This article is a response to an article by Alon Harel and Ariel Porat, recently published in the Michigan Law Review. In the article, the authors argue that, under certain conditions, courts should be permitted to convict a defendant in an unspecified offense. This possibility is meant to address situations in which there is no reasonable doubt that the defendant committed an offense, even though the prosecution failed to prove beyond reasonable doubt that the defendant committed any specific offense of which he was accused. The authors term this new decision principle as the Aggregated Probabilities Principle – APP.
In this response, I have tried to demonstrate the inadmissibility or the undesirability of the APP approach. The article shows that APP is premised upon a Pascalian mathematical approach to the standard of proof. In that framework, the authors’ treatment of interdependence is mathematically in error. A correct mathematical approach to probabilistic interdependence among offenses would render APP impractical.
Further, the response raises doubts as to whether adopting APP would actually lead to a reduction of enforcement costs or to minimizing adjudication errors, and shows that adopting APP would require that we choose between giving away equality or hurting deterrence.
- unspecified offense,
- reasonable doubt,
- rove beyond reasonable doubt,
- specific offense,
- Aggregated Probabilities Principle,
- retributive justice,
- inductive probability,
- L. J. Cohen,
- Inductive Reasoning,
Available at: http://works.bepress.com/mahdi_naamnee/1/