When individuals violate the law, detection and verification of the violation are rarely, if ever, perfect. Before the state can dole out punishment, it must first identify a suspect and then produce sufficient evidence to persuade a judge and/or jury beyond some threshold level of confidence that the suspect, in fact, violated the law. The court might be uncertain that the state has the right person. If the suspect is undoubtedly the one who caused the harm, the court might be unsure about whether his act constitutes a violation of the law (e.g., whether the suspect was, in fact, speeding). The state, given the level of resources allocated to law enforcement, might not be able to produce a suspect. Limitations on enforcement resources lead to imperfect detection. Evidence production and proof problems cause both mistaken convictions and mistaken acquittals. Errors have many sources, including hindsight bias, lack of complete information about the defendant’s possible options and chosen action, untrustworthy eyewitness testimony, the admission of impartial evidence in trials, and unwillingness or inability to expend resources on detection, among others. We focus here not on the sources of errors, but rather on their effects on deterrence and punishment policy. Our purpose is to briefly summarize the theoretical literature that studies the effects of legal errors on crime and punishment rates, and to critically review studies that report on experiments conducted to test such theories. The theoretical literature includes analyses of both criminal law and civil law violations, and so we cover both here. Part 2 summarizes theories offered to explain and predict how imperfect detection and guilt-determination errors affect crime and punishment rates. Part 3 summaries, synthesizes and critiques experimental studies designed to test the theories. Part 4 catalogs, in broad terms, where we are and offers ideas for potentially fruitful avenues for continued exploration in the lab.
Read full abstract