Base-rates of negative traits: instructions for use in criminal trials

Decision-makers in institutional and non-institutional contexts are sometimes confronted with the issue of whether to use generalisations expressing the statistical incidence of a negative trait in a disadvantaged and discriminated-against social group in order to draw an inference concerning a member of that group. If a criminal court were confronted with such a question, what answer should it give? First, the paper argues that, our qualms notwithstanding, morality does not demand that these generalisations be disregarded. In doing so, the paper addresses the relationship between factual accuracy and the demands of morality in criminal trials. Second, the paper considers the implications of this conclusion for the legal question as to whether the evidence at issue should be excluded, in particular, on grounds of unfairness – pursuant to section 78(1) of the Police and Criminal Evidence Act 1984.