« Two Types of Content Verbs | Main | Fregean Anti-realism »



You write that:
"But notice that, since Frege is explicit that syntactically distinct sentences can express the same proposition, it follows that Fregean logical form can mirror at most the syntactic structure of one of those sentences."

Why is that? Suppose somebody gives five syntactic structures of five sentences in Swahili. They all express exactly the same proposition. Why can't those syntactic structures have five separate Fregean logical forms? Those logical forms will be interderivable, and they will be true or false in exactly the same conditions. But can't they be five instead of one?

Tony Marmo

To make one long story brief, because in such case you are translating into or mapping natural language utterances onto a logical language utterance= proposition.

So, under this concept of logical form, when you say five sentences express one proposition, this proposition is the very endpoint of the translation from a natural language to a logical language.

On the other hand, if you think of LF in terms of May's and Chomsky's works, then it is a little bit tricky to say that five different syntactic derivations will yield the same conceptual output.


"They all express exactly the same proposition. Why can't those syntactic structures have five separate Fregean logical forms?"

Good question. I think I was too quick on this one. But here is a partial response.

On the Fregean conception, logical form is a characteristic of a proposition. In fact, it is an analysis of that proposition. So your question boils down to this two part question: (i) can a proposition have multiple, logically equivalent analyses, (ii) one corresponding to the LF of each synonymous sentence?

It seems to me that the answer to (i) for at least some propositions is, "Possibly". That is, I think it is an open question about the granularity of propositions (and I don't think that all propositions must have the same granularity). So you are right to push here.

As to the second part of the question: does the LF of some sentence which expresses the proposition p correspond to an analysis of p? Well, I hope this much is clear: not obviously.

One reason for being skeptical up front is that LF is a language dependent concept, whereas Fregean logical forms are language independent. I guess I would find it surprising (even counterintuitive) if the LF of every sentence in every possible language that expresses the proposition that p entails a corresponding analysis of p. At any rate, I hope that is enough to make the two concepts of logical form look sufficiently different to make their identification (or mere equivalence) appear nontrivial.

Thanks for the comment.

(Incidentally, Frege, it turns out, thought that propositions didn't have unique analyses (see his "On concept and object"). But even on Frege's view the distinct analysis wouldn't, in any straightforward way, correspond to distinct LFs.)

The comments to this entry are closed.

August 2011

Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31      
Blog powered by Typepad