Epistemic Contextualism: A Defense and Analysis

Epistemic contextualists maintain that the truth conditions of knowledge ascriptions and denials change according to the context of utterance. In this dissertation, I defend this view against one of its main rivals, classic invariantism, which holds that the contents of such statements remain fixed across contexts. While epistemic contextualists provide a straightforward semantic account of the variability in our knowledge-ascribing behavior, classic invariantists cannot, and therefore must offer some explanation as to why it seems as though the standards for ‘knowing that p’ shift from one context to the next. To this end, classic invariantists draw a distinction between what speakers convey (the pragmatic content of their assertions) and the proposition literally expressed by their utterance. They maintain then that the standards of ‘knowing that p’ seem to vary, not because the meaning of ‘knows’ shifts, but rather because what is conveyed or grasped by these utterances vary from one context to the next. The main thrust of classic invariantism comes from the claim that invariantists can account for fluctuations in uses of ‘knows’ without adopting a context-sensitive analysis of the term. My aim in the dissertation, then, is two-fold: First, I chip away at the general strategy of invoking pragamatic implications to explain away unfavorable linguistic intuitions, thus undermining the classic invariantist’s capacity to explain the phenomenon in question: the variability of our knowledge ascriptions and denials. Without a proper explanation of our use of ‘knows’, one that clarifies why we believe that ‘knows’ applies in a given context when it strictly does not, the entire invariantist enterprise fails. After a brief introductory chapter, I discuss, and argue against, a version of moderate invariantism in the second chapter, skeptical invariantism in the third and low-standard invariantism in the fourth. In the fifth chapter, I show the context-sensitivity of ‘knows’ is linguistically supported by identifying ‘knows’ within the class of gradable context-sensitive expressions such as ‘tall’ and ‘flat’. Advisor: Albert Casull