The Shifting Content of Knowledge Attributions

Abstract

In Contextualism and Knowledge Attributions, Keith DeRose defends a contextualist theory of knowledge.' He claims that his theory is superior to some earlier relevant alternatives theories in respect of its proper handling of issues concerning the meaning of knowledge attributions. I think that some of DeRose's key claims on this score are mistaken. DeRose wants to distinguish two sorts of factors that can affect the truth values of knowledge attributions: subject factors are features of the putative knower's situation, and attributorfactors concern the linguistic and psychological of the attributor of knowledge, who may well be distinct from the putative knower. (918-19) Attributor factors, but not subject factors, affect the truth values of knowledge attributions by affecting the truth conditions-in particular, the content, in David Kaplan's technical sense-of the knowledge-attributing sentences. The sentence 'This room is hot' has a constant character in different contexts of use, though its content shifts when I first utter it in my bedroom and later utter it in my kitchen. The character of a sentence is a function from contexts of utterance to contents, and the content is, what is said in a context of utterance (what I say differs from room to room in the above example).2 According to DeRose, the context-invariant character of 'S knows that p' is, roughly, that S has a true belief that p and is in a good enough epistemic position with respect to p. (922) However, the content of a knowledge-attributing sentence can shift when the context of attribution shifts in certain ways. DeRose says, What the context fixes in determining the 'content' of a knowledge attribution is how good an epistemic position S must be in to count as knowing that p. (922)