Knowledge and Its Limits

A technical but influential work that reverses the usual order of epistemology. Since Gettier, many philosophers had tried to analyse knowledge as true belief plus some further condition. Williamson argues that this project fails and that knowledge should be taken as basic, an approach now called 'knowledge first'. Knowing is itself a mental state, the most general factive mental state, which one can be in only with respect to truths. He argues that no interesting mental state is luminous, meaning that being in it always puts one in a position to know one is in it, using margin-for-error principles drawn from his work on vagueness; the same principles count against the KK principle, that knowing implies knowing that one knows. He proposes that one's evidence is exactly one's knowledge, summed up as E = K, and builds a theory of evidential probability on it. Other chapters treat scepticism, the claim that knowledge is the norm of assertion, and the paradox of knowability.

It is the founding text of knowledge-first epistemology, and its arguments on luminosity, evidence and assertion are among the most discussed in the field.