

So, clearly, there are not, on one side, verbs like
"to run" with 2 arguments, and the the other side,
verbs like "to dance" with only 1 argument. How comes
that J. apparently (except in case I find a new point
of view later on in his book) didn't understand, or at
least refused to consider, this ? Especially if you
consider another example he gives further in ch.5.2,
about Spanish:
(5) La botella floto a la cueva (the bottle floated to
the cave)
which he comments:
"The translations must rather include both a verb of
external motion (GO) and a participle expressing
manner of motion (MOVE):
(6) La botella entro a la cueva flotando > The bottle
moved-in to the cave floating.

This is very exactly what I thought when I read his
passage about "to run"  ! "To float" and "to float
into" are two different concepts, the secund one being
a compound of the first one ! Just like "to run" and
"to run toward" for ex.

I also disagree about his example "The car hit the
tree", where "the tree" is considered as a Goal. To
me, such a sentence essentially means "the car
smashed", or at least "there was a problem on the
trajectory of the car", "the car had a bump", and "the
tree" is just something like a circumstancial
complement, explaining more closely how it happened.
If we wanted to talk from the point of view of the
tree, we would have said "the tree was hit by the
car". I can't see any Goal in "the car hit the tree",
while there surely is one when saying "I flew to New
York". The tree was an obstacle, not a goal.

Also, he doesn't seem to clearly see that motional
concepts don't work like other action concepts. For
ex, he mentions (ch.7.1), that, while one can say:
"What Pete did to the ball was hit it into the field"
one cannot say:
*What Bill did to the room was enter it".
Well, in my opinion, such a comparison is pointless.
Suffices to compare to French: "Bill est entre dans la
piece" (not "Bill a entre la piece"), so the
transitive form in English in this case is just a
syntactic peculiarity. Of course, "to enter a room"
has nothing to do with "to hit a ball", except
english-syntactically. J. also wonders somewhere why
one cannot say "the room was beentered by Bill". Well,
in French you can say for ex "cette piece a vu entrer
Bill", which is kind of the passive concept
corresponding to "Bill est entre dans cette piece".
Also, "Armstrong a marche sur la Lune" vs "la Lune a
ete foulee par Armstrong". "L'explorateur a penetre la
foret vierge" vs "La foret vierge a ete penetree par
l'explorateur". The fact that there is apparently no
such form as "to be beentered" in English (or "etre
entre par" in French) doesn't mean a thing.

Also, assertions like "to kill somebody" = "to be the
cause of the death of somebody" are not very
satisfactory. Suffices to compare:
Bertrand Cantat killed the actress Marie Trintignant
Bertrand Cantat caused the actress Marie Trintignant
to be dead
to feel the difference (the advocate for M.Trintignant
would probably use the 1st formulation, while
B.Cantat's advocate would rather use a formulation
close to the 2nd one). One cannot just think in terms
of causation. "To kill" implies much more that just a
causation. This is poor semantic analysis, IMO.

--- Jonathan Knibb <j_knibb@...> wrote:
> This is an extremely interesting thread, and I only
> wish I could
> contribute more intelligently to it.  However, while
> I'm halfway
> through Jackendoff's 'Foundations', Bouchard's
> sitting on the shelf
> and Lakoff's in the post, I shall content myself
> with asking one
> question, following on from Philippe's question
> about the difference
> between semantics and cognition.  The question is:
>
> Is there any formal, qualitative difference between
> linguistic
> semantics and other biological semiotic systems?
> For example, can
> semantics be said to be generative (in and of
> itself, aside from its
> relationship to syntax)?
>
> Jonathan.

=====
Philippe Caquant

"High thoughts must have high language." (Aristophanes, Frogs)


	
		
__________________________________
Do you Yahoo!?
Yahoo! Photos: High-quality 4x6 digital prints for 25¢
http://photos.yahoo.com/ph/print_splash

Reply