Mark Colyvan Colloquium

Abstract: Various real and imagined criminal law cases provoke the intuition that there is something wanting with purely statistical evidence in the courtroom. But in the cases in question, the probabilities of guilt or culpability are very high—high enough to meet the relevant standard of evidence. This problem is known as the ‘proof paradox’. An oft-expressed position is that legal verdicts should be based on ‘specific’ rather than ‘general’ evidence of guilt. But it is unclear what this distinction amounts to. Moreover, we need to know whether general evidence is supposedly problematic for epistemic or for moral reasons. We argue that, all other things being equal, moral considerations should not influence the import of legal evidence. And given the salient ways of distinguishing specific and general evidence, there is no good epistemic reason for downgrading the latter. Finally, we argue that in the problem cases, the statistical evidence is indeed wanting but the deficiency in question is epistemic, not moral, in character.