Abstract

“eyes are surer witnesses than ears,” says Heraclitus, deploying the term ‘witnesses’ metaphorically to steer us toward what we can see for ourselves, and away from depending literally on the witness of others.1 Much ancient epistemology leans the same way. The tendency from pre-Socratic times on is to distinguish between doxa and epistêmê, and to say that ordinary human testimony on its own can give us no more than doxa.2 Some ancient philosophers have what we might call ‘rationalist’ reasons for not recognizing testimonially based knowledge: for them knowledge requires intellectual insight into some truth.3 Others have ‘empiricist’ reasons: for them, one’s knowledge must ultimately trace...