Aune, Bruce - An Empiricist Epistemology Ch 3 "Empiricism and the A Priori"

05/18/2007

Unpublished Manuscript

In this chapter author gives a positive account of analyticity and how it might work in relation to so-called propositional attitudes. The primary aim is to show that cases of supposed a priori knowledge are just cases of analyticity in language or in concepts.

The chapter begins with a brief survey of the origins of analytic truths as conceived by Kant. For Kant, an analytic truth (similar to Leibniz) is a truth where the predicate being ascribed to the object is already contained in the concept of the object itself. Author considers Kant's work to be acceptable, but only applicable to a limited class of judgments. (pg 82) Frege attempted to build on Kant by claiming that an analytic truth is a truth derivable only from general logical laws and definitions.

Frege and other empiricist philosophers were generally considered to be refuted by Quine in his paper "Two Dogmas of Empiricism". Author describes the problem was that we cannot use the idea of synonymy without using 'analytic', and vice versa. The two are involved in a vicious circle of definition. Further, any attempt at lining up all the sentences containing supposedly synonymous terms will contain the term 'analytic'. Quine later retreated from his earlier attack and admitted that there was some common-sense appeal to analyticity. He gives a rough definition: something is analytic for a native speaker S if S learns the truth of P by learning the usage of one or more of the words in P. This truth must be deductively closed-- so the steps to analyticity must 'count as analytic in turn'. (pg 88) But still, as with Kant, Quine relegates analytic truths to logic ones and linguistic tautologies.

Author describes his version of analytic truth as one developed from Carnap's, which consists of the specification of a formal language-system that has semantic rules and definitions. Author uses the example of specifying how 'if...then' applies, a usage that is separate from common usage. Once these specifications are considered, then the supposed counterexamples to modus ponens and modus tollens are clarified and dismissed (pg 97-98).

Author lastly turns to the problems with necessary truths that Kripke has shown. One example is the origin claim: an item cannot have it's origins in a different hunk of matter than it does now. (pg 112-) Author goes about showing what the proof for this would be. Assuming that 'distinctness' is defined properly; author claims this Kripkean claim can be considered analytic.

Lastly, author considers psychological states, which are not the same as propositions. If propositions can be analytic, can psycholgical states like beliefs? Author review the classical notion of propositions as expressing the sense of a sentence, with words having conceptual content. With the theory of names being rigid designators this classical notion is undone. Author discusses the ramifications of this failure and also discusses conceptualism, which says that propositional attitudes have 'contents' rather than 'objects'. Using this method at least, the empiricist can have analyticity in psychological states.