Avoiding Bayes in DNA Cases

Abstract

In a recent article David Hodgson suggests that where DNA evidence is presented in statistical terms it would be helpful for juries to be directed how to apply Bayes’ Theorem. I argue that such direction involves applying a mathematical conception of proof that has never been accepted by courts, that courts approach proof inductively, and that the inductive approach can be better preserved by presenting DNA evidence as a frequency rather than as a likelihood ratio. This avoids the need to direct juries in Bayesian terms.