اثر خط 1 نوع در بازدارندگی

According to a conventional view, type-1 error (wrongful conviction) is as detrimental to deterrence as type-2 error (wrongful acquittal), because type-1 error lowers the pay-off from acting within the law. This view has led to the claim that the pro-defendant bias of criminal procedure, and its high standards of evidence, can be explained by the negative effect on deterrence of type-1 error. Adding new insights to some points made in the literature, we present a different view of the effect of type-1 error on deterrence, which seems incompatible with the claim. We thereby show that type-1 errors may not only lead to over-deterrence or to the chilling of socially benign acts, rather than to under-deterrence, but also to a socially desirable increase in deterrence or to a socially desirable lowering of activity levels. Moreover, since type-1 and type-2 errors are defined as court errors, i.e. as conditional on deterrence, their effects on deterrence depend on the likelihood of adjudication. For harm-based sanctions, harm is a condition for adjudication, and so for tortious acts or for certain criminal acts, the effect of type-1 error is proportional to the likelihood of harm. Since harm occurs more rarely and for some crimes not at all when the lawful act is chosen, type-1 error conditional on adjudication affects deterrence less than type-2 error, and for some crimes not at all. For the latter crimes, type-1 errors concerning the identity of the offender may still be thought to either affect deterrence or to chill socially benign activities. However, we show, contrary to claims in the literature, that type-1 errors concerning identity do not in themselves lower deterrence and we argue that chilling of socially benign activities is not a general phenomenon and so cannot explain the high evidentiary standards of criminal law that apply uniformly.