Is There A Priori Knowledge by Testimony?

Abstract

In a series of recent papers, Tyler Burge (1993, 1997, 1999) draws an analogy between the perceptual processes normally involved in the acquisition of belief from others, and the memory processes normally involved in deductive reasoning.Both processes, he argues, are purely preservative, where a purely preservative process is one that "introduces no subject matter, constitutes no element in a justi cation, and adds no force to a justi cation or entitlement.It simply maintains in justi cational space a cognitive content with its judgmental force" (Burge 1993, 465).A purely preservative process, then, makes a belief (or other attitude) available, for later use or use by others; moreover, it does so without adding to, or altering, the original epistemic status of the belief (see, for example, ibid., 466, 486-87 and 1997, 37).It just transfers the original warrant (if any) along with the belief, from one's earlier self to one's later self, or from the source to the recipient of testimony.Importantly, the process does not affect the warrant's being a priori or a posteriori; nor does it affect the warrant's strength.So, on Burge's view, if one's earlier self has a priori (or a posteriori) warrant for the belief that p, then, if all goes