In 'The Problem of Global Justice' Nagel defends a political conception of (socioeconomic/redistributive) justice. A political conception of justice holds that justice is a political virtue: the existence of sovereign states '...is precisely what gives the value of justice its application, by putting the fellow citizens of a sovereign state into a relation that they do not have with the rest of humanity, an institutional standard of fairness and equality that fill out the content of justice' (p. 120). I define this type of view as a relational approach to justice, because it holds that justice considerations arise through relationships between persons, rather than prior to them.
Nagel contrasts the political conception with the cosmopolitan conception, under which justice is pre-political: '...the demands of justice derive from an equal concern or a duty of fairness that we owe in principle to all our fellow human beings, and the institutions to which standards of justice can be applied are instruments for the fulfillment of that duty' (p. 119). This conception of justice is non-relational, because justice considerations arise prior to the existence of relationships between persons. The political conception is typified by Rawls' view that justice is a political value.
Nagel agrees with the Hobbesian view that justice requires sovereignty (pp. 115-117), and so concludes that for both the cosmopolitan and political conceptions, global sovereignty is required for global justice (p. 122). My inclination is to dispute this conclusion, because I need further convincing that justice is impossible to achieve without sovereignty. I'll not go into this further here though, perhaps another time.
Under Nagel's political conception socioeconomic justice is 'fully associative' (p. 127). The relationship that we stand with to each other as fellow citizens is the basis of us having duties of justice towards each other. Why? For Nagel it is because the societal rules which govern our relationships with each other as fellow citizens are coercively imposed - the state is not a voluntary organisation (p. 128).
'Without being given a choice, we are assigned a role in the collective life of a particular society... thereby supporting the insitutions through which advantages and disadvantages are created and distributed. Insofar as those institutions admit arbitrary inequalities, we are, even though the responsibility has been simply handed to us, responsible for them, and we therefore have standing to ask why we should accept them' (p. 129).
Nagel argues that there are seperate humanitarian duties and basic rights which do not depend on this type of coercive community, but apply universally (p. 130-1). He maintains however that the obligations of socioeconomic justice remain restricted to within this type of community. He also rejects the claim that this type of coercive relationship now obtains at the international level, and so remains committed to socioeconomic justice being a matter for within states (p. 137).
This conclusion is clearly non-cosmopolitan. There are two different routes that a cosmopolitan can take to dispute it.
The first route involves accepting his relational conception of justice. There are then two seperate lines of argument that can be pursued. The first strategy for a relational cosmopolitan involves arguing that the coercive conditions under which considerations of justice arise do actually exist at a global level, and so socioeconomic obligations are relevant globally. The second strategy for a relational cosmopolitan involves arguing that Nagel's conception of coercive institutions giving rise to considerations of justice is false, and putting forward an alternative conception. This is a course that has been pursued by many of the commentators on Nagel whom I will be discussing at a later date.
The alternative route, which is the one I find most appealing, is to reject Nagel's relational approach to justice. If we pursure a non-relational, or 'cosmopolitan' approach, then as Nagel says, we will be committed to global redistributive justice. Nagel admits that this conception 'has considerable moral appeal' (p. 126) but does not spend any time attempting to defend he. He chooses the relational, or 'political' approach 'partly because I believe that the political conception is accepted by most people in the privileged nations of the world, so that, true or false, it will have a significant role in determining what happens' (p. 126). To me this seems like a reason not to go with this conception, not a reason to go with it - it's plausible that such a conception is accepted by people in the privileged nations because such a conception supports their continued privilege.