I wrote:
I wonder if anyone has thought of this. Pre-theoretically, we want to say
that 1) "The King of France is bald." "implies" or "pre-supposes" that there
is a King of France or that "There is a King of Fran ce." is true. Grice
offers up a theory where the utterance standardly implicates, but does not
*entail*, that there is a King of France or that "There is a King of
France." is true. But could we not argue instead that 1) *entails* that
probably there is a King of France or that "Probably there is a King of
France." is true? One advantage of looking at things this way is you don 't
have to invoke any extra-logical notation like `~>'. One supporting line of
argument for this way of looking at things is, even assuming the vocabulary
of implicature (the wavy arrow and etc.), it is surely the case that we
would count a piece of discourse as *Rational* (and therefore, maybe, as
truth valued), only if the standard implications of the utterances contained
in that piece of discourse were, *mostly* , in place (not cancelled).
Randy responded:
Another way to look at this is that the sentence "The present king of France
is bald" doesn't imply or presuppose anything, but the USE of the sentence
in a context where "The present king of France" is taken as a topic would
lead to the presupposition that there is a present king of France, as we
normally would not make something a topic (subject to some predication) if
it did not exist (in some world). The suggestion of talking about probable
existence reflexts a kind of pragmatic accomodation that we would do if we
heard such a sentence, i.e., assume that if the speaker talks about
something as a topic, then it probably exists, assuming the speaker is
rational (CP). If the sentence is used in a context where "The present king
of France" is not a topic, but focal (e.g. "Which of the kings are bald?"),
then it does not evoke the existential presupposition and can have a truth
value. (Georgia Green talks about this in her pragmatics textbook).
----Thanks for the response, especially since it seems that I bunged up my own example. The origonal sentence that Grice makes so much hay out of is "The present King of France is *not* bald." where, he argues the linguistic meaning of the sentence does not entail the existance of the King of France, but in most contexts the speaker will implicate that there is a King of France (though this implication may be cancelled). I know that the same kind of arguments come up with def. descs. in positive statements (referential vs. attributive uses), but I haven't really thought the issue through there and in fact *meant* to write all of my examples above in the negative form. In any case, that's all I'll talk about below.
You write:
"The suggestion of talking about probable existence reflexts a kind of pragmatic accomodation that we would do if we heard such a sentence, i.e., assume that if the speaker talks about something as a topic, then it probably exists, assuming the speaker is rational (CP). "
It sounds like you are saying that, for each utterance, we decide that CP holds on the basis of that particular context of utterance, which I don't think will work. Whereas I would say that the assumption that CP holds is one we bring to every context (other than maybe fictional ones). Its a precondition for further interpretation. So we bring the assumption that CP holds plus the linguistic meaning of the utterance to the context, and calculate implicatures from the context.
If so, then the calculation that "There probably exists a King of France." isn't something we have to calculate from the particular context. Its something that we can calculate merely from the background (a priori?) assumption that CP holds and the linguistic meaning of the utterance, outside of any particular context. Something like "If CP holds then the standard implications associated with a phrase/word etc. are probably "in place"." and "If CP holds then in "The current King of France is not bald." the standard implicatures associated with the sentence, in particular with the definete description it contains, are probably in place. CP holds, therefore there probably exists a King of France.
What I am thinking is that a derivation from an utterance, according to Grice, goes roughly as follows. Assume CP and the linguistic meaning of an utterance being what it is, then calculate inferences based on the maxims, each of which inference is likely, to some implicated proposition: "The King of France is not bald and there is a King of France." However, the last half of the conjunct is not *entailed* because the inferences, each likely, may nevertheless be incorrect (because of the phenomenon of cancellability). On the other hand, I would think that, had the derivation gone from CP plus linguistic meaning to "Probably x, and probably y, therefore "The King of France is not bald and probably there is a King of France." then the latter conjunct would be *entailed*. Even if there turns out to be not King of France, the inference is okay: "Probably x." is not false if ~x.
Hope this is clear.
Incidentally, was your message to the list or private? cc'd messages don't appear.
Cheers,
M.J. Murphy
The shapes of things are dumb. -L. Wittgenstein
The shapes of things are dumb. -L. Wittgenstein
This archive was generated by hypermail 2b29 : Thu Jun 06 2002 - 22:25:47 GMT