Bayesian Litigation (An Application of Bayess Theorem to Law)

Abstract

Ron Allen and Mike Pardo have criticized probabilistic theories of legal proof and have presented an alternative “relative plausibility” theory of legal proof. Alas, both Allen and Pardo’s critique and their new approach are wrong. The remainder of this paper will describe their theory of relative plausibility, restate Allen and Pardo’s objections to subjective or Bayesian probability, and then answer each one of these objections. This paper will then explain why a Ramsian or subjective view of proof provides a better explanation of juridical proof than Allen and Pardo’s relative plausibility theory. At the end of the day, the key question is this: which of the two competing theories of juridical proof--i.e. subjective probability or relative plausibility--makes more sense, either as a descriptive or normative manner? In summary, because relative plausibility models can be Dutch-booked, and because juror evaluation of proof is an essentially subjective exercise, this paper shall defend a subjective conception of legal proof. Lastly, I shall describe a simple method for operationalizing subjective verdicts in civil cases.