Bayesian verdicts

By F.E. Guerra-Pujol

degree of belief probability juries voting bayes

Why do juries emit binary verdicts, i.e. guilty or not guilty? Why not allow each juror to emit a "Bayesian vote" reflecting his/her subjective degree of belief in the probability of the defendant’s guilt? Under this alternative method of Bayesian voting, jurors would "score" the evidence presented by the parties at trial on a scale of 0 to 1 or some other scale. (Whatever scale is used, the higher one’s score, the greater the one’s subjective degree of belief in the defendant’s guilt, while a midpoint score, such as 0.5, means the juror is undecided about the question of guilt.) Such a simple system of scoring would then produce a numerical verdict, a Bayesian verdict, consisting of an average value or sum total of the jurors’ individual scores. Under this method of voting, the moving party would prevail only if the average value or sum total, as the case may be, of the jury’s collective score exceeds some critical threshold value.

Comments

Please log in to add a comment.