Polity . Volume 41, Number 4 . October 2009 r 2009 Northeastern Political Science Association 0032-3497/09 www.palgrave-journals.com/polity/ Cosmopolitanism, Nationalism and Moral Opportunity Costs* Carmen E. Pavel University of Virginia R C O PY Cosmopolitanism and nationalism are at war, if the criticism they relentlessly direct at each other is any guide. The current debates between defenders of these two views tend to solve their disagreements by showing that one view is incoherent and assigning victory to the other. I argue instead that cosmopolitanism and nationalism do not fail on their own, but are rather incomplete facets of the truth, because each reflects demands of morality that are in permanent tension with one another. Moreover, there may be no way in principle to establish a binding order of priority between nationalistic and cosmopolitan claims, even if in practice we will find various ways to negotiate between them. Polity (2009) 41, 489–513. doi:10.1057/pol.2009.1; published online 30 June 2009 Keywords TH O cosmopolitanism; nationalism; moral responsibility; deontic pluralism; opportunity costs; justice U Carmen E. Pavel is a postdoctoral fellow in the program in Political Philosophy, Policy and Law at University of Virginia. She specializes in liberal theory, international justice and the morality of international institutions. She can be reached at pavel@virginia.edu. A Cosmopolitans and nationalists display distinctive attitudes about international justice. On the one hand, cosmopolitans say that national boundaries cannot mark the limits of our moral obligation toward our fellow humans. One’s substantive duties also extend beyond the borders of one’s political community, to include equally all human beings. On the other hand, defenders of the primacy of patriotic duties claim that one’s compatriots are entitled to special *I am indebted to John Tomasi, Sharon Krause, Charles Larmore, and William Galston, for kindly reading several drafts. In addition, I would like to acknowledge Jason Brennan, Adam Tebble, and the participants at the Political Philosophy Workshop at Brown University for constructive feedback. I am grateful for suggestions from Loren Lomasky, Thom Brooks, Andrew Polsky, and two anonymous referees for Polity during the final stages of writing this paper. 490 COSMOPOLITANISM, NATIONALISM A U TH O R C O PY consideration.1 As the national political community embodies the highest good, our moral allegiance goes first to honor the demands of political membership. The moral primacy of patriotic duties necessarily subordinates an individual’s other allegiances, including those owed to the wider community of human beings. These two views are at war, if the criticism they relentlessly direct at each other is any guide. Cosmopolitans see patriotism as narrow and parochial. They look down on it for failing to take seriously the equality of all human beings. Advocates of patriotism counter that cosmopolitanism fails to accommodate special ties, especially patriotic ones, and therefore it must be rejected as incoherent. My argument is a radical departure from the current tenor of the debates between cosmopolitans and nationalists, which tend to resolve the disagreement by attempting to show that one view is incoherent and assigning victory to the other. I claim instead that cosmopolitanism and nationalism do not fail on their own, but are rather partial, incomplete facets of the truth, because both incorporate objective demands of morality that are in permanent tension with one another. To get a more accurate picture of our moral responsibilities, we need to keep the tension in mind without suppressing either partial view. If this evaluation is correct, then we need not only to redraw our map of international justice in a manner that accommodates both views, but also to reevaluate accordingly the solutions to the various philosophical and institutional problems that this new map raises. The next two sections argue that cosmopolitanism and nationalism have traditionally underestimated the dilemma of competing moral allegiances. The subsequent section explains why this is a problem in terms of opportunity costs. In the following section, I offer a pluralist moral view that is best positioned to help us understand why the tension between cosmopolitanism and nationalism is real. Pluralism has wider implications for global justice. It explains not only what gives rise to the tension between cosmopolitanism and nationalism, but also why we cannot hope to provide a unified theoretical account that incorporates both views. The penultimate section focuses on the argument that Kok-Chor Tan offers for incorporating nationalist duties within a broader cosmopolitan framework. I show precisely why the tension between the two kinds of duties cannot be resolved this way. Finally, I clarify the implications of this analysis for theory and practice. 1. For the purpose of this paper I use patriotism and nationalism interchangeably here. Clearly there are cases where the national community and the political community have different boundaries; however, the concepts overlap significantly in both political practice and ordinary language. Carmen E. Pavel 491 Cosmopolitanism and the Place of Special Ties A U TH O R C O PY Moral cosmopolitanism draws on several distinct moral traditions, and as such it does not represent one unified view, but rather a family of diverse views with common threads. Cosmopolitan perspectives share the idea that individual human beings are the objects of ultimate moral concern and equally so, regardless of their membership in particular communities, cultures, or political structures. This idea has important implications for the scope of rights and obligations individuals have. The universal moral equality of human beings entails substantive duties that extend across national boundaries. Moral cosmopolitanism is distinguished from political cosmopolitanism, which is sometimes, although not always, taken as a direct consequence of the former.2 Political cosmopolitanism is a view about the appropriate global institutions that should govern the behavior of individuals around the world, be it a world government, a federal system of states, or some other system of international institutions. In this paper, I will engage moral cosmopolitanism. It is not my purpose to describe and make sense of the variety of moral cosmopolitan views, which cannot be usefully contained within the scope of this paper. I will rather identify a representative thread within liberal cosmopolitanism committed to the moral equality of all individual human beings and highlight a particular difficulty with it. Peter Singer has made a distinctive contribution to moral cosmopolitanism by offering a principled argument for a global duty of assistance. Appealing to our deep aversion to human suffering, he presents troubling images of disease, illiteracy, lack of shelter, high mortality rates, and developmental deficits due to malnutrition among the world’s poor.3 The cosmopolitan responds to this condition by urging each of us to act so as to prevent the bad and promote the good: ‘‘if it is in our power to prevent something bad from happening, without thereby sacrificing anything of comparable moral importance, we ought, morally, to do it.’’4 In positing a duty of assistance, the core principle does not discriminate between people who are close to us and people who are far away. It is irrelevant if 2. In addition to moral and political cosmopolitanism, cultural cosmopolitanism represents a distinctive view. Cultural cosmopolitanism rejects the idea that a person’s well-being depends on membership in a defined culture. This view emphasizes openness to cultural diversity and mingling, immersion in different cultural milieus, and the plasticity of an individual’s personal identity. See Samuel Scheffler, ‘‘Conceptions of Cosmopolitanism,’’ in Boundaries and Allegiances (Oxford, UK: Oxford University Press, 2001), 111–30. For a proponent, see Jeremy Waldron, ‘‘Minority Cultures and the Cosmopolitan Alternative,’’ University of Michigan Journal of Law Reform 25 (1992): 751–93. For a critical perspective, see Pratap Mehta, ‘‘Cosmopolitanism and the Circle of Reason,’’ Political Theory 28 (2000): 619–39. 3. Peter Singer, ‘‘Famine, Affluence and Morality,’’ Philosophy and Public Affairs 1 (1972): 229–43; also Peter Unger, Living High and Letting Die: Our Illusion of Innocence (New York: Oxford University Press, 1996), which is an elaboration and defense of Singer’s position. 4. Singer, ‘‘Famine,’’ 231. 492 COSMOPOLITANISM, NATIONALISM A U TH O R C O PY the person who needs our help is a neighbor’s child or a stranger in a distant country on the other side of the world.5 Other cosmopolitan defenses follow this general line of argument. One such prominent version is based on the theoretical framework set up by John Rawls in A Theory of Justice. Rawls argues that a person’s race, gender, talents, and wealth are arbitrary from a moral point of view.6 Decisions on the principles that govern a society have to abstract from such morally irrelevant characteristics accordingly. Rawls imagines a decision device, the original position, in which people deliberate under a ‘‘veil of ignorance’’ that ensures the hypothetical decision-makers are not aware of their contingent features and beliefs such as race, gender, status in society, and conception of the good life.7 The principles of justice that result from deliberation in the original position reflect this condition of fairness, and incorporate the fundamental values of liberty and equality in two principles of justice, lexically ordered. The first principle ensures the rights of persons to an extensive liberty, compatible with the liberty of all, and the second ensures substantive economic redistribution to correct existing inequalities, inequalities that can be tolerated only to the extent that they are beneficial to the least advantaged group in society. Although Rawls himself resisted the extension of justice as fairness on a global scale,8 some Rawlsians such as Charles Beitz and Thomas Pogge argued that just like race, gender, and natural talent, membership in a nation is just another morally irrelevant characteristic.9 Principles of justice should apply to individuals across societies and not just within one society. A strong case can be made on grounds of justice that ‘‘persons of diverse citizenship have distributive obligations to one another analogous to those of citizens of the same state.’’10 This is because the moral status of individuals is the same no matter what their political membership happens to be. In the same vein, Leif Wenar, who appropriates Scanlon’s general theory of moral duties to justify cosmopolitan obligations, says that all our actions are subject to a moral requirement: we need to act in ways that can be justifiable to others on grounds that they cannot reasonably reject, regardless of our 5. Singer, ‘‘Famine,’’ 231. 6. John Rawls, A Theory of Justice (Cambridge: Harvard University Press, 1971), 15. 7. Rawls, A Theory of Justice, 11, 18. 8. John Rawls, The Law of Peoples (Cambridge: Harvard University Press, 1999). 9. Thomas Pogge, Realizing Rawls (Ithaca, NY: Cornell University Press, 1989), 247. See also Allen Buchanan, ‘‘Rawls’s Law of Peoples: Rules for a Vanquished Westphalian World,’’ Ethics 110 (July 2000): 697–721. 10. Charles Beitz, Political Theory and International Relations (Princeton: Princeton University Press, 1979), 128. See also 127–83. Carmen E. Pavel 493 A U TH O R C O PY relationship with them.11 This desire to act in ways that others cannot reasonably reject comes from the value that one perceives in other human beings. Wenar thinks that such an approach gives us good reason to deliberate about our duties in global terms, taking into account all moral agents.12 The reasons we have to justify our actions to other human beings take precedence over more particular reasons we have to engage those closer to us.13 The moral duties that respond to universal reasons are consequently prior to our more particular duties. Because of this priority, the acceptable form of our individual, diverse, and particular allegiances will be shaped by the framework created by our more general obligations. Despite their differences, the different foundational accounts of cosmopolitanism share a view of justice according to which our moral obligations carry over the boundaries of particular political communities. Cosmopolitanism thus enlarges the moral world of each person to include all other human beings equally and globally in an ethically significant relationship.14 This ethical relationship is fundamental to the cosmopolitan outlook, in the sense that it generates the map of our essential moral responsibilities and the contours of our primary moral allegiances in unambiguous terms. As citizens of the world we have a strong, overriding, prima facie moral duty ‘‘to the moral community made up by the humanity of all human beings.’’15 Our moral horizon needs to be reshaped in a way that allows us to give preeminence to this primary allegiance, so that in our everyday lives we do not get sidetracked by ties that are parochial and narrow. Understood this way, cosmopolitanism paints a pristine picture of moral life, purged of the dilemmas caused by competing moral allegiances. Both critics and defenders note that cosmopolitanism does not give adequate consideration to moral claims that spring from legitimate special ties to one’s family, community, or country. This is ‘‘the philosophical weakness most characteristic of cosmopolitan theories,’’ writes Charles Beitz, in a call for revision of the cosmopolitan moral outlook.16 Not only is this neglect ubiquitous, but it also involves at times an active effort to crowd out special ties and obligations. Some cosmopolitans regard these 11. Leif Wenar, ‘‘What We Owe to Distant Others,’’ Politics, Philosophy & Economics 2 (2003): 283–304. Thomas Scanlon, What We Owe to Each Other? (Cambridge: Belknap Press, 1999). 12. Wenar, ‘‘What We Owe to Distant Others,’’ 284. 13. Wenar, ‘‘What We Owe to Distant Others,’’ 285. 14. Catherine Lu, ‘‘The One and Many Faces of Cosmopolitanism,’’ Journal of Political Philosophy 8 (June 2000): 244–67. 15. Martha Nussbaum, ‘‘Patriotism and Cosmopolitanism,’’ in For Love of Country? ed. Joshua Cohen (Boston: Beacon Press, 1996), 9. Nussbaum offers a much more nuanced view in Women and Human Development: The Capabilities Approach (Cambridge: Cambridge University Press, 2000). 16. Charles Beitz, ‘‘International Liberalism and Distributive Justice: A Survey of Recent Thought,’’ World Politics 51 (1999): 269–96, 291. 494 COSMOPOLITANISM, NATIONALISM U TH O R C O PY partial ties as positively problematic. To respond to the moral claims of people with whom one stands in close relationship amounts to a form of pernicious prejudice.17 Partial concern is subversive, Nussbaum says, ‘‘because it substitutes a colorful idol for the substantive universal values of justice and right.’’18 When someone has claimed devotion to a local culture, a particular group of people, or system of beliefs, that person has defined herself by a morally irrelevant characteristic. We have to transcend the divisions that subvert the values that hold us together, and ‘‘give our first allegiance to what is morally good.’’19 The most important moral good is our common humanity, and our moral resources should be channeled first and foremost to its support. It is unclear, though, why people’s particularistic attachments are morally irrelevant characteristics. Nussbaum herself gives no reason other than equating such attachments to political extremism (as in the case of nationalism) or questionable social hierarchy (as in the case of a Hindu who chooses to define himself primarily as a member of the upper caste). But her examples are far from exhaustive. Local affiliations, whether chosen or ascriptive, represent meaningfully defining human traits. Theorists have extensively documented the values and valences of associations for their members and the societies of which they are part.20 Groups provide intrinsic satisfaction and value for members by sustaining human flourishing, self-development, and self-affirmation. Their economic, political, and cultural impact makes them worthy objects of investment of time and energy. Groups become sites of moral commitment, and sources of benefits and obligations that shape the identity of their members and define what gives their lives meaning and value. Nonetheless, apart from the occasional nod, special ties receive a hostile treatment by and large in the cosmopolitan literature. A The Priority of National Duties As is the case with cosmopolitanism, nationalism unites under its banner a group of theorists with diverse moral outlooks.21 Communitarian, conservative, 17. Robert E. Goodin, Protecting the Vulnerable (Chicago: The University of Chicago Press, 1985). 18. Nussbaum, For Love of Country? 5. The strong language she uses to disparage particularistic concerns prompt Nathan Glazer, Hilary Putman, and Michael Walzer to criticize this facet of her view in their response essays included in the volume. Nussbaum herself resists this interpretation of her view in ‘‘Reply’’. However, her critics’ interpretation is plausible, if not entirely on the mark. 19. Nussbaum, For Love of Country? 5. 20. Nancy Rosenblum, Membership and Moral: The Personal Uses of Pluralism in America (Princeton, NJ: Princeton University Press, 2000). 21. Roger Scruton, ‘‘The First Person Plural,’’ in Theorizing Nationalism, ed. Ronald Beiner (Albany NY: SUNY Press, 1999); David Miller, Citizenship and National Identity (Cambridge: Polity Press, 2000); Alasdair MacIntyre, ‘‘Is Patriotism a Virtue?’’ The Lindley Lecture, University of Kansas, 1984. Carmen E. Pavel 495 A U TH O R C O PY and liberal nationalists share skepticism of the idea of international justice. Nourished by a commitment to the primacy of the bounded political community, nationalists resist the suggestion that we have significant moral responsibilities to all human beings in the world. This diverse group does take special ties, or rather a certain kind of special tie (ties to one’s country), very seriously. Communal, patriotic ties demand that one puts the interests of one’s own compatriots ahead of the interests of distant strangers. With the exception of Alasdair MacIntyre, who claims that reconciling the morality of patriotism with that of cosmopolitanism is incoherent, proponents of nationalism do grant that individuals may show concern to people outside of their political communities.22 Importantly however, patriotism is for them not just one of many sources of moral allegiance, but often the supreme form. National identity is the primary source of identity, and as such it commands our highest allegiance. National identity is important for a variety of reasons. According to David Miller, it cultivates social solidarity otherwise missing in large and anonymous communities, encourages the protection of the national culture, and fosters a concern with social justice through common institutions.23 There is an important parallel here with cosmopolitanism. Both cosmopolitans and defenders of patriotic ties postulate one morally relevant community that contours our moral space—the worldwide community of human beings for cosmopolitans and the nation for defenders of patriotism. The respective morally relevant communities define our primary moral responsibilities that subordinate all our other moral or non-moral concerns. But there is another important similarity. As I noted, cosmopolitanism is insufficiently attuned to the plural and conflicting nature of competing moral demands. In particular, cosmopolitans fail to accommodate special ties and responsibilities to our family, community, or country. However, cosmopolitans are not the only ones neglecting the difficulty of plural moral allegiances. The defenders of nationalism are also guilty of this neglect. Nationalists are ostensibly more attuned to the plurality of moral obligations. Both Miller and MacIntyre admit that national identity is not the sole duty-generating special tie. Miller says that ‘‘there may be other, perhaps smaller and more intense, communities whose members we owe duties that are more stringent than those we owe to Britons, Swedes, etc. at large.’’24 And MacIntyre claims that ‘‘patriotism is one of a class of loyalty-exhibiting virtues . . . , other members of which are marital fidelity, the love of one’s own family and kin, friendship and loyalty to such institutions as schools or cricket and baseball clubs.’’25 Still, for Miller and MacIntyre these duties 22. 23. 24. 25. MacIntyre, ‘‘Is Patriotism a Virtue?’’ 12. Miller, On Nationality (Oxford: Clarendon Press, 1995). Miller, Citizenship and National Identity, 27. MacIntyre, ‘‘Is Patriotism a Virtue?’’ 4. 496 COSMOPOLITANISM, NATIONALISM A U TH O R C O PY are secondary to those generated by patriotic ties. However, this priority is not explicitly defended. We can attempt to reconstruct a case for the primacy of patriotic partiality. The reason for patriotic priority, nationalists argue, is that we cannot forge our other loyalties without a foundational national identity that creates the stable preconditions necessary for their thriving. Key among preconditions is public institutions that guarantee a safe space in which we develop meaningful human interactions with other human beings. This is a complex claim about the relationship of national identity to stable political institutions. And it may very well be true that stable political institutions are indeed necessary for people to be able to express and develop their personal loyalties. But a strong national identity, although helpful for such stability, is neither empirically necessary nor morally required. People can develop meaningful human relationships, ones that are duty-generating in the strongest sense, both inside and outside of the political boundaries they inhabit, without having the experience of national fellow-feeling. There is nothing uniquely necessary about national commitment to either a fulfilling individual life or the proper functioning of political institutions. This is not to say that political institutions do not depend on conditions within the larger culture that ensure their stability. While these conditions may overlap with a strong national identity, they can be also be understood as something entirely different: commitments to a set of principles that should regulate public life independent of any special allegiance of citizens to one another. In light of this, the puzzling feature of the nationalist view comes more forcefully to light if we take for granted Miller’s underlying belief that people may legitimately show special concern for their compatriots. ‘‘It may be properly part of someone’s identity that they belong to this or that national grouping; . . . national identity may, but need not be, a constitutive part of personal identity.’’26 The upshot is that not everybody will have good reason to show such concern. We do not need to go so far as Kateb, who sees patriotism as a sign of ‘‘moral and mental obtuseness,’’ to realize that for many patriotism bears no significant moral weight.27 Certain social or historical conditions might elicit patriotic fellow-feeling, which can be enlisted for good causes, such as fighting a common oppressor. Nonetheless, Kateb is right to caution us against the morally stultifying force of patriotic allegiance. More often than not, patriotism has been used to engage in actions that are irrational or irresponsible. People have been ready to die and to kill in the name of patriotism for causes that were both 26. Miller, Citizenship and National Identity, 27. 27. George Kateb, ‘‘Is Patriotism a Mistake?’’ Social Research 67 (Winter 2000): 901. Carmen E. Pavel 497 A U TH O R C O PY immoral and far distant from their own personal concerns.28 It is not entirely unreasonable, then, to consider one’s allegiance to one’s nation as secondary to that which springs from familial ties and small cultural communities, or to a moral community that transcends national boundaries. On some accounts, the view that patriotic ties are morally prior is also problematic for failing to appreciate the normative force of the cosmopolitan impulse. MacIntyre believes that our deeply embedded moral life makes it ‘‘impossible’’ to have any regard for people who might be suffering in the distance; such a stance would require an implausible conception of human beings, one that is ‘‘abstract and artificial,’’ disconnected from the local commitments that give shape to their everyday existence.29 MacIntyre may rightly insist that cosmopolitans fail to understand and make room for our special attachments by claiming to judge all moral matters from an impartial point of view. However, he cannot claim that because individuals must work through and live with their partial attachments, they are not capable of abstracting from them to consider the plight of outsiders. If he did believe that compassion for strangers is always implausible given the necessarily circumscribed nature of the human moral space, Mother Teresa would represent not just an anomaly but a morally deficient person, for failing to focus primarily on her own family or compatriots. Additionally, defenders of nationalism miss the extent to which countries are not, if they ever were, isolated and self-sufficient units. Individuals and groups within a state are increasingly engaged in interactions across borders. The mobility of labor, capital, and people in recent decades prove that borders are becoming more porous. Those interactions themselves generate special relationships, and are therefore significant sources of moral claims. The recent waves of immigration from less developed to more developed countries, for instance, put immigrant citizens in a closer relationship with the family they left behind in their home country rather than the fellow citizens of their adoptive country. Thus obligations of citizenship must give way both to special obligations that extend across borders and to obligations to distant strangers. 28. George Kateb, ‘‘Is Patriotism a Mistake?’’ 907, also 910–913. 29. MacIntyre argues that cosmopolitanism ‘‘requires of me to assume an abstract and artificial— perhaps even an impossible—stance, that of a rational being as such, responding to the requirements of morality not qua parent or farmer or quarterback, but qua rational agent who has abstracted him or herself from all social particularity, who has become not merely Adam Smith’s impartial spectator, but a correspondingly impartial actor, and one who in his impartiality is doomed to rootlessness, to be a citizen of nowhere. How can I justify to myself performing this act of abstraction and detachment?’’ in ‘‘Is Patriotism a Virtue?’’ 12. 498 COSMOPOLITANISM, NATIONALISM Opportunity Costs A U TH O R C O PY The unarticulated assumption behind the nationalist resistance to the cosmopolitan view can be best understood in terms of opportunity costs. MacIntyre implicitly assumes compassion for strangers is difficult because it takes away from the compassion one should show for close ones. In other words, compassion for strangers is costly, and prohibitively so. Opportunity cost, a term often used by economists, represents the benefits that are associated with an opportunity forgone. When a town decides to build a new road, it means it cannot use that money for a new hospital. Similarly, if a country decides to spend more on its military, the opportunity cost is some other program that might have been supported with that amount of money, such as welfare programs or medical research. Opportunity costs need not be expressed in monetary terms. They can be anything that is of value for the person or institutions doing the evaluation. For instance, the opportunity costs of a day at work is not just more leisure time, but also less stress, perhaps less physical and intellectual exertion, which are not always quantifiable in terms of money. Opportunity costs are distinguished from accounting costs, which represent the amount of money exchanged to acquire the good or service in question and are always monetary. Opportunity costs are essential for knowing the true cost of an action, beyond its monetary expression. They are often hidden costs, whose neglect can create the impression that an action has only benefit and no cost at all. When a new road is built, we realize its benefit, but we may not ‘‘see’’ that a plan for a hospital was set aside because of it. Opportunity costs help us better understand what is at stake when cosmopolitan or nationalist proposals are advanced. Whenever individuals avert their moral gaze from duties to their immediate fellows and respond to universal duties of assistance, there is an opportunity cost for their compatriots. This is measured in terms of affective moral capacities, physical and material resources that could have benefited the needy compatriots, but for their diversion to more distant, though perhaps more pressing, causes. Moral opportunity costs work both ways. Whenever individuals pay more attention to people who are closer to them, they forego the chance to respond to those more distant, whose lives are less entangled with their own in visible and immediate ways. Samuel Scheffler briefly describes how opportunity costs operate in the case of competing moral duties: Suppose that you have recently become my friend and that I have therefore acquired special responsibilities to you. Clearly, these responsibilities work to your advantage, inasmuch as I now have a duty to do things for you that would not previously have been required to do. At the same time, there are at least two different ways in which my responsibilities to you work to the Carmen E. Pavel 499 disadvantage of those people with whom I have no special relationship. First, in the absence of my responsibilities to you, I might have done certain things for them even though I had no duty to do so. Now, however, discharging my responsibilities to you must take priority over doing any of those things for them. Second, there may also be situations where my responsibilities to you take priority over the responsibilities that I have to them simply as human beings. For example, there may be times when I must help you rather than helping them, if I cannot do both, even though I would have been required to help them but for the fact that you too need help.30 A U TH O R C O PY The inattention to opportunity costs may be rooted in a certain understanding of the relationship between cosmopolitan and patriotic duties. On this understanding, one type of responsibility does not weaken other types of responsibilities. So, it might be argued, the addition of patriotic responsibilities does nothing to the nature and strength of cosmopolitan ones and vice versa. Different kinds of responsibilities do not come into conflict with one another, they merely pile up to increase one’s total share of obligations. Some responsibilities that we have as human beings are fixed and independent of any additional responsibilities that we might take.31 Their combination is thus frictionless. Discharging one kind of responsibility does not really affect our ability to discharge other kinds. If this is the case, then there is no need to make theoretical accommodations for multiple moral responsibilities. If conflicts among them do not occur, no negotiations need to be made, and the necessity for trade-offs among conflicting claims disappears. Although interesting, this argument is ultimately unconvincing. Human beings inhabit a universe of scarce resources. They have limited time, energy, money, and moral attention. When they are required to discharge their moral responsibilities, they must always choose how best to allocate those scarce resources. The use of material, motivational, and institutional resources to the satisfaction of one kind of moral responsibility leads to a reduction in the resources devoted to the other kind. Additional responsibilities raise the issue of opportunity costs, and so the necessity of negotiating between conflicting claims always arises. Consider an exchange between David Miller and Thomas Pogge. Pogge argues that understood in a certain way, cosmopolitan duties do not compete with national obligations, and therefore do not raise the question of opportunity costs. In a departure from standard cosmopolitan views that regard duties to outsiders 30. Samuel Scheffler, ‘‘Justice and Responsibility,’’ in Boundaries and Allegiances, 192. 31. For a good discussion and a critique of this view, see Scheffler, ‘‘Justice and Responsibility,’’ especially 86–87. 500 COSMOPOLITANISM, NATIONALISM TH O R C O PY as positive redistributive duties, Pogge claims that our most important duties to distant strangers are negative duties—duties not to cause harm.32 It is true that we may have more reasons to help family and fellow citizens than outsiders, but ‘‘compatriotism makes no difference to our most important negative duties.’’33 The stringency of negative duties is not affected by the special obligations to compatriots. Miller challenges cosmopolitan morality for denying the force of special obligations. He accepts the idea that every human being is equally an object of moral concern, whether a compatriot or not.34 But it does not follow from this that we have to treat all human beings equally, in that we have the same moral duties to each.35 We have special responsibilities for fellow citizens because political communities engage in institutionalized reciprocity and in common projects that reflect their particular values and beliefs of their members. This does not mean that we have no global duties at all, but ‘‘our duties need to be differentiated: we owe more to some than to others.’’36 Pogge concedes the moral weight of special obligations to compatriots. But if one understands cosmopolitan duties as negative duties not to cause harm, Miller’s objection can be sidestepped. Pogge portrays negative duties in a particular way, as duties not to impose unjust social institutions on other people. It is important to note that this is a more indirect way of understanding negative duties than the standard view. In the standard view, negative duties refer to the duty of one person to refrain from directly harming another person (by physically injuring that person) or to the duty of political institutions to refrain from harming individuals (by wrongly imprisoning them). It is less clear what the negative duty not to impose unjust institutions entails, or that it exists at all.37 A U 32. Thomas Pogge, ‘‘Cosmopolitanism A Defense,’’ Critical Review of International Social and Political Philosophy (CRISPP) 5 (2002): 86–91. See also World Poverty and Human Rights (Cambridge: Polity/ Blackwell, 2002). 33. Pogge, ‘‘Cosmopolitanism: A Defense,’’ 87, emphasis in original. 34. David Miller, ‘‘Cosmopolitanism: A Critique,’’ Critical Review of International Social and Political Philosophy 5 (2002): 80–85. 35. Miller, ‘‘Cosmopolitanism: A Critique,’’ 82. 36. Miller, ‘‘Cosmopolitanism: A Critique,’’ 83. 37. If one is a citizen of a country where another individual suffers some institutional wrong, such as lack of due process, for instance, is one responsible for the unjust institutional outcome? Does this mean that one has failed in one’s negative duty not to impose unjust institutions on others? Even if a more systematic injustice occurs, such that a group is mistreated in some objectionable way, say by not having access to equal political rights, does that imply that the other citizens are responsible for the unjustness of the institutional order? This seems like a difficult point to settle a priori. Responsibility depends on the kind of relationship between the citizens and the institutions that govern them. It matters whether those institutions are representative (they are the product of the citizens’ will in a broad sense) or are authoritarian, what each citizen’s place in the institutional hierarchy is, and whether one has the power to affect the decisions of those institutions. In any case, the issues here are complicated, and at the very least the examples I just offered show that it is problematic to move from a negative duty not to cause harm to another person to a negative duty not to impose unjust political institutions upon that person. Carmen E. Pavel 501 A U TH O R C O PY But let us assume that we do have negative duties not to impose unjust institutions on others. The international institutional order is unjust, Pogge continues, because it does violence to some people’s basic needs. These people lack the ability to live free from physical torture, meet their basic economic needs, or practice their cultural commitments. It follows, then, that people everywhere have a duty not to participate in this institutional order that is unjust to some people around the world, although it may be neutral or benign relative to others. If mere abstention is all that is required for those who take cosmopolitanism seriously, then Pogge is indeed right that there are no opportunity costs in terms of obligations to compatriots. The citizens of a nation could stop participating in unjust international institutions but continue to give preference to each others interests.38 No allocation of resources between competing ends is involved. But Pogge does not think the story ends here. Indeed the whole point of casting cosmopolitan obligations this way is precisely to point out the many ways in which people in wealthy countries have failed to uphold their negative duties not to impose unjust institutions upon the rest of the world. As a consequence, these well-off countries have incurred obligations to pay for the wrongs inflicted.39 The citizens of country X should not only refrain from participating in an international institution that harms the citizens of country Y. In cases where harm has been inflicted on whole populations through unjust international institutions, negative duties entail positive duties of redress. And here is where the positive duties of redress country X owes to country Y due to the injustice of international institutions conflict with the positive duties citizens of country X owe to each other. This creates a problem for both cosmopolitans and nationalists. As Thom Brooks clarifies in the same debate, this point can be easily missed by nationalists focused exclusively on defending the obligations of nationality. When states are causally and morally responsible for suffering in other states, they must be held remedially responsible for their actions.40 When the general obligations are positive duties, special responsibilities do conflict with more general ones. The duty to correct global institutional injustices will add on to one’s already existing responsibilities to compatriots. Contrary to Pogge, then, persons living together in political communities can decrease what 38. There may be opportunity costs of the citizens of the country in question in terms of the benefits they have purportedly received by participating in the international institutions, which they now have to give up, but this is a different opportunity costs than that involving reciprocal obligations of membership in a nation. 39. Thomas Pogge, World Poverty and Human Rights. 40. Thom Brooks, ‘‘Cosmopolitanism and Distributing Responsibilities,’’ Critical Review of International Social and Political Philosophy 5 (2002): 92–97. 502 COSMOPOLITANISM, NATIONALISM they owe to foreigners, and what is owed to foreigners can decrease what citizens of a country owe to each other.41 Pluralism and Special Responsibilities A U TH O R C O PY The failure of cosmopolitanism and nationalism to account for competing duties and the opportunity costs they induce does not reflect weaknesses in the particular views they advance, but rather a problem with an underlying assumption shared by both. The assumption is that there is a fundamental coherence to the moral world. This assumption is challenged by a rich, yet overlooked, account of morality. The view that I describe in this section is moral pluralism. As I do not have space to defend pluralism, those who have reservations toward this view might find the suggestions offered below incomplete. I hope, however, to show even skeptics that pluralism is well positioned to explain the deep tension between cosmopolitanism and nationalism. In the final two sections I will suggest that pluralism, if true, has important implications for global justice. Moral pluralism opposes a certain way of viewing the moral world that either reduces goods to a single measure of value or creates comprehensive orderings among values. Isaiah Berlin made this view popular in the second half of the last century.42 Bernard Williams, Joseph Raz, Stuart Hampshire, Thomas Nagel, Michael Stocker, Charles Larmore, and John Kekes are just a few of the philosophers who have joined Berlin in affirming moral pluralism.43 Pluralism maintains the existence of an objective moral universe. Unlike monists, pluralists say that fundamental human values are many, conflicting, rarely harmonious, and incommensurable, such that there is no single, common standard of measurement or arbitration for all. I focus on pluralism of duties— deontic pluralism—as opposed to a more general pluralism of values.44 Central to this version of pluralism is the idea that the sources of our moral obligations spring primarily (but not exclusively) from the relationships in which we stand 41. Pogge, ‘‘Cosmopolitanism: A Defense,’’ 91. 42. Isaiah Berlin, ‘‘Two Concepts of Liberty,’’ in Liberty, ed. Henry Hardy (New York: Oxford University Press, 2002). 43. Bernard Williams, Moral Luck (Cambridge: Cambridge University Press, 1981); Steven Lukes, Moral Conflict and Politics (Oxford: Clarendon Press, 1986); Joseph Raz, The Morality of Freedom (Oxford: Clarendon Press, 1986); Stuart Hampshire, Morality and Conflict (Cambridge: Harvard University Press, 1983); Thomas Nagel, Mortal Questions (Cambridge: Cambridge University Press, 1979); Michael Stocker, Plural and Conflicting Values (Oxford: Clarendon Press, 1990); Charles Larmore, Patterns of Moral Complexity (Cambridge: Cambridge University Press, 1987); Martha Nussbaum, The Fragility of Goodness: Luck and Ethics in Greek Tragedy and Philosophy (Cambridge: Cambridge University Press, 1986); John Kekes, The Morality of Pluralism (Princeton: Princeton University Press, 1993). 44. For a brief description of deontic pluralism, see also W.D. Ross, The Right and The Good (Oxford: Clarendon Press, 1930), especially Chapter II. Carmen E. Pavel 503 A U TH O R C O PY with other people. The point is not to provide a complete list of distinct duties, but to show that sources of moral obligations are plural and normatively independent.45 I will proceed in two steps. The first, uncontroversial step is to show that we have a variety of relationships that generate morally significant commitments. The second and more contentious move is to prove that these morally significant commitments are normatively independent and irreducible to one another. This second step will be the main focus of the section. Let us begin with the plurality of moral obligation. It is a feature of commonsense morality that, as individuals, we have responsibilities for our families, friends, and other people with whom we stand in special relationships. As a consequence we experience our everyday life as a competition between claims that pull us in different directions. I have made a promise to meet someone for lunch, but there is a new deadline at work I have to meet by tonight so I am faced with a choice. I am called to make voluntary contributions to both my church and my local library. As I have limited funds, I need to decide which of these organizations should receive my support. Situations such as these populate our daily ethical life and raise dilemmas that each of us has to resolve. Sometimes these decisions are easy, sometimes they are heart wrenching, but they reflect choices between obligations that we incur by virtue of being who we are: individuals uniquely situated within a web of personal relationships and commitments that define our moral space. Friendships, familial and professional ties, relationships with neighbors, sports clubs, religious and ethnic groups, and ties of membership in a nation are all relevant to the moral life of individual people. Of course, one can also have general obligations to strangers—not to cause harm or to help in cases of natural disasters—that are not explained by any special relationship that connects us and the strangers in question. Crucially, though, whether they come from special relations or are more general in kind, our moral responsibilities are many, incommensurable, and irreducible to one another. Special ties ground duties that may conflict, and their priorities will be decided according to the circumstances of the case. W.D. Ross has called them ‘‘prima facie’’ duties or ‘‘conditional’’ duties.46 We do not agree which of the many different obligations should carry weight, but we do agree that there is more than one. Family and professional commitments are relatively uncontroversial, as are those that spring from promises/contractual obligations. But what if, normatively speaking, these multiple duties arising out of our special relationships are the 45. Ross and other deontic pluralists do provide a systematized account of distinctive duties, by separating primary, non-derivative duties from duties that are derivative from the primary ones. For such a list see Ross, The Right and The Good, 21. 46. Ross, The Right and The Good, 19. 504 COSMOPOLITANISM, NATIONALISM A U TH O R C O PY expression of a single general moral duty? What if the conflicts that we experience in ordinary moral practice are illusory, because the many different duties that we think we have are merely different expressions of one moral obligation? By this view, a proper understanding of their relationship to the overarching moral obligations would dissolve the apparent conflicts. There are strong reasons to think that this is not the case. The conflict of duties is an actual one, grounded in the reality of multiple normative commitments with irreducible normative bases. Duties to our family conflict with our professional duties because each set of duties is based on relationships whose value is not reducible to one another, which cannot be made commensurable, and are not substitutable in any simple sense. The conflict we experience in our daily life is grounded in objective features of moral action, and it is not simply a figment of our imagination. Andrew Mason and Samuel Schefler have offered two different but related justifications for the normative independence of obligations that flow from special ties, and each of them is useful for getting a better picture of morality that fits with the actual facts of our moral experience. One way to ground special obligations is to argue that they are justified by the good involved in these relationships. Mason offers an illuminating account of special obligation to compatriots by building on Joseph Raz’s view of how friendship justifies special obligations.47 Friendship is a relationship that is not valued instrumentally but rather for its own sake, Raz argues. People in friendships have certain obligations to one another, which are justified by the good of friendship.48 These special obligations are internally related to the good of friendship, because they allow the expression of mutual concern, sympathy, and compassion.49 Obligations of friendship are partially constitutive of the good of friendship, even if they do not represent the whole of the good. They cannot be conceived as means of realizing the good of friendship. Rather, friendship and fulfillment of obligations are in a broad sense equivalent and coextensive. Mason argues that a comparable account could be developed of the value of membership in a nation. Citizens owe to each other special consideration, which is both grounded in and an expression of the unique good that patriotism fosters. Mason’s account of the special ties of friendships and national membership and the responsibilities they entail can be usefully applied to a variety of special ties that color and delineate the moral boundaries of individual persons. We can recognize that, just as friendship does, membership in religious, cultural, or professional groups, and even in a nation, generates special obligations that are 47. Andrew Mason, ‘‘Liberating Duties,’’ Law 48. Mason, ‘‘Special 49. Mason, ‘‘Special ‘‘Special Obligations to Compatriots,’’ Ethics 107 (1997): 427–47; Joseph Raz, and Philosophy 8 (1989): 3–21. Obligations,’’ 439. Obligations,’’ 440. Carmen E. Pavel 505 A U TH O R C O PY internally related to the good that justifies membership in the respective community. The special responsibilities are partially constitutive of the good. The wide-ranging nature of particularistic associative ties suggests a similarly wideranging span of moral obligations, grounded in the ties themselves and not reducible to a higher good or value. The more diverse our reasons for affirming special responsibilities are, the less likely that they will be amenable to integration into a well-ordered hierarchy with permanently established normative priorities. Scheffler also suggests that special responsibilities may have a variety of irreducible grounds.50 Obligations arise for different considerations. Discreet interactions, such as contracts or promises, might explain special obligations, but so too does the nature of the relationship to friends, neighbors, and family.51 Scheffler thus provides a non-reductionist account of special obligations. People see their special relationships as a source of special duties when they have relationships they have reason to value. To attach value to relationships is to see the other people as a source of special responsibilities in virtue of the special ties that connect one to them.52 People can make claims on individuals whether one has special relationships with them or not. A stranger in need that a passer-by can help without undue costs has a claim on the passer-by. However, persons that are part of a special relation can make additional claims, beyond those that people in general can make. This is true, Scheffler says, because individuals have special relationships with others valued for their own sake and not instrumentally: ‘‘insofar as we have good reasons to value our interpersonal relations, we have good reasons to see ourselves as having special responsibilities. And, accordingly, skepticism about such responsibilities will be justified only if we are prepared to deny that we have good reason to value our relationships.’’53 What Mason and Scheffler’s accounts have in common is a commitment to the view that the source of our special obligations lies in the relationships that generate them. The varied nature and justification of the relationships, as well as the variety of goods these relationships embody, indicate a similar variety of justifications for the obligations that are constitutive of those relationships. Moral obligations are irreducibly plural because the nature of the good is plural. 50. Samuel Scheffler, ‘‘Relationships and Responsibilities,’’ Philosophy and Public Affairs 26 (1997): 189–209. 51. Scheffler, ‘‘Relationships and Responsibilities,’’ 189. 52. Scheffler, ‘‘Relationships and Responsibilities,’’ 196. 53. Scheffler, ‘‘Relationships and Responsibilities,’’ 200–201. 506 COSMOPOLITANISM, NATIONALISM The Priority of Cosmopolitanism as a Moral Framework A U TH O R C O PY Recently cosmopolitan theorists have begun to come to terms with special obligations, and have advanced views that incorporate them into a cosmopolitan framework. Like the earlier dismissal of special ties, these attempts are also problematic. Pluralism, if taken seriously, not only explains why the tension between cosmopolitanism and nationalism is real, but also why we cannot hope to provide a unified theoretical account that incorporates both views. To make that point, I will focus on a particular argument advanced by Kok-Chor Tan. Two broad responses to incorporate special obligations into the cosmopolitan moral view have emerged, both to increase the appeal of the cosmopolitan view and to make it correspond more closely to commonsense morality. The first argues that cosmopolitan morality can incorporate partial responsibilities such as patriotic ties for instrumental reasons. Allowing people to pursue special relationships and fulfill the obligations they incur is a good way of accomplishing cosmopolitan ideals, cosmopolitans have argued. This position takes particularist ties to be morally reducible to the ultimate moral ideals of cosmopolitan thought. I will call this the reductionist thesis. The second view recognizes the normative independence of special ties and relationships, and argues that cosmopolitan morality can accommodate them as long as it subordinates their claims to the universal claims of cosmopolitanism.54 Partial commitments and allegiances can only be satisfied once the requirements of cosmopolitan duty are satisfied, as the latter respond to more basic requirements of moral impartiality. This is the conditionality thesis. I believe both these responses fail. The reductionist thesis recognizes special ties and commitments because they provide efficient means of meeting global obligations.55 Martha Nussbaum, in an attempt to moderate her claims on behalf of cosmopolitanism, defends partiality as a ‘‘sensible way to do the good.’’ The supreme good, for her, is allegiance to the worldwide community of human beings, so her accommodation of partiality takes place on the cosmopolitan terrain. When special allegiance violates the more fundamental cosmopolitan concerns, they lose their moral ground.56 The approach has the apparent advantage of solving the opportunity cost problem because it allows for special obligations to be fulfilled instead of obligations to distant others. On the surface, at least, it avoids the problem of conflicting duties that more extreme forms of cosmopolitanism raises. However, this concession to special obligations is not persuasive. As others have shown, attaching independent 54. Kok-Chor Tan, Justice without Borders: Cosmopolitanism, Nationalism and Patriotism (Cambridge: Cambridge University Press, 2004). 55. This is, for instance, Robert Goodin’s view in ‘‘What’s So Special About Our Fellow Countrymen?’’ Ethics 98 (1988): 663–86. 56. Nussbaum, For Love of Country? 136. Carmen E. Pavel 507 A U TH O R C O PY moral value is inherent in any meaningful account of the special relationships from which these attachments spring.57 The special relationships have a value not reducible to the requirements of the cosmopolitan moral view, and as such pose a more serious challenge to this view than the above-mentioned cosmopolitans recognize.58 Kok-Chor Tan rejects the reductionist thesis, too. He does so on the grounds that treating particularistic ties as only instrumental to the realization of cosmopolitan ideals amounts to morally impoverishing the former. So what might a cosmopolitanism that respects special ties as normatively independent look like? He presents a more sophisticated defense of cosmopolitan accommodation of special ties that I call the conditionality thesis. Tan believes we can distinguish between claiming the worth of a relationship is reducible to an impartial principle or good and claiming the legitimacy of that relationship is conditional on its not violating this principle.59 We need not reduce a person’s conception of the good to an impartial principle. All we need to say is that the pursuit of that conception of the good needs to be limited by certain standards of justice. The practice of patriotism, for instance, must be constrained by cosmopolitan duties, even if the worth of patriotism is not reducible or explained by these considerations.60 Impartial principles, Tan claims, make it possible to balance the requirements of cosmopolitan duty of assistance to distant others with special duties. Marrying the universalistic, impartial concerns of cosmopolitanism to the particularistic concerns of everyday life is possible because impartial principles do not demand impartiality across the whole of people’s lives. Following Brian Barry, Tan distinguishes between first-order and second-order impartiality. Cosmopolitan morality does require impartiality at the foundational level but not at the substantive, first-order level of everyday action. The impartiality demanded by cosmopolitan morality is second-order impartiality, that is, the rules and principles of social action need to be impartial with respect to people’s choices and preferences.61 Cosmopolitanism need not demand that all our actions and our everyday lives conform to impartiality. The purpose of cosmopolitan morality is then not to eliminate all forms of particularistic associative obligations but to 57. David Miller thinks that the reductionist view misdescribes and undervalues the nature of particularist commitments, because it denies that they have ‘‘intrinsic ethical relevance.’’ Miller, ‘‘Bounded Citizenship,’’ in Cosmopolitan Citizenship, ed. Kimberly Hutchings and R Dannreuther (New York: St Martin’s Press, 1999), 165; quoted in Tan, Justice without Borders, 148. 58. See also Tan Justice without Borders; Scheffler, ‘‘Relationships and Responsibilities.’’ 59. Tan, Justice without Borders, 150. 60. Tan, Justice without Borders, 150. 61. Brian Barry, Justice as Impartiality (Oxford University Press, 1995); Tan, Justice without Borders, 157. 508 COSMOPOLITANISM, NATIONALISM A U TH O R C O PY determine ‘‘the global context and rules within which such concerns and interests may be legitimately pursued.’’62 The conditionality thesis recognizes that people’s different duties might flow from relationships that have irreducible moral grounds but argues that the shape of those particularistic duties is generated by the constraints provided by impartial cosmopolitan principles. The distinction between first-order and second-order impartiality seems to capture something intuitively true about how we conceive of the general context of rules. General social norms such as traffic rules provide an example of second-order impartiality that provides standards without limiting people’s pursuit of different conceptions of the good. Tan’s view accommodates special obligations by offering a rank ordering of different responsibilities. He preserves the moral priority of cosmopolitan duties, which he views as quite extensive, and argues unconvincingly that they should consistently trump patriotic duties and other duties arising out of special relationships. Tan’s purpose is to show that there is a fundamental coherence between nationalist duties properly understood and cosmopolitan ones if we accept that nationalist duties need to be framed by the impartial requirements of cosmopolitan morality. His ambition is to clean up the moral landscape and reconcile seemingly conflicting moral claims by shaping and fitting them into a more systematic moral view. The conditionality thesis takes into account the pluralistic nature of moral obligations but fails to account for opportunity costs. To better understand why this is so, I will show that Tan fails to distinguish between two related but different ideas. He confuses impartiality, the idea that rules apply equally to all without any regard for their preferences and choices, with neutrality of effect, the idea that certain rules have no impact with respect to people’s conception of the good. In other words he does not distinguish between rules that are both impartial and neutral, such as traffic rules, for instance, and rules that are impartial but nonneutral, such as a requirement that all members of a community participate ingroup prayers five times a day. Traffic rules are neutral because they do not ask individuals to expend their resources and therefore do not raise questions about competing uses of resources, but the group prayer requirement does. Although we would not be mistaken to consider the former group of rules as merely setting standards within which the pursuit of goods can take place, we would be in error to say the same about the second. The latter category of impartial but non-neutral rules cannot be construed as a case of second-order impartiality but needs to be seen as an example of first-order impartiality, impartiality at the substantive level, which affects people’s pursuit of their goals. Tan’s failure to distinguish between 62. Tan, Justice without Borders, 157. Carmen E. Pavel 509 A U TH O R C O PY the two undermines his ability to make a good case for cosmopolitan duty of assistance as being a case of second-order impartiality. The example that Kok-Chor Tan provides is very useful to illustrate this point, because it shows not only that cosmopolitan duties can have the ability to substantially limit the pursuit of particularistic goals, but that at a certain point the cosmopolitan duties can exclude these goals altogether. Let us suppose, with Tan, that the cosmopolitan duty of assistance requires a global resource tax (GRT).63 Each country will pay a tax to a global developmental fund. A country will be able to favor its citizens ‘‘only after doing its share as required by the terms of the GRT,’’ just as in the domestic case, families may favor their children only after the requirements of national taxation are met.64 He argues that once the requirements of global justice are met this way, partial concern for compatriots or members of one’s community is not just expected but unreasonable to deny. But herein lays the problem. If we prioritize duties that way, duties with higher priority are going to get satisfied first and duties with lower priority last or not at all. Imagine a Mormon, living in U.S., who greatly cares about his family and religious community. He contributes a percentage of his income to the GRT, another portion as a national tax, 10 percent to his church, and the rest of his income he can use for himself and his family. However, little or much this person has left to meet his special obligations to his family, it is likely that successive waves of taxation are going to diminish his ability to meet them. Therefore, it is misleading to regard cosmopolitan duties as a framework of rules or standards by which people can satisfy the moral claims arising out of their particularistic ties, when the ‘‘framework’’ is effectively crowding out their ability to meet them. Contrary to Tan’s argument, the cosmopolitan moral requirement does act as a kind of firstorder impartiality, which is a pervasive constraint that handicaps people’s ability to meet their special obligations. Defenders of the conditionality thesis are making a substantial concession to partiality by allowing it to share the moral space previously considered the monopoly of impartialist, universalist principles. Nonetheless, as cosmopolitans recognize that different affiliations may be grounded in reasons that are not commensurable or reducible to one another across the spectrum, they start losing the firm, unitary moral ground that informed much of their thought. As they acknowledge the pluralistic nature of the moral universe, one in which different duties are grounded in reasons or values that are incommensurable, they need to acknowledge that conflict arising between duties with normatively independent grounds cannot be solved by positing rules of priority. Consequently, the account that cosmopolitans provide for dealing with the problem of 63. Tan, Justice without Borders, 158. 64. Tan, Justice without Borders, 159. 510 COSMOPOLITANISM, NATIONALISM A U TH O R C O PY conflicting duties does not seem persuasive. On what grounds do they argue for the priority of cosmopolitan duties? In other words, why do cosmopolitans claim that the moral legitimacy of special obligation is conditional on not violating the impartial requirements of universal principles? Unless they say something about the relative value of the grounds of particularistic moral claims, it is not clear why cosmopolitan morality should determine and limit the structure of particularistic duties and not the other way around. One way to answer this objection is to say that the moral grounds of cosmopolitanism reflect the general demands of morality. Indeed, cosmopolitans might argue that cosmopolitanism, by abstaining from incorporating any contingent facts about individual’s associative ties, represents most ideally the impersonal perspective from which, it is supposed, moral judgments are to be made. This is the path taken by Leif Wenar. He makes concessions to special ties while maintaining the priority of universal duties. Our closer relations to friends, family, and fellow citizens need to be considered within the framework set by the requirements of universal, impartial morality. The requirements of universal morality shape the acceptable forms of special ties. The universal is prior to the particular.65 Wenar recognizes that we tend to act in response to reason and motivational aspirations that draw our attention to more ‘‘local’’ values, goods, and persons. This ‘‘habitual drawing of our attention to nearer horizons,’’ as he calls it, is deeply problematic in his view, because it does not reflect the requirements of the universalist reasons we have to engage with others. We are morally and motivationally deficient for sidestepping the impartial demands required by cosmopolitanism. In this sense, cosmopolitanism is identical with impartial morality, and therefore whatever we defend as part of that morality has in principle priority over other claims. However, this response faces difficulties. Although impartiality may play some role in moral judgment, it cannot, by the cosmopolitan’s own account, exhaust the demands of morality. It is plausible to regard some forms of partiality—loyalty to one’s family or community—as morally admirable, even as morally required. Moral duties arising out of these relationships involve partiality in an irreducible manner. As both partialists and impartialists agree that moral partiality is legitimate and irreducible to any form of impartiality at a more fundamental level, then something needs to be said about the grounds of adjudicating conflicts where conflicts arise. The cosmopolitans who shield their priority claim based on the idea that cosmopolitanism more closely approximates the demands of impartial morality simply beg the question of why impartial considerations should trump partial ones and consequently why duties to assist distant others should take precedence over duties to those close by. 65. Leif Wenar, ‘‘What We Owe to Distant Others?’’ 285. Carmen E. Pavel 511 Conclusion A U TH O R C O PY The tension between cosmopolitanism and nationalism is real, and cannot be dismissed by sophisticated attempts to reduce one kind of responsibility to instrumental considerations or integrate both into one unified account. This way to conceive the relationship between cosmopolitanism and nationalism leads to false reductions and a refusal to recognize parts of what is valuable in the opposing view. Both reach simplicity at the cost of moral clarity. This tension notwithstanding, proponents of both cosmopolitanism and nationalism make important contributions to our understanding of the bounds of justice and the kinds of moral demands that contour our ethical landscape. Identifying and explicating the tension is itself helpful to understand what is at stake in debates between cosmopolitans and nationalists but the analysis offered here has profound implications for global justice for at least three reasons. First, cosmopolitan obligations and patriotic ones clearly involve tradeoffs and opportunity costs. Helping those with whom one stands in a special relationship means that one’s resources, energy, and attention are no longer available for others who are not part of that special relationship. In favoring compatriots, we pass up the opportunity to help distant strangers. The reverse also holds true. Meeting moral responsibilities toward those affected by poverty and disease in other countries means that less is available for projects that involve special relationships. Those who defend a minimalist understanding of cosmopolitan duties as negative duties argue that, conceived this way, cosmopolitanism does not impose opportunity costs. Such a minimalist view would require that we merely refrain from harming distant others, without asking to make positive efforts to improve the plight of the world’s poor.66 The duty not to cause harm applies to everyone and protects the interests of all human beings in the world equally. Abstentions from acts that cause harm typically involve no opportunity costs in terms of helping those close to us, if the former is conceived as a negative duty. And this is precisely how Pogge has recast his cosmopolitan view. But he also makes an important empirical point: international arrangements have already inflicted massive harm on the world poor. This empirical fact invokes the responsibility of redress by cash transfers.67 Responsibilities of redress are positive duties, and do impose opportunity costs on individuals and groups who see themselves as having special obligations. In any event, the representatives of the main strand of cosmopolitanism that I have identified in this paper are all 66. For a cosmopolitan defense of negative duties see Loren Lomasky, ‘‘Liberalism beyond Borders,’’ Social Philosophy and Policy 24 (2007): 206–33. 67. Thomas Pogge, ‘‘‘Assisting’ the Global Poor,’’ in The Ethics of Assistance: Morality and the Distant Needy, ed. Deen K. Chatterjee (Cambridge: Cambridge University Press, 2004), 260–88. 512 COSMOPOLITANISM, NATIONALISM A U TH O R C O PY committed to positive, redistributive duties across borders, even if they differ with Pogge on what justifies those duties. Second, some loss is inevitable and there is no account of justice that will allow us to simultaneously and completely satisfy all our moral responsibilities. Cosmopolitans are often not aware of the opportunity costs cosmopolitan duties impose, and argue as if duties to distant others merely add on to our existing duties, without causing any friction. But cosmopolitan duties impose real opportunity costs. People are constrained in what they can do in the name of distant strangers, even when those strangers face situations of dire need. This is not to say that nationalists can justify exclusive preference for the interests of their compatriots. Nations define one kind of moral community, but it is not the only one. Exclusive preference for compatriots not only obscures the fact that historically some nations have been involved in causing injustice elsewhere and bear responsibility for correcting it, but also the fact that national boundaries are increasingly more porous, and individuals engage in interactions across borders that generate significant responsibilities to which the responsibilities of citizenship have to give way. Third, cosmopolitanism and nationalism are not mutually incompatible ideals, as the existing debates have suggested so far. There is no reason why cosmopolitanism and nationalism cannot embrace in principle the pluralist critique I have advanced here. Any compelling account of global justice will have to be sensitive to the values that underlie each position and accommodate the competing demands advanced on their behalf. Even though the relationship between cosmopolitanism and nationalism defies an easy theoretical solution, this does not mean that accommodation cannot be achieved in practice. But crucially, that accommodation, in addition to recognizing the validity of both types of claims, depends on empirical facts that create the conditions for accommodation. Can patterns of historical relationships explain the condition of deprivation of some people around the world?68 Do international institutions produce patterns of harmful policies that worsen the condition of those who are badly off globally? Are there effective institutional responses available, both regionally and internationally, when countries in the affluent west are committed to helping the rest of the world? Even if the answer to all these empirical questions is clear, people will disagree on the proper way to balance all the moral and empirical dimensions of global justice. That is because there is no overriding perspective from which to rank-order and prioritize these dimensions. We are encumbered then with moral indeterminacy which cannot be simply set aside or solved through some sort of 68. For instance, are some countries responsible, due to their history of colonial domination, for the internal collapse of their former colonies? What kinds of responsibilities do these countries incur? Carmen E. Pavel 513 A U TH O R C O PY priority procedure that promises to clear the moral landscape of conflicting moral demands. The varied nature of the special relationships into which individuals enter and the complex historical web of international relations among countries make it difficult to prescribe ‘‘one size fits all’’ solutions for the dilemmas of multiple moral duties. A more contextual approach is in order, one that gives due to the various special relationships in which people participate, but one that also heeds the claims of outsiders, simply as human beings entitled to equal moral consideration.