Semantic holism vs. semantic atomism

Jerry Fodor and Ernest Lepore have argued that semantic holism is incompatible with the in-principle possibility of languages that consist of a single meaningful expression. Should the traditional Quinean/Davidsonian holist be worried by this apparently weird possibility? In this paper, I offer some reasons for why he should be. My argument focuses upon Davidson's account of how a hypothetical interpreter might come to understand an unfamiliar human language. Davidson's discussion of the methodology of so-called “radical interpretation” appears to rule out the possibility of semantically atomic languages upon purely a priori grounds. But this appearance turns out to be deceiving.