Philip Pettit

Imagine a very fi ne grid or graph on which dots are placed at various coordinates so that, as a consequence, this or that shape materializes there. Depending on the coordinates of the dots, different shapes will appear, and for every shape there will be a pattern in the coordinates that guarantees its appearance. Take, for example, the diagonal line that slopes rightward and upward at an angle of 45 degrees from the origin. This line is bound to make an appearance so long as the coordinates satisfy the condition or pattern that as they move away from the origin, (0,0), the coordinates are progressively larger pairs of equal numbers: (1,1), (3,3), and so on. In the world of such dots and shapes, it is going to be in principle possible, for any array of dots that realizes a relevant shape, to derive the presence of the shape from the numerical coordinates of the dots. More particularly, it is going to be possible to derive that shape without reliance on anything other than, fi rst, the empirical fact that the given array of coordinates instantiates this or that pattern; and second, the a priori knowable fact that the pattern guarantees the presence of the shape in question. The nature of the shapes on any grid—if indeed there are any relevant shapes present—is going to be a priori derivable from the positions of the dots; it is going to be possible in principle to derive the one from the other. The simplest and most appealing version of physicalism parallels this sort of doctrine about dots and shapes (Pettit 1994, 1995). It holds that just as the positions of the dots determine the nature of the shapes a priori, so the way the natural world is physically organized a priori determines the way it presents itself in psychological and other terms. The way things are physically confi gured entails the presence of psychological and other realities, and it does this without reliance on anything other than what a..

Amartya Sen has argued the impossibility of the Paretian liberal. While his abstract argument is compelling, the concrete significance of the conclusion is in some doubt. This is because it is not clear how important liberalism in his sense is; in particular it is not clear that the sort of liberalism required for the impossibility result is a compelling variety. We show that even if the argument cannot be used to establish the inconsistency of Paretianism and common-or-garden liberalism, it can be adapted to prove a parallel impossibility. This is the impossibility of combining the Pareto criterion with a loyalty constraint involving certain claim-rights rather than liberty-rights. The impossibility of the Paretian loyalist is of interest in itself but it is also interesting for the light it throws on the source of Sen-style impossibilities.

We have two ways of talking about liberty or freedom, one in the singular, the other in the plural. We concern ourselves in the singular mode with how far someone is free to do or not to do certain things, or with how far someone is a free person or not a free person. But, equally, we concern ourselves with the plural question as to how far the person enjoys the liberties that we take to be important or basic. What are those plural liberties, however? What does it take for something to count as a basic liberty? e usual approach to this question is to give a list of some presumptive basic liberties — say, those of thought, speech, and association — and then to add a gestural ‘and so on’. My aim in this paper is to do a little better in elaborating a conception of the sorts of liberties at which the ‘and so on’ gestures. I argue that the basic liberties can be usefully identifi ed as the liberties required for living the life of a free person or citizen and I spell out that requirement in three constraints, which I describe as feasible extension, personal signifi cance and equal co-enjoyment. ere are many candidate sets of basic liberties that might be proposed for protection, whether in general or for a particular society. e claim that I defend is that in order to count as a set of basic liberties, the types of choice protected under any proposal should be capable of being equally enjoyed at the same time by everyone (equal co-enj oyment), should be important in the life of normal human beings (personal signifi cance), and should not be unnecessarily restricted: they should be as extensive as the other constraints allow (feasible extension). e aim of the paper being quite limited, I abstract from many important issues. I do not provide an argument for why it is important that some set of basic liberties be protected, nor do I rate the importance of such protection against other social goals. I say nothing on how far it should be a requirement of democracy that certain sorts of liberties are entrenched—whether constitutionally or otherwise—and how far democratic process should be allowed to vary the specifi - cation and protection of basic liberties..