A new argument for skepticism

The traditional argument for skepticism relies on a comparison between a normal subject and a subject in a skeptical scenario: because there is no relevant difference between them, neither has knowledge. Externalists respond by arguing that there is in fact a relevant difference-the normal subject is properly situated in her environment. I argue, however, that there is another sort of comparison available-one between a normal subject and a subject with a belief that is accidentally true-that makes possible a new argument for skepticism. Unlike the traditional form of skeptical argument, this new argument applies equally well to both internalist and externalist theories of knowledge.