Since 1967, preventive theories of punishment (whether strictly utilitarian or more loosely consequentialist) have entered a long decline, beginning with the virtual disappearance of reform theory in the 1970s. Crowding them out are various alternatives generally categorized as "retributive."

All preventive theories treat punishment as (primarily) a means of controlling objectionable behavior. Insofar as they propose to justify punishment on the assumption that penalties can be tuned to achieve a certain degree of social control, they are empirically vulnerable (as well as morally vulnerable for ignoring justice). What had become clear by the 1970s was that social science could not then, or in the foreseeable future, give preventive theories much empirical content. Social science could not, that is, say what effect, if any, statutory penalties, rehabilitation, exemplary punishments, or even the incapacitation of criminals would have on the crime rate. If even relatively crude tuning of penalties...