Wolfgang Schwarz

Analyticity

I keep wavering between two different uses of "analytical". This entry is meant to remind me of the difference and of why I should prefer the one over the other.

On the first use, a sentence is analytical if it has a universal A-intension. On the second, a sentence is analytical if one can't understand it unless one believes it (this is what I, unoriginally, proposed last year). The first is the better explication.

One problem with the second is that understanding a sentence might require a lot of substantial knowledge. For a trivial example, perhaps you can't understand "I understand this sentence" without believing that it's true, yet it isn't intuitively analytical. Or suppose understanding "there is water" requires knowing that "water" denotes watery stuff. Then "there is water or 'water' denotes watery stuff" wrongly comes out analytical on the second explication. More generally, some self-declared rationalists like Chris Peacocke argue that substantial a priori knowledge can be somehow explained in terms of the 'possession conditions' of the relevant concepts. On the second explication, such knowledge would always be knowledge of analytical truths, which seems to contradict the claim that it is substantial.

The other (more Quinean) problem with the second notion is that it is very hard to say what believing analytical truths amounts to. (Robert Stalnaker has completely convinced me of that.) Believing that all bachelors are unmarried can hardly amount to behaving in a way that would be rational if all bachelors were indeed unmarried. For that condition is empty. For the same reason we can't say that it amounts to having an internal representation that is correct iff all bachelors are indeed unmarried. Presumably, believing or disbelieving analytical truths must have something to do with asserting or not asserting sentences: you don't believe that all bachelors are unmarried if you're not disposed to assent to "all bachelors are unmarried". But then only people who understand English can believe that all bachelors are unmarried. So we should allow translations. But what kinds of translations are allowed? And if assenting to strict synonyms is allowed, then how is it possible that one can believe that all eye doctors are eye doctors without believing that all eye doctors are ophthalmologists?

The best way out of this mess is to assume that what appear to be attitude attributions with analytic content are usually attitide attributions with synthetic, meta-linguistic content. "Paul believes that ophthalmologists are ear doctors" attributes to Paul a contingently and synthetically false belief about the word "ophthalmologist".

From this point of view, the confusion motivating the second explication of "analyticity" is nicely explained: understanding a sentence ordinarily involves possessing knowledge about language; and knowledge about language is what we really attribute when we appear to attribute knowledge of analytical truths. Hence the mistaken idea that knowing an analytical truths requires (just) knowledge about language. But look at how strange this is: analytical truths like "all bachelors are unmarried" are not about language at all. So why would you have to know something about language in order to know that all bachelors are unmarried? (Even worse, you will presumably have to know something about the English language in order to know that all bachelors are unmarried.) The second conception of analyticity confuses analytical truths with the meta-linguistic truths belief in which we usually attribute when we appear to attribute belief in analytical truths.