The project deals with the ways in which generalizations, both quantified and unquantified, are used in legal fact-finding, either explicitly as pieces of evidence or implicitly as support for certain inferences. The project begins by identifying confusing intuitions and practices, according to which some types of generalization are used as a matter of course (e.g. compensation for loss of earnings relies on the average life expectancy), whilst other types of generalization raise strong objections, such as the case of Sally Clark, in which an expert witness relied on a quantified generalization according to which the probability of two cot deaths in one family like the Clarks is 1 per 73 million.
The project develops an original account to aid in distinguishing between acceptable and objectionable generalizations. The new account is grounded on Kant’s concept of moral autonomy. According to the account, a generalization is objectionable when it requires the legal fact-finder to presuppose that an individual’s behaviour was determined by a property he shares with other people and when the evidence is used in the context of attributing personal culpability to that individual. For example, using racist generalizations in legal fact-finding is objectionable not only because these generalizations might be factually wrong or offensive to the accused and to other people. Using racist generalizations also undermines the very basis of attributing moral responsibility to the accused, because their use requires the fact-finder to presuppose that members of the group in question share some property that causally influences their criminal behaviour and this presupposition is inconsistent with attributing personal culpability to the accused. The account also emphasises that our approach to the use of generalizations in court should be derived from answers to deeper questions about what “culpability” is, how it should be determined and what role it should have within the law.
Field of science
- /social sciences/law
Call for proposal
See other projects for this call