Fairness is a central issue in our scientific understanding of the law and criminal behavior. The fairness of legal proceedings has long been of interest in a variety of disciplines, but only recently have theory and research in criminology begun to focus upon the effects of fairness on illegal behavior. Procedural justice theory suggests that offenders who believe that they have been treated fairly will also be less likely to reoffend in the future. Empirical research has been generally supportive of this prediction, but is rather limited in volume and has been unable to establish an unambiguous direction of causality. The analysis presented here is a preliminary effort to bring the power of randomized experimentation to the criminological study of fairness. An Australian field experiment has randomly assigned 900 persons, all arrested for intoxicated driving, to one of two methods for dealing with their offense. Half of these drivers have had their cases referred to the courts in the traditional manner, while the other half have been offered the opportunity to attend a diversionary conference instead of going to court. Among other things, these conferences--which are broadly based upon the restorative justice ideals outlined by Braithwaite (1989; 1998)--are thought to be substantially different from court in terms of the fairness they offer to the offenders who participate in them.