Peter Vallentyne

This is the second volume of Oxford Studies in Political Philosophy. Since its revival in the 1970s political philosophy has been a vibrant field in philosophy, one that intersects with jurisprudence, normative economics, political theory in political science departments, and just war theory. OSPP aims to publish some of the best contemporary work in political philosophy and these closely related subfields. The papers in this volume address a range of central topics and represent cutting edge work in the field. They are grouped into three main themes: ideal theory, the moral assessment of states, and issues in social reliations.

Left-libertarianism, like the more familiar right-libertarianism, holds that agents initially fully own themselves. Unlike right-libertarianism, however, it views natural resources as belonging to everyone in some egalitarian manner. Left-libertarianism is thus a form of liberal egalitarianism. In this article, I shall lay out the reasons why (1) left-libertarianism holds that (a) private discrimination is not intrinsically unjust and (b) it is intrinsically unjust for the state to prohibit private discrimination, and (2) that, nonetheless, a plausible version of left-libertarianism holds that it is unjust for the state (and many private individuals) to take no steps to offset the negative effects of systematic private discrimination. The basic line is not new. It is simply that there is nothing unjust in principle with private discrimination, but there is (at least typically) something unjust about doing nothing to promote equal life prospects.

We address the question of how finitely additive moral value theories (such as utilitarianism) should rank worlds when there are an infinite number of locations of value (people, times, etc.). In the finite case, finitely additive theories satisfy both Weak Pareto and a strong anonymity condition. In the infinite case, however, these two conditions are incompatible, and thus a question arises as to which of these two conditions should be rejected. In a recent contribution, Hamkins and Montero (2000) have argued in favor of an anonymity-like isomorphism principle and against Weak Pareto. After casting doubt on their criticism of Weak Pareto, we show how it, in combination with certain other plausible principles, generates a plausible and fairly strong principle for the infinite case. We further show that where locations are the same in all worlds, but have no natural order, this principle turns out to be equivalent to a strengthening of a principle defended by Vallentyne and Kagan (1997), and also to a weakened version of the catching-up criterion developed by Atsumi (1965) and by von Weizsäcker (1965). Footnotes1 For valuable comments, we would like to thank Marc Fleurbaey, Bart Capéau, Joel Hamkins, Barbara Montero, Tim Mulgan, and two anonymous referees for this journal.

In this nicely written book, Dale Murray critically discusses the moral rights posited by Robert Nozick in Anarchy, State, and Utopia. His focus is on these rights and not on Nozick's arguments about the justness of the state. He argues that Nozick's rights to compensation give rise to rights to government-financed health care and that Nozick should recognize a natural right to enough goods to ensure a reasonable chance of living a decent and meaningful life (if feasible for all). Murray also discusses issues such as the role of invisible hand arguments, moralized conceptions of freedom, and the issue of whether just steps (transactions) preserve the justice of situations.

I argue that the liberty condition of Sen's important impossibility of a Paretian liberal result is not a condition that liberals (or libertarians) would accept. The problem is that an appropriate liberty condition must be formulated in terms of consent - not in terms of preference. To formulate an adequate condition the framework needs to expand from collective choice rules (which only take information about preferences as input) to rights-based social choice rules (which also take as input information about which options have been consented to and which would violate someone's rights). I formulate a more adequate liberty condition based on the notion of consent that is acceptable to liberals, and then show that Pareto optimality is incompatible even with that condition. I then show how the liberty condition can be weakened in a plausible manner, and describe an interesting class of theories - rights-based Paretian theories - that satisfy the Pareto optimality requirement while being sensitive to liberty considerations.