Justice and morality beyond naïve cosmopolitanism

Lea Ypi
Justice and morality beyond naïve cosmopolitanism

Justice and morality beyond naïve cosmopolitanism

Lea Ypi*

Nuffield College, Oxford, UK

A human being counts as a human being only in virtue of his humanity alone, not because he is a Jew, Catholic, Protestant, German, Italian etc. This is an assertion which thinking ratifies and to be conscious of it is of infinite importance. It is defective only when it is crystallized, e.g. as cosmopolitanism in opposition to the concrete life of the state. (G.W.F. Hegel, Philosophy of Right, 209)

Abstract

Many link their moral defence of specific principles of justice to a critique of the normative standing of states. This article explores some conceptual distinctions between morality and justice by focusing on the nature of claims they entail, the obligations they generate and the distribution of agency that they require. It then draws out some implications of these distinctions so as to illustrate how states play a non-arbitrary role in the process of both rendering determinate the principles of global justice and allocating agency compatibly with strict rather than large obligations. Contrary to many existent defences of the state, it shows how a similar conception promotes rather than undermining the ideal of global justice.

Keywords: cosmopolitanism; statism; equality; justice; morality

*Correspondence to: Lea Ypi, Nuffield College, New Road, Oxford OX1 1NF, UK. Email: lea.ypi@nuffield.ox.ac.uk

Published: 1 September 2010

Citation: Ethics & Global Politics, Vol. 3, No. 3, 2010, pp. 171–192. DOI: 10.3402/egp.v3i3.5487

©2010 L. Ypi. This is an Open Access article distributed under the terms of the Creative Commons Attribution-Noncommercial 3.0 Unported License (http://creativecommons.org/licenses/by-nc/3.0/), permitting all non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.


PRELIMINARY REMARKS

What kind of relationship does the idea of the state entertain with the principles of global justice? Cosmopolitans often take that relationship to be arbitrary from a moral point of view. They believe that every individual in the world is a valid source of equal moral claims; that such claims generate clearly identifiable principles of global justice, and that states play a very limited, if any, role in establishing the content of these principles and their mode of application.1 This article challenges a similar view. It articulates some normative distinctions between the spheres of morality and justice and it explores their relevance for understanding the fundamental (i.e. non-arbitrary) role of states in relation to the problem of global justice.

In some respects, this defence of the state is not new. Several authors have insisted on the normative relevance of states with reference to global justice debates. However, to the extent that they have done so, the intent has often appeared critical rather than constructive. States have been invoked to undermine rather than contribute to a proper understanding of cosmopolitan principles. To put it more precisely, many critics of cosmopolitanism have appealed to justice-based political relations to argue how obligations between fellow-citizens are different, and potentially more demanding, compared to those between citizens and strangers. They have also argued that these obligations lead to different kinds of principles, principles of egalitarian justice in the case of domestic societies and principles of assistance to those in dire need in the case of the global sphere.2

This article makes a different attempt. On the one hand, it emphasises the relevance of the state by exploring the distinction between moral and justice-based obligations. On the other hand, it criticises a number of commonplace arguments regarding the implications of this distinction for the principles of global justice. More specifically, it illustrates how the principles of assistance advocated by statist theories are no less demanding than egalitarian ones—if anything they seem more frustrating. It also tries to show how these principles might not necessarily be more efficient or more feasible than those reflecting the conception of global equality that cosmopolitans advocate. Finally, the article insists, by taking seriously the normative standing of states, and the civic agency expressed through state-mediated institutions, justice is able to expand its reach as an inherently political enterprise.

To sum up, associative political relations can be taken seriously without sacrificing cosmopolitan ambitions. Emphasising the distinction between morality and justice, and insisting on the specificity of relations involved in each case is extremely important. This argumentative strategy leads to a normative justification of the state which does not undermine the cause of global justice; on the contrary, it promotes it. Against cosmopolitans, it serves to show that states are non-arbitrary. Against non-cosmopolitans, it serves to show how the non-arbitrary role of states does not necessarily limit the scope of global justice.

UNDERSTANDING THE CLAIMS

Cosmopolitans often tend to treat moral obligations as equivalent to obligations of justice. For them each individual's moral claims are sufficient to generate specific principles of justice that apply both within and across state boundaries.3 However, there are fundamental differences between the spheres of morality and justice. Grasping these differences is of the foremost importance in understanding the kinds of obligation at stake between citizens of different states and for articulating the scope of these obligations. Let me start by presenting some key assumptions to which political theorists frequently refer in analysing the distinctive tasks we face when we make claims about morality and claims about justice. And let me then further explain how these distinctions serve a more appropriate understanding of the claims of cosmopolitan justice.

Morality is normally concerned with a non-comparative assessment of individual behaviour. We usually rely on moral principles to qualify actions as ‘good’ or ‘bad’. On the other hand, references to justice are usually conveyed as ‘right’ and ‘wrong’ or ‘fair’ and ‘unfair’ and often apply to particular relations between individuals, namely relations that involve a comparative dimension. In these cases, what we tend to call ‘just’ or ‘unjust’ is a share. An important part of understanding claims of justice then is concerned with understanding proportional equality or ‘equality of ratios’.4 Justice invokes comparative concepts, morality is one-dimensional.

To say that justice applies to comparative relations between agents is not to say that it is entirely separate from morality. In fact, the two concepts are importantly intertwined with each other. To understand their relation we need to consider instances in which agents’ moral demands over specific goods may collide. It is precisely when we are confronted with agents making comparative demands with potentially conflicting implications that the need for justice arises. When we talk about justice, we implicitly refer to the adjudication of conflicting requirements over particular shares in a way that can be considered authoritative and binding for all affected parties. In doing so, justice is served by the idea of a collective political authority designed to guarantee that the claims of everyone are equally taken into consideration. The pages that follow will articulate this notion in greater detail. But before proceeding with the argument, let us give an example to illustrate the plausibility of some of the claims made concerning the distinction between morality and justice.

Most theorists accept that everyone is, as a matter of moral principle, entitled to certain standards of material well-being. But consider this general moral claim expressed in a particular context, say that of former socialist countries in Eastern Europe. Following the collapse of socialism in Eastern Europe, a similar moral principle was endorsed by many former land and property owners to require the restitution of at least part of the goods (especially houses) of which they had been deprived by the socialist governments that came to power after the Second World War. Hence, during the 1990s, the governments in charge of liberal reforms in such countries had to consider seriously the claims of this part of the population. This not merely because respect for property rights became a fundamental principle of the newly established democracies but more importantly because such families, after having suffered from class-discrimination and years of imprisonment or isolation, found themselves in particularly impoverished material conditions and became vulnerable to political instability and market fluctuations. Recognising their entitlement to certain standards of well-being implied trying to address their claims to enjoy at least some of the benefits previously earned by their families. However, in previous years most of the properties belonging to these former property owners had been allocated by socialist governments to other families. These families also had no other sources of housing available and could not be easily removed from the places where they used to live. So the same claim to well-being led to a conflict on the allocation of certain external objects and prompted the need to think about a principled way of resolving this distributive conflict.

Of course if the governments in question had disposed of enough flats and lands to satisfy everyone, no need to compare the claims of these agents to one another would arise. But as already emphasised, the question here was no longer satisfying separate moral claims but involved thinking about distributive justice. In all cases, the outcome of the process was neither the result of the proved moral superiority of one discourse over the other (both kinds of claims sounded sufficiently plausible to start with) nor the result of the reflexive identification of a superior authoritative norm. It emerged from particular political processes, social conflicts, legal negotiations, as well as from the specific historical tradition in accordance with which the required principles of justice were identified.

Of course all this might not be sufficient if we need to compare the position of citizens across different jurisdictional boundaries. The challenge in this latter case is to identify a principled way of reflecting on similar relationships able to capture both the relevance of domestic political authority in mediating the demands of fellow-citizens and to modify the scope and function of that authority when conflicting claims between agents are not clearly confined to the territory of a particular state. Notice, however, that even if the kinds of relations identified in the example above are all domestic political relations, the case remains useful to remind us of the difficulty of mediating between conflicting claims by appealing to non-comparative moral principles. Invoking such principles was in the example we examined no more than a starting point. It was precisely when the requirements of morality collided with each other, that the need for comparative principles of justice seemed to arise. Justice provides the kinds of collective normative constraints according to which different shares between agents are normally allocated. Its principles constitute a result of political processes where conflicting entitlements are balanced against each other and eventually mediated.

Teasing out some of the implications of the distinction between morality and justice suggests that the process of articulating non-comparative moral constraints is at least in part a normatively distinct enterprise from that of addressing comparative claims of justice. It also suggests that the path from uni-dimensional moral principles to the justice-based allocation of relative shares is necessarily linked to an understanding of particular relations mediated by a universal authority able to administer justice impartially. To say it with Aristotle, ‘justice is the political good’.5 The meaning of this good is negotiated through political processes and democratic procedures in virtue of which particular interpretations of that good are confronted, modified and impartially negotiated.

But what role exactly does the institution of the state play in this process? Why can't we articulate claims of justice by taking into account relations between individuals considered simply as individuals? One could object here that there are many instances of inter-personal comparative conflict in which it is possible to address the different demands of agents without invoking the idea of a universal political authority. The family provides a relevant case in point; relations between friends might be another example. In both cases it seems intuitively plausible to think about ways of accommodating conflicting requirements without recurring to an idea of political authority similar to the one I have sketched out. If, say, my brother and I disagree over how much of the Christmas pudding each of us should have, surely, an objector might argue, we could solve that conflict without invoking such a grand idea as the authority of the state.

But let us grant that it is occasionally possible to solve certain disagreements between individuals (e.g. members of a family or a circle of friends) without recurring to the idea of justice as administered by a state-based political authority. The fact that alternative principles of distribution (e.g. principles based on generosity, family loyalty or personal attachment to particular individuals) might occasionally orient our judgements in special associative circumstances, does not wholly remove the need for comparative principles of justice administered in a certain way. To see this point it is enough to reflect on cases involving the distribution of non-trivial goods (e.g. the share of a common family property) in which none of these alternative principles seems to get us very far in solving potential conflicts between relatives. Indeed in a significant range of conflicting circumstances, a final say concerning the distribution of shares between family members is obtained by invoking the authority of allegedly impartial institutions. It is not at all uncommon, even for family members, to bring their distributive disagreements in front of courts of justice, for example. Instances such as these suggest that even if in some cases we might be able to refer to non-comparative principles in addressing the requirements of particular individuals (e.g. principles based on charity or love), considerations of justice as administered by an impartial political authority remain fundamentally important.

All this, however, does not yet explain why exactly we should take seriously state-mediated political relations in order to address conflicting claims between citizens of different states. This is the cosmopolitan challenge. In order to address that challenge we need to focus on the distinction between the kinds of obligations that respond to claims of morality and those that respond to claims of justice. Understanding this difference introduces us to the non-arbitrary role of the state. Neglecting the difference, as cosmopolitans often seem to do, runs the risk of undermining the distribution of political agency and weakening the conditions upon which cosmopolitan justice can be effectively grounded.

UNDERSTANDING OBLIGATIONS

So far, we have underlined how morality and justice articulate distinctive kinds of claims and how justice is required to mediate conflicting comparative demands. In this section I would like to explain how the kinds of obligations responding to each type of claim are also strikingly different. To see this point, let us start by considering in how far the spheres of morality and justice resemble each other, and where they part ways.

Both the concept of morality and that of justice are linked to rule-guided processes and refer to a set of norms that place constraints on peoples’ actions. Yet if claims over the distribution of relative shares collide, finding a way out of these conflicts is clearly irreducible to favouring one perspective over the other. We cannot both accept the plausibility of two or more distinctive but equally valid points of view and then arbitrarily privilege one over the other. If we did so, in at least one of the cases under consideration, the moral agency initially expressed in articulating normative constraints would end up being undermined. One particular will would prevail, thus imposing on all other agents’ decisions that would fail to reflect their reasons for action. The favourite part would end up forcing the rest to endorse an individual point of view for which no plausible justification could be found. What is required to exit this impasse is a collective institution able to represent both moral sources of valid claims yet make decisions in a non-unilateral and impartial manner. It is this arbiter that will ultimately be the source of necessary principles for adjudicating conflicting moral claims, in short, of principles of justice. But, how can this more impartial source of valid norms be constructed?

Kant's Metaphysics of Morals provides a good introduction to this topic by clarifying the distinctive requirements of morality and justice, and by distinguishing between ‘duties of justice’ and ‘duties of virtue’.6 Both the concepts involve a notion of constraint and a right ‘to act’. However, a duty of virtue, i.e. a moral duty, is of indeterminate obligation, whereas a duty of justice is of strict one. According to Kant, conforming to a duty of virtue requires benevolence and acting against it is morally reprehensible. However, it is not unjust. This is because the final arbiter on the appropriate course of action remains the individual and the normatively relevant relation is in this case the one entertained with oneself. Other individuals do not appear to be affected in a unilaterally inappropriate way. However, things look different once we enter into a relational dimension in which conflicting comparative claims arise. Here an agent is already wronged by the possibility of having to endorse a unilateral point of view that was not initially his own. The need for collective arrangements arises because claims over distributive shares can conflict with each other and individuals as such might not be able to converge on the satisfaction of each other's demands. These collective arrangements ought therefore to be able to unify choices in a way that directs the exercise of moral agency in the external world in a non-arbitrary and non-partial manner. This is why Kant refers to this source of authority as one construed compatibly with the sum of conditions according to which ‘the freedom of each can coexist with the freedom of everyone in accordance with universal laws’.7 The requirement of universality ensures that no agent is privileged and no particular will can dominate the others.

Now, this source of universal constraint can only be grounded upon a civil–political union. As Kant (and Hobbes as well as Rousseau before him) emphasised, no unilateral will (one could also say no single moral perspective) could serve as a coercive authority, imposing certain normative constraints on others. That would infringe upon everyone else's freedom to act similarly. The flaws attached to the prevailing of an individual perspective and the need to avoid domination render necessary the construction of a universal authority administering justice in an impartial manner. As Kant puts it, ‘it is only a will putting everyone under obligation, hence only a collective general (common) and powerful will, that can provide everyone this assurance’.8 Only in the presence of the state, can principles of interaction that are non-unilateral and non-partial arise.

It is important to notice that the authority of the state is not needed merely to implement already agreed-upon general principles. The point is not simply that the state is needed to enforce justice. The point is that without the existence of a collective authority able to adjudicate conflicting comparative claims, no principles acceptable for everyone could even be formulated in a non-arbitrary way.9 This is because an agent's particular interpretation of what their share might be does not naturally coincide with that of every other. The concern does not necessarily arise because that agent might be in bad faith, though that of course will occasionally also turn out to be a problem. The question might simply be that each particular agent has no access to perfect information over the status of others, their number, the way they came to make claims over specific goods, how they relate to each other, and so on. Without a collective mechanism to coordinate requirements and determine what exactly is due to everyone, we have no way of establishing what normative claims each of us should authoritatively endorse. We know what we think, but we have no reason to assume that what we think coincides with the views of others. Indeed if that certainty had been available, many disagreements would have been avoided from the start.

The idea of a general, powerful and collective will embodied in the authority of the state is articulated to resolve the impasse in which the existence of conflict originating from the clash between various particular claims leads. As a result of the establishment of that institution, each individual is asked to merge part of his moral self with that of all others. But each individual is also required to submit to the rules arising from this general will—a will that is constitutionally unable to express the moral agency of anyone in particular. The rules enforced by the general will are therefore considered to be universally accessible, collectively effected and binding for everyone.

This is also why only a similar collective authority can effectively allocate duties of justice. Duties of justice contain a form of obligation which is not simply reprehensible from a moral point of view. The contrary of a duty of justice is not mere vice, as is the case of moral duties; it leads to transgression. Duties of justice bring with them a title to force others into accepting one, if not necessarily their own, emergent point of view. However, this point of view can only avoid being imposed in an arbitrary manner if it results from something that has initially been set up through the contribution of everyone.10

It should by now be clear why only duties of justice imply a corresponding possibility to be compelled by an external, inter-subjective, authority.11 The infringement of personal moral agency could only be justified by subsuming part of one's individual moral self under a collective political body representing people as members of a civic whole. The modern principle of popular sovereignty reflects precisely this concept; only political institutions representative of citizens as part of the collective, general will, could impose strict or narrow obligations to comply. Only if norms originate from a will suitably transformed to take into account the claims of everyone, can they be considered to have been formed autonomously.

This argument partly serves to clarify the distinctive nature of interactions when we consider the claims of human beings as claims of morality or claims of justice. The notion of a common humanity, sharing a general conception of welfare or relying on certain universal attributes of the person might well be sufficient for agents to come up with a number of plausible moral commitments. However, that people stand in certain moral relations towards each other (as they always do) is not enough to do justice to their distributive claims. In the absence of collectively representative political institutions, such claims would always present themselves as unilateral and fail to place specific responsibilities on relevant agents.

Justice then relies on the exercise of political authority as its general enabling condition.12 It is only through the collective recognition of one another as members of a political society and the public acknowledgement of authoritative comparative principles that claims of justice can be effectively formulated and legitimately applied. But what does this imply for the naïve cosmopolitan defence of global justice and the related emphasis of the arbitrariness of the state?

A first consequence of the distinction between morality and justice with regard to the claims of individual human beings is the following. Even if some fundamental cosmopolitan claims are presented as weighty enough from a moral perspective, this is not sufficient to qualify such claims as claims of justice. Taking seriously the authority of the state seems very relevant for establishing in a non-arbitrary way the comparative implications of these general moral claims. It is also crucial to determine which agents should be involved in this distribution and how responsibilities compatible with it ought to be allocated. So even if we insist, as many cosmopolitans do, that each individual in the world is entitled to an equal share of natural resources, or to equal standards of well-being, or that equality is a universal value, for these claims to have effective purchase on agents, for them to result in strict rather than loose obligations to comply, the comparative principles they give rise to ought to be formulated and applied through a collective political authority. Otherwise, however well-intended, these cosmopolitan claims will always result unilateral and consequently fail to be binding.

However, the fact that obligations of justice are mediated by the authority of the state and therefore differ from moral obligations does not necessarily imply that duties between fellow-citizens necessarily take precedence over duties towards foreigners. In establishing which duties take precedence we might well have to take into account the urgency of the claims on which they are based, and not merely who makes these claims.13 But what this does imply is that our relationship with foreigners is always institutionally mediated. We do not engage with the ‘poor’ of the world as separate individuals, we engage with citizens of specific states, as citizens of other states. Even our global participation in transnational networks or civil society structures, however much celebrated, ultimately relies on political resources, political structures and an institutional distribution of responsibilities that is state-generated, state-based or state-oriented.14 It is undeniable that the relationship with our fellow-citizens is in some way special: through joint authoritative institutions we are able to have them endorse even principles stemming from normative commitments that some fellow-citizens might initially not share. This is something we could hardly obtain from foreigners. But the point simply proves that it is misleading to assume, as many sceptics of cosmopolitanism do, that the special nature of the political relationship always works at the advantage of fellow-citizens. To whose advantage it works, depends on the development of internal political processes. It is impossible to rule out that in some cases fellow-citizens might end up being burdened. What matters for our purposes is that they will be burdened in a legitimate way. This is why the transition from morality to justice through the idea of the state is of foremost importance.

Hence, insisting on the arbitrariness of political boundaries does very little to serve the cause of global justice. If anything, it undermines the assignment of institutional responsibilities; it leads to a mere unilateral endorsement of certain claims without examining the political constraints that these claims face. But if we take seriously the state, and what is made possible by the political relationship reflected in its institutions, principles of justice (even those going in a cosmopolitan direction) can be effectively formulated and successfully established.

Although, as discussed earlier, the concepts of morality and justice have often not been adequately distinguished, significant differences emerge if we consider the distinctive nature of their claims and the obligations to which they give rise. These differences have frequently been emphasised by critics of cosmopolitanism to argue against the idea of global equality and in favour of global principles of assistance to those in dire need. In the next section, I try to show that it is perfectly possible to preserve the distinction between morality and justice without conceding to the sceptic that a similar distinction ends up undermining the scope of global egalitarian justice.

AFTER STATISM, BEYOND NAÏVE COSMOPOLITANISM

Those who tend to insist on the distinctive nature of claims of justice, as opposed to morality, typically do so in order to undermine the global scope of egalitarian justice. Because of the distinctive demands justice makes, statists argue, outside the state egalitarian concerns should not arise. We need to worry instead about building a fair structure of cooperation at home and about absolute deprivation and severe violations of basic human rights abroad. Before examining more in detail the reasons for such principled opposition to global egalitarianism and see if they may be circumscribed, it is important to be clear on what this critique is and is not about. The statist scepticism towards egalitarianism and the consequent favouring only of principles of assistance is due to specific methodological concerns. Similar concerns target the particular form that existent defences of global egalitarianism have taken: individualist cosmopolitanism. But global egalitarianism need not rely on individualist cosmopolitanism in order to be defended. There may be an alternative. Before exploring this alternative, it may be worth examining in some detail a number of (weak) arguments that have often been provided against the statist line.

Firstly, it seems rather simplistic to dismiss the statist agenda as a mere attempt to defend the status quo.15 Statists envisage a world of internally egalitarian societies who also recognise obligations of assistance to those burdened with absolute deprivation outside state boundaries. It does not take much to see that a world in which internally just states collectively recognise a universal responsibility to supply goods to people in need is a far better world than the one we presently have; indeed it is a world in which Karl Marx and John Lennon would have been content in equal measure. To see this point, imagine ‘international solidarity’ rather than ‘national security’ being one of the keywords of a typical US presidential campaign, then imagine a heated discussion between two future presidents on who offers to spend more in foreign aid. And then compare this to the last electoral debate you heard. Whatever one may say, the measures that all statist would favour to combat global injustice are significantly more advanced than anything we currently have. Political realism does not seem an appropriate explanation for the statist unwillingness to replace a global principle of assistance with a global egalitarian one.

Secondly, principles of assistance are not favoured because embracing egalitarianism would be much more demanding in economic terms.16 It is true that a concern for relative deprivation implies an ongoing (re)distribution of primary goods, whereas a concern for absolute deprivation only applies for as long as poor people fall below a certain threshold. Indeed obligations of assistance, of the kind that the statist logic envisages, are valid until poor people are able to build and develop their own political institutions; such principles have, in the words of Rawls, ‘a target or cut-off point’.17 However, statists are usually also committed to principles of political autonomy for states.18 This means that it is not those who discharge obligations of assistance to poor societies that have discretion to set the target or determine the cut-off point. We are bound by obligations to give ‘enough’ to people in need but hardly ever in a position to participate in the political processes that would accelerate events in the direction that we hope. This means that, in practice, obligations of assistance will be as demanding as (and perhaps more frustrating than) egalitarian ones. Thomas Nagel, for example, recognises that, in extreme circumstances, ‘even a nation's immunity from the need to justify to outsiders the limits on access to its territory is not absolute’. Indeed, he continues, in those cases ‘denial of the right of immigration may constitute a failure to respect human rights or the universal duty of rescue’.19 But it is not clear that a politics of flexible borders, of the kind that discharging obligations of assistance in this case would require, is less demanding or might prove more acceptable to domestic publics than participation in egalitarian schemes for the distribution of goods across countries.

Thirdly, assistance is not preferred to egalitarianism by virtue of its efficiency. A principle of assistance requires you to help people in extreme need every time the need arises. An egalitarian principle instead requires you to engage in certain kinds of distribution so as to make sure that individuals are always in a position to fulfil their needs, should they arise. But consider the following example. In your neighbourhood there are two kinds of houses: the red and the blue houses. They are equally comfortable, except the red houses have a sophisticated fire-alarm system and the blue houses don't. When a party called Egalitaria rules the neighbourhood it tries to equalise housing standards so that the Blues also have access to a sophisticated fire-alarm system and it does so by asking the Reds to contribute monthly to a collective fire-alarm fund. When a party called Assistenza rules the neighbourhood it simply asks the Reds to help the Blues if their houses catch on fire. But it is hard to see how the policy of assistance is a more efficient one than the egalitarian one. Surely if the blue houses went on fire the Reds would much rather have contributed to the fire-alarm system than have to host the Blues in their own homes while waiting for the reconstruction work to take place. Again, preventing the houses from catching on fire in the first place is a more efficient measure than having to deal with the impact of its consequences. Egalitarian principles in this case seem to serve efficiency much better than the principles of assistance in cases of extreme need.

Finally, principles of assistance are not to be preferred to egalitarian ones by virtue of their feasibility. There are no reasons to think that policies aiming to assist people with the basic resources needed to relieve absolute poverty are more feasible than global egalitarian policies. If we take the idea of international assistance seriously, there has to be some coordination among the wealthy contributors on how one goes about discharging their responsibilities compatibly with such principles. Setting up those coordination mechanisms will inevitably require concerted institutional reforms trying to achieve the required international transfers of resources. This is by no means favoured by the status quo where each country is left free to decide what part of its revenues it is going to devote to overseas assistance, regardless of the initiative of other states and accountable only to the pressure of its own citizens. Such a scenario is hardly different from a domestic situation in which individuals have complete discretion to decide how much of their wealth ought to be given to the domestic poor. Taking seriously principles of assistance requires systematic, informed and enforceable institutional measures aiming to distribute responsibilities for poverty relief across a range of states. We therefore need to think about the international implementation of collective schemes of foreign aid, about how we can render contribution proportional to the population size or wealth of each country, about how to differentiate the role of countries according to their geographical position and history of relationships with developing states, about the international procedures needed to persuade all potential candidates to take up responsibilities for participating in similar schemes, and so on. There are no reasons to suppose that similar schemes are much more feasible or easy to achieve than other schemes of egalitarian distribution.

But if considerations of realism, moderation, efficiency or feasibility hardly provide strong arguments for explaining the statist preference for principles of assistance, what motivates the cautious nature of this approach to the issue of global justice? Overcoming naïve cosmopolitanism requires attempting an answer to this question. In order to articulate that answer, we need to return to the distinction between morality and justice, and consider if some objections against existent forms of cosmopolitanism can be appropriately circumscribed even if what emerges from that distinction is duly taken into account.

THE QUESTION OF JUSTIFICATION

As we already emphasised, cosmopolitans have been frequently reproached for failing to distinguish appropriately between the normative requirements of morality and those of justice. The arguments advanced are linked to a methodological stance often also referred to as ‘justificatory monism’.20 Justificatory monism is the kind of normative strategy deployed by all those who treat the spheres of morality and justice as equivalent, with regard to the claims being made, to the obligations they generate and to the agents from which such claims arise.

Statists, on the other hand, take a different view on the matter. The principles designed to reform particular institutions, they argue, are principles of justice. They reflect specific associative political relations and therefore have a different nature from moral principles applying to individual conduct. This means that while confronting the problem of global justice, we need to think about the distinctive kind of associative relationship involved in each of the cases under consideration. The political nature of relations within the state gives rise to a specific set of principles, which is hardly applicable without mediation at the international level. If on the other hand, we focus on the international realm where this form of associative political relationship is absent, we have to articulate alternative principles, appropriate to agents who interact with each other on a voluntary basis. The principles regulating interactions between citizens of different political communities, statist argue, cannot be obtained by simply extending to the international domain the kind of reflection we apply to the analysis of individual moral relations. We need to take into account the distinctive features of institutional cooperation characterising each particular associative context.

However, it is important to notice that the rejection of justificatory monism upon which this particular argument rests does not necessarily entail the rejection of egalitarianism. It also does not explain why we shouldn't be concerned with relative deprivation on a global scale. The fact that different types of relational practices require focusing on alternative subjects (either individuals as moral agents or individuals as citizens) does not immediately explain why this differentiation of subjects must also lead to a differentiation of principles. The best we can obtain by following the statist logic here is that any principle of justice at the international level will have to be conceived as an unavoidably political principle, tested first in circumstances where the political relation is stronger (e.g. between citizens of particular states) and from there brought to bear in the international sphere where individuals interact with each other as citizens of particular states. Some forms of cosmopolitanism might be uncomfortable with this claim. But a version of the theory attuned to the distinction between justice and morality need not. A modified version of cosmopolitanism might grant that the principles of justice at the international level will require reforms compatible with existent political institutions and particular political relations. This means that while requirements to equalise the position of individuals worldwide might be harder to justify, the same could not be said once these individuals relate to each other as citizens of different states, with reciprocal demands on each other mediated by their states.21

Statists, however, reject even this modified version of global egalitarianism. They do not simply claim that there are no plausible grounds for asserting distributive equality between individuals worldwide; they emphasise that there are no plausible grounds for asserting distributive equality between anyone. Egalitarian justice and the distribution of relative shares simply ought not concern us outside particular state boundaries. To see this point, consider the two standard arguments that statists offer in an attempt to defend the domestic limitation of global egalitarianism. The first emphasises how, given the distinction between morality and justice, agents within the state cooperate politically with one another in the public provision of collective goods (cooperation/reciprocity argument).22 The second emphasises how given the distinction between morality and justice, agents within the state subject themselves to the coercive power of its political institutions (the coercion/autonomy argument).23

Both versions of the argument emphasise how justice-based relations of agents within the state give us specific normative reasons for rejecting egalitarianism as a global conception of justice. For purposes of this article it is not essential to distinguish between reciprocity and coercion-based versions of the argument. Both seem to succeed and fail on similar points. Where they have some plausibility is the critique of justificatory monism and the emphasis on the distinction between moral relations and relations of justice. As already pointed out, both versions of the argument emphasise that since the terms of interaction between individuals within the state are different from broader moral relations, domestic institutions face distinctive demands of justification.

But notice that to say that state-based institutions face distinctive demands of justification is not the same as saying that such demands are higher compared to those faced when individuals only share moral relations with each other. As I already pointed out, egalitarian principles are not necessarily more demanding than principles of assistance. Cooperation and coercion constitute relevant associative features to take into account, but not because global egalitarianism would be harder to justify to citizens of wealthy states than global principles of assistance. The specific sense in which ‘distinctive demands of justification’ might pose a problem is with regard to our attempt to conceive of relations between citizens of different states as potentially conducive to justice-based interactions. In other words, the real target of the critique is our argument that global relations of justice should be politically constructed, even if they are not already there. The risk is that emphasis on the distinctive associative relationship of agents within the state, might also commit us to saying that such justice-based relations are hard or even impossible to extend when we take into account relations between states.

Consider the following objection that a statist might advance. Only where agents take part in a political process through which rules determining everyone's position in society are established, statists might argue, do we have reasons for holding any one of them responsible for the relative inequalities produced by the system as a whole.24 So if citizens of different states are not responsible for each others’ relative inequalities, why insist that they should enter into political relations of justice? An example might clarify what is at stake here. If US citizens interact with one another in establishing political institutions upon which the position of each American in society depends, then we have no reason to ask the Cubans to bear the cost of remedying for any relative inequality that the American public system of rules may produce in the long run. This issue is, from the point of view of statist, independent from the fact that Cubans ought to take measures to supply food to Americans if they were on the verge of starvation (as principles of assistance require). Indeed, statists would object, our attempt to construct appropriate political institutions able to satisfy egalitarian principles does not take into account the distinction between what the claimant may reasonably claim (to be assisted in the relief of absolute deprivation) and what the provider may reasonably refuse to provide (entering into a political relation which would require an ongoing distribution of certain goods). So, the statist might continue, if we were to accept the revised cosmopolitan proposal suggested in this article, we would not be able to distinguish between: (a) cases of absolute and relative deprivation for which remedial responsibilities fall on every individual in the world and (b) cases of relative but not absolute deprivation for which only those who cooperate within the same political system may be held accountable for any arbitrary inequalities the system produces. Even though from the point of view of the recipients it might be desirable to construct political institutions realising global justice, from the point of view of those contributing to relieve poverty the same may not be the case. This, of course, unless it could be shown that the claimant and the addressee of the claim stand in some particular justice-based relation (of ‘responsibility’, e.g.) triggering distributive obligations.

Individualist cosmopolitans respond to this argument by emphasising how the existence of specific associative relations in which individuals simply happen to find themselves upon birth is arbitrary from a moral point of view. The statist argument, they claim, starts with already existing institutions and asks how they could be justified to all affected parties. But why not start with a global world, and from there consider which specific schemes of cooperation are most suited to support particular claims of individuals?

This answer, sceptics point out, does very little to show that the statist perspective of starting with existing institutions is wrong. It merely opens up a different question. Of course the fact of being born in a particular family rather than another is also arbitrary, yet the special obligations following from this relation do not necessarily extend to everyone else in the world. For example, I have a special obligation to care about my drug-addicted brother and the fact that he happens to be my brother is as arbitrary as the fact that I happen to have green eyes. But the arbitrariness of that state of affairs does not imply that I also have a special obligation to care in an equal way about all the drug-addicts in the world. To take a different example, the fact that I happen to pay a monthly rent to one particular landlord rather than another is undeniably arbitrary—I could have rented the flat above and owed that amount of money to someone else. But recognising this arbitrariness does not commit me to pay rent to whoever claims my monthly contribution on these ‘arbitrariness’ grounds. Once we recognise the normative commitments that are implied in certain special relations (even if those relations appear arbitrary at the start) we see that the same commitments are not easily extended. A similar thought, for some statist, applies to global distributive justice. As Thomas Nagel has emphasised, even though obligations of justice ‘arise as a result of a special relation, there is no obligation to enter into that relation with those to whom we do not yet have it, thereby acquiring those obligations toward them’.25 The cosmopolitan emphasis on the arbitrariness of special relations fails to provide additional ground to convert the ‘arbitrariness’ premise into a ‘positive obligation’ applying independently from already existing relations of justice.

There may well be ways of responding to this objection, of developing the cosmopolitan argument from arbitrariness so that special obligations of distributive justice are extended to every individual in the world. But I shall not pursue this strategy here. As an alternative, in the remainder of this article, I shall sketch a defence of global egalitarianism, which grants statists the premise that egalitarian obligations apply in the presence of particular associative relations but denies their conclusion that egalitarian justice can only have domestic scope. This defence follows from the distinction between morality and justice I introduced at the beginning. It suggests that once the agents affect each other in certain normatively relevant ways, they already find themselves in special associative circumstances. Developing these circumstances to allow conflicts between distributive claims to be resolved, necessitates recognising the normative authority of states as the sole source for constraining citizens of different jurisdictions to enter into politically relevant relations.

THE QUESTION OF AGENCY

Unlike many familiar cosmopolitan views, the revised account I have presented, need not start with an abstract question on how the world's political structure would look like if its institutions were designed from scratch. It may not ask whether there ought to be states at all, it may endorse the statist exhortation to start with the institutions that we have. As we saw in the beginning of this article, the idea of a political authority is indispensible in considering how individuals ought to resolve conflicting normative requirements. The current concept of popular sovereignty is what allows us to articulate an authoritative solution to this problem. But, we need to ask, what if conflict is no longer contained within the boundaries of the state? In this case states need not, indeed cannot, limit the expansion of political relations. Taking the need for justice seriously requires placing political constraints on agents so as to allow them to act in concert in the normatively relevant ways described. Where that imperative is not taken seriously, all our political decisions, both those affecting individuals within specific states and those in different states, display the same degree of unjustifiable unilaterality that we initially identified within the state.

But why should we assume that individuals understood as citizens of different states affect each other in their exercise of moral agency and make conflicting comparative claims? To understand this point, consider how the version of cosmopolitanism I have introduced in this article modifies the account of cooperation and coercion with which the sceptic starts. The sceptic argument proceeds as follows. Firstly, it claims that distributive equality is a demand of justice when individuals are in a special associative political relation determining the opportunities and positions in society of each. Secondly, it emphasises that no such special associative relation between individuals exist at the global level. From there it is usually concluded that distributive equality is not a demand of justice.

But it is important to notice that the second premise of this argument restricts the understanding of special relations to imply special relations between individuals. Since cooperation or coercion of individuals is always state-mediated, remedial responsibilities for relative inequalities generated by a shared political system only apply to fellow-citizens. This is enough to defeat certain simplistic forms of cosmopolitanism. However, it is possible to modify the cosmopolitan argument in such a way that principles of distributive equality retain validity in the global sphere. Following that revision, a cosmopolitan might agree that distributive equality is a demand of justice where agents are in a special associative relation determining the opportunities and positions in society of each. However, states are the kinds of agents that we may find in a similar special associative relation in the global sphere. The way some states conduct domestic political affairs, how they pursue policies of say, migration, resource use or military management has an immensely important impact on the decisions of all others.26 But even more importantly, the nature of interactions between states in the international arena also modifies the comparative position of citizens in each. Think, for example, about unequal access to legal expertise in international negotiations, and how the outcome of these negotiations turns out to affect the policies of particular states in a number of key areas (e.g. trade).27 Or think about the global impact of the intellectual property regime and its influence on the price and manufacturing of, for example, generic medicines, leaving those with less resources available unable to afford essential health treatment.28 With regard to a number of increasingly important areas, the positional nature of global goods implies that citizens of different states find themselves making comparative claims with a transnational dimension.

Given these circumstances, the need for authoritative mechanisms to determine and adjudicate conflicting comparative claims seems imperative. Entering into political relations able to overcome the unilaterality of self-centred points of view, appears as important as overcoming the unilaterality of individual interactions in the domestic case we initially examined. It allows us to move beyond a purely moral understanding of the relations between individuals to one in which constructing an impartial authority adjudicating conflicting claims of justice becomes a political enterprise.

There may be objections to this proposal. One is to say that states (and consequently their citizens) do not find themselves in particular associative relations, and therefore to deny that they can affect each other in such a degree that the creation of a political authority administering justice universally becomes a political imperative. The second objection is to contend that even though states may develop special associative relations these are not relations of a relevant kind to require the distribution of relative shares. Let me consider them in turn.

The first version of the critique is also the easiest to deal with. States do not exercise their capacity, including their coercive capacity, in isolation from each other. Kant observed that fact as early as in 1796, noting that ‘a violation of justice in one place of the world’ was now felt ‘everywhere’. An increasing body of by-now-familiar literature on liberalisation, path-dependence and global governance shows that the growth of a global market economy, the increased volume of capital flows across boundaries, the development of rapid international exchange and investment, the delocalisation of production and the evolution of multinational corporations have created conditions encouraging the institutionalisation of cooperation between states. The consolidation of a number of regional and international economic and political agreements (NAFTA, GATT, the founding treaties of the European Union—EU), international financial regimes (including the International Monetary Fund, the World Bank and other mechanisms regulating currency exchange), a global system of intellectual property rights as well as other international and regional legal agreements for the exploitation of natural resources are all well-studied phenomena. Moreover, states increasingly rely on each other for the production of collective goods such as security or environmental protection and even participate in networks facilitating the exercise of their coercive capacity. So, for example, the international police organisation (Interpol) or the EU criminal enforcement network (known as Trevi), both designed to ensure the sharing of intelligence information or the coordination of police activity with regard to local and international pressure (such as terrorist threats) provide only two of the most relevant examples.29

That states participate in institutional schemes from which the exercise of their role as states, as well as the position of each in the global society of states, depends is a statement that very few people in a globalised world would find contentious. Yet it may be a mistake to interpret the political relations objection to global egalitarian justice as one that attempts to reject the claim that states are mutually dependent or cooperate in one another in the so-called global society of states. Perhaps the force of the criticism emerges more clearly if we emphasise that the issue is not so much whether states form relevant associative relations with each other but whether their willingness to develop such special associative relations is of a type that goes beyond self-interest. But what does this really imply for the rejection of global egalitarianism?

In an attempt to disentangle the domestic–international analogy with regard to global distributive justice, Brian Barry has pointed out that while individuals within the state cooperate with one another in the production of goods that collectively enjoyed similar conditions of reciprocity fail to obtain in the international sphere. More specifically, even though states might be engaged in trade agreements and joint institutional schemes for collective security or environmental protection, such associative relations lack reciprocity and may not be of a kind that triggers distributive obligations.30 The distinction that comes to mind to further clarify this point is one that Rawls has traced between ‘social union’ and ‘private society’. A private society is a sphere in which individual interactions are guided by private ends, either competing or independent, but not in any case complementary.31 A social union is instead a set of interactions where agents pursue the shared political end of creating and preserving political institutions responsible for the just distribution of relative shares.32 Special relations between states, a critic might observe, are perhaps more similar to the former than to the latter. Indeed the international sphere resembles Hegel's or Smith's civil society where agents simply develop interdependent needs and interests prompting them to work for mutual advantage.33 The special political relations between states in the global sphere lack the element of reciprocity that allows the ascription of responsibility in a shared cooperative enterprise and this might limit the scope of egalitarian justice.34

However, it is hard to believe that this is what is going on, even within the state. As already observed at the beginning of the article, what really triggers the need for the creation of a joint political authority is the existence of controversy over the comparative claims particular agents can authoritatively make, given the unavoidability of them affecting each other in relevant contexts. Justice is far from needing benevolence to come about. But if a similar account of the establishment of political relations within the state has some initial plausibility, why deny that it can ever be extended? Why not concede that, in accordance with particular historical developments, agents will need to find ways of including more and more relevant others in their political relation therefore expanding the scope of justice? A similar reading allows us to see the relationship between citizens within the state and citizens across states as politically dynamic rather than static. As the scale of conflicts increases, so does the range of inclusion. While it is true that justice requires political authority as a general enabling condition, constructing an authority able to impartially administer justice starting from where we are is an inherently political enterprise.

Now, it is clear from the earlier analysis of this article that a similar political authority will need to satisfy certain democratic criteria in order to be able to adjudicate impartially between conflicting comparative requirements. To prevent the emergence and imposition of individual interpretations on claims concerning relative shares, the general will should be constructed in accordance with democratic procedures, where the voice of everyone is equally taken into account. This applies both to citizen-relations within the state and to those across different states. But to insist on the democratic nature of institutions able to administer justice in an impartial way, does not necessarily mean that the imperative to expand political relations with relevant others involves an obligation to coercively bring about democracy where it is absent. A similar move would be as unilateral as the state of affairs it is trying to undermine. If politics is to be taken seriously, external intervention to democratise seemingly undemocratic states might simply weaken the political agency expressed in civic efforts to transform institutions from within.

This is why naïve cosmopolitans are wrong to dismiss the state. We cannot ignore existent political relations: it would be normatively mistaken and it would be practically counterproductive. As long as states interact with one another in ways that affect their citizens’ capacity to occupy specific social roles, and as long as their institutional involvement in the global sphere is such that it generates a concern for the comparative position of each, this provides solid ground to assert our interest in their distributive equality. It is precisely the existence of this interest that allows new associative political relations to develop on the basis of old ones. Citizens can interact politically with each other to channel their moral demands, so as to transform these moral demands into constraints of justice. Coercion and cooperation do not limit global justice—they rather specify the normatively relevant context on the basis of which justice can expand its reach.

CONCLUSION

Throughout the article I have tried to show that the kinds of moral principles to which we often appeal in defending cosmopolitan commitments are at least in part distinguished from principles of justice. I based this argument on what seems to be an important distinction between comparative and non-comparative claims, and I tried to clarify what kind of relationship the concept of justice and the idea of the state entertain with such comparative demands. Emphasising the distinction between morality and justice has allowed us to show that in the absence of legitimate collective authorities able to produce determinate claims and to enforce compliance, cosmopolitanism can produce only wide rather than strict obligations.

As I also hope to have shown, it is possible to shift from a moral to a justice-based dimension by understanding the allocation of cosmopolitan obligations as a political task. Placing appropriate emphasis on the state as providing the relevant context in which citizens can politically interact with each other is fundamental for the effective, strict, undertaking of an attempt to build cosmopolitan justice. The burden for those who assert the contrary is to show how it might be possible to allocate strict obligations of egalitarian justice sidestepping the collective forms of political institutionalisation that the state provides. The distinctions between morality and justice that I have underlined require a shift from a normative strategy of justifying cosmopolitan principles on the basis of individual moral entitlements to a more sophisticated account emphasising the role of citizens of particular states as agents of political progress. Understanding the specificity of circumstances under which a political will takes shape, is the first important and unavoidable step in the promotion of global justice.

ACKNOWLEDGEMENTS

I am grateful to Christine Chwaszcza, Bob Goodin, Jonathan White and, especially, the editors and referees of Ethics & Global Politics for extremely helpful comments on earlier versions of this article.

NOTES

1. Here I focus on cosmopolitan principles that aim to reduce global inequalities between individuals, i.e. principles of an egalitarian kind. The kind of cosmopolitan argument addressed in this article is present in the work of various prominent scholars such as Charles Beitz, Brian Barry, Simon Caney, Darrel Moellendorf, Thomas Pogge, and others. For detailed references to their work and a reconstruction of the arguments, see Lea Ypi, ‘Political Membership in the Contractarian Defence of Cosmopolitanism’, The Review of Politics 70, no. 3 (2008): 442–72.

2. I have in mind here authors such as Michael Blake, David Miller, Thomas Nagel, Mathias Risse and others. I have introduced and discussed the views of these authors in Lea Ypi, ‘On the Confusion between Ideal and Non-Ideal in Recent Debates on Global Justice’, Political Studies 58, no. 3 (2010): 536–55.

3. For a good review of the literature and a recent articulation of this cosmopolitan position see, for example, Gillian Brock, Global Justice: A Cosmopolitan Account (Oxford: Oxford University Press, 2009), esp. Chap. 12. It should be noted that even though Brock's argument often oscillates between a complete endorsement of cosmopolitan egalitarianism and creating space for legitimate forms of nationalism, she is very keen to emphasise that the principles of global justice follow from our recognition of the equal moral standing of all human beings. If the argument presented here is correct, that claim deserves a more thorough defence than has been given.

4. Aristotle, Nicomachean Ethics (London: Penguin, 2004), 119.

5. Aristotle, The Politics (London: Penguin, 1981), Chap. 12.

6. On the distinction between duties of justice and duties of virtue, see Immanuel Kant, ‘The Metaphysics of Morals’, in Practical Philosophy, ed. Mary Gregor (Cambridge: Cambridge University Press, [1797] 1996), 521.

7. Ibid.

8. Ibid., 409.

9. I have discussed this issue at length in Lea Ypi, ‘Self-Ownership and the State: A Democratic Critique’, Ratio (forthcoming).

10. The contribution need not have been explicitly consented to at one particular time. In enjoying civic benefits and taking up civic responsibilities, in participating in the political life of the state, in teaching their children collectively acquired rights and obligations, the activity of members of the state is already expressive of their involvement in collective institutions without which no binding collective decision could be made.

11. One should be cautious, however, with the use of terms such as ‘internal’ or ‘external’ authority. Kant, for one, shared with Rousseau the idea that the political sphere is the only one where individuals fully realise themselves. As he puts it, ‘one cannot say: the human being has sacrificed a part of his innate outer freedom for the sake of an end, but rather, he has relinquished entirely his wild lawless freedom in order to find his freedom as such undiminished, in a dependence upon laws, that is in a rightful condition,’ see Kant, ‘The Metaphysics of Morals’, 521.

12. A similar argument is also made, although in anti-cosmopolitan key by Thomas Nagel, ‘The Problem of Global Justice’, Philosophy & Public Affairs 33, no. 2 (2005): 113–47. I shall examine the anti-cosmopolitan part of the argument in a moment.

13. I have discussed this issue in Lea Ypi, ‘Statist Cosmopolitanism’, Journal of Political Philosophy 16 (2008): 48–71.

14. See for a discussion of these issues also Thomas Risse, Stephen C. Ropp, and Kathryn Sikkink, eds., The Power of Human Rights: International Norms & Domestic Change (Cambridge: Cambridge University Press, 1999); Margret Keck and Kathleen Sikkink, Activists Beyond Borders: Advocacy Networks in International Politics (Ithaca, NY: Cornell University Press, 1998).

15. See for the status quo critique to statism A.J. Julius, ‘Nagel's Atlas’, Philosophy & Public Affairs 34, no. 2 (2006): 176–92; Joshua Cohen and Charles Sabel, ‘Extra Rempublicam Nulla Justitia?’, Philosophy & Public Affairs 34, no. 2 (2006): 174–75.

16. John Rawls, for example, emphasises that rich countries might not agree to principles of this kind because they appear too demanding, see John Rawls, The Law of Peoples (Cambridge, MA: Harvard University Press, 1999), 106.

17. See Ibid., 106–18; Michael Blake, ‘Distributive Justice, State Coercion, and Autonomy’, Philosophy & Public Affairs 30, no. 3 (2001): 257–96; Nagel, ‘The Problem of Global Justice’, 135–6.

18. See, for one prominent example, David Miller, National Responsibility and Global Justice (Oxford: Oxford University Press, 2007).

19. Nagel, ‘The Problem of Global Justice’, 130.

20. I borrow the term from Liam Murphy, ‘Institutions and the Demands of Justice’, Philosophy & Public Affairs, no. 27 (1998): 253–4; see also Nagel, ‘The Problem of Global Justice’, 112.

21. I explore more in detail the concrete demands of statist cosmopolitanism in chapter five of my book manuscript ‘Global Justice and Avant-Garde Political Agency’ (under contract with Oxford University Press).

22. See Brian Barry, ‘Humanity and Justice in Global Perspective’, in Contemporary Political Theory. An Anthology, ed. Robert E. Goodin, and Philip Pettit (Oxford: Blackwell, 1982), 525–40; Andrea Sangiovanni, ‘Global Justice, Reciprocity, and the State’ Philosophy & Public Affairs 35, no. 1 (2007): 3–39.

23. See Blake, ‘Distributive Justice, State Coercion, and Autonomy’; Nagel, ‘The Problem of Global Justice’; Mathias Risse, ‘What to Say About the State’, Social Theory and Practice 32, no. 4 (2006): 671–98.

24. I am framing the issue here slightly differently from the way in which defenders of the cooperation argument frame it. I believe it is the ascription of responsibility for the inequalities produced by the system rather than the simple fact that individuals cooperate with one another in producing specific goods that is relevant for tracing the distinction between intra-national and trans-national instances of cooperation. On the issue of responsibility see also Miller, National Responsibility and Global Justice, cit.

25. Nagel, ‘The Problem of Global Justice’, 121.

26. The argument that decisions of particular states affect citizens living outside their boundaries is now commonplace in the literature and has led to a number of theoretical contributions consistent with the argument of this article. See for some prominent examples, Iris Marion Young, Inclusion and Democracy (Oxford: Oxford University Press, 2002); John S. Dryzek, Deliberative Democracy and Beyond—Liberals, Critics, Contestations (Oxford: Oxford University Press, 2002); Nancy Fraser, Scales of Justice: Reimagining Political Space in a Globalizing World (Cambridge: Polity Press, 2008).

27. See in this issue Joseph Stiglitz, and Andrew Charleton, Fair Trade for All (Oxford: Oxford University Press, 2005).

28. For discussions of this topic see the essays contained in Thomas Pogge, Matthew Rimmer, and Kim Rubenstein, eds., Incentives for Global Public Health: Patent Law and Access to Essential Medicines (Cambridge: Cambridge University Press, 2010).

29. For a more detailed analysis of how states engage in networks of information sharing, enforcement of existing national laws and harmonisation of regulatory standards, see Anne-Marie Slaughter, A New World Order (Princeton, NJ: Princeton University Press, 2004).

30. Barry, ‘Humanity and Justice in Global Perspective’, 531–2.

31. John Rawls, A Theory of Justice, rev. ed. (1971, Cambridge, MA: Belknap Press of Harvard University Press, 1999), 457.

32. Ibid., 458.

33. Ibid., 457.

34. Barry, ‘Humanity and Justice in Global Perspective’, 532; Nagel, ‘The Problem of Global Justice’, 127–8.

About The Author

Lea Ypi
Nuffield College
United Kingdom