William A. Edmundson

John Rawls argued in A Theory of Justice that “justice as fairness…is likely to have greater stability than the traditional alternatives since it is more in line with the principles of moral psychology”. In support, he presented a psychology of moral development that was informed by a comprehensive liberalism. In Political Liberalism, Rawls confessed that the argument was “unrealistic and must be recast”. Rawls, however, never provided a psychology of moral development informed by a specifically political liberalism, leaving it at a disadvantage with respect to comprehensive liberalism itself. I argue that no coherent account is available. But, because Rawls’s Liberal Principle of Legitimacy, along with its implied stricture against “indoctrination in the pejorative sense,” is a creature of ideal rather than non-ideal theory, the deficiency is far less significant than many would assume. In non-ideal circumstances -- such as ours -- political liberalism does not disallow, and in fact requires, compulsory, free early education informed by the best available comprehensive liberal philosophy.

The late Jerry Cohen struggled to reconcile his egalitarian political principles with his personal style of life. His efforts were inconclusive, but instructive. This comment locates the core of Cohen’s discomfort in an abstract principle that connects what we morally ought to be compelled to do and what we have a duty to do anyway. The connection the principle states is more general and much tighter than Cohen and others, e.g. Thomas Nagel, have seen. Our principles of justice always put our personal integrity to the test, unless those principles are designed not to. But to craft principles with an view to avoiding that test is, as Cohen argued, itself to undermine both justice and our integrity.

The question, "Why should I obey the law?" introduces a contemporary puzzle that is as old as philosophy itself. The puzzle is especially troublesome if we think of cases in which breaking the law is not otherwise wrongful, and in which the chances of getting caught are negligible. Philosophers from Socrates to H.L.A. Hart have struggled to give reasoned support to the idea that we do have a general moral duty to obey the law but, more recently, the greater number of learned voices has expressed doubt that there is any such duty, at least as traditionally conceived. The thought that there is no such duty poses a challenge to our ordinary understanding of political authority and its legitimacy. In what sense can political officials have a right to rule us if there is no duty to obey the laws they lay down? Some thinkers, concluding that a general duty to obey the law cannot be defended, have gone so far as to embrace philosophical anarchism, the view that the state is necessarily illegitimate. Others argue that the duty to obey the law can be grounded on the idea of consent, or on fairness, or on other ideas, such as community.

It is commonly supposed that citizens of a reasonably just state have a prima facie duty to obey its laws. In recent years, however, a number of influential political philosophers have concluded that there is no such duty. But how can the state be a legitimate authority if there is no general duty to obey its laws? This article is an attempt to explain how we can make sense of the idea of legitimate political authority without positing the existence of a general duty to obey the law. The explanation makes use of a distinction between laws of general application, on one hand, and on the other the particularized, directed efforts by state officials to channel and resolve disputes (including those arising from violations of the law). A state's legitimate authority entails a general duty to cooperate in the latter type of effort, rather than upon a dubious general duty to obey the law.

Contextualism answers skepticism by proposing a variable standard of justification, keyed to the context of utterance. A lawyer's situation with respect to a criminal defendant's factual guilt is a special one. The argument here is that in this special context an especially high standard of epistemic justification applies. The standard is even more exacting than the proof-beyond-reasonable-doubt standard that juries are sworn to follow. The upshot is that criminal defense lawyers normally cannot know that a client is factually guilt.

This is a critical discussion of the Baldus study of capital sentencing in Georgia. It concludes that the Baldus finding of a "race-of-the-victim" effect is less robust than capital-punishment abolitionists have claimed. But the flaws in the Baldus study should not comfort death-penalty advocates, for they reveal an epistemological barrier to the US Supreme Court's ever being able to satisfy itself both that the sentence reflects particularized consideration of the circumstances and character of the defendant and that it is not the product of racial bias