Abstract
Appraising the worth of others’ testimony is always complex; appraising the worth of expert testimony is even harder; appraising the worth of expert testimony in a legal context is harder yet. Legal efforts to assess the reliability of expert testimony—I’ll focus on evolving U.S. law governing the admissibility of such testimony—seem far from adequate, offering little effective practical guidance. My purpose in this paper is to think through what might be done to offer courts more real, operational help. The first step is to explain why the legal formulae that have evolved over the years may seem reassuring, but aren’t really of much practical use. The next is to suggest that we might do better not by amending evidentiary rules but by helping judges and attorneys understand what questions they should ask about expert evidence. I focus here on (i) epidemiological testimony, and (ii) the process of peer review.