ABSTRACT
The “first-generation” logicism of Frege and Dedekind differs from subsequent versions by claiming a purely rational status for mathematics. An updated version of this claim is defensible. Mathematics is a priori in the sense of needing empirical data only to compensate for human limits of memory and attention. It is analytic in the sense that any rational beings satisfying minimal conditions would have reason to develop forms of arithmetic and set theory. This version of logicism is immune to standard objections. In particular, it avoids Philip Kitcher’s critique. My notion of analyticity, however, is more general than its traditional counterparts. One point is that analytic knowledge may be conjectural. Logicism is, ultimately, a thesis about human cognitive capacities and mechanisms.
