Legitimate Punishment in Liberal Democracy

Abstract

What are the terms of legitimate punishment in a liberal democracy? Traditional approaches to this question tend to focus on the purposes punishment is supposed to serve (deterrence, retribution, rehabilitation, moral education, etc.) while giving little if any consideration to the coercive deployment of state power punishment represents. In this article, I take the coercive nature of state punishment as my starting point. My aim is to determine what normative constraints, if any, exist on the state's power to punish criminal offenders in a liberal democracy - a determination, I argue, that is especially urgent given the current size of America's prison population. To answer this question, I draw on the work of John Rawls. I do so because I share Rawls' view that, if the exercise of state power in a liberal democracy is to be legitimate, it must be justifiable in terms that all members of society subject to that power would accept as just and fair. Rawls' deliberative model was originally intended for questions of ideal theory, on which all members of society are assumed to act justly towards others. The first task of the paper is thus to render Rawls' model applicable to problems of partial compliance, of which punishment is one. Ultimately I argue that, assuming conditions of partial compliance, deliberating parties would approach the task of selecting principles of punishment by considering the implications of various alternative principles as if they could end up as either crime victim or punished offender once they enter society as citizens. Having established this perspective, and its consistency with the basic liberal ideals of moral equality and individual sovereignty, I then go on to determine the principles of punishment that would be selected by parties deliberating under these conditions and would thus constitute the terms of legitimate punishment in liberal democracy. I identify five such principles, at the heart of which is what I call, following Braithwaite and Pettit, the "parsimony principle." The basic idea of this central principle is that the punishment of convicted offenders must be no more severe than necessary to yield an appreciable deterrent effect on the commission of serious offenses. Finally, I consider how the principles of legitimate punishment might be translated into actual criminal justice policy. Here, I concede that the inevitability of reasonable disagreement, even among legislators deliberating in good faith over what punishments the principles allow, means that in practice we can never be fully confident of the legitimacy of any punishments imposed. As I show, however, the principles of legitimate punishment I identify still provide the basis from which to call into question the legitimacy of a range of criminal justice policies currently in force in the United States, including mandatory minimums, California's "three strikes" law, the under-funding of indigent defense, and the widespread overcrowding and sexual violence in the nation's prisons and jails. In this way, the theoretical analysis I offer provides a basis for challenging the legitimacy of many criminal sentences being served right now in American prisons.