Williamson on Evidence and Knowledge

Timothy Williamson’s Knowledge and its Limits is a treasure trove of philosophical riches. By reversing the usual explanatory order and putting “knowledge first” in epistemology (p. v) Williamson reconceives the relationships between knowledge, belief, justification and evidence. While there is much to be said about Williamson’s ambitious project, I will focus on his imaginative account of evidence (Ch. 9). Most epistemologists explain knowledge in terms of evidence: knowledge requires justification, and justification is a matter of having sufficient evidence for one’s beliefs. This approach assumes that evidence can be understood independently of knowledge. Williamson denies this, and analyses evidence in terms of knowledge. He defends the “E = K thesis” that “knowledge, and only knowledge, constitutes evidence. . . . S ’s evidence [is] S ’s knowledge, for every individual or community S in any possible situation” (p. 185). While E = K does capture some aspects of our concept of evidence, I will argue that it fails to secure others. Williamson ends up characterising an austere ‘high church’ evidence that leaves out many core features of our cognitive and social practices: among them, the fact that evidential status is generally a matter of degree, the fact that we often speak of a person’s evidence when presenting her own subjective reasons for believing what she does, and the fact that what counts as evidence varies with context. Let me begin by saying that I agree with Williamson on a number of points of methodology and substance. First, he correctly recognises the central role for a theory of evidence in epistemology (pp. 188–190). Second, he sees that a theory of evidence should be judged on the basis of its ability to explain core features of our evidential practices. Third, he is right that only something propositional can do the job that evidence does. (For those tempted by the idea of non-propositional evidence I can recommend no better tonic than Section 9.6 of Knowledge and its Limits.) Finally, Williamson is one of the few authors who explicitly recognises that a proposition e’s ability to serve as evidence for an hypothesis h depends both on e’s relationship to h and on e’s own epistemic status. This divides the theory of evidence into what I call a theory of evidential relevance, which seeks to determine whether and how much “e speaks in favour of h”, and a theory of evidential status, which specifies the sort of “creditable standing” that e must have in order to count as evidence for anything (p. 186). While many epistemologists focus on only part of the job, Williamson addresses both issues. Williamson’s theory of evidence can be stated in two theses: