Crime: Accident or Disease: An Exploration Using Probability Models for the Generation of Macro-Criminological Data

Abstract

One approach in practical criminology, which some call social work, is based on a disease model for most crimes. We have adopted another approach: for certain classes of crime, we have likened the occurrence of an offense to an accident or random event.

Thus, we have drawn heavily on studies of the occurrence of accidents and of accident proneness, which have used probabilistic models to account for these phenomena. In this way we should be able to investigate different accounts of the processes of crime and social control which generate criminological data.

Although aggregate information on offenders and their offenses is usually sparse, we have attempted to test various theories using these data. Without such an attempt, we cannot evaluate the general applicability of conclusions drawn from the more usual research conducted on easily accessible samples which may be a collection of extreme cases.

We find, however, that the different theories are inadequately specified, and that the collection of aggregate data is not normally designed to test them. Unless there are improvements in both theory and data, we shall not be able to distinguish between the various theories of crime and its social control.