Global Justice: An defense of institutional cosmopolitanism

This dissertation argues first that social and economic institutions, rather than one's character or individual transactions, are the proper site of distributive justice-an account called "institutionalism." The dissertation makes a case for thinking that distributive justice is the solution to the conflict in claims to rights, resources, opportunities, and powers based on peoples' pluralistic and incompatible ideas of the good life, including their religious and ethical beliefs. Because justice is the solution to a practical interaction problem, the reasoning for a conception of justice needs to accurately represent the problem. Once our reasoning reflects the problem, it is argued that only institutionalism is justified. The reasoning for anti-institutionalist conceptions, such as utilitarianism and act-egalitarianism, does not accurately represent the interaction problem arising from distinct persons with their own ideas of the good. The dissertation uses the framework supporting institutionalism to argue second that distributive justice is global in scope-a view called "cosmopolitanism." It examines the main cases for distributive justice, compatible with institutionalism, that restrict the scope of justice to the institutions of a state. Conceptions of distributive justice are based on a universal conception of person, which people share across borders. So these cases hold that the social institutions of a state organize economic cooperation to conclude that distributive justice applies only among compatriots. This is because distributive justice organizes a reciprocal division of labor for a socioeconomic product, which only occurs iv with economic cooperation. The dissertation shows that if justice is a response to the conflicting claims we make on resources, then the main cases for distributive justice imply instead that justice is global in scope. Principles of distributive justice determine the functional roles of a socioeconomic system, that is, they determine what the division of labor in an economy is and who can participate. Thus, being members of an existing division of labor, organized by state institutions, cannot restrict the scope of justice. Instead, only the capacity to cooperate or be in a division of labor is morally relevant to the practical interaction problem of social justice. Distributive justice therefore applies to all persons as such.

GLOBAL JUSTICE: A DEFENSE OF INSTITUTIONAL COSMOPOLITANISM by Edward Andrew Greetis A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Philosophy The University of Utah May 2019 Copyright © Edward Andrew Greetis 2019 All Rights Reserved The University of Utah Graduate School STATEMENT OF DISSERTATION APPROVAL The dissertation of Edward Andrew Greetis has been approved by the following supervisory committee members: , Chair Cynthia Stark 02/20/2019 Date Approved , Member Leslie Francis 11/26/2018 Date Approved , Member Erin Beeghly 11/26/2018 Date Approved Chrisoula Andreou , Member 11/26/2018 Date Approved Darrel Moellendorf , Member 02/20/2019 Date Approved and by , Chair/Dean of Matthew Haber the Department/College/School of and by David B. Kieda, Dean of The Graduate School. Philosophy ABSTRACT This dissertation argues first that social and economic institutions, rather than one’s character or individual transactions, are the proper site of distributive justice—an account called “institutionalism.” The dissertation makes a case for thinking that distributive justice is the solution to the conflict in claims to rights, resources, opportunities, and powers based on peoples’ pluralistic and incompatible ideas of the good life, including their religious and ethical beliefs. Because justice is the solution to a practical interaction problem, the reasoning for a conception of justice needs to accurately represent the problem. Once our reasoning reflects the problem, it is argued that only institutionalism is justified. The reasoning for anti-institutionalist conceptions, such as utilitarianism and act-egalitarianism, does not accurately represent the interaction problem arising from distinct persons with their own ideas of the good. The dissertation uses the framework supporting institutionalism to argue second that distributive justice is global in scope—a view called “cosmopolitanism.” It examines the main cases for distributive justice, compatible with institutionalism, that restrict the scope of justice to the institutions of a state. Conceptions of distributive justice are based on a universal conception of person, which people share across borders. So these cases hold that the social institutions of a state organize economic cooperation to conclude that distributive justice applies only among compatriots. This is because distributive justice organizes a reciprocal division of labor for a socioeconomic product, which only occurs with economic cooperation. The dissertation shows that if justice is a response to the conflicting claims we make on resources, then the main cases for distributive justice imply instead that justice is global in scope. Principles of distributive justice determine the functional roles of a socioeconomic system, that is, they determine what the division of labor in an economy is and who can participate. Thus, being members of an existing division of labor, organized by state institutions, cannot restrict the scope of justice. Instead, only the capacity to cooperate or be in a division of labor is morally relevant to the practical interaction problem of social justice. Distributive justice therefore applies to all persons as such. iv TABLE OF CONTENTS ABSTRACT ....................................................................................................................... iii ACKNOWLEDGMENTS ................................................................................................ vii Chapters 1. INTRODUCTION .......................................................................................................... 1 1.1 Part 1: The Site of Distributive Justice ................................................................ 9 1.1.1 Chapter 2: Rescuing Rawls’s Institutionalism and Incentives Inequality ........................................................................................................ 12 1.1.2 Chapter 3: An Act-Egalitarian Worry for Tan’s Institutional Approach 14 1.1.3 Chapter 4: The Separateness of Persons: Defending the Institutional Approach to Justice ......................................................................................... 16 1.2 Part 2: The Scope of Distributive Justice........................................................... 19 1.2.1 Chapter 5: Reciprocity and the Scope of Distributive Justice ............... 21 1.2.2 Chapter 6: Against Practice-Dependence: Social Justice and Constructivism ................................................................................................ 22 1.2.3 Chapter 7: Cosmopolitan Constructivism: A Humanity-Centered Approach to the Economic Division of Labor ................................................ 24 2. RESCUING RAWLS’S INSTITUTIONALISM AND INCENTIVES INEQUALITY .................................................................................................................. 28 2.1 Cohen’s Case against Incentives Inequality and Institutionalism...................... 30 2.2 The Choice/Necessity Claim and the Subjective Circumstances of Justice ...... 34 2.3 Justificatory Community and a Shared Conception of Justice .......................... 37 2.4 Concluding Remarks .......................................................................................... 45 2.5 References……………………………………………………………………...46 3. AN ACT-EGALITARIAN WORRY FOR TAN’S INSTITUTIONAL APPROACH ..................................................................................................................... 48 3.1 The Institutional Approach to Equality.............................................................. 51 3.2 Tan’s Institutional Approach ............................................................................. 55 3.3 A Worry for Tan ................................................................................................ 59 3.4 Toward a Defense of the Institutional Approach ............................................... 67 4. THE SEPARATENESS OF PERSONS: DEFENDING THE INSTITUTIONAL APPROACH TO JUSTICE .............................................................................................. 69 4.1 Constructivism and the Problem of Conflicting Claims .................................... 75 4.2 Rawls, Utilitarianism, and the Separateness of Persons .................................... 83 4.3 Act-Egalitarianism and the Separateness of Persons ......................................... 94 5. RECIPROCITY AND THE SCOPE OF DISTRIBUTIVE JUSTICE ....................... 104 5.1 Part 2: Institutional Constructivism and the Scope of Justice.......................... 104 5.2 Chapter 5: Reciprocity and the Scope of Distributive Justice ......................... 110 5.3 Reciprocity Arguments for Statism ................................................................. 112 5.4 Reciprocity Arguments and Perversity ............................................................ 119 5.5 Against Reciprocity as Triggering Social Justice ............................................ 124 5.6 Conclusion ....................................................................................................... 136 6. AGAINST PRACTICE DEPENDENCE: SOCIAL JUSTICE AND CONSTRUCTIVISM ..................................................................................................... 138 6.1 Practice Dependence ........................................................................................ 142 6.1.1 Reciprocity, Status Quo Bias, and Perniciousness............................... 146 6.2 Practice Dependence and Institutional Social Justice ...................................... 148 6.3 Social Justice is not Practice Dependent .......................................................... 155 7. COSMOPOLITAN CONSTRUCTIVISM: A HUMANITY-CENTERED APPROACH TO THE ECONOMIC DIVISION OF LABOR………………………...160 7.1 Anticosmopolitan Rawlsian Constructivism and the Division of Labor ......... 166 7.1.1 Coercion ............................................................................................... 168 7.1.2 Reciprocal Cooperation ....................................................................... 170 7.1.3 Pervasive Impact .................................................................................. 172 7.2 The Modern Concept of Distributive Justice ................................................... 173 7.3 Organizing Property and Functional Roles, the Fruits of Functional Roles, and Opportunities for Functional Roles ........................................................................ 179 7.3.1 The Systemic Concept of Distributive Justice Versus Anticosmopolitans ........................................................................................ 188 7.4 A Global Difference Principle? ....................................................................... 191 REFERENCES ............................................................................................................... 197 vi ACKNOWLEDGMENTS In writing this dissertation I benefited greatly from discussions with, and comments from, my colleagues and my dissertation committee. I would like to thank the following people for helpful comments and discussion: Chrisoula Andreou, Kyle Barrett, Erin Beeghly, Daniel Callies, Thomas Christiano, Rainer Forst, Leslie Francis, Darrel Moellendorf, and Cynthia Stark. I would also like to thank the audience at the Forschungskolleg Humanwissenschaften and Normative Orders at Goethe University. I am especially grateful to my advisor Cindy Stark for her guidance. She has helped my writing and research more than anyone. I am also grateful to Darrel Moellendorf for his guidance, support, and help with my research over the years. I also thank Dustin Stokes who has helped me greatly with his mentorship. Finally, I would like to thank my parents, Gary and Annette, my siblings, and Courtney for their encouragement and support. CHAPTER 1 INTRODUCTION Many recent social and political movements are focused on reducing economic inequality and poverty. Consider Bernie Sanders’ presidential campaign, Thomas Piketty’s recent best seller, the Occupy Wall Street movement, support in parts of Europe for a guaranteed basic income, and protests against international trade agreements that disadvantage the global poor. Each movement cites inequality and poverty as matters of injustice. Throughout much of human history, economic relations were seen as the natural order ordained by God, poverty was seen as the consequence of the vices and sins of the poor, and some humans were seen as having a higher moral status than others.1 Given these beliefs, help to the poor was more often seen as a matter of charity rather than justice. These beliefs have largely been overturned with the help of social science, literature, and philosophy. For example, we now know that economic systems and poverty are to a large extent amenable to human control, and most people now see each other as moral equals rather than as belonging to morally superior or inferior classes. Because of these developments—especially the fact that ending poverty is no longer seen as impossible—we can assess economic arrangements as one of the ways in which we treat each other. The following example from Rousseau helps to illustrate these points. 1 For an explanation of the historical development of distributive justice, see Samuel Fleischacker, A Short History of Distributive Justice (Cambridge: Harvard University Press, 2004). 2 He says, The first man, who, after enclosing a piece of ground, took it into his head to say, “This is mine,” and found people simple enough to believe him, was the true founder of civil society. How many crimes, how many wars, how many murders, how many misfortunes and horrors, would that man have saved the human species, who pulling up the stakes or filling up the ditches should have cried to his fellows: Be sure not to listen to this imposter; you are lost, if you forget that the fruits of the earth belong equally to us all, and the earth itself to nobody!2 Rousseau’s point is not that we should do away with property and civil society and go back to a state of nature but that the institution of property is a social convention that involves some having power over others, and as such, it is an important way in which we treat each other. Further, because civil society is founded on human convention, the system of property is not fixed; it can be arranged in multifarious ways. The social movements mentioned above are dedicated to assessing economic arrangements as ways of being treated either justly or unjustly, with the aim of making them more just. This way of assessing economic relations is often called “economic justice” or “distributive justice.” Distributive justice is a virtue of economic arrangements. I hold that it is a practical virtue concerned with solving the concrete problem of adjudicating conflicting claims to economic resources, rights, and powers. On my account, conceptions of distributive justice attempt to solve the problem by providing guidelines for a fair or justifiable distribution of income and wealth, the division of economic powers such as the ownership and control over the means of production, positions of economic responsibility, and the opportunities to obtain these powers. One of the practical aims of a conception of justice is to justify and provide 2 The First and Second Discourses, ed. Roger D. Masters, trans. Roger D. Masters and Judith R. Masters (New York: St. Martin’s Press, 1964), 141–42. 3 direction for political and social movements and policies, such as the ones mentioned above.3 Once we agree that economic relations can be just or unjust, we need to decide to whom and to what distributive justice applies. In other words, we must answer two related questions: (1) what is the “scope” of principles of distributive justice? and, (2) what is the “site” of principles of distributive justice?4 By the “scope” of principles of distributive justice, I mean the set of agents to whom a principle applies. “Cosmopolitans”5 hold that egalitarian principles of distributive justice apply to all 3 Theories differ on how they do this. Most theories of distributive justice offer distributive principles to specify an ideally just arrangement. The ideal is then used to evaluate current economic arrangements and to provide a goal to work toward. Some “nonideal theorists” come up with principles to apply directly to nonideal economic arrangements, or they argue that ideal theory principles of justice should apply directly to nonideal circumstances (e.g., Thomas Pogge, Realizing Rawls (Ithaca: Cornell University Press, 1989)). On the ideal/nonideal theory distinction see, for example, John Simmons, “Ideal and Nonideal Theory,” Philosophy & Public Affairs 38, no. 1 (2010): 5–36; Zofia Stemplowska, “What's Ideal About Ideal Theory?,” Social Theory and Practice 34, no. 3 (2008): 319–40. I will not address this debate. I only argue for some idealizations when discussing arguments for principles of justice. This is unavoidable; we must use some abstraction to argue for normative principles. 4 For the definition of the site and scope of justice see Arash Abizadeh, “Cooperation, Pervasive Impact, and Coercion: On the Scope (Not Site) of Distributive Justice,” Philosophy & Public Affairs 35, no. 4 (2007): 323; Kok-Chor Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality (Oxford: Oxford University Press, 2012), 1. 5 Those holding cosmopolitan accounts include Abizadeh, “Cooperation, Pervasive Impact, and Coercion: On the Scope (Not Site) of Distributive Justice”; Brian Barry, “Humanity and Justice in Global Perspective,” Nomos 24 (1982): 219–52; Charles R. Beitz, Political Theory and International Relations (Princeton: Princeton University Press, 1979); Charles R. Beitz, “Cosmopolitan Ideals and National Sentiment,” The Journal of Philosophy 80, no. 10 (1983): 591–600; Gillian Brock, Global Justice: A Cosmopolitan Account (Oxford: Oxford University Press, 2009); Allen Buchanan, “Rawls’s Law of Peoples: Rules for a Vanished Westphalian World,” Ethics 110, no. 4 (2000): 697–721; Simon Caney, Justice Beyond Borders: A Global Political Theory (Oxford: Oxford University Press, 2005); Simon Caney, “Humanity, Associations, and Global Justice: In Defence of Humanity-Centred Cosmopolitan Egalitarianism,” The Monist 94, no. 4 (2011): 506–34; Joshua Cohen and Charles Sabel, “Extra Rempublicam Nulla Justitia?,” Philosophy & Public Affairs 34, no. 2 (2006): 147–75; Robert E. Goodin, “What Is So Special about Our Fellow Countrymen?,” Ethics 98, no. 4 (1988): 663–86; A. J. Julius, “Nagel’s Atlas,” Philosophy & Public Affairs 34, no. 2 (2006): 176–92; Darrel Moellendorf, Cosmopolitan Justice (Boulder, CO: Westview Press, 2002); Darrel Moellendorf, Global Inequality Matters (New York: Palgrave Macmillan, 2009); Martha Nussbaum, Frontiers of Justice: Disability, Nationality, Species Membership (Cambridge: Harvard University Press, 2006); Onora O’Neill, Bounds of Justice (Cambridge: Cambridge University Press, 2000); Miriam Ronzoni, “The Global Order: A Case of Background Injustice? A Practice-Dependent Account,” Philosophy & Public Affairs 37, no. 3 (2009): 229–56; Philippe Van Parijs, “International Distributive Justice,” International Economic Law 1 (2011): 62–79; Pogge, Realizing Rawls; Samuel Scheffler, Boundaries and Allegiances: Problems of Justice and Responsibility in Liberal Thought (Oxford: Oxford University Press, 2001); Samuel Scheffler, “Cosmopolitanism, Justice & 4 persons as such while “statists”6 hold that principles of distributive justice apply only among fellow compatriots.7 By the “site” of principles of distributive justice I mean to which entities principles of distributive justice apply. The “institutional approach”8 holds Institutions,” Daedalus 137, no. 3 (2008): 68–77; Amartya Sen, The Idea of Justice (Cambridge: Harvard University Press, 2009); Henry Shue, Basic Rights: Subsistence, Affluence, and U.S. Foreign Policy (Princeton: Princeton University Press, 1980); Kok-Chor Tan, Justice Without Borders: Cosmopolitanism, Nationalism, and Patriotism (New York: Cambridge University Press, 2004); Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality; Laura Valentini, Justice in a Globalized World: A Normative Framework, 2011; Lea Ypi, Global Justice and Avant-Garde Political Agency (Oxford: Oxford University Press, 2012). 6 Those holding statist accounts include Michael Blake, “Distributive Justice, State Coercion, and Autonomy,” Philosophy & Public Affairs 30, no. 3 (2001): 257–96; Samuel Freeman, “The Social and Institutional Bases of Distributive Justice,” in Cosmopolitanism Versus Non-Cosmopolitanism Critiques, Defenses, Reconceptualizations, ed. Gillian. Brock (Oxford: Oxford University Press, 2013); Samuel Freeman, “The Law of Peoples, Social Cooperation, Human Rights, and Distributive Justice,” Social Philosophy and Policy 23, no. 1 (2006): 29–68; Aaron James, Fairness in Practice: A Social Contract for a Global Economy (New York: Oxford University Press, 2012); Saladin Meckled-Garcia, “On the Very Idea of Cosmopolitan Justice: Constructivism and International Agency,” Journal of Political Philosophy 16, no. 3 (2008): 245–71; David Miller, National Responsibility and Global Justice (Oxford: Oxford University Press, 2012); Richard Miller, “Cosmopolitan Respect and Patriotic Concern,” Philosophy & Public Affairs 27, no. 3 (1998): 202–24; Richard Miller, “Rawls and Global Justice: A Dispute over Kantian Legacy,” The Philosophical Forum 43, no. 3 (2012): 297–309; Thomas Nagel, “The Problem of Global Justice,” Philosophy & Public Affairs Philosophy Public Affairs 33, no. 2 (2005): 113–47; John Rawls, The Law of Peoples, 1999; Mathias Risse, On Global Justice (Princeton: Princeton University Press, 2012); Andrea Sangiovanni, “Global Justice, Reciprocity, and the State,” Philosophy & Public Affairs 35, no. 1 (2007): 3–39; Michael Walzer, Spheres of Justice: A Defense of Pluralism and Equality (New York: Basic Books, 1983). I group the authors into “cosmopolitan” and “statist” for simplicity. However, there are varying degrees of cosmopolitan or statist views. For example, James argues for more robust principles of global justice than Nagel and other statists but less robust principles than the cosmopolitans. Cosmopolitans also hold different views about why principles of justice apply beyond state borders. Moellendorf (2009), for example, holds that we owe duties of distributive justice to others when we share a certain kind of association. 7 The debate is over “egalitarian” principles or principles of distributive justice, that is, principles of relative deprivation. All those involved in the debate agree that we at least owe some minimum humanitarian duties sufficient for the basic needs of the global poor. 8 Elizabeth Anderson, “Rawls’s Difference Principle,” in The Cambridge Companion to A Theory of Justice, ed. Matthew Clayton and Andrew Williams, Forthcoming; Freeman, “The Law of Peoples, Social Cooperation, Human Rights, and Distributive Justice”; Samuel Freeman, Rawls (New York: Routledge, 2007); Freeman, “The Social and Institutional Bases of Distributive Justice”; Louis-Philippe Hodgson, “Why the Basic Structure?,” Canadian Journal of Philosophy 42, no. 3–4 (2013): 303–34; A. J. Julius, “Basic Structure and the Value of Equality,” Philosophy & Public Affairs 31, no. 4 (2003): 321–55; Moellendorf, Cosmopolitan Justice; Thomas Nagel, Equality and Partiality (New York: Oxford University Press, 1991); Pogge, Realizing Rawls; John Rawls, A Theory of Justice (Cambridge: Harvard University Press, 1971); John Rawls, Political Liberalism Expanded Edition (New York: Columbia University Press, 1993); Miriam Ronzoni, “What Makes a Basic Structure Just?,” Res Publica: A Journal of Moral, Legal and Social Philosophy 14, no. 3 (2008): 203–18; Samuel Scheffler, “Is the Basic Structure Basic?,” in The Egalitarian Conscience: Essays in Honor of G. A. Cohen, Christine Sypnowich (Oxford: Oxford University Press, 2006): 102–29; Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality; Andrew Williams, “Incentives, Inequality, and Publicity,” Philosophy & Public Affairs 27, no. 3 (1998): 5 that basic social institutions are the site of distributive justice. That is, it holds that certain distributive principles apply to institutions that do not also apply to individuals’ choices within the institutional framework. The account is based on the idea that certain regulative principles depend on the nature of the thing they regulate. For example, Rawls claims, “the correct regulative principle for anything depends on the nature of the thing.”9 Many of those who support the institutional approach are also statists because they see domestic institutions, such as the apparatus of the state, as the proper site of justice. Those opposed to the institutional approach10 hold that all distributive principles apply both to institutions and individual economic transactions. For example, utilitarians hold that the principle of utility should regulate all of agency. As Samuel Scheffler says, If one knows what principle a consequentialist uses to evaluate individual actions, then one also knows what principle he uses to evaluate social policy. This is not true with a non-consequentialist, for although the principles of personal and institutional morality are certainly closely related on any plausible non-consequentialist view, they need not be identical.11 Those who reject the institutional approach also often hold that applying these principles to individual economic transactions is as important as applying the principles to institutions.12 225–47; Andrew Williams, “Justice, Incentives and Constructivism,” Ratio 21, no. 4 (2008): 476–93. I get the name “institutional approach” from Tan (2012). 9 A Theory of Justice, 29. 10 Including utilitarians and Brian Berkey, “Against Rawlsian Institutionalism about Justice,” Social Theory and Practice 42, no. 4 (2016): 706–32; Brian Berkey, “Double Counting, Moral Rigorism, and Cohen's Critique of Rawls: A Response to Alan Thomas,” Mind 124, no. 495 (2015): 849–74; Paula Casal, “Occupational Choice and the Egalitarian Ethos,” Economics and Philosophy 29, no. 01 (2013): 3–20; G. A. Cohen, Rescuing Justice and Equality (Cambridge: Harvard University Press, 2008); Liam Murphy, “Institutions and the Demands of Justice,” Philosophy & Public Affairs 27 (1999): 251–91; Serena Olsaretti, “The Inseparability of the Personal and the Political: Review of G.A. Cohen’s Rescuing Justice and Equality,” Analysis 72, no. 1 (2012): 145–56. 11 The Rejection of Consequentialism (Oxford: Oxford University Press.), 33. 12 For instance, Cohen, Rescuing Justice and Equality. 6 Adequately answering the site and scope questions is of great practical importance because each answer leads to vastly different conceptions of distributive justice and assessments of economic schemes. For example, if statism is the most reasonable answer to the scope question, then we owe (merely) humanitarian duties to the poor in foreign countries (or other duties of structural equity between trading countries). We do not owe noncompatriots egalitarian duties of justice. Or—to take another example—if the institutional approach to the site question is unreasonable, then we must guide our occupational choices, purchases, and other daily economic transactions according to the same principles of justice we use to organize our institutions. So, for example, if just institutions are those that are arranged so that they minimize economic inequality, persons are required by justice to act in ways that minimize economic inequality. It follows that whether or not the social and political movements I alluded to above are justified in their aims depends on answers to the site and scope questions. My dissertation defends a kind of cosmopolitan institutional approach, which I call “cosmopolitan constructivism.” I think that once the institutional approach is given an adequate defense as the solution to the site problem, we have strong reason to accept cosmopolitanism as the most reasonable solution to the scope problem. In Part 1, I argue for the institutional approach to distributive justice. I begin my case, in Chapter 2, by defending the institutional approach against G. A. Cohen’s charge that it is internally inconsistent. I defend the institutional approach in general and Rawls’s institutional application of his “difference principle” (explained below) in particular. In the next two chapters, I determine whether intuitionalism is the most reasonable approach to distributive justice. My defense of the internal consistency of the approach points to 7 “constructivism” (explained below) and “value pluralism” as potential grounds for accepting the approach, so I explore these ideas. In Chapter 3, I examine what I take to be the strongest case for intuitionalism to date, given by Kok-Chor Tan. He holds that the institutional approach is needed to resolve a tension between egalitarianism and value pluralism—the idea that in liberal societies there will be a plurality of ideas of the good, including moral and religious beliefs. I show that Tan’s argument is still susceptible to an anti-institutionalist argument, so the case for institutionalism is incomplete. In Chapter 4, I argue that institutionalism has a rational basis in an interpretation of constructivism. Constructivism is the idea that conceptions of justice and morality are solutions to practical interaction problems. I hold that distributive justice is the solution to the problem of conflicting claims among people who cannot be expected to share ideas of the good life. That is, the problem is framed by assuming a roughly Kantian conception of the person as a moral agent. This conception of the person leads to value pluralism since agents will have their own ideas of the good. I argue that the institutional approach is the only solution to the problem that is justifiable to moral agents who hold a plurality of values. Thus, we should accept institutionalism as the most reasonable approach to distributive justice if we hold that justice protects individuals or is fairness between persons—for example, we might agree that “Each person possesses an inviolability founded on justice.”13 However, I only claim that the ground for institutionalism is intuitive; I do not offer a robust defense of my interpretation of constructivism. I call my institutionalist account “Rawlsian constructivism” since it is based on an interpretation of Rawls. I end Part 1 by explaining that if we accept 13 Rawls, A Theory of Justice, 3. 8 Rawlsian constructivism, then we also have grounds to reject the anti-institutionalist reply to Tan that I outline in Chapter 3. Thus, my argument also lends support to Tan’s account. I conclude that there are two reasonable cases for institutionalism, mine and Tan’s, if we accept Rawlsian constructivism. In Part 2, I aim to work out the implications of Rawlsian constructivism with respect to the scope of distributive justice. Rawlsian constructivism holds that distributive justice is a response to a problem among moral persons who make conflicting claims to rights and resources. This problem is global in scope, so the account seems to support cosmopolitanism. However, Rawls and most Rawlsians argue that distributive justice is restricted to the state. In Part 2, I show that once we accept basic institutions as the proper site of justice, as proposed by Rawlsian constructivism, we are not in fact led to statism as many Rawlsians have thought. Instead, I defend the claim that Rawlsian constructivism supports cosmopolitanism. In Chapter 5, I argue that the main Rawlsian reciprocity arguments for state-based distributive justice instead imply that the distributive principles in question apply to all persons who are capable of reciprocal cooperation. Thus, reciprocity arguments support my claim that Rawlsian constructivism supports cosmopolitanism. In Chapter 6, I show that the adoption of a new methodology, “practice-dependence constructivism,” will not save the restriction of reciprocity arguments for distributive justice to the state. Finally, in Chapter 7, I argue that most Rawlsian constructivist accounts share a key premise, which I call the “division of labor” thesis. The thesis says that distribute justice only applies to existing divisions of labor for a social product because distributive justice is meant to organize cooperation for production and divide the fruits of production. I show that only the capacity to be a 9 member of a division of labor is relevant for the scope of distributive justice. Thus, according to these Rawlsian arguments, distributive justice applies to all persons with this capacity, or distributive justice is “humanity-centered,” which implies that justice is cosmopolitan. Finally, I discuss whether Rawls’s difference principle in particular should apply globally. I offer some reason to think that it should be the principle of reciprocity that governs global economic interaction. My approach—determining the answer to the scope problem by defending a certain answer to the site problem—is not new. In the current literature, the scope debate often employs Rawls’s institutional approach to the site problem (explained below) to either answer or shed light on the scope problem. However, much of the scope debate has taken the site answer for granted without adequately justifying it. My contributions to the debates are 1) to propose a novel defense of the institutionalist answer to the site problem, which I call “Rawlsian constructivism,” and 2) to show that if we accept this answer, and the reasoning behind it, it should influence our answer to the scope problem, giving us reason to be cosmopolitans. As such, my dissertation proposes a novel justification for cosmopolitanism. 1.1 Part 1: The Site of Distributive Justice To better understand the site debate, and make sure it is not confused with another debate, I will briefly contrast it with the public/private feminist critique of liberalism. The latter is the criticism that liberalism’s “state-society” distinction sections off part of society, the family, as “private” and thus insulates it from political and social critique. It further associates women with domestic life in the “private” sphere. As Carole Pateman 10 claims, liberalism’s state-society distinction is a distinction “within the world of men”— men’s public roles and men’s roles in civil society—since women are assumed to be naturally suited to the domestic sphere.14 This assumption devalues the work often done by women in civil society, and it devalues the work done in the “domestic sphere.” Will Kymlicka responds that the distinction between the “public” world of men and the “private” world of women far predates liberalism and is not an essential part of the theory. I think liberalism’s distinction between state-society is best explained by Rawls’s idea of “background justice.” Background justice is the idea that the institutional background institutions, against which our individual interactions take place, are just if they establish and maintain fair background conditions such that our interactions are free and fair over time.15 Within civil society then, individuals and associations are free to pursue their own ends knowing that justice over time is being maintained. According to Kymlicka, “feminists and liberals share a basic commitment to viewing public power as a means for the protection of particular interests, needs, and social relationships.”16 The family is a social institution, part of civil society, which means that it should be made free and fair by the state, or background justice. For example, there must be laws making sure one partner is not wholly financially reliant on the other such that the marriage is not truly voluntary. The family is also not a truly voluntary association for children since they are born into one. Hence, there must be laws necessary for democratic participation, among other things, which require children to be educated, nourished, and so on. As a 14 Cited in Will Kymlicka, Contemporary Political Philosophy: An Introduction (Oxford: Oxford University Press, 2002), 390. 15 Rawls, Political Liberalism Expanded Edition, 266. Background justice can have social aims in mind, such as minimizing gender inequality and all class inequalities. See Ronzoni, “What Makes a Basic Structure Just?”; Anderson, “Rawls’s Difference Principle.” 16 Contemporary Political Philosophy: An Introduction, 392. 11 social institution, rights also cannot be abrogated within the family—for example, spousal rape is a crime. Liberals and the feminists in question seem to agree on these points. There is still a debate, however, between some feminists and liberals on the place of the family and care work. (This issue is important for any account of distributive justice, so I discuss paid and unpaid care work, especially in Chapter 7.) I only briefly canvas the public/private debate here to show that it should not be confused with the site debate. The site debate is instead about whether morality is itself “fragmented,” in Thomas Nagel’s words, or whether the same principles are suitable for all aspects of life.17 In other words, the site debate is whether the idea of “background justice” is reasonable or whether the distinction collapses. An act-utilitarian, for example, will make no such distinction. Each action, whether it is by state institutions or an individual economic transaction, should adhere to the principle of utility. All action should seek to maximize aggregate satisfied interests. Institutionalism, on the other hand, holds that background institutions are organized by one set of standards. These standards maintain background justice, which makes people’s actions free and fair within the institutional structure. This does not mean, however, that justice and morality are finished once background justice is established. Other standards will apply within the background conditions. Thus, arguments for institutionalism, including mine, do not imply that any area of society is “sectioned off” from scrutiny. Instead, they imply that distinct standards of scrutiny are required. As I explain below, my support for institutionalism relies on Rawls’s claim that 17 Mortal Questions (Cambridge: Cambridge University Press, 1979), chaps. 8–9. 12 principles of justice depend on what they regulate. I hold that background justice is a response to a practical interaction problem. Society is made up of moral agents, and moral agents cannot universally agree on values, including morality and religion. According to Rawlsian constructivism, social justice is the fair response to this problem. Other standards apply to distinct problems. This does not mean, however, that institutionalists think that the family is an entirely distinct problem from background justice. For instance, background conditions are enforceable and must be acceptable to those in a democratic society. Because of this, Paula Casal argues that institutionalists should promote a gender-egalitarian ethos today—so long as it does not restrict liberty— because background conditions are not free and fair for the family.18 This means that familial relations are still scrutinized from the standpoint of background justice, and action may be taken within the family in hope of moving society closer to free and fair background conditions. 1.1.1 Chapter 2: Rescuing Rawls’s Institutionalism and Incentives Inequality G. A. Cohen’s criticisms of the institutional approach are both challenging and influential. He famously criticizes Rawls’s restriction of his difference principle (explained below) to institutions as internally inconsistent. In Chapter 2, I defend the institutional approach against the claim that it is an internally inconsistent view. My arguments have two aims: (1) to defend the institutional approach in general, and (2), to defend the institutional site of the difference principle in particular. If my defense is “Marx, Rawls, Cohen, and Feminism,” Hypatia 30, no. 4 (2015): 825. cf. Ronzoni, “What Makes a Basic Structure Just?” 18 13 successful, it means that at least one theory of distributive justice can consistently apply distributive principles to social institutions without also applying them to choices within the institutional framework. This is important since I will provide a more general argument for the institutional approach to the site question in Chapter 4. My arguments also defend the site restriction of the difference principle to institutions in particular. This is because I will use Rawls’s case for the difference principle to make a comparative argument supporting his institutional approach over the reasoning for anti-institutionalist accounts in Chapter 4. I will also defend the global scope of the difference principle in Chapter 7. So my focus on the site restriction of the difference principle is not simply because Cohen focuses on it but because I accept the arguments for it and aim to defend them and extend their scope.19 Cohen (2008) claims to find a deep conflict within Rawls’s theory of justice. On the one hand, Rawls endorses incentives inequality—more pay for certain professions is just when that pay benefits everyone. On the other hand, Rawls says that citizens in a just society accept the principles of justice that govern them. So a society is just by Rawls’s lights only if its citizens accept his difference principle as just. The difference principle holds that inequalities are just only if they are necessary for maximizing the wealth of the least well off. According to Cohen, citizens cannot both accept the difference principle and demand incentives inequality. This is because incentives inequality is not necessary for maximizing the wealth of the least well off. Consequently, when citizens demand incentives pay, they are acting contrary to the difference principle. It is also important for me to offer a different defense of the institutional approach’s internal consistency than the common Rawlsian “basic structure response” since I will explain in Chapters 2 and 3 that this is not an adequate response to Cohen and other anti-institutionalists. 19 14 Demanding incentives is therefore incompatible with accepting the difference principle as just. To act justly, the citizens must guide their economic transactions by the difference principle, including foregoing economic incentives, rather than leaving the difference principle to exclusively structure social institutions.20 Thus, Cohen concludes, Rawls’s institutional approach to the site of the difference principle fails. I defend Rawls’s site restriction of the difference principle to basic institutions. I show that Cohen’s “internal criticism” of Rawls requires the rejection of two parts of his account of a just society. It is therefore not an internal criticism of the theory. Citizens can consistently accept the difference principle as just while not using it to constrain their daily economic choices. I then explain how my response to Cohen shows that the Cohen/Rawls dispute comes down to their more fundamental dispute over constructivism. The parts of Rawls’s theory that Cohen’s arguments assume are false are important for Rawls’s constructivism (because of this, I return to constructivism in Chapter 4). I conclude that at least one fundamental moral principle meant to apply to institutions, the difference principle, need not apply also to choices. 1.1.2 Chapter 3: An Act-Egalitarian Worry for Tan’s Institutional Approach Tan provides what I take to be the strongest argument for the institutional approach to date. He builds on Rawls’s motivation for the institutional approach. Rawls held that principles of justice organize the overall institutional structure. Within the institutional rules, people are permitted to freely and equally pursue their diverse and conflicting ideas of the good while justice is maintained by their shared institutions. That 20 For this part of the argument to be persuasive, Cohen needs to respond to the “basic structure response.” 15 is, people have space to pursue their idea of the good consistent with justice, rather than having to structure their actions by justice. Tan adds to Rawls’s argument that social justice must make space for diverse ideas of the good because of value pluralism. People have many differing nonegalitarian ideas of the good. So there is an apparent tension between egalitarian justice and personal pursuits. Justice cannot require egalitarianism across the board because of pluralism; but justice has priority and requires equality. For Tan, institutionalism is the way to solve this apparent tension. In the name of justice, our institutions must maintain conditions of equality; in the name of pluralism, we are freely and equally permitted to pursue our idea of the good within just background rules. According to Tan, institutionalism is the only way to draw the bounds of justice and personal pursuits while maintaining the priority of justice over personal pursuits. In other words, justice needs to clearly set ample space aside for personal pursuits, but it still must have priority over the good in that it determines what personal pursuits are permissible. Tan’s argument is forceful since even critics of institutionalism are usually not willing to deny pluralism or that justice has priority over the good.21 At the very least, his argument shows the rationality of the approach—it comes from taking pluralism and the priority of justice seriously. In Chapter 3, I argue that a division of labor to make space for value pluralism is not a good reason to accept the institutional approach; the institutional approach can be demanding. Tan’s idea that justice must frame ideas of the good is more reasonable. For example, Cohen and Berkey seem to accept pluralism, see Berkey, “Against Rawlsian Institutionalism about Justice,” e.g., 713; Cohen, Rescuing Justice and Equality, 394–98. Also see footnote 24 below. Critics of the institutional approach could reject pluralism or the priority of justice. I think this strategy is unpromising since there are independently strong reasons to accept these premises. Thus, my act-egalitarian worry will accept Tan’s premises. 21 16 However, I pose an “act-egalitarian” worry for Tan’s account. Act-egalitarians, including Cohen, hold that the egalitarian evaluative standards that we apply to institutions should also apply to daily interactions and even to one’s character.22 For example, Cohen advocates an egalitarian ethos where people constrain their actions by the difference principle and use it to shape their character.23 Advocates of the approach attempt to provide people space to pursue their ideas of the good by accepting “personal prerogatives,” which allow for justifiable actions that do not promote equality.24 I argue that act-egalitarianism can respond to Tan’s apparent tension while both drawing the bounds of justice and personal pursuits and maintaining the priority of justice. Actegalitarianism can do this by holding that the personal prerogative defines a list of types of permissible actions as part of the egalitarian principle advocated. I outline an act-egalitarian worry for institutionalism in this chapter to show that the case for the approach is incomplete. I largely agree with Tan’s and Rawls’s case for institutionalism, but I think a stronger case comes from focusing on the constructivist justification of principles of justice, which I aim to do in the next chapter. 1.1.3 Chapter 4: The Separateness of Persons: Defending the Institutional Approach to Justice In Chapter 4, I offer a novel argument for the institutional approach, which is not susceptible to the act-egalitarian response to Tan outlined in the previous chapter. Like 22 For example, Cohen, Rescuing Justice and Equality, 3. He denies to have a fully worked out account, however, and instead takes himself to simply be questioning the actual reasons for institutionalism (e.g., 398.). But to do this, he argues that act-egalitarianism is a plausible alternative. 23 However, he ends up rejecting the difference principle in Cohen, Rescuing Justice and Equality, chap. 4. 24 The fact that act-egalitarians accept a personal prerogative also indicates that they are not willing to deny pluralism. In other words, they agree that egalitarianism should not permeate all aspects of life since there are other important and diverse values. They seem to accept Tan’s apparent tension and attempt to use a prerogative to address it. 17 most philosophers who accept the institutional approach, I accept it because of value pluralism. To explain why pluralism supports the institutional approach, I provide a framework for thinking about the character of social justice, which is a sort of constructivism. Christine Korsgaard holds that “constructivism” is the view that ethics and justice attempt to solve practical problems.25 She explains that Rawls’s concept/conception distinction in A Theory of Justice illustrates how constructivism with respect to justice works. She says, [T]he concept of justice refers to the solution to a problem. The problem is what we might call the distribution problem: people join together in a cooperative scheme because it will be better for all of them, but they must decide how its benefits and burdens are to be distributed. A conception of justice is a principle that is proposed as a solution to the distribution problem, arrived at by reflecting on the nature of the problem itself.26 So principles of justice are “constructed” by human reason for human practical interaction problems. I hold that social justice is a practical virtue meant to address a practical interaction problem, which I call “the problem of conflicting claims.” Briefly, the problem is that our incompatible ideas of the good lead to conflicting claims to economic resources, powers, and opportunities. The problem is framed by assuming a conception of the person as a moral agent. This conception of the person leads to value pluralism and conflicting claims based on our incompatible values. I argue that because we should think of conceptions of justice as responses to the problem of conflicting claims, on my account, the reasoning for conceptions of justice must accurately reflect the problem. Christine Korsgaard, “Realism and Constructivism in Twentieth-Century Moral Philosophy,” in The Constitution of Agency: Essays on Practical Reason and Moral Psychology (Oxford: Oxford University Press, 2010), 302–28. 26 Christine Korsgaard, The Constitution of Agency: Essays on Practical Reason and Moral Psychology (Oxford: Oxford University Press, 2010), 322. 25 18 I offer a comparative argument among extant theories. The idea is to use constructivism to expose the assumptions and reasoning that underlie institutional and anti-institutional conceptions of justice. In other words, “The procedure of contract theories [or constructivism] provides … a general analytic method for the comparative study of conceptions of justice.”27 For this comparative study, I take Rawls’s theory as the main representative of institutionalism and utilitarianism and act-egalitarianism as the main representatives of anti-institutionalism. I show that Rawls’s reasoning accurately reflects the problem of conflicting claims, whereas, utilitarian and act-egalitarian reasoning idealizes away aspects of moral agency used to frame the problem of conflicting claims. I argue for my claim by interpreting Rawls’s charge that utilitarianism does not take seriously the distinction between persons. I interpret this claim as saying that utilitarianism does not accurately reflect the problem of conflicting claims. I show that act-egalitarianism similarly does not reflect the problem. Instead, utilitarianism and act-egalitarianism are derived by reflecting distinct problems and conceptions of the person. I then show that Rawls’s reasoning—particularly his condition that the parties in his contract argument are mutually disinterested—accurately reflects the problem of conflicting claims. Thus, only the reasoning for institutionalism respects persons as separate moral agents with their own ideas of the good. If we think distributive justice is a response to conflicting claims, then the case for Rawls’s institutionalism is more reasonable than anti-institutionalist accounts. Although my arguments compare extant theories, I take them to show that institutionalism has a rational basis in constructivism, where the problem is framed on the assumption that 27 Rawls, A Theory of Justice, 121–22. 19 persons are moral agents. I believe that there are good reasons to hold that distributive justice is the fair response to conflicting claims, and I provide some reason to accept the constructivist basis. However, my main aim is to provide a rational basis for institutionalism and work out its implications in Part 2. That is, my claim is that if we think that distributive justice is a response to the problem of conflicting claims, then we should accept the institutional approach. Finally, I explain how my case offers a response to the act-egalitarian worry in the previous chapter and thus supports Tan’s account. Act-egalitarianism does not represent value pluralism with its reasoning, so it does not offer a reasonable solution to the tension Tan outlines between egalitarianism and pluralism. Neither Tan’s nor my argument offers “knockdown” cases for institutionalism, but together they provide a strong case for the view. Further, each argument relies on a certain conception of persons as having their own ends that they value. This provides a way to contrast the commitments that lead to institutionalism and the commitments that lead to utilitarianism and act-egalitarianism. In Part 2, I aim to work out the implications of Rawlsian constructivism, so my arguments for cosmopolitanism will rely on these institutionalist commitments regarding justice and persons. 1.2 Part 2: The Scope of Distributive Justice In Part 1, I argue that Rawlsian constructivism, with its conception of moral persons, supports the idea that distributive justice only applies to basic socioeconomic institutions. In Part 2 of the dissertation, my aim is to work out the implications of Rawlsian constructivism with respect to the scope debate. My case for the institutional 20 approach is based on the idea that distributive justice is a response to the problem of conflicting claims. This problem is global in scope. However, many Rawlsians think there are reasons to restrict the problem of distributive justice to the state. In Part 2, I examine the main broadly Rawlsian constructivist accounts for why distributive justice applies only to state institutions. I conclude that these accounts of distributive justice actually support cosmopolitanism. My aim in Part 2 is to show that Rawlsian constructivism supports the idea that distributive justice is “humanity-centered.” The main arguments for state-based distributive justice use a universal conception of moral personality. They then claim that distributive justice is restricted to state institutions because there is a special relationship, such as reciprocal cooperation, among moral persons that only occurs among compatriots. I claim that these arguments can only reasonably hold that one must have the capacity to be in these relationships for distributive justice to apply. I reject what is known as “membership-dependence,” with respect to Rawlsian constructivism, which is the view that duties of distributive justice depend on being a member of an association. I conclude that the main Rawlsian arguments for distributive justice apply to all persons; Rawlsian distributive justice is “humanity-centered.” This is a common argument-form for supporting cosmopolitanism.28 What is novel about my approach is that my case relies upon my constructivist case for institutionalism with respect to the site of justice. I end Part 2 by providing a limited case for holding that the difference principle should apply globally. 28 For example, see Caney, Justice Beyond Borders: A Global Political Theory. 21 1.2.1 Chapter 5: Reciprocity and the Scope of Distributive Justice In Chapter 5, I examine reciprocity arguments for maintaining that distributive justice only applies among compatriots. Reciprocity arguments claim that a certain form of reciprocity is the distinct moral ideal that should regulate social cooperation among members of a state. Proponents of the argument hold that nondenizens of a state do not cooperate within a state’s basic structure institutions and so have no distributive claim to our cooperative surplus based on reciprocity. In other words, distributive justice is a type of reciprocal cooperation, so nondenizens, who do not do their fair share in cooperation, are not owed any of the cooperative surplus. International institutions constitute a different form of cooperation and reciprocity that requires less than distributive equality. For example, international cooperation might require a form of reciprocity with respect to supplementing national income via trade. I show that the use of moral relationships, such as reciprocity, to define the scope of justice conflicts with how justice is social. I claim that the account unjustifiably uses ideas that operate within society to apply to society as a whole and the idea of social justice. It also attributes to people a fixed nature theoretically prior to socialization. These charges are roughly Hegel’s case against the contract doctrines of his day, such as Locke’s. The charges are problematic for proponents of reciprocity accounts since they are forms of Rawlsian constructivism. They aim to show that social justice, with its particular principles of distribution, only apply to state social and political institutions. The upshot of my argument is that reciprocity arguments do not offer a reason to accept statism. Indeed, I show that their arguments can only reasonably require the capacity to reciprocally cooperate as restricting the scope. Persons share this capacity 22 across borders, so these Rawlsian arguments imply that distributive justice is global in scope. In the next chapter, I determine whether practice dependence constructivism can help reciprocity accounts restrict distribute justice to the state. 1.2.2 Chapter 6: Against Practice-Dependence: Social Justice and Constructivism Practice-dependence constructivist accounts of justice began with an influential interpretation of Rawls by Aaron James.29 James explains that the practice-dependence account that comes from his interpretation of Rawls is a kind of “constructivist methodology.” Principles of justice are “constructed” by human reason for human practical interaction problems. What makes the practice-dependence account a distinctive form of constructivism is its restriction regarding the construction, or justification, of principles and their scope. Proponents claim that a social interpretation of the goals or function of a practice must be used to justify and determine the kind of principles of regulation for the practice. The goals or function of the practice also must determine the scope of the principles by determining the scope of the practice. Thus, the scope of distributive justice, on this account, is determined by determining the scope of the practice it regulates given a social interpretation of the practice’s function. The human interaction problem or practice that justice responds to on this version of constructivism is delimited by a social interpretation of the moralized notions that define the practice’s goal or function—for example, the practice of social cooperation has the moralized goal of reciprocal benefit. Statists use this method to argue that distributive Aaron James, “Constructing Justice for Existing Practice: Rawls and the Status Quo,” Philosophy & Public Affairs, 2005, 69–104. 29 23 justice principles are the proper response to the practice of social cooperation within a state given the function of political/social cooperation. International cooperation is a distinct practice that requires less than distributive equality—for example, it might require sufficiency or principles of procedural fairness in trade.30 I closely examine the practice-dependence part of this form of constructivism, which determines the scope of a practice, and I argue that it should be rejected. This is because the moralized social interpretation of a practice conflicts with how justice is social. I claim that the account unjustifiably applies ideas that operate within society to society as a whole and the idea of social justice. The charges are problematic for proponents of practice-dependence and reciprocity accounts since they aim to show that social justice, with its principles of distribution, only applies to state social and political institutions. That is, they accept the institutional approach and aim to provide a Rawlsian constructivist account of social justice. I conclude that practice dependence constructivism should be rejected. My arguments support the idea that constructivism should be conceived in Korsgaard’s sense as a response to a practical interaction problem without considering the social goals of practices. Thus, my conclusion of the previous chapter—that only the capacity to cooperate is relevant for reciprocity accounts of distributive justice—still stands. Both claims lend support to my main Part 2 argument that Rawlsian distributive justice is humanity-centered. 30 For example, see James, Fairness in Practice: A Social Contract for a Global Economy; Sangiovanni, “Global Justice, Reciprocity, and the State.” 24 1.2.3 Chapter 7: Cosmopolitan Constructivism: A Humanity-Centered Approach to the Economic Division of Labor In the final chapter, I complete my case for the claim that Rawlsian constructivism supports the global scope of distributive justice. I examine a premise common to most Rawlsian constructivist accounts of distributive justice. I explain that the main anticosmopolitan Rawslian argument is a “basic structure argument” with respect to the scope of distributive justice. There are several accounts of what constitutes a basic structure, including the institutions that organize reciprocal cooperation, coercive institutions, and institutions that have a pervasive effect on life-chances. Anticosmopolitan Rawlsians argue that there are only state basic structures, and therefore distributive justice only applies among compatriots. I explain that there are three main challenges to the Rawlsian basic structure argument. First, Rawlsians need to show that the basic structure relation is not global in scope; that is, they must show that there is no global basic structure. Second, Rawlsians need to explain why the basic structure relation, such as coercion or reciprocity, makes distributive justice (rather than allocative, retributive, etc.) the relevant regulative concept. Anticosmopolitans admit that the basic structure relationship occurs to some extent across the globe. Since these relations exist across borders, Rawlsians must explain what type of basic structure relation makes distributive justice relevant only among compatriots. I explain that Rawlsian constructivists support what I will call the “division of labor thesis” to answer the first two challenges. Briefly, the division of labor thesis holds that distributive justice only becomes relevant where members of a division of labor jointly produce some socioeconomic product. The thesis allows Rawlsians to claim that the basic structure relation, such as coercive institutions, only requires distributive justice as its correct 25 regulative principle if the set of institutions organizes a division of labor among cooperators. Distributive justice is the relevant concept because it is meant to fairly organize economic production. However, and this is the third challenge, Rawlsians need to offer a reasonable concept of distributive justice. They seem to accept what Elizabeth Anderson calls the “modern systemic concept of distributive justice”—the idea that justice organizes entire economic systems with respect to their distributive consequences or the relationships they maintain.31 I argue that anticosmopolitan Rawlsians cannot hold both that an existing division of labor among members for a social product occasions distributive justice as its proper regulative principle and the modern systemic concept of distributive justice. If existing divisions of labor make distributive justice relevant as the proper regulative concept for the practice, then distributive justice cannot assess the nature and scope of divisions of labor. For anticosmopolitan Rawlsians, the nature and scope of a division of labor is socially determined. Distributive justice is for the existing social practice. However, I show that it is part of the concept of systemic distributive justice to fairly determine the roles and opportunities for a division of labor. These roles and opportunities determine who participates in the division of labor; that is, they determine de facto membership of the socioeconomic system distributive justice organizes. The systemic concept assesses the nature and scope of divisions of labor. Thus, Rawlsians cannot hold that the systemic concept is the relevant regulative principle for existing divisions of labor. I assume that Rawlsians should accept the systemic concept of distributive justice; otherwise, they do not meet the third challenge. If they accept the systemic concept, they can only hold that 31 Anderson, “Rawls’s Difference Principle.” 26 the capacity to be a member of a division of labor is relevant for the scope of distributive justice. I therefore conclude my Part 2 argument: the main anticosmopolitan Rawslian constructivist arguments for distributive justice support the idea that the principles of distributive justice are global in scope. I end the dissertation by asking whether the difference principle should apply globally. My arguments throughout the dissertation are mostly general. However, Chapter 2 defends the site of Rawls’s difference principle, and Chapter 4 offers a comparative case for Rawls’s reasoning for his egalitarian principles over antiinstitutionalist accounts of justice. Since I argue that Rawlsian constructivism is humanity-centered by arguing that the case for state-based reciprocity arguments applies to all people, the question arises whether the difference principle, which is a principle of reciprocity, specifically applies globally. I offer some reasons to support a global difference principle, but I note that the claim faces many more challenges that cannot be addressed in the dissertation. CHAPTER 2 RESCUING RAWLS’S INSTITUTIONISM AND INCENTIVES INEQUALITY Republished with permission of Springer Science and Business Media, from “Rescuing Rawls’s Institutionalism and Incentives Inequality,” E. A. Greetis, forthcoming, 2019; permission conveyed through Copyright Clearance Center, Inc. 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 CHAPTER 3 AN ACT-EGALITARIAN WORRY FOR TAN’S INSTITUTIONAL APPROACH Rawls famously held that the primary subject of social justice is the basic structure of society—the main social and political institutions of society. In other words, the basic structure requires distinct normative principles of regulation that do not also apply to individual actions or one’s character.32 This is one of his most influential and widely accepted ideas. It is also novel in the history of philosophy. It represents a large departure from theorizing about justice, especially from the long dominant utilitarian tradition, which holds that one impartially derived principle ought to organize all of life.33 Rawls’s view on social justice became known as the “institutional approach” and is now the standard approach to the site of social justice.34 However, Brian Berkey, Simon Caney, Paula Casal, G. A. Cohen, Liam Murphy, Serena Olsaretti, and those in the utilitarian tradition have questioned the motive behind applying special principles of social justice exclusively to social institutions. 32 Rawls JAF, pp. 7, 10, 162–68; PL pp. 261 and 283; TJ pp. 54-55. For explanations of the novelty of Rawls’s approach, see Fleischacker, A Short History of Distributive Justice; Anderson, “Rawls’s Difference Principle”; Elizabeth Anderson, “Thomas Paine’s ‘Agrarian Justice’ and the Origins of Social Insurance,” in Ten Neglected Classics of Philosophy, ed. Eric Schliesser (Oxford University Press, 2017) ; Scheffler, “Cosmopolitanism, Justice & Institutions,” 72; and cf. Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, 82–83. 34 For example, see Murphy, “Institutions and the Demands of Justice,” 252, and Berkey, “Against Rawlsian Institutionalism about Justice,” 706–7. 33 49 Kok-Chor Tan provides what I take to be the strongest argument for the institutional approach to date. He builds on Rawls’s motivation for the institutional approach. Rawls held that principles of justice organize the overall institutional structure. Within the institutional rules, people are permitted to freely and equally pursue their diverse and conflicting ideas of the good while justice is maintained by their shared institutions. That is, people have space to pursue their idea of the good consistent with justice, rather than having to structure their actions by justice. Tan adds to Rawls’s argument that social justice must make space for diverse ideas of the good because of value pluralism. People have many differing nonegalitarian ideas of the good. So there is an apparent tension between egalitarian justice and personal pursuits. Justice cannot require egalitarianism across the board because of pluralism, but justice also requires equality. For Tan, institutionalism is the way to solve this apparent tension. In the name of justice, our institutions must maintain conditions of equality; in the name of pluralism, we are freely and equally permitted to pursue our idea of the good within just background rules. According to Tan, institutionalism is the only way to draw the bounds of justice and personal pursuits while maintaining the priority of justice over personal pursuits. In other words, justice needs to clearly set ample space aside for personal pursuits, but it still must have priority over the good in that it determines what personal pursuits are permissible. Tan’s argument is forceful since even critics of institutionalism are usually not willing to deny pluralism or that justice has priority over the good.35 At the very 35 For example, Cohen and Berkey seem to accept pluralism, see Cohen RJE 394-398, and Berkey, “Against Rawlsian Institutionalism about Justice,” e.g., 713. Also see footnote 38 below. Critics of the institutional approach could reject pluralism or the priority of justice. I think this strategy is unpromising since there are independently strong reasons to accept these premises. Thus, my act-egalitarian worry will accept Tan’s premises. 50 least, his argument shows the rationality of the approach—it comes from taking pluralism and the priority of justice seriously. In this chapter, I pose an “act-egalitarian” worry for Tan’s account. Actegalitarians, including Cohen and Murphy, hold that the egalitarian evaluative standards that we apply to institutions should also apply to daily interactions and even to one’s character.36 For example, Cohen advocates an egalitarian ethos where people constrain their actions by the difference principle and use it to shape their character.37 Advocates of the approach attempt to provide people space to pursue their ideas of the good by accepting a “personal prerogative,” which allows for justifiable actions that do not promote equality.38 I argue that act-egalitarianism can respond to Tan’s apparent tension while both drawing the bounds of justice and personal pursuits and maintaining the priority of justice. Act-egalitarianism can do this by holding that the personal prerogative defines a list of types of permissible actions as part of the egalitarian principle advocated. However, I largely agree with Tan’s and Rawls’s case for institutionalism. I outline an act-egalitarian worry for institutionalism in this chapter to show that the case for the approach is incomplete. In Chapter 4, I propose a novel argument for the institutional approach, which is not susceptible to the act-egalitarian response to Tan. My account comes in part from responding to the act-egalitarian reply that I outline in this chapter. I show that the problem with act-egalitarianism is the justification for the view. 36 For example, Cohen, Rescuing Justice and Equality, 3. He denies to have a fully worked out account, however, and instead takes himself to simply be questioning the actual reasons for institutionalism (e.g., RJE, 398). But to do this, he argues that act-egalitarianism is a plausible alternative. 37 However, he ends up rejecting the difference principle in Cohen, chap. 4. 38 The fact that act-egalitarians accept a personal prerogative also indicates that they are not willing to deny pluralism. In other words, they agree that egalitarianism should not permeate all aspects of life since there are other important and diverse values. They seem to accept Tan’s apparent tension and attempt to use a prerogative to address it. 51 I then compare the anti-institutionalist reasoning with the reasoning for an institutional approach. I argue that Rawls’s institutional approach is more reasonable than the natural justification for act-egalitarianism. My account also shows why the act-egalitarian reply to Tan, outlined in this chapter, should be rejected. It is therefore also a defense of Tan’s case that the institutional approach is the only reasonable way to reconcile egalitarianism and personal pursuits. Thus, my argument shows that there are two strong cases for institutionalism, mine and Tan’s, justified from different starting points. 3.1 The Institutional Approach to Equality As we saw above, the institutional approach holds that social institutions require certain distributive principles that do not also apply to individuals’ choices within the institutional framework. This does not mean that ethical principles, or principles of justice of another kind, do not apply to actions or one’s character. To the contrary, the institutional approach requires certain ethical principles for individuals. For instance, citizens have a duty of justice to establish, support, and maintain just social institutions, according to Rawls among others.39 Instead, the distinguishing feature of the institutional approach is that it holds that basic social institutions require distinct evaluative principles of justice. As Rawls claims, “the correct regulative principle for anything depends on the nature of the thing” (TJ, 29). Rawls stipulates that the concept of social justice is the proper balance between competing claims and a conception of social justice is the set of principles that identify 39 Rawls TJ, p. 115, and see pp. 333-350 for individual duties. Also see Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, 28, and Aaron James, “Power in Social Organization as the Subject of Justice,” Pacific Philosophical Quarterly 86, no. 1 (2005): 33. 52 the proper balance.40 Rawlsian institutionalism, again, as we saw above, says that the “primary subject” of social justice is the basic structure. The basic structure, for Rawls, is most important for social justice, and whether the basic structure is just determines whether a society is just. Three of Rawls’s reasons for why the basic structure is the primary subject of justice are the following. First, it largely shapes the interests and aspirations of people.41 Second, its effects on inequalities among persons’ rights and entitlements are profound and present from the start of life.42 Finally, when principles of social justice organize the institutional structure, ideas of the good are “framed” by just institutional rules and people are then left to effectively pursue their idea of the good consistent with justice. Rawls held that justice has primacy over ideas of the good. This means that justice determines what ideas of the good are worthy of pursuit. He also held that justice is primary in that it is the first virtue of institutions—it is primary over other institutional virtues, such as efficiency. By holding that justice organizes the basic structure—and is the first virtue of it—Rawls’s theory makes justice primary in that justice determines what interactions and ideas of the good are permissible. This is because the just basic structure rules determine what interactions are permitted, and then citizens are free to pursue their own ends within these rules. In other words, a just basic structure constrains what actions and ways of life are permissible; it “maintains background justice,” while citizens are left free to effectively and equally pursue their ends.43 Background justice is the set of social conditions necessary for individual 40 A Theory of Justice, 3–10. For example, TJ pp. 259-260. 42 For example, TJ p. 7. 43 For example, PL pp. 265-269, and TJ pp. 548, 563, and 31-32. Scheffler explains these three reasons in “Is the Basic Structure Basic?” For more on these on reasons for the institutional approach, see Elizabeth Anderson, “Rawls’s Difference Principle,” and Ronzoni, “What Makes a Basic Structure Just?” 41 53 transactions to be free and fair. We need to know whether the background conditions of society, the basic structure institutions, are just to know whether an individual action, or association among people, is really free or fair. If the background social conditions allow for huge accumulations of wealth to be passed down each generation, for instance, individual transactions will not be free or fair between the rich and poor. Caney (2005), Cohen (2008), and Murphy (1999), among others, have posed many questions and problems for the institutional approach beyond the charge of internal inconsistency, which I responded to in the previous chapter. An adequately justified and motivated account will give reasonable answers to the following two questions taken from their criticisms. First, how do we distinguish between the basic institutions that are claimed to be the site of justice and the choices made within that institutional framework? Cohen, for example, thinks that if Rawls’s institutional approach distinguishes institutions from choices by holding that only institutions have profound effects on life prospects, then his site distinction “collapses” since daily economic choices can also have profound effects on life prospects. The problem is that if the site of distributive justice is not clearly demarcated, then the institutional approach does not have a clear subject matter.44 The second question is, why do basic institutions require distinct principles of distributive justice that do not also apply to individual conduct within those institutions? In other words, what justifies the institutional approach’s special and exclusive focus on 44 Rawls (JAF, p. 12) purposely starts with a rough delineation of the basic structure. He claims that if it is too specific we may prejudge what future social conditions call for and thus his theory will not be able to adapt to other social circumstances in the future. Instead, he holds that the theory should establish a “framework of thought within which these questions can be approached” (Ibid.). However, an account should at least be able to give guidelines for rough boundaries, at least enough to give guidance for applying the theory’s principles. 54 basic institutions? For example, Cohen claims that if Rawls holds that basic social institutions are the site of justice because they are coercive and so need justification, then Rawls gives a “purely arbitrary delineation of his subject matter” since Cohen argues that there are strong independent reasons for regarding noncoercive relationships like the family as part of the subject of justice (2008, 137).45 And Caney (2005, 111; 2011) argues that institutions are not morally relevant from the point of view of distributive justice.46 The second question is the most important for establishing the motivation and justification of the institutional approach because it is the distinguishing feature of the approach. Accordingly, my focus in the next two chapters is on the second question.47 What critics object to is the exclusive focus on institutions. It is clear that political philosophy ought to be focused on our main complex social institutions. Critics maintain, however, that we ought to also apply the evaluative egalitarian principles to actions, and perhaps even one’s character, as well as institutions. In other words, critics agree with the necessary conditions for focusing on institutions for social justice but disagree that principles for institutions are sufficient for a just egalitarian society. As Tan explains, Rawls’s reasons go some way toward justifying the institutional approach but do not completely support principles for institutions as sufficient for social See also Susan Okin (1989, 100). Even though Cohen cites Okin for support, Okin’s point is different. Okin argues that Rawls’s account of moral development in a just society requires families to be just. So, social justice requires family justice. She also argues later, in Chapter 6, that there is no strict dichotomy between the personal and the political. However, it is unclear whether this conflicts with institutions having a certain kind of priority. For example, see Rawls’s response (JAF, 166-167). 46 Murphy questions the motivation and justification of the institutional approach. He asks, “If people have a duty to promote just institutions [on Rawls’s institutional approach], why do they lack a duty to promote whatever it is that just institutions are for [their aim given by a principle of justice]?” Murphy, “Institutions and the Demands of Justice,” 280. 47 I will assume that Tan’s response to the first question is adequate. He holds that the basic structure is made up of the main social, economic, and political institutions that “are subject to public–political regulation.” Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, 34–38. Also, see on this question Julius, “Basic Structure and the Value of Equality”; Hodgson, “Why the Basic Structure?”; Ronzoni, “What Makes a Basic Structure Just?”; Scheffler, “Is the Basic Structure Basic?” 45 55 justice. Rawls’s first two reasons only show that focus on the basic structure is necessary for social justice, which critics of the institutional approach agree with.48 His third reason, Tan claims, gets us closer to the idea that focus on institutions is sufficient for social justice because it says that exclusive focus on the basic structure is warranted since it frames a space for citizens’ nonegalitarian personal pursuits. That is, principles of social justice organize the basic structure rules. The just basic structure rules frame what forms of life, including many nonegalitarian pursuits, can be pursued consistent with justice and they determine what interactions are free and fair. However, Tan claims that we still need to know why a theory of social justice must preserve space for nonegalitarian pursuits that are consistent with justice. 3.2 Tan’s Institutional Approach Tan aims to fill the gap in Rawls’s argument for institutionalism by explaining why a conception of justice needs to free up space for nonegalitarian personal pursuits. Tan begins his argument by explaining the fact of value pluralism. He notes that the problem of justice arises because people have legitimate and competing interests that often conflict.49 The point of life for most people is not justice and need not be justified by a conception of justice. Nor do people need to justify their way of life to others as long as it is consistent with justice (except perhaps on grounds of rationality). As Rawls says, people “put forward competing claims, and while they are willing to act justly, they are not prepared to abandon their interests” (TJ, 281). This means that justice can only Berkey, “Against Rawlsian Institutionalism about Justice,” 712 and 731; Casal, “Marx, Rawls, Cohen, and Feminism,” 819; Murphy, “Institutions and the Demands of Justice,” 262–64. 49 Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, 26. Tan cites Rawls, A Theory of Justice, 189. 48 56 “draw the limit” of permissible ways of life; it cannot show us the point of life.50 Tan concludes that a conception of justice cannot require egalitarianism in all aspects of life because of the many conflicting nonegalitarian ideas of the good. For example, familial special concern or devoting one’s life to serving one’s God are not derivable from some universal master principle of ethics. In other words, all of people’s values are not reducible to one value such as egalitarianism—without resulting in a kind of despotism or rejecting pluralism. According to Tan, value pluralism leads to an apparent conflict between egalitarian justice and nonegalitarian personal pursuits or ideas of the good. In the name of egalitarian justice, we ought to strive toward equality, but in the name of personal pursuits, we ought to live a life we deem valuable which need not be egalitarian.51 To resolve the apparent tension, Tan says an egalitarian theory “(i) has to be able to provide a means of approximating the bounds of justice and personal pursuits, and do that in a way that, (ii) accords justice primacy over personal pursuits.”52 According to Tan, the institutional approach solves the apparent tension while meeting his two conditions. The approach holds that egalitarian justice applies only to the institutional structure. Institutional rules act to frame ideas of the good. So when egalitarian justice organizes the institutions, just institutional rules define the bounds within which people can pursue their ideas of the good compatible with justice. Justice is primary since it determines what ideas of the good are compatible with justice and so are permitted. The principles of justice do this by organizing the institutional structure, which determines permissible Rawls PL, 174. Also see “The Priority of Right and Ideas of the Good” in John Rawls, Collected Papers, ed. Samuel Freeman (Cambridge: Harvard University Press, 1999), 449. 51 Cf. Rawls JAF, 22. See Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, 28. 52 Tan, 30. 50 57 interactions and forms of life. Institutionalism approximates the bounds of justice and personal pursuits because citizens are permitted to pursue their idea of the good within the institutional rules. Egalitarianism is relegated to institutions, which constrain publicly enforceable actions,53 leaving clear and ample space for people to freely and equally pursue what the point of life is for them. And according to Tan, justice is meant to establish fair background conditions which draw the limit of action to define free and fair interaction among persons. Thus, institutionalism solves the apparent tension while meeting Tan’s two conditions for a reasonable conception of egalitarian justice.54 The institutional structure for Tan and Rawls is analogous to the rules of a game.55 Players are mostly permitted to do as they please within the rules of the game—as long as they are supporting the rules of the game and the rules have cultivated certain virtues that act to preserve the game. They know that the rules, if followed, will accomplish the Tan responds to Cohen’s argument that the basic structure is “fatally ambiguous” in part because “publicly enforceable system of rules” leaves out the family. Tan argues that the family is a complex institution. We think that some parts of it are publicly enforceable while enforcing rules for other parts would violate freedom of association. Tan, 37–38. Similarly, Paula Casal argues that Rawlsians ought to reject some egalitarian ethoses because they are incompatible with certain liberties, such as freedom of association. She claims that when the family, however, cannot be considered a truly voluntary association, rejecting a gender egalitarian ethos is less plausible. See Casal, “Marx, Rawls, Cohen, and Feminism,” 825. A Rawlsian institutionalist can offer the following reply. If institutionalism can reasonably demarcate the basic structure, which will include parts of the family as a free association in ideal theory, then one can still hold that social justice principles only apply to the basic structure. The institutionalist might then agree with Casal that in nonideal theory one should promote a gender egalitarian ethos. However, the institutionalist does this as a way of making the family a truly voluntary association, which would make the basic structure more just. The institutionalist could also hold that the gender egalitarian ethos is justified as a separate principle of justice for individuals—it is not simply applying the difference principle or fair equality of opportunity to one’s actions. See, for example, James, “Power in Social Organization as the Subject of Justice,” 30–33. Furthermore, the prevalence of a misogynistic ethos could also be seen as evidence of unjust institutions for the institutionalist. For example, Rawls, TJ p. 32, thinks that institutions must not encourage attitudes that are contrary to his two principles of justice. A misogynistic ethos is contrary to equality of opportunity, so an institutionalist can argue that institutions are neither just nor stable that promote a misogynistic ethos. 54 It should be noted that Tan’s account is for ideal theory. It outlines a just basic structure. For an explanation of what his account requires in nonideal theory, see Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, 79–81. 55 Anderson puts the point as justifying the rules of a practice versus justifying actions within the practice. “Rawls’s Difference Principle.” 53 58 goals of the game, including fair competition. They may wish to promote many different virtues and styles of play within the rules, but the way of playing that they promote must be within the rules of the game—the rules “draw the limit.” Similarly, citizens know that under just institutional rules, they can freely and equally pursue their idea of the good knowing that background justice is being maintained by their shared institutions. They know that their transactions with other citizens, and the entitlements that may result from the transactions, are free and fair because the just institutional rules maintain background justice. Again, this still requires certain principles and virtues for individuals who will help to uphold just institutions, but the principles of social justice for the basic structure are exclusively suited to the basic structure. The point of political virtues and principles for individuals is different; the principles that apply to individuals require them act to uphold a just basic structure or establish one where one is not yet in place,56 whereas, the point of social justice, applied to the basic structure, is to fairly adjudicate between the conflicting claims of pluralist peoples. Tan considers the possible objection that institutionalism is not the only way to solve the tension while meeting his two conditions. A critic might hold that actegalitarianism can use a “personal prerogative” to do this. Recall that act-egalitarianism with a personal prerogative is the view that institutions and people must apply egalitarian justice to their actions as well as their institutions. However, some actions count as justifiable departures from promoting equality—they are justified personal prerogatives. In this way, act-egalitarians accept the primacy of justice—the departures from egalitarian actions must be justified, that is, consistent with justice—and they accept 56 Cf. James, “Power in Social Organization as the Subject of Justice,” 33. 59 pluralism by leaving space for personal pursuits consistent with justice. Tan claims, however, that by proposing a personal prerogative, act-egalitarians just restate the apparent tension that the institutional approach is meant to address: we need to properly balance our commitment to both egalitarian and personal aims. Tan agrees that egalitarianism needs something like a personal prerogative. In other words, it needs to define the bounds of justice and make room for legitimate personal pursuits. He thinks that the institutional approach provides the legitimate bounds of personal prerogative. Just institutions frame a space within which people can freely and equally pursue their idea of the good. Act-egalitarianism does not have an account of how justice can frame the permissible space of ideas of the good while maintaining the primacy of justice.57 Stating that there should be such space is, as Tan notes, simply a restatement of the apparent tension between egalitarianism and personal pursuits. 3.3 A Worry for Tan I agree with Tan that justice must allow for adequate space for ideas of the good that people deem worthy of allegiance. I also agree that justice must be primary in that it determines worthy ideas of the good and that a conception of justice must clearly “frame” the bounds of personal pursuits consistent with the demands of justice. However, we can question whether the special and exclusive focus on institutions for egalitarian justice is what clearly provides the space for legitimate nonegalitarian ideas of the good. There is 57 It should be noted that the personal prerogative for act-egalitarianism must be part of the conception of justice. That is, the personal space framed by justice must not be a compromise of justice or something that makes society less just. This would be to pit justice against reasonable pluralism. Cf. Norman Daniels, “Democratic Equality: Rawls’s Complex Egalitarianism,” in The Cambridge Companion to Rawls, by Samuel Freeman (Cambridge: Cambridge University Press, 2006), 268. As I will argue in the next chapter, social justice is meant to be a virtuous response to the problem that arises because of pluralism; it should not stand against it. 60 reason to think that other parts of a conception of justice, namely, the content of the egalitarian principles, can do much of this work. The priority of justice means in part that there can be no independent criterion of what ideas of the good are permissible or how much space for ideas of the good there ought to be. This is because justice determines what ideas of the good are permissible.58 So an argument that a conception of justice is too constraining or demanding, in that it does not leave enough space for the good, is an objection to the conception itself.59 It is an objection that the conception does not match our intuitions. Thus, to be reasonable, Tan’s claim must be that egalitarian institutional conceptions of justice are the only way to clearly demarcate justice and personal pursuits and provide “enough” room for personal pursuits to match our intuitions. However, different institutional egalitarian approaches will provide drastically different space for nonegalitarian ideas of the good. Consider applying Cohen’s “strict difference principle”—the interpretation of the difference principle that only allows strictly necessary inequality—to the basic structure. If this is done, the basic structure rules would disallow incentives inequality.60 This is an institutional approach that intuitively does not provide “enough space” for personal pursuits. If incentive pay is not permitted, many opportunities for worthy personal pursuits are lost. For instance, suppose that a conceptual artist will only work as a commercial artist for significantly higher pay because she prefers conceptual art.61 She prefers the higher pay, however, See Rawls “Priority of Right and Ideas of the Good” in PL, especially §§5 and 6, and TJ pp. 548, 563, and 31-32. Also see Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, 29–30. 59 Cf. Rawls PL, 198. 60 Scheffler, “Is the Basic Structure Basic?” pp. 115-117 61 The example is modified from Williams, “Incentives, Inequality, and Publicity,” 236–37. Murphy also gives examples of how institutions can drastically restrict personal space. For example, he claims that 58 61 because she can use it to support faith-based causes that are important to her and to support the best education in art school possible for her children. If incentive pay is disallowed, then she will work as a conceptual artist. She does not have the opportunity to pursue her religious and family related personal pursuits as she wants.62 It is also possible that act-egalitarianism can have a very permissive personal prerogative that leaves “sufficient” space for nonegalitarian pursuits. Samuel Scheffler proposes an agent-centered personal prerogative that gives greater weight to one’s own interests than an exclusive appeal to impersonal calculus would allow.63 For example, a principle of justice might require people to promote the well-being of others while giving greater weight to the worst off.64 The personal prerogative would allow the agent to give her or his interests some proportionately greater weight than the interests of others. That is, It would…allow the agent to promote the non-optimal outcome of his choosing, provided only that the degree of its inferiority to each of the superior outcomes he could instead promote in no case exceeded, by more than the specified proportion, the degree of sacrifice necessary for him to promote the superior outcome.65 On this hybrid consequentialist act-egalitarianism, personal pursuits are those that are permitted when one determines the expected aggregate well-being caused by a possible action—giving greater weight to the worst off—and determines that promoting one’s positive law can be enacted that requires citizens to aim at justice. Murphy, “Institutions and the Demands of Justice,” 262. 62 For a discussion of freedom of occupation and the egalitarian ethos, see Casal, “Occupational Choice and the Egalitarian Ethos.” 63 Scheffler, The Rejection of Consequentialism. 64 The principle comes from Derek Parfit, Equality or Priority? (Lawrence: University of Kansas, 1995). Murphy holds that the principle is egalitarian. See Murphy, “Institutions and the Demands of Justice,” 263. 65 Scheffler, The Rejection of Consequentialism, 20. This a particular form of hybrid theory. When Scheffler defends the rationality of a personal prerogative, he does not commit to a particular hybrid theory Scheffler, 71. 62 own interests is permissible. We determine that promoting our own interests is permissible when the degree of sacrifice required to support the superior outcome does not exceed, by some specified proportion, the well-being promoted by pursuing our own interests. In short, if the action that best promotes one’s own interests wins out in the calculus, then one’s personal pursuit is permitted by justice. If enough weight is given to personal pursuits, then this conception of justice will provide ample space for personal pursuits. Consequently, the content of the egalitarian principles plays a large role in determining how much space is left for personal pursuits.66 Thus, making space for personal pursuits is not a good reason to accept institutionalism. Tan emphasizes the need to make space for personal pursuits as the strongest reason supporting the institutional approach. Although he is right that a theory of justice needs to make space for personal pursuits, I think the real force of Tan’s argument is his explanation of why justice needs to “frame” ideas of the good (in a way that provides enough space for worthy ideas of the good).67 He wants to resolve the apparent tension between egalitarianism and personal pursuits while still maintaining the primacy of justice. This means that a conception of justice must not only make space for personal 66 For similar points about the demands of justice for institutionalism and act-egalitarianism, see Berkey, “Against Rawlsian Institutionalism about Justice,” 726–32. And Murphy, “Institutions and the Demands of Justice,” 262–64, 288–91. 67 Here I disagree with Berkey, “Against Rawlsian Institutionalism about Justice,” 714–15. He argues that Nagel’s, Scheffler’s, and Tan’s point about pluralism must be about the demands of justice. However, “framing” can be quite demanding and so can a duty of justice to establish just institutions. Also see Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, sec. 2.6. on the demandingness of the institutional approach. I showed above that act-egalitarianism and institutionalism can be more or less demanding. So, instead, I interpret Tan’s point to be about framing pluralist ideas of the good while maintaining the primacy of justice. If justice is not primary, then the concern with pluralism is about the demandingness of justice. But if justice must be primary among pluralist peoples to adjudicate their claims, then the main point of Tan’s argument is not about the demands of justice. It is instead about how justice can determine what actions and ideas of the good are worthy of pursuit among pluralist peoples. That is, egalitarian justice and personal pursuits conflict, so a conception of justice must make space for personal pursuits while still making egalitarian justice determine, in a clear way, what personal pursuits are worthy of pursuit. Cf. Rawls’s discussion in “Priority of Right and Ideas of the Good” in PL, and TJ, 31-32. 63 pursuits to ease the demands of justice. Instead, it must also make the bounds of personal pursuits locatable68—it must draw the limit of permissible action—and also define what permissible action is. This is part of the primacy of justice—justice determines what forms of life are worthy of pursuit. But justice cannot determine what forms of life are worthy too narrowly or use one form of life to define the bounds of worthy personal pursuits. This is because the aim of justice is fair adjudication between conflicting claims among pluralist peoples. With the hybrid consequentialist view with a personal prerogative given above, it is unclear how justice frames personal pursuits—that is, it is unclear how justice constrains them and defines what ideas of the good are consistent with justice. It does not clearly determine what forms of life are compatible with justice. Any personal interest gets weight in the utility calculation as part of one’s good and is potentially consistent with justice. Rather than putting initial bounds on what kind of pursuits are worthy and what kind of people we ought to be, the theory must use current forms of life and current desires to determine what actions are just. In other words, on this account we can refer to the good without any reference to the right or what is just (Rawls TJ, 25). It therefore does not make justice prior to the good in the sense I have described above.69 Consider how an act-egalitarian account might not only provide “enough” space for personal pursuits but also can clearly specify the bounds of egalitarian justice and nonegalitarian pursuits while maintaining the primacy of justice. One way an actegalitarian might do this is by holding that personal prerogatives are justified departures 68 Nagel, Equality and Partiality, 18. Cf. Rawls TJ, 32. Consequentialists reject the priority of justice, so they can respond to Tan’s tension in the above way. However, as I said above, I think there are independently strong reasons to accept Tan’s two conditions for egalitarian justice, so I will not avail myself of the consequentialist response. 69 64 from strict equality. This is similar to how the difference principle specifies justifiable departures from equality. The view I have in mind attempts to resolve the tension Tan describes by having the personal prerogative as part of the content of a distributive principle. I argued above that the content of principles of justice in part determines the space for personal pursuits, so the content of a distributive principle might be able to address Tan’s apparent tension. The act-egalitarian could argue that a certain list of personal prerogatives is justified as part of the egalitarian principle being defended. Just people and just institutions will be constrained by a principle of equality except for a list of types of actions permitted by justice since they are justified personal prerogatives. If this actegalitarian account could be worked out, it could demarcate personal space framed by justice by offering a list of permissible types of action. This seems to be what Cohen has in mind when he grants types of actions that can permissibly result in wealth inequality and rejects others, such as the incentive effect of inequality.70 The personal prerogative space is determined or “framed” by a principle of distributive justice. This means that justice has priority over ideas of the good and determines which ideas of the good can exist consistent with justice. Unlike Scheffler’s personal prerogative examined above, the list of personal prerogatives as justifiable departures from strict equality offers a way of clearly defining justice and personal space in a way that makes justice primary. The egalitarian principle places initial bounds on what pursuits are worthy. The principle is also being complied with when someone is following a personal pursuit, so justice 70 Cohen concedes that certain actions can result in justified inequality in wealth. In response to objections by Joshua Cohen, David Estlund, Andrew Williams, and others, he simply allows that some actions result in justifiable inequality of wealth because they do not constitute an inequality in well-being. For example, see Cohen, Rescuing Justice and Equality, 387–94. 65 determines what personal pursuits are worthy and compatible with justice. The list that defines personal prerogative space can also be more or less permitting of personal pursuits depending on the way we justify egalitarian principles of justice. If the list of personal space is permissive enough and demarcates the space clearly enough, then actegalitarianism has met Tan’s two conditions and resolves his apparent tension. There are many ways to justify principles of justice. For example, Martha Nussbaum offers a list of capabilities to give content to the idea of human dignity.71 Each condition on her list must be met for us to consider the life one worthy of human dignity. She argues that the list can be the object of an “overlapping consensus”—it can be accepted by many differing ideas of the good. Cohen seems to have something similar in mind with his intuitive examples of justified personal prerogatives. For example, he accepts personal prerogatives given by David Estlund and Andrew Williams among others.72 He accepts, for instance, that especially arduous work should receive higher compensation and can therefore lead to justifiable wealth inequality. This is one way of generating a list of justified departures from strict equality. Another way to generate a list is a type of justificatory test, such as Scanlon’s contractualism73 or Cohen’s “interpersonal test.” Cohen’s interpersonal test, discussed in the previous chapter, can be used to determine a list of personal prerogatives. If the proposed prerogative leading to wealth inequality can be justified to the rest of the community, then it is a justified policy. Using the test can lead to a list of justified personal prerogatives as part of an egalitarian 71 Frontiers of Justice: Disability, Nationality, Species Membership, 76–78. He accepts them as not causing inequality of well-being. He then offers an interpersonal test to justify policies. His view seems to be that only claims that lead to equality of well-being can pass the interpersonal test. Thus, there are justified inequalities of resources, but there are not justified inequalities of well-being. 73 What We Owe to Each Other (Cambridge: Harvard University Press, 2000). 72 66 principle. The principle would hold, for example, that just actions are those that can be justified to the rest of the community and only actions promoting wealth equality, save some personal prerogative actions, can be so justified. No matter how the list is justified as part of an egalitarian principle, justice will have priority over personal pursuits since they are defined as part of the egalitarian principle of justice, which determines permissible personal pursuits. Within this space for personal pursuits, people are free to pursue their interests consistent with justice. The bounds of egalitarian justice and personal pursuits are also clearly demarcated since there is a list of justifiable departures from equality—or a procedure to generate such a list when one is unsure whether an action is a justified personal prerogative. Thus, it is unclear why act-egalitarianism cannot meet Tan’s two conditions and solve the tension he outlines. To return to the game analogy, just institutional rules act to draw the limit of permissible action for the institutionalist. Principles of justice organize the institutional structure, and the just structure determines what actions are permissible. The actegalitarian can rely instead on the egalitarian principles themselves to draw the limit of permissible action. The “rules of the game” for the act-egalitarian are determined by the principles of justice rather than relying on institutional rules that are determined by principles of justice. Institutions are important for establishing the rules of the game on this account. However, individuals should act to constrain their actions by the rules of the game—the egalitarian principles of justice—as well. 67 3.4 Toward a Defense of the Institutional Approach Tan claims that there is a tension between egalitarian justice and personal pursuits. To solve the tension, an egalitarian theory must determine the bounds of justice and personal pursuits while maintaining the priority of justice. Tan concludes that institutionalism is the only way to do this. Egalitarian justice organizes the institutional structure, which determines permissible forms of life and what interactions are free and fair. So justice is primary because it determines what forms of life are worthy of pursuit. Further, the bounds of personal space and justice are clear. Egalitarian justice is relegated to organizing the institutional structure. Within the institutional structure, people are permitted to pursue their idea of the good consistent with justice. I argued that Tan’s justification of institutionalism is susceptible to an actegalitarian reply. A plausible act-egalitarian account could be worked out that solves Tan’s tension by determining the bounds of justice and personal pursuits while maintaining the primacy of justice. Act-egalitarianism can do this by including a list of justifiable departures from equality, or a procedure to generate such a list, as part of the proposed egalitarian principle. I present this worry to show that Tan’s and Rawls’s defense of institutionalism needs further argument—it is not a “knockdown” case for institutionalism. I think that another case for the institutional approach can be made by comparing the reasoning for act-egalitarian principles with the reasoning for an institutional approach to justice. The arguments in this chapter and the last point to the reasoning for institutional and actegalitarian conceptions of justice as an important factor in determining whether we accept institutionalism or anti-institutionalism. I present my case in the next chapter. I largely 68 agree with Tan’s and Rawls’s case for institutionalism. Part of my argument for institutionalism will show that, given certain assumptions, act-egalitarianism is unreasonable—when compared to an institutionalist account—so my argument will also defend Tan’s account. The result is that both my argument and Tan’s support institutionalism from different starting places. Neither argument provides a “knockdown case” for the institutional approach, but both converge to provide a comparatively strong case for the institutional approach over act-egalitarianism. CHAPTER 4 THE SEPERATENESS OF PERSONS: DEFENDING THE INSTITUTIONAL APPROACH TO JUSTICE The “institutional approach” holds that basic social institutions are the “site” of distributive justice.74 That is, it holds that certain distributive principles apply to institutions that do not also apply to individuals’ choices within the institutional 74 For the definition of the site of justice see Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, 1; Abizadeh, “Cooperation, Pervasive Impact, and Coercion: On the Scope (Not Site) of Distributive Justice,” 323. I get the name “institutional approach” from Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality. For other defenses of the approach, see for example, Kenneth Baynes, “Ethos and Institution: On the Site of Distributive Justice,” Journal of Social Philosophy 37, no. 2 (2006): 182–96; Joshua Cohen, “Taking People as They Are?” Philosophy & Public Affairs 30, no. 4 (2002): 363–86; Norman Daniels, “Democratic Equality: Rawls’s Complex Egalitarianism,” in The Cambridge Companion to Rawls, by Samuel Freeman (Cambridge: Cambridge University Press, 2006), 241–76; David Estlund, “Liberalism, Equality, and Fraternity in Cohens Critique of Rawls,” Journal of Political Philosophy 6, no. 1 (1998): 99–112; Samuel Freeman, Rawls (New York: Routledge, 2007); Louis-Philippe Hodgson, “Why the Basic Structure?” Canadian Journal of Philosophy 42, no. 3–4 (2013): 303–34; Aaron James, “Power in Social Organization as the Subject of Justice,” Pacific Philosophical Quarterly 86, no. 1 (2005): 25–49; A. J. Julius, “Basic Structure and the Value of Equality,” Philosophy & Public Affairs 31, no. 4 (2003): 321–55; Kasper Lippert-Rasmussen, “Inequality, Incentives and the Interpersonal Test,” Ratio 21, no. 4 (2008): 421–39; C.M. Melenovsky, “The Basic Structure as a System of Social Practices,” Social Theory and Practice 39, no. 4 (2013): 599–624; Thomas Pogge, “On the Site of Distributive Justice: Reflections on Cohen and Murphy,” Philosophy & Public Affairs 29, no. 2 (2000): 137–69; John Rawls, A Theory of Justice (Cambridge: Harvard University Press, 1971) (TJ); John Rawls, Political Liberalism Expanded Edition (New York: Columbia University Press, 1993) (PL); John Rawls, Justice as Fairness: A Restatement, ed. Erin. Kelly (Cambridge: Harvard University Press, 2001) (JAF); Miriam Ronzoni, “What Makes a Basic Structure Just?” Res Publica: A Journal of Moral, Legal and Social Philosophy 14, no. 3 (2008): 203–18; Samuel Scheffler, “Is the Basic Structure Basic?,” in The Egalitarian Conscience: Essays in Honor of G. A. Cohen, Christine Sypnowich (Oxford: Oxford University Press, 2006), 102–29; Alan Thomas, “Cohen’s Critique of Rawls: A Double Counting Objection,” Mind 120, no. 480 (2011): 1099–1141; Andrew Williams, “Incentives, Inequality, and Publicity,” Philosophy & Public Affairs 27, no. 3 (1998): 225–47; Andrew Williams, “Justice, Incentives and Constructivism,” Ratio 21, no. 4 (2008): 476–93. 70 framework. Applying distributive justice only to institutions rather than individual characters is a relatively new idea that likely did not get a clear and systematic representation until Rawls.75 According to Rawls, the institutional approach is based on the idea that “the correct regulative principle for anything depends on the nature of the thing” (TJ, 29). Based on the nature of social institutions and social justice, Rawls concludes that his principles “do not apply directly to or regulate internally the institutions and associations within society” (JAF, 10) and that they are “plainly not suitable for a general theory” (PL, 261).76 Proponents of the institutional approach often claim that relegating egalitarianism to institutions creates a “division labor” in that citizens are free to pursue their own ideas of the good while their institutions maintain background justice.77 For example, as we saw in Chapter 3, Kok-Chor Tan argues that we need this division of labor to solve a tension between equality and value pluralism.78 People have many differing nonegalitarian ideas of the good, but justice requires equality and has priority over the good. According to Tan, the institutional approach is the only way to reconcile the two values. Institutionalism makes space for people to pursue their ideas of the good within just institutional rules while the institutional structure maintains equality and “frames” permissible interactions and forms of life. Those opposed to the institutional approach hold that all distributive principles Fleischacker, A Short History of Distributive Justice, 12; Scheffler, “Cosmopolitanism, Justice & Institutions,” 72. Also see Anderson, “Thomas Paine’s ‘Agrarian Justice’ and the Origins of Social Insurance.” 76 See also (TJ, 54–55, PL, 261 and 283, JAF, 7, 162–168). 77 Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality. Also see Rawls PL, 25785. 78 For instance, 30. 75 71 apply both to institutions and individual economic transactions.79 As Samuel Scheffler notes, “If one knows what principle a consequentialist uses to evaluate individual actions, then one also knows what principle he uses to evaluate social policy.”80 For instance, utilitarianism holds that individual and institutional actions ought to maximize satisfied interests. There are nonconsequentialist egalitarian rejections of the institutional approach as well. For instance, as we saw in Chapter 2, G. A. Cohen argues that there is a conflict within Rawls’s institutional egalitarianism.81 He concludes, recall, that to solve the tension, Rawls’s difference principle must apply both to institutions and individual transactions. Besides Cohen’s inconsistency argument, anti-institutionalists claim that the reasons supporting the institutional approach are inadequate.82 For instance, actegalitarians claim that egalitarianism does not need a division of labor. Actegalitarianism can adopt “personal prerogatives”—justifiable actions that do not promote equality—which makes space for diverse personal pursuits.83 In what follows, I take Rawls’s approach as the main representative of institutionalism and argue that it is more reasonable than the two main anti-institutionalist approaches to justice, utilitarianism and act-egalitarianism. Like most philosophers who accept the institutional approach, I accept it because of “value pluralism” or “reasonable Brian Berkey, “Against Rawlsian Institutionalism about Justice,” Social Theory and Practice 42, no. 4 (2016): 706–32; Brian Berkey, “Double Counting, Moral Rigorism, and Cohens Critique of Rawls: A Response to Alan Thomas,” Mind 124, no. 495 (2015): 849–74; Paula Casal, “Occupational Choice and the Egalitarian Ethos,” Economics and Philosophy 29, no. 01 (2013): 3–20; G. A. Cohen, Rescuing Justice and Equality (Cambridge: Harvard University Press, 2008); Liam Murphy, “Institutions and the Demands of Justice,” Philosophy & Public Affairs 27 (1999): 251–91; Serena Olsaretti, “The Inseparability of the Personal and the Political: Review of G.A. Cohen’s Rescuing Justice and Equality,” Analysis 72, no. 1 (2012): 145–56; and consequentialists. 80 The Rejection of Consequentialism, 33. 81 Cohen, Rescuing Justice and Equality. 82 For example, see Cohen, chap. 3; Murphy, “Institutions and the Demands of Justice.” 83 See Scheffler, The Rejection of Consequentialism; Berkey, “Against Rawlsian Institutionalism about Justice”; Cohen, Rescuing Justice and Equality. 79 72 pluralism.”84 However, my argument is not that the institutional approach is the only way to make space for pursuing pluralistic ideas of the good. To explain how value pluralism supports the institutional approach, I provide a framework for thinking about the concept of social justice. I hold that distributive justice is “constructivist.”85 According to Christine Korsgaard, “constructivism” is the view that ethics and justice attempt to solve practical problems. She explains that Rawls’s concept/conception distinction in A Theory of Justice illustrates how constructivism with respect to distributive justice works. She says, [T]he concept of justice refers to the solution to a problem. The problem is what we might call the distribution problem: people join together in a cooperative scheme because it will be better for all of them, but they must decide how its benefits and burdens are to be distributed. A conception of justice is a principle that is proposed as a solution to the distribution problem, arrived at by reflecting on the nature of the problem itself.86 Korsgaard compares constructivist concepts with human artifacts. The concept "chair" does not make sense without the problem it solves, which is something like needing a place to sit. Different types of chairs will be different solutions to the sitting problem. Some artificial concepts, including “medicine,” may refer to the solution to a problem Besides Tan and Rawls, also see Freeman, Rawls, 124; Nagel, Equality and Partiality; Scheffler, “Is the Basic Structure Basic?”; Iris Marion Young, Justice and the Politics of Difference (Princeton: Princeton University Press, 1990), 36. I discuss a possible difference between pluralisms in footnote 88. 85 One might also accept constructivism simply as a useful analytical method for comparatively studying the reasoning for conceptions of distributive justice (cf. Rawls TJ, 121-122). Christine Korsgaard thinks that most of modern political philosophy, including Hobbes, Kant, and Rawls, thinks of justice as a solution to a practical problem (e.g., Korsgaard, The Constitution of Agency: Essays on Practical Reason and Moral Psychology, 318.). Many philosophers also hold something similar to my claim that distributive justice is the solution to the problem of conflicting claims, or value pluralism. For example, see Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, 26 and 83. Kant discusses justice and pluralism in Kant political writings, ed. Hans Reiss, trans. H. B. Nisbet (New York: Cambridge University Press, 1996), 73–74. Also see Elizabeth Anderson’s argument that two modern conceptions of distributive justice—Rawls’s, which has its roots in Thomas Paine, and luck egalitarians, which stem from Fabian socialists—attempt to transcend a class society to solve class conflict. See Anderson, “Rawls’s Difference Principle,” in The Cambridge Companion to A Theory of Justice, ed. Matthew Clayton, Matthew and Andrew Williams, Forthcoming 86 Korsgaard, The Constitution of Agency: Essays on Practical Reason and Moral Psychology, 322. 84 73 over which there is disagreement. People disagree about “medicine” because they disagree about the definitions of “health” and “disease.” Similarly, philosophers disagree on both the concept and conceptions of distributive justice. For neo-Hobbesians, the concept "distributive justice" refers to the solution of the problem of conflicting interests. Their conception is a proposed solution to their interpretation of the distribution problem. Rawls, on my interpretation, thinks that the concept of distributive justice is the solution to the “problem of conflicting claims” (explained below). His conception is a proposed solution to the problem. Furthermore, neither conception obviously follows from reflecting on the nature of an interpretation of the distribution problem, so philosophers will also disagree about conceptions of justice. For instance, I show below that utilitarianism is also a response to conflicting interests. I hold that distributive justice is the response to the “problem of conflicting claims.” The problem is framed by assuming a conception of the person as a moral agent. This conception of the person includes the idea that people accept their own ideas of the good and believe they have claims to a fair share of the rights and resources they need to pursue those ideas of the good. This conception of the person, then, gives rise to the problem of conflicting claims to rights and resources. Hence, “value pluralism” or “reasonable pluralism” engenders the problem of conflicting claims. I use my interpretation of constructivism to argue for the following two claims. First, if the concept of distributive justice refers to the solution to conflicting claims and the reasoning for conceptions of justice represents conflicting claims, then Rawls's institutionalism is justified, and the two anti-institutionalist conceptions are not justified. Second, if we examine the natural justifications for utilitarian and act-egalitarian 74 conceptions, we find that their natural justifications do not represent conflicting claims; a different conception of the person is represented instead. To support these claims, I explain and defend Rawls’s argument that “utilitarianism does not take seriously the distinction between persons” (TJ, 27). I claim that Rawls is charging utilitarianism with not accurately reflecting the problem of conflicting claims. That is, the natural reasoning for the principle of utility—the “impartial spectator,” explained below—does not represent separate people with conflicting values. Instead, a different conception of the person and society is represented. Similarly, I argue that the natural reasoning for actegalitarianism reflects a different conception of the person and therefore a different problem of distributive justice. When we reflect conflicting claims in our reasoning— which I will argue Rawls’s mutual disinterest condition does—then Rawls’s institutional principles are justified over the two anti-institutionalist conceptions. My arguments are comparative among extant theories, but I take them to illustrate a more general point about how my interpretation of constructivism supports the institutional approach. My aim is to expose the underlying assumptions of institutionalism and anti-institutionalism, which will show that institutionalism has a rational basis: principles of justice depend on the nature of what they regulate—separate moral agents—which supports the idea that social justice only properly regulates social institutions. This basis—separate moral agents with their own ideas of the good—is also something that most egalitarians are not willing to reject, including critics of institutionalism such as Cohen and Brian Berkey. 87 So my arguments also provide a case for thinking that egalitarianism is institutional. See Cohen, Rescuing Justice and Equality, 394–98; Berkey, “Against Rawlsian Institutionalism about Justice,” 713. The fact that anti-institutionalists accept a “personal prerogative” also indicates that they are not willing to deny pluralism. In other words, they agree that justice should not permeate all aspects of life since people have other important and diverse values. 87 75 4.1 Constructivism and the Problem of Conflicting Claims In this section, I explain my interpretation of constructivism and how I use it to compare institutionalism and anti-institutionalism. I hold that social justice refers to the response to what I call the “problem of conflicting claims.” The problem of conflicting claims is the conflicting claims and moderate scarcity circumstances of justice according to Rawls. For Hume, the two most important circumstances are confined generosity—the fact that people are not perfect altruists—and moderate scarcity of resources. Rawls replaces confined generosity with the fact that people have incompatible ideas of the good, which he calls “conflicting claims,” or as he later calls it, “reasonable pluralism.”88 Social justice is occasioned, for Rawls, not because people are not perfect altruists as Hume thought (Rawls questions whether the idea of perfect altruism is even coherent),89 but because people have fundamental disagreements about what is of ultimate value in life and are not willing to give up their “ideas of the good” since they are what makes their lives worthwhile. An idea of the good or “plan of life” is “an ordered family of final Rawls calls it “conflicting claims” in (TJ, 257). Rawls may be read as adding to the conflicting claims circumstance when he calls it “the fact of reasonable pluralism” in his later works (e.g., PL and JAF). The circumstance may be interpreted as a deeper conflict in his later works since he abandons his “congruence argument”—the argument for the stability of a just society in Part III of (TJ) that aims to show the congruence of justice and the good. Instead, he comes to accept that people have irreconcilable “comprehensive moral or religious doctrines” that lead to conflicts in claims to liberties and opportunities (e.g., JAF, 84-85). One’s idea of the good is normally “set within” and “interpreted by” one’s comprehensive doctrine (Rawls, JAF, 19). These doctrines are incommensurable, which leads Rawls to interpret his theory as a “political” theory and abandon the theory’s “partially comprehensive” elements (elements that presupposed the truth of parts of a comprehensive doctrine), such as congruence. On congruence and Rawls’s “political turn,” see for example, Samuel Freeman, “Congruence and the Good of Justice,” in The Cambridge Companion to Rawls (Cambridge: Cambridge University Press, 2003), 277– 315.; and Paul Weithman, Why Political Liberalism?: On John Rawls’s Political Turn (New York; Oxford: Oxford University Press, 2013). Rawls (PL, 64-65) also notes that “pluralism as such” and “reasonable pluralism” can be assumed as a circumstance of justice at the stage of principle choice—what Rawls calls the “original position.” Reasonable pluralism assumes that the conflicting claims come from reasonable comprehensive religious and moral doctrines and that these incompatible doctrines will always result from free institutions and democratic culture (JAF, 84). For my purposes, the circumstance can be expressed either as the conflicting claims circumstance, reasonable pluralism, or as pluralism as such. 89 TJ, 189. 88 76 ends and aims which specifies a person's conception of what is of value in human life” (Rawls JAF, 19). For example, people have different religious and moral beliefs that make their lives worthwhile, which lead them to advance distinct interests. Rawls explains that these interests are “interests of a self that regards its conception of the good as worthy of recognition and that advances claims in its behalf as deserving satisfaction” (TJ, 127). In other words, our ideas of the good are what make our lives worthwhile, so we feel entitled to advance claims on its behalf. Because people think that their ideas of the good are deserving of satisfaction, they prefer and claim a greater as opposed to a lesser share of the social product. And because there are moderately scarce resources and each person claims more, their claims will conflict.90 This does not mean that people are selfish. Instead, they seek to protect the effective pursuit of their separate ways of life through rights, resources, social relations, etc. (cf. JAF, 85). Because of this, Rawls holds that justice would still be necessary among “saints and heroes” since their interests can be as opposed as any interests (TJ, 129). It is common to hold that justice is occasioned by the circumstances of justice, which means that justice is neither necessary nor possible when the circumstances do not obtain.91 It also means, as Simon Hope argues, that the circumstances of justice have a minimal justificatory role for a conception of justice. If a conception of justice idealizes away a circumstance of justice, it can straightforwardly be rejected because it cannot be action-guiding. The circumstances of justice are a list of problems that must be 90 Rawls claims that this circumstance results when people claim a greater rather than a smaller share of the social product, rather than their claiming a greater share because they want more resources than others (TJ, 144). 91 Rawls (TJ, 126). 77 overcome for our inclination for a peaceable society.92 So any practical reasoning for a conception of justice must accurately reflect the salient norms and facts about the world to be effectively action-guiding.93 This does not mean that we need to perfectly reflect the world or avoid all falsehood in theorizing about justice. What needs to be avoided, according to Laura Valentini, are those idealizations that “build into a normative theory a false, i.e., idealized, account of the social phenomena the theory itself aims to put under moral scrutiny, in this way severely undermining its potential for guiding action in the real world.”94 As I explained above, however, philosophers disagree about the social phenomena under scrutiny, namely, the subjects of justice—for example, moral persons—because the way in which we conceive of the subjects of justice is a normative issue. Therefore, whether a conception should be rejected depends on what we accept as the concept of distributive justice. I hold that we should think of a conception of social justice as not only “occasioned” by the problem of conflicting claims but as the virtuous or right way of responding to the problem of conflicting claims. People make conflicting claims to protect their ultimate aims in life. They are willing to act justly but not willing to give up what makes life worthwhile.95 A conception of justice, as a virtue that pluralistic people can share, must offer a reasonable response to this problem. Recall that Korsgaard uses 92 Simon Hope, “The Circumstances of Justice,” Hume Studies 36, no. 2 (2010): 133. Hope, e.g., 140. 94 Laura Valentini, “On the Apparent Paradox of Ideal Theory,” Journal of Political Philosophy 17, no. 3 (2009): 352. For example, she claims that Rawls’s Law of Peoples holds that “peoples” are relatively selfcontained political groups, and they are “well-ordered,” so fair background conditions are in place. She says, “real-world societies are not well-ordered, such a background does not exist, and it is precisely its absence that gives rise to the question of international justice in the first place” (351). The resulting principles, therefore, do not reasonably guide action for our world since they idealize away the social phenomena the conception aims to put under moral scrutiny. 95 Cf. Rawls (TJ, 281). 93 78 Rawls’s distribution problem to explain constructivism. I further explicated his distribution problem by explaining the problem of conflicting claims. Korsgaard thinks that Rawls’s distribution problem is part of a larger problem for him within liberalism. Liberalism holds that policies that structure cooperation must be justified to the cooperators. However, under liberal institutions, people have many differing reasonable ideas of the good. “So there is the problem: how are we to give reasons that everyone can accept, in a society where people derive their reasons from radically different conceptions of the good?”96 I will continue to use the more general “conflicting claims” problem since it is the more general circumstances of justice that occasion the virtue of justice. All that matters for the purposes of this chapter is that justice is understood as a response to conflicting claims based on differing ideas of the good, which is part of the problem Korsgaard describes. Since I am discussing three liberal theories of justice, I will also assume a liberal criterion of justification, namely, that principles of justice must be acceptable to those regulated by them. The reasoning for a conception of justice must accurately reflect the circumstances of justice. According to my constructivist account, this is important for a conception of justice because the concept of distributive justice is the response to the problem of conflicting claims, and conceptions are arrived at by reflecting on the nature of the problem of distributive justice. At minimum, one premise, or part of a representation/thought experiment (for example, the original position or impartial spectator (explained below)), should accurately reflect or imply each aspect of conflicting 96 Korsgaard, The Constitution of Agency: Essays on Practical Reason and Moral Psychology, 319. 79 claims.97 In other words, the reasoning should not build into a conception an idealized, i.e., false, idea of conflicting claims. According to Valentini’s condition on idealizations, this would make the normative theory inadequate for guiding action. Indeed, such a conception would fail to meet the criteria for the concept of social justice since its principles fail to be a response to the problem of conflicting claims. To idealize away parts of conflicting claims is to idealize away parts of the concept of distributive justice. Further, if we accept liberalism’s justification criterion, we should agree that reasoning for principles that idealizes away parts of conflicting claims will not be acceptable to those regulated by justice. The problem of conflicting claims is morally relevant to justice. So if any part of the problem is idealized away, it will not be acceptable to those who constitute the problem since it will fail to respect whatever aspect of the concept of justice that is idealized away. Constructivist theories arrive at a conception of justice by reflecting on the nature of the problem of distributive justice.98 The conflicting claims problem arises, given moderate scarcity, because there are multiple moral agents, which can reason about, affirm, reform, and act on their own idea of the good. Individual agents may achieve relatively internally consistent final ends. However, universal agreement among separate agents cannot be reached on final ends—without political oppression—because our normative concepts are somewhat vague and subject to hard cases, our normative I claim that utilitarian and act-egalitarian reasoning idealizes away conflicting claims while Rawls’s reasoning accurately reflects it. I do not want my argument to rest on a narrow interpretation of “accurately reflecting or implying” the problem of conflicting claims, so I intentionally leave it indefinite. 98 There are further conditions regarding what facts a theory must account for according to “practicedependence” constructivism. My argument does not rely on these conditions—it is not “practicedependent”—so I will not discuss them. See James, “Constructing Justice for Existing Practice: Rawls and the Status Quo”; Andrea Sangiovanni, “Justice and the Priority of Politics to Morality,” Journal of Political Philosophy 16, no. 2 (2008): 137–64. 97 80 judgements are subject to weighing the moral importance of things based on distinct values, and so on.99 Reasoning about final ends involves diverse value judgements and many other obstacles that prevent universal agreement, which means that our diverse final ends will lead to conflicting claims based upon them. I will call this part of the problem of conflicting claims the “minimal conception of moral agency.” It is this conception of moral agency, attributed to the people subject to justice, which distinguishes the problem of conflicting claims from Hume’s problem of limited altruism. Hume’s problem could occur among people who share interests and so do not meet the conditions of minimal moral agency. Conflicting claims, on the other hand, arise because separate moral agents inevitably have diverse ideas of the good. The minimal conception of agency also distinguishes the problem of conflicting claims from the conflicting interests within one person. The problem of conflicting claims is not simply a problem of conflicting claims or interests. Instead, the problem of conflicting claims is framed by assuming there are distinct agents who cannot be expected to share final ends because of their separate agency. Because of this, Rawls, for instance, says that society will have no shared ends other than those that are part of the conception of justice (JAF, 20). His theory only assumes that the moral agents subject to justice share “higher-order” interests in developing and exercising their “two moral powers.” The two moral powers of free and equal moral persons are a capacity for a sense of justice and a capacity for a conception of the good (JAF, 18-19; cf. TJ, 12, 19, and 120). The first expresses the desire to live according to fair terms of cooperation and the second is an interest in having, revising, and rationally pursuing one’s idea of the good. Both express, and aim 99 I am referring to Rawls’s “burdens of judgement” (JAF, §11). 81 to protect, one’s moral agency, so no universal idea of the good is assumed. In short, the nature of the problem of conflicting claims is primarily framed by assuming people meet the conditions of the minimal conception of moral agency, so it is important that our reasoning accurately reflects it, for example, by not assuming we can share ideas of the good. If the conception of moral agency is idealized away, the conception of justice will fail to be action-guiding; indeed, it will fail to be a conception of social justice. The resulting principles will also fail the liberal criterion of justification. Moral agency is morally relevant for the concept of justice, so in idealizing it away, a conception fails to respect persons—with respect to the minimal conception of agency that is used to frame the problem of distributive justice. Similarly, if reasoning idealizes away our moral equality, and moral equality is relevant to justice, then the resulting principles fail to respect our equal status. In both cases, the resulting principles of justice will fail to be acceptable since moral equality and the conception of moral agency are morally relevant to social justice.100 I will argue below that the natural reasoning for utilitarianism and actegalitarianism do not accurately reflect or imply the minimal conception of agency; they instead imply a distinct conception of the person. To complete my comparative case for institutionalism over anti-institutionalism, I will also show that Rawls’s contract theory accurately represents the moral agency of each person constituting the problem of conflicting claims with his mutual disinterest condition. My comparative argument relies on accepting my interpretation of constructivism. Utilitarians might already “jump ship” at my formulation of what social justice is. However, I think justice has to do with agency, particularly having an idea of the good, and this leads to the conflict in claims that justice addresses. I explain below that this is why I think utilitarianism has struggled to account for justice. 100 82 The constructivist framework captures the intuition that justice involves respecting moral agency, which results in value pluralism. It also explains the following ideas associated with the concept of justice. Historically, the concept has been a secular, rational, practical, and usually enforceable virtue that people are expected to share across different ideas of the good.101 For example, in the Thomist tradition, faith, or accepting the divine law, is necessary for having all the virtues, but all people can grasp the “natural law” or justice. Constructivism explains how justice is a “practical virtue” by interpreting social justice as aiming to address the practical interaction problem—conflicting claims—that occasions the virtue. It also explains why justice is a virtue that can and should be shared across differing ideas of the good. Justice can be accepted by pluralistic people because the virtue is a response to the conflict among differing ideas of the good. If a conception of justice does not allow pluralistic people to live peaceably with one another and cannot be shared among them, it is not a reasonable solution to the problem. The liberal criterion of justification is also captured by constructivism since pluralistic people cannot agree on ideas of the good or a moral authority, so the fair adjudication of their conflicting claims must be what they each can accept.102 In the next section, I will use this framework to help explain and support Rawls’s well-known argument for accepting his theory over utilitarianism. And, as I explain below, my interpretation of Rawls’s argument explains why utilitarianism has struggled to account for justice. Finally, since 101 Fleischacker, A Short History of Distributive Justice, 5–7, 10–12. As Tan explains, the more modern notion of justice, such as Kant’s, is more concerned with responding to pluralism, whereas the ancient concept dealt more with changing people’s characters to conform to the same norms. See Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, 82–83. Similarly, Elizabeth Anderson shows that modern conceptions of justice, particularly conceptions of distributive justice, are concerned with transcending class society. Previous conceptions merely extolled one’s class specific virtues, which is incompatible with pluralism. See Anderson, “Rawls’s Difference Principle.” 102 Cf. Rawls JAF, 15. 83 the comparative case similarly applies to act-egalitarianism, I claim, the framework provides a plausible comparative case for institutionalism over anti-institutionalism.103 4.2 Rawls, Utilitarianism, and the Separateness of Persons Rawls argues that “Utilitarianism does not take seriously the distinction between persons” (TJ, 27). Utilitarianism holds that aggregate utility, the sum total of satisfied desires of people in society, or “net balance of satisfaction,” ought to be maximized.104 The reasoning for the conception implicates both individual and institutional action. Without an external argument for only applying to institutions, the reasoning for utilitarianism leads to an anti-institutionalist conception of justice. The principle of utility organizes all of social cooperation such that it mirrors the rationality of one person. It is rational to maximize the satisfaction of your own interests over your lifetime. To make the extension of the rationality of one person to all of society work, utilitarianism uses the impartial sympathetic spectator. The spectator is a thought experiment where we imagine that a perfectly rational and benevolent spectator considers, or sympathizes with, all interests in society as if they were its own. A social system is right, on this view, if the impartial spectator would approve of it with full knowledge of the requisite information (TJ, 184-5).105 The principle of utility results Korsgaard offers other reasons for constructivism in “Realism and Constructivism in Twentieth-Century Moral Philosophy.” She also argues (325) that the real difference between moral realism and constructivism is that the realist sees moral philosophy as a theoretical enterprise where the philosopher attempts to find ethical knowledge to apply in action. The constructivist sees moral principles as the solution to practical problems. The only knowledge to be gained is the solution to the problem, and the only way to gain it is to attempt to solve the problem. My framework, therefore, is constructivist in her sense, so those who reject constructivism will reject my account. 104 Rawls’s argument is against classical utilitarianism as found in Sidgwick. See TJ p. 24 and §5. He offers different arguments against “average utilitarianism.” I only discuss the classical version. 105 Also see John Rawls, Lectures on the History of Political Philosophy, ed. Samuel Freeman (Cambridge: Harvard University Press, 2008), sec. Hume II. 103 84 from this way of reasoning since the spectator is the one person who identifies with all of the desires of people in society (TJ, 187). Since the spectator is rational, it will maximize the net balance of satisfied desires. Thus, impartial spectator reasoning provides a way to apply the reasoning of one person to all of society to derive the principle of utility. Rawls claims, however, that it makes the extension of the rationality of one person to society work by conflating the interests and desires of separate persons into one and therefore does not take the distinction between persons seriously. What does Rawls mean by his claim that utilitarianism does not take seriously the distinction between persons, and why is this wrong for a theory of social justice? Alastair Norcross, for example, calls it a “dogma of deontology.”106 He claims that it is either a metaphysical claim about conflating numerically distinct persons or a moral claim that trade-offs between separate people are wrong. He responds that the first is implausible,107 and the second is either implausible or applies to all theories—every plausible theory involves some trade-offs. Norcross, however, does not attempt to interpret Rawls’s argument. He instead replies to the “general” deontological argument that he claims to find in many authors including Rawls.108 I think this is a mistake. Rawls is clear that he is not offering a general moral argument against utilitarianism.109 It is an argument against the natural way of reasoning for the principle of utility as a conception of social justice. Norcross Alastair, “Two Dogmas of Deontology: Aggregation, Rights, and the Separateness of Persons,” Social Philosophy and Policy 26, no. 1 (2009): 76–95. 107 77–79. 108 76–77. 109 I will argue for this claim below. However, it should also be noted that Rawls often makes it clear that he is interested in utilitarianism as a political and social doctrine. For instance, see Rawls TJ, §5, and Rawls, Lectures on the History of Political Philosophy, 162. 106 85 Rawls gives his argument twice in A Theory of Justice. He first outlines the argument (TJ, §6) to contrast his contract theory with utilitarianism and motivate the original position to replace the impartial spectator. There are two standard readings of Rawls’s argument. First, many authors, including Norcross, take Rawls’s point to be that utilitarianism allows for unacceptable tradeoffs. Second, others, including Scheffler, take Rawls’s point to be that utilitarianism is indifferent about distribution.110 However, Rawls in fact makes three separate contrasts with utilitarianism. Before he outlines his three contrasts, he notes that the principle of utility is not directly concerned with distribution. It is only concerned with maximizing utility and so distribution only matters to the extent that it contributes to aggregate utility (TJ, 26).111 Rawls uses this feature of the principle of utility to explain that the natural way of arriving at the principle is to apply the rationality of one person to all of society. For instance, he concludes his discussion of distribution by saying that The most natural way, then, of arriving at utilitarianism … is to adopt for society as a whole the principle of rational choice for one man. Once this is recognized, the place of the impartial spectator and the emphasis on sympathy in the history of utilitarian thought is readily understood. [TJ, 26-27; my emphasis] A rational person will maximize her or his interests without direct regard for the distribution of satisfied interests over a lifetime. This is why the natural way of arriving See Scheffler, The Rejection of Consequentialism, chap. 1; Richard Arneson, “Rawls Versus Utilitarianism in the Light of Political Liberalism,” in The Idea of a Political Liberalism: Essays on Rawls, ed. Victoria Davion and Clark Wolf (Lanham: Rowman & Littlefield, 2000). The more general way to put both criticisms is that the principle of utility “aggregates” and therefore ignores the separateness of persons. The tradeoffs and lack of concern for distribution occur as part of the principle of utility’s aggregation of separate individual’s desires. See Iwao Hirose, “Aggregation and the Separateness of Persons,” Utilitas 25, no. 02 (2013): 182–205. 111 He also notes that this lack of concern with distribution contradicts justice precepts concerning desert, liberties, and so on. He returns to this with his first contrast since the distribution of liberty, for a utilitarian, only matters to the extent that it contributes to net satisfaction. 110 86 at the principle of utility is to apply the rationality of one person to society. The historical use of the impartial spectator in the utilitarian tradition is explained since it is used to apply the rationality of one person to all of society. Further, the place of sympathy in the utilitarian tradition is explained by the fact that the principle of utility is derived from a benevolent spectator who considers everyone’s interests together. We must sympathize with the interests of others in order to follow the dictates of the spectator and the principle of utility. Thus, Rawls uses the principle of utility’s lack of direct concern for distribution to explain the natural reasoning for the principle rather than as an argument against it. His first actual contrast (TJ, 27-28) is that the principle of utility does not give justice priority, whereas his theory does. He says, “Justice denies that the loss of freedom for some is made right by a greater good shared by others. The reasoning which balances the gains and losses of different persons as if they were one person is excluded” (TJ, 28). It is a precept of justice that persons have inviolable liberties. Rawls’s theory accounts for these common convictions by giving liberty priority.112 The principle of utility and the reasoning for it, however, allow tradeoffs of liberty for social welfare. Liberties are only valid to the extent that they contribute to the net balance of desire satisfaction. Rawls concludes that “while the contract doctrine accepts our convictions about the priority of justice as on the whole sound, utilitarianism seeks to account for them as a socially useful illusion” (TJ, 28). The second comparison (TJ, 28-30) is between Rawls’s original position and the 112 On the priority of liberty see TJ, §26 and Ch. 4. 87 reasoning for utilitarianism.113 The reasoning for the principle of utility “extend[s] the principle of rational prudence to the system of desires constructed by the impartial spectator”; and it is reasoning this way that “does not take seriously the plurality and distinctness of individuals” (TJ, 29 and 27). Rawls claims that “the second contrast is related to the first, since it is this conflation [of the desires of separate persons], and the principle based upon it, which subjects the rights secured by justice to the calculus of social interests” (TJ, 29-30; cf. 27). In other words, tradeoffs and a lack of direct concern with distribution—especially of liberties—are distinct implausible consequences of the second contrast with utilitarianism. Thus, the two standard interpretations of Rawls’s argument are mistaken. Rawls’s second contrast is not an argument against the principle of utility itself. It is an argument against the natural reasoning for the principle of utility. He argues that we should not expect the rationality of one person to straightforwardly provide principles of social justice. Rawls says, On the contrary: if we assume that the correct regulative principle for anything depends on the nature of that thing, and that the plurality of distinct persons with separate systems of ends is an essential feature of human societies, we should not expect the principles of social choice to be utilitarian. [TJ, 29; my emphasis; cf. 187] Rawls explicitly cites the problem of conflicting claims, rather than the lack of attention to distribution or the permissibility of tradeoffs, as the reason why conflating persons with their separate and conflicting ultimate aims is wrong for choosing principles of social justice. His second contrast is between his contract theory, which holds that principles of justice are the result of a hypothetical agreement between mutually The third comparison is between utilitarianism as a teleological theory and Rawls’s theory as a deontological theory. See TJ, 30-32. 113 88 disinterested trustees of moral persons (the original position), and the rationality of one person applied to all of society via the impartial spectator. Because the second contrast concerns the reasoning for principles, Rawls notes that it does not show that the principle of utility should be rejected. Indeed, he concedes that for what he says so far, the principle of utility could still be derived from his contract (TJ, 29). He concludes that “from the standpoint of the contract doctrine anyway,” the impartial spectator should be rejected (TJ, 29). The contract doctrine seeks to fairly adjudicate between conflicting claims in a way that all can accept. Rawls seeks to provide an impartial and fair agreement that represents people as separate and not willing to sacrifice their final aims for others. That is, he aims to have the parties express peoples’ “conduct and motives in cases where questions of justice arise” (TJ, 129). The impartial spectator, however, fails to take seriously the distinction between persons—we have our own desires, plans of life, religions, etc.—that leads to the conflict in our claims. The spectator combines these final ends and assesses them together. And because the spectator only uses one point of reference to consider all final ends—the spectator who considers all desires as if they were its own—to adjudicate the conflicting claims of people in society, it also fails to “recognize as the basis of justice that to which men would consent” (TJ, 29). So the reasoning for utilitarianism does not take seriously conflicting claims and the liberal condition that justice ought to be acceptable to free and equal persons. He goes on to argue that if conflicting claims and acceptability are properly represented in our reasoning—by the original position—the principle of utility will not be justified. But this final conclusion cannot be supported until the original position reasoning is presented. Accordingly, Rawls returns (TJ, §30) to the “distinct persons argument” after 89 giving the original position arguments. In his original position arguments, he shows that the parties would choose his two principles of justice over rival principles. He argues that the principle of utility would clearly not be chosen in the original position because the original position models the conflicting interests of distinct persons (TJ, 188-191 and 163-164). The separate parties disinterestedly seek to protect the ends of the person they represent (TJ, 180). Since the parties are mutually disinterested, they have no desire to maximize total satisfaction of desires (TJ, 184). So to consider the merits of utilitarianism, Rawls returns to a closer examination of its natural derivation, the impartial spectator. The impartial spectator is often taken to be an interpretation of impartial benevolence, which derives the principle of utility. However, Rawls explains that justice is not benevolence. Questions of justice arise with conflict, and benevolence cannot settle conflicts. You may be motivated to be just because of benevolence (TJ, 191), but benevolence cannot adjudicate the conflicting claims of separate moral persons (TJ, 190-191). Utilitarianism uses the impartial spectator to make benevolence impartially adjudicate between conflicting claims. However, in so doing, “utilitarianism mistakes impersonality for impartiality” since the reasoning conflates the desires of separate moral persons into one system of desire (TJ, 188 and 190). Rather than being sympathetic, Rawls explains that the parties of the original position are mutually disinterested, which is how his theory takes the distinction between persons seriously while still modeling impartiality (TJ, 187). He says, Instead of defining impartiality from the standpoint of a sympathetic observer who responds to the conflicting interests of others as if they were his own, we define impartiality from the standpoint of the litigants themselves. It is they who must choose their conception of justice once and for all in an original position of equality. They must decide by which principles their claims against one another are to be settled…. [TJ, 190] 90 Because our plans of life and religious and moral doctrines are incommensurable (e.g., TJ, 127), it would be a strong assumption to stipulate that the parties are benevolent, and it would not represent the circumstances of justice.114 Instead, the parties are mutually disinterested, they know that the interests of those they represent are opposed (TJ, 14), and they know that society is characterized by the circumstances of justice (e.g., TJ, 128). Finally, Rawls examines ways that the impartial spectator could be modified to represent separate persons in the circumstances of justice. He shows, however, that it is unclear that these other interpretations of the spectator can derive the principle of utility. He concludes, “this suggests that the conflation of persons into one is indeed at the root of the classical [utilitarian] view” (TJ, 191).115 Using the constructivist framework, Rawls’s argument is the following. Principles of justice depend on the nature of what they regulate, which is distinct moral agents who cannot be expected to share ideas of the good. His contract reasoning accurately represents the problem of the conflicting claims of separate persons and represents what principles moral agents would accept if they were fairly and impartially situated. The mutually disinterested parties of the original position are an idealization, but they accurately represent distinct moral agents with their incompatible ideas of the good. They ensure that each agent’s own good and distinct agency is represented in the argument for Rawls’s principles. It follows that Rawls’s reasoning is particularly suitable for adjudicating the conflicting claims problem. This is the conflict that arises from 114 Rawls argues, however, that making the parties in the original position benevolent would not change the original position. As long as they are treated as separate persons whose good conflicts, benevolence cannot decide anything. TJ, 149 and 191. 115 Rawls is using his method of reflective equilibrium here. He considers the impartial spectator as the kind of contract or constructivist argument that could derive the principle of utility. In so doing, we can see the assumptions that underlie the view. He says, “The procedure of contract theories provides, then, a general analytic method for the comparative study of conceptions of justice” (TJ, 121-122). 91 separate moral agents rather than conflicts that arise due to other features of the world. For instance, if distinct moral agents are represented, the reasoning cannot adjudicate between one person’s conflicting interests or Hume’s problem of limited altruism. Each of these problems allows the possibility that the person(s) regulated by justice are not moral agents in the relevant sense. These possibilities, such as everyone’s values converging, are blocked when separate moral agents with distinct ideas of the good are represented. Once distinct moral agents are represented, the conflict being addressed can no longer be limited altruism, it can only be the conflict arising from separate moral agents with plural ideas of the good. The reasoning for utilitarianism, on the other hand, does not accurately reflect or imply the minimal conception of agency. It evaluates the interests of distinct persons together from one perspective to determine distributive shares, which represents or implies a distinct conception of the person. Consider that the spectator could suitably adjudicate Hume’s problem of limited altruism. The problem can consist of persons who all share final ends. Since the spectator considers all ends together, it can adjudicate between the conflicting claims of persons who all share ends. It allows that all people’s values could converge and so does not accurately represent the conflicting claims of agents. As Rawls shows, spectator reasoning is the reasoning of one person, so it could also suitably adjudicate the conflict in one person’s interests. Spectator reasoning allows for the possibility that justice applies to one person. Again, the minimal conception of agency is idealized away. Conflicting claims arise when there are multiple moral agents whose separate agency leads to distinct moral judgments and conflict based on them. The reasoning for utilitarianism implies a conception of the person, with respect to 92 justice, as thing(s) with interests since only interests are represented as morally relevant. This is why the reasoning can adjudicate the claims of people who share ends or the conflict within one person. According to constructivism, the plurality of moral agents and their separate interests are morally relevant; they constitute the problem of distributive justice. Utilitarian reasoning idealizes away this conception of the person, replacing it with an idea of persons as thing(s) with interests, which is at odds with the concept of social justice. Since the reasoning for utilitarianism does not take seriously the distinction between persons, it is inadequate for guiding action—with respect to justice—in our world of moral agents who cannot agree on ideas of the good. It does not represent the minimal conception of moral agency, which primarily constitutes conflicting claims— that is, it idealizes away precisely what justice aims at assessing—which makes it inadequate for guiding action. In fact, it does not meet the criteria for the concept of social justice. The principle of utility is not a response to conflicting claims. The reasoning for it instead implies a distinct conception of the person, and so it is not the virtue of distributive justice according to my version of constructivism. It offers a different way of conceiving of justice and persons. As Rawls says, on the utilitarian “conception of society separate individuals are thought of as so many different lines along which rights and duties are to be assigned and scarce means of satisfaction allocated in accordance with rules so as to give the greatest fulfillment of wants” (TJ, 27). Finally, spectator reasoning fails to provide an acceptable basis for principles of justice. Moral agents will not accept this reasoning because it idealizes away their separate agency, which is precisely what social justice is a response to. A distinct conception of 93 the person is built into the theory instead, which implies that the separate agency of persons is not morally relevant to justice, only their interests are. Thus, the reasoning fails to respect persons.116 Even if utilitarianism is modified to allow for “personal prerogatives,” the reasoning for it still does not take seriously the distinction between persons.117 Personal prerogatives are designed to allow room for ideas of the good that do not maximize utility, which helps reply to the charge that utilitarianism is too demanding. The spectator might also support personal prerogatives as part of maximizing interests, which seems to better respect agency. However, this reasoning still idealizes away the minimal conception of moral agency. Spectator reasoning holds that all final ends can be evaluated from one perspective to determine aggregate wants. This represents a conception of the person as thing(s) with interests rather than separate moral agents who reason about their own ends. On my reading, rather than a “dogma of deontology,” Rawls offers a powerful argument for why throughout the history of political philosophy utilitarianism has struggled to account for justice. Utilitarianism struggles to account for justice because the idea that justice always involves a conflict between the moral judgments of persons is widely held.118 Unlike utilitarianism, Rawls’s theory accounts for the conflict of interests arising from separate moral agents and for our precepts of justice, such as the priority of 116 My reading of Rawls makes sense of his description of a conception of social justice as the fair balance of conflicting claims (TJ, 5-6, 129, 133-35, 189; JAF, 50). It explains why mutual disinterest is the main motivational condition in the original position (TJ, 189) and is the main point of his contrasts with utilitarianism. It also avoids Norcross’s charge of dogma. 117 See, for example, the hybrid consequentialism from Scheffler, The Rejection of Consequentialism. 118 For example, Korsgaard explains that Hobbes, Kant, and Rawls (e.g., TJ, 129) hold this view. Korsgaard, “Realism and Constructivism in Twentieth-Century Moral Philosophy.” Also see Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, 82–83. 94 justice over social welfare. Utilitarianism must explain these convictions about justice as “socially useful illusions.” Finally, my constructivist framework and Rawls’s arguments fit with the historical use of the concept of justice that I explained above. Utilitarians cannot reasonably say that justice is a practical virtue aimed at solving the conflicting claims arising from pluralism, so they need to explain, or explain away, parts of this concept of distributive justice. 4.3 Act-Egalitarianism and the Separateness of Persons My framework explains and supports Rawls’s case for his original position and two principles over the impartial spectator and utilitarianism. If we agree that distributive justice is a response to conflicting claims and represent it in our reasoning, then Rawls’s institutional egalitarianism is justified over utilitarianism. I will now argue that the framework offers a similar comparative case for Rawls’s institutionalism over actegalitarianism, which, like utilitarianism, holds that a principle of justice ought to regulate all aspects of life. My argument follows the structure of Rawls’s. I will show that the natural derivation of act-egalitarianism fails to take seriously the distinction between persons. I will also briefly argue that the original position cannot justify actegalitarian principles because it models distinct persons using mutual disinterest. Like utilitarianism, act-egalitarianism is incompatible with mutual disinterest. Thus, if we accept conflicting claims as the problem of distributive justice and represent it in our reasoning, Rawls’s institutionalism is justified over act-egalitarianism. And when we consider the natural reasoning for act-egalitarianism, it does not take seriously the separateness of persons; it represents instead a distinct conception of the person. 95 Consider Thomas Christiano’s well-known argument for egalitarianism. Christiano claims that his “impersonal reasoning” for his principle of equality is particularly suited to evaluating institutions but can be applied beyond institutions.119 Further, Christiano’s argument for egalitarianism is put in terms of treatment of other persons. There is no reason in principle why the argument does not apply to actions as well, which is why he offers an argument for focusing on institutions after giving his argument for his “abstract” egalitarian principle.120 So for my purposes, I will take it as a possible way of deriving act-egalitarianism. I examine Christiano’s argument because it offers the strongest foundation for egalitarianism, that I am aware of, that can potentially support act-egalitarianism. It also includes many common premises often thought to provide a foundation for egalitarianism.121 Finally, I argue below that the common premises in the argument are the natural way of deriving act-egalitarianism. Christiano grounds certain basic principles of justice in human dignity, and these principles ground his principle of equality. He argues first that the nature of well-being implies that each person’s well-being ought to be promoted. His argument for this claim 119 For example, Thomas Christiano, The Constitution of Equality: Democratic Authority and Its Limits (Oxford: Oxford University Press, 2012), 25, 30–31. He also notes that he thinks his principle applies beyond institutions. Thus, he mustn’t see the reasons he gives for focusing on institutions (after giving his abstract principle of equality) as supporting the claim that just institutions are sufficient for egalitarian justice. That is, his reasons may support focusing on institutions, but they do not support institutionalism. His reasons for focusing on institutions are that persons have their own lives to lead, which is a claim about the proper demands of justice (he also cites Nagel’s argument about the proper demands of justice), Christiano, 30–31. For a response to this type of argument, see Berkey, “Against Rawlsian Institutionalism about Justice.” 120 Christiano claims that his principle of equality is abstract. His argument only implies that people are due certain treatment, not that individuals have a responsibility to treat other individuals in a certain way. However, this difference in responsibility of treatment requires an independent argument for institutionalism, which he provides. Christiano, The Constitution of Equality: Democratic Authority and Its Limits, 31. So the abstract principle can be used to support act-egalitarianism. 121 See, for example, Cohen, Rescuing Justice and Equality; Ingmar Persson, “In Defense of Extreme Egalitarianism,” in Egalitarianism: New Essays on the Nature and Value of Equality, ed. Nils Holtug and Kasper Lippert-Rasmussen (Oxford: Clarendon Press, 2011). 96 is as follows. Human dignity is grounded in our humanity, or our “capacity to recognize, appreciate, engage with, harmonize with, and produce intrinsic goods.”122 Humans all have essentially the same unique capacity with respect to appreciating and producing intrinsic value, that is, we are “authorities in the realm of value.”123 Our equal capacity in this realm means that we have equal status. Christiano then explains that well-being consists in the “happy exercise of the distinctive authority [in the realm of value] of human beings.”124 Well-being has fundamental value: it is both good for the person who has it and it is an intrinsic good, it constitutes the flourishing of a person. With respect to justice, each person is due this exercise of the capacity for being an authority in the realm of value, so each person is due well-being. In general, more well-being for a person is better than less. The next step in the argument requires the introduction of two principles of justice. First, he holds that the “principle of propriety” is true. This principle says that the concept of justice is each person receiving her or his due. “What is due a person is grounded in some quality of the person that gives the person a certain status or merit.”125 Given his previous conclusion that the nature of well-being implies that it is fitting that it is promoted, he concludes that what each person is due “is that that person’s well-being be advanced.”126 The second principle is the “generic principle” of justice, which holds that relevantly like cases ought to be treated alike. This principle applies to the reasons given by the principle of propriety. Together the principles require that people are treated 122 Christiano, The Constitution of Equality: Democratic Authority and Its Limits, 14. Christiano, 17. 124 Christiano, 19. 125 Christiano, 20. 126 Christiano, 20. 123 97 according to their similar or distinct features that confer a certain status or merit. Since each person is due well-being, the distribution of well-being must be in proportion to the relevant differences in status or merit of persons.127 Finally, Christiano argues that there are no relevant differences between persons such that one person merits more well-being than others. He argues that the traditional reasons, such as desert, reciprocity, productivity, and need are unpersuasive. Thus, only equal status, our dignity grounded in our humanity, is relevant in determining the level of well-being that one merits.128 He concludes that “only equality of well-being is compatible with the fundamental value of well-being, the generic principle of justice, the equality of persons, and the absence of relevant differences between persons.”129 The generic/proprietary principles are common elements in arguments for egalitarianism. For example, Ingmar Persson thinks that the principles, along with the no relevant differences claim, are sufficient to derive egalitarianism.130 Cohen, although he does not offer an argument for egalitarianism, similarly holds that justice is giving each person her due and that we are due equal well-being.131 Indeed, I think the principles are the natural way of deriving act-egalitarianism. All liberal political philosophy starts with the claim that persons are morally equal. So it makes sense that the case for actegalitarianism would require treating people based on each person’s morally relevant features that confer merit. In this way, our treatment of others is attached to the liberal ideal of equal status as the only morally relevant feature that confers merit, which leads to 127 Christiano, 24. Christiano, 26. 129 Christiano, 26. 130 Persson, “In Defense of Extreme Egalitarianism.” 131 For example, Cohen, Rescuing Justice and Equality, chap. Introduction. Cohen assumes that equality is a fundamental moral principle. 128 98 act-egalitarianism. It is the natural way of deriving act-egalitarianism because the principles start with what individuals are due, which implicates both institutions and actions. It focuses our treatment of others on the morally relevant features of each individual person. This is an ancient way of conceiving of justice which focused on shaping one’s character to treat others with what they deserve.132 So there is no reason, given in the derivation, why the just treatment of individuals would only apply to institutions. If the generic/proprietary principles were used to support an institutional egalitarianism, an external argument would be needed to show why egalitarianism only applies to institutions. Thus, I take it that since the argument focuses on the treatment of individuals in general, it is the natural way to derive act-egalitarianism. The generic/proprietary principles can also support a list of personal prerogatives, which allows space for personal pursuits that can lead to inequality. We might, for example, hold that special concern for one’s family is a justifiable departure from promoting equal well-being. If we agree that one’s child has a different moral status than a stranger, then it is permissible, when treating like cases alike, for everyone to promote their child’s well-being over a stranger’s well-being. So we can use the principles that support act-egalitarianism to generate a list of justifiable departures from strict equality to make space for nonegalitarian ideas of the good.133 However, I argue that the natural reasoning for act-egalitarianism does not take seriously the distinction between persons. It does not do so in the same way that utilitarianism does. Recall that utilitarian reasoning exclusively evaluates our conflicting 132 For example, see Fleischacker, A Short History of Distributive Justice; Tan, Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, 82–83. 133 This is a way for act-egalitarians to respond to Tan’s tension. See Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality, 31. 99 interests and aggregates them at the cost of idealizing away the minimal conception of agency. Christiano explains that the generic/proprietary principles imply that the principle of utility is not a principle of justice.134 The principle of utility aggregates wellbeing. In principle, it allows tradeoffs between the well-being of separate persons. According to the generic/propriety principles, what each person is due is grounded in relevantly similar qualities of the person that confer merit. What someone is due cannot be influenced by how it would further the well-being of others or aggregate well-being. The treatment of each person must be based on what that specific person merits and it cannot be different from how others are treated unless there is a morally relevant difference in status. Thus, the reasoning for act-egalitarianism prevents impartial spectator type reasoning. However, act-egalitarian reasoning is, similar to impartial spectator reasoning, impersonal rather than impartial.135 The generic/proprietary principles solely evaluate one’s status that confers merit to justify distributive shares. This is impersonal in that the principles do not accurately reflect or imply the minimal conception of agency; the reasoning instead reflects a distinct conception of the person. Consider that the principles can suitably adjudicate Hume’s problem of limited altruism. They can adjudicate between the claims of people who all share final ends because they do not consider people’s separate ends. People have the same status that confers merit whether they share ideas of the good or not. Further, one person has this status whether other people 134 He claims that other well-known principles, such as prioritarianism and the difference principle are also ruled out. See Christiano, The Constitution of Equality: Democratic Authority and Its Limits, 23. 135 As I noted above, Christiano is aware that his reasoning is impersonal. For example, see Christiano, 30– 32. However, he thinks this way of reasoning is fitting for institutions (and beyond institutions). However, if justice is a response to conflicting claims, all impersonal reasoning with respect to justice is unreasonable. 100 exist with him or her in the circumstances of justice. Isolated persons will still have the same status that confers merit, which means that their well-being should be promoted equally with others. The reasoning therefore does not accurately reflect the minimal conception of agency. It allows the possibility that we all share ends and that justice applies to isolated persons because it only evaluates one’s merit as morally relevant. This means that it implies a conception of the person where one counts as a person, with respect to justice, if one has this status, which Christiano thinks is our dignity grounded in humanity. However, according to my interpretation of constructivism, the minimal conception of the person cannot be idealized away as morally irrelevant to justice. On my account, distributive justice is the fair response to the conflicting claims to moderately scarce resources made by multiple moral agents. Even if the status that confers merit is the minimal conception of agency, the generic/proprietary principles determine distributive shares by evaluating this status alone. Separate and conflicting interests are not represented, which implies that people’s separate and conflicting interests are morally irrelevant for justice. By attaching distributive shares to the minimal conception of the person, only the unrealized capacity for an idea of the good is represented. In other words, utilitarian reasoning only represents interests as morally relevant, whereas, act-egalitarianism only represents merit conferring capacities of persons without representing the conflicting interests that result from these realized capacities. The minimal conception of agency, which leads a plurality of people to incompatible moral judgments and conflicting claims based on them, is idealized away. Thus, the natural reasoning for act-egalitarianism, and all meritbased conceptions of justice, does not take the distinction between persons seriously. 101 Consequently, act-egalitarianism has all the problems associated with utilitarianism, such as failing to be action-guiding—for agents in the problem of conflicting claims—and failing to respect persons as agents with distinct interests. Both anti-institutionalist theories fail to take seriously the distinction between persons. Rawls’s contrast with utilitarianism highlights the need for the mutual disinterest condition in the original position. This condition, he claims, also prevents the parties of his contract from choosing the principle of utility. I think that Rawls’s contrast can be made with act-egalitarianism as well (I made this case in Chapter 2). After giving incentives inequality as an example of an inequality that the parties in the original position would accept, Rawls says, One might think that ideally individuals should want to serve one another. But since the parties are assumed not to take an interest in one another’s interests, their acceptance of these inequalities is only the acceptance of the relations in which men stand in the circumstances of justice. They have no grounds for complaining of one another’s motives. [TJ, 151] In other words, given mutual disinterest, the representatives cannot complain or judge motives. They do not know what the represented person’s interests will be (because they are behind a “veil of ignorance”) and they are not concerned with the interests of others. They therefore cannot complain about preferences for incentives to work certain jobs, which leads to inequality. So they will reject Cohen’s version of the difference principle, which applies to institutions and actions and so prevents motives that make incentives inequality necessary for bettering the worst off.136 Mutual disinterest also prevents the parties from being concerned with the treatment of others based on what they deserve. The basis for act-egalitarianism—and as I explained above, utilitarianism—is therefore 136 Cohen, Rescuing Justice and Equality. 102 blocked by mutual disinterest. So according to the comparative argument, at least one institutional egalitarian theory, Rawls’s, is more reasonable than the natural way of arriving at act-egalitarianism and utilitarianism. In other words, if we accept that distributive justice is the response to conflicting claims and represent it in our reasoning, then only Rawls’s institutionalism is justified. The two representative anti-institutionalist accounts cannot result from the original position; nor do their natural derivations reflect conflicting claims. Finally, if my arguments are persuasive, they also show that the act-egalitarian reply to Tan, outlined in the previous chapter, should be rejected. My case for institutionalism shows that the reasoning for act-egalitarianism does not represent value pluralism. Even if act-egalitarianism uses personal prerogatives to solve Tan’s tension between egalitarianism and value pluralism, it still does not reflect pluralism in its reasoning. Therefore, my framework also offers a defense of Tan’s case that the institutional approach is currently the only reasonable way to reconcile egalitarianism and value pluralism. Tan is concerned with addressing the apparent tension between egalitarianism and personal pursuits. My argument is concerned with reasonably addressing the problem of conflicting claims. Neither argument offers a “knockdown” case for institutionalism. However, both arguments converge from different starting points to support the plausibility of institutionalism. Following Martha Nussbuam, I think that convergence from different procedures on the same conclusion should be a source of confidence in the conclusion.137 Furthermore, I think that Tan’s, Rawls’s, and my arguments help to show the premises used in deriving institutional egalitarianism, act- 137 Frontiers of Justice: Disability, Nationality, Species Membership, 81. 103 egalitarianism, and utilitarianism, which offers a way to comparatively assess each conception of justice as a whole. Chapters 2-4 show that the case for institutionalism is arrived at by taking seriously a certain aspect of persons. Tan defends the requirement that justice must “frame” pluralistic ideas of the good. My argument focuses on the requirement that the reasoning for a conception of justice must accurately represent the problem of conflicting claims, which arises because people have pluralistic and conflicting ideas of the good. Both arguments claim that the plausibility of the institutional approach stems from taking seriously this aspect of persons. We are not simply built to promote justice; we have our own final aims.138 As such, conceptions of justice must only frame, or constrain, ideas of the good, and the reasoning for conceptions of justice must accurately reflect the problem that arises from our differing ideas of the good. 138 This conclusion is important because many authors have assumed—mistakenly, I think—that the case for institutionalism rests on the demandingness of justice. I discussed this in the previous chapter. CHAPTER 5 RECIPROCITY AND THE SCOPE OF DISTRIBUTIVE JUSTICE 5.1 Part 2: Institutional Constructivism and the Scope of Justice I argued in Part 1 of the dissertation that the institutional approach to the site of justice is supported by Rawlsian constructivism—a constructivist methodology, which uses a certain conception of persons as having their own ideas of the good apart from justice. In Part 2 of the dissertation my aim is to determine what Rawlsian constructivists should believe with respect to the scope debate. The debate about the scope of distributive justice often begins with an apparent tension in liberal thought. Michael Blake describes the apparent tension as follows, Liberals are, on the one hand, committed to moral equality, so that the simple fact of humanity is sufficient to motivate a demand for equal concern and respect. Liberal principles, on the other hand, are traditionally applied only within the context of the territorial state, which seems to place an arbitrary limit on the range within which liberal guarantees will apply.139 My focus for Part 2 of the dissertation is Rawlsian liberal thought with respect to the scope of distributive justice. What I call “statist” accounts of the scope of distributive Blake, “Distributive Justice, State Coercion, and Autonomy,” 257. Thomas Pogge asks, after also describing the apparent arbitrariness, “Why should our moral duties, constraining what economic order we may impose upon one another, be so different in the two cases?” World Poverty and Human Rights: Cosmopolitan Responsibilities and Reforms (Cambridge: Polity Press, 2016), 95. 139 105 justice seek to resolve the tension in liberal thought by holding that robust principles of distributive justice—concern for the relative share of goods or wellbeing of individuals— only properly regulate the main institutions of a state. This is because state institutions place their members in a special relationship, which requires distributive justice. International institutions put noncompatriots in a distinct relationship, which at most requires principles of structural equity, principles of sufficiency (a concern for an absolute level wellbeing), or principles of humanity. “Cosmopolitans” attempt to resolve the tension in liberal thought by arguing that our distributive duties ought to apply to all people who share morally relevant properties, that is, “humanity-centered egalitarianism,” or that the special association that occasions distributive justice is global in scope.140 In Part 2, I argue that Rawlsian constructivism supports humanity-centered egalitarianism. Simon Caney offers the most general and well-known argument for humanitycentered egalitarianism. He reasons as follows. If a moral principle is valid, it applies to all persons who share morally relevant properties.141 Common to all the standard arguments for distributive justice is a “universalist conception of moral personality.”142 All persons share moral personality. If the standard rationales for applying distributive justice to some people—standardly compatriots—are plausible, then they apply to all people who share morally relevant properties. Thus, there are cosmopolitan egalitarian “Membership-dependent” cosmopolitan accounts hold that distributive justice duties only arise in special relationships or associations but that this relationship is global in scope. See, for example, Moellendorf, Global Inequality Matters; Pogge, Realizing Rawls; Ronzoni, “The Global Order: A Case of Background Injustice? A Practice-Dependent Account.” Humanity-centered cosmopolitans hold that principles of distributive justice apply to all persons as such, so justice is global in scope. See, for example, Caney, Justice Beyond Borders: A Global Political Theory. 141 Caney, Justice Beyond Borders: A Global Political Theory, 37. 142 Caney, 122. 140 106 principles in both form (the same principle applies uniformly regardless of membership) and scope (the principle applies to all).143 While supporting his argument, he rejects the main reasons for “membership dependence”—the view that human association occasions distributive justice. For example, Caney argues that institutional membership is irrelevant for distributive justice given the normal reasons for entitlements. He asks us to consider two people from separate systems of interaction that never interact with one another. The people are identical in needs and abilities, but one is wealthy and lives in a wealthy system of interaction and the other is impoverished and lives in an impoverished system. He says, But it is difficult to see why—concentrating on any possible and reasonable criteria for entitlement—this is fair. Ex hypothesi, she is not more hard-working or more gifted or more needy. In all respects they are identical (bar one, namely that one is lucky to live in the prosperous society and one is not) and yet an institutionalist approach confers on one many more benefits … Institutional schemes do not track any properties that would generate entitlements and as such they treat people unfairly, denying some their entitlements.144 Caney concludes that economic interaction is not relevant to distributive justice, only moral personality is relevant. He seems to share with Christiano the belief that justice is merit or entitlement attached to human characteristics. Recall that I instead assumed Rawlsian constructivism to support institutionalism. My view is that principles of distributive justice adjudicate between conflicting ideas of the good by organizing background socioeconomic institutions to make interaction free and fair. However, my argument for humanity-centered egalitarianism is similar in form to Caney’s. So it is important to note that his argument does not imply that distributive justice is about 143 144 Caney, 36. Caney, 111. 107 entitlements or merit attached to an individual’s character. Humanity-centered egalitarianism can also be institutional, including the version I defend. My claim will be more closely aligned with those who have argued that Rawls’s principles should apply globally because his egalitarian argument is individualistic.145 They argue that only moral personhood is relevant for the scope of distributive justice given Rawls’s arguments. As Charles Beitz says, Rawls’s view that persons are “self-authenticating sources of valid claims” is sufficient to represent all people in the original position.146 These arguments have the same logical structure as Caney’s,147 but they hold that distributive justice is institutional and in large part procedural—it organizes rules for land, capital, and labor to make individual interactions free and fair. This means that entitlements to goods are determined by systemic rules organizing economic interaction rather than being based on each individual’s character. Thus, we can be humanitycentered egalitarians and institutionalists. Humanity-centered arguments, if correct, only show that existing institutions are not relevant for distributive justice. As Arash Abizadeh shows, institutions may be instrumental for achieving distributive justice or constitutive of a just society without making distributive justice relevant—that is, they need not be existence conditions or triggering conditions of distributive justice (I explain his distinction further below). Including Abizadeh, “Cooperation, Pervasive Impact, and Coercion: On the Scope (Not Site) of Distributive Justice”; Charles R. Beitz, Political Theory and International Relations (Princeton: Princeton University Press, 1999); Pogge, Realizing Rawls; Tan, Justice Without Borders: Cosmopolitanism, Nationalism, and Patriotism. Darrel Moellendorf argues that Rawls’s conception of the person is universal; it does not only apply to a liberal democratic society (Cosmopolitan Justice, 16–26.). But he claims that social justice—principles and institutions that dominate political and social life—is associational (ch. 3). Citing Hume, Moellendorf claims that associations affect moral interests, so they raise moral questions that do not arise between nonassociates (32.) 146 Beitz, “Cosmopolitan Ideals and National Sentiment,” 595. Also see Political Theory and International Relations, 1999, 202–4. 147 Caney, Justice Beyond Borders: A Global Political Theory, chap. 4. 145 108 Instead, with Tan, cosmopolitans can hold that distributive justice requires an institutional structure that establishes and maintains background justice. Tan argues that starting with existing institutions and schemes of cooperation would arbitrarily favor the status quo, which is contrary to the aim of justice. He also notes that Rawls is often misinterpreted as requiring existing cooperative arrangements before justice can apply to people. However, Rawls’s justice as fairness holds that we have a duty of justice to “further just arrangements not yet established, at least when this can be done without too much cost to ourselves” (1971, p. 115).148 So if establishing the institutions required by justice is feasible and can be done without too much cost, we may have a duty to establish new institutions.149 I offered a Rawlsian constructivist case for the institutional approach in Part 1. Rawlsian constructivism is based on a conception of personhood, so it only implies that institutions are a constitutive condition of a just society. This is how my cosmopolitan constructivism will avoid holding that an existing institutional framework is a precondition or an “existence condition” of distributive justice. My humanity-centered arguments will only apply to certain people because I am considering the implications of the Rawlsian constructivist case for institutionalism offered in Part 1. Rawlsian constructivism holds that distributive justice is a response to the problem of conflicting claims. This means that social justice does not apply to all persons as such. Instead, as Tan says, “so long as others are vulnerable to our actions or omissions” we have duties of justice to them. Tan is supporting Kant’s claim that justice, or the concept of right, applies “only with the external and indeed practical relation of 148 Tan, Justice Without Borders: Cosmopolitanism, Nationalism, and Patriotism, 60. On feasibility, and how ideology can affect what is feasible, see Lisa Herzog, “Distributive Justice, Feasibility Gridlocks, and the Harmfulness of Economic Ideology,” Ethical Theory and Moral Practice 18, no. 5 (2015): 957–69. 149 109 one person to another, insofar as their actions, as facts, can have (direct or indirect) influence on each other”150 Since I reject Caney’s and Christiano’s entitlement conception of distributive justice in favor of Rawlsian constructivism, any humanitycentered Rawlsian approach will apply to those in the circumstances of justice— interpreted as the problem of conflicting claims—rather than all persons as such.151 It is a morally relevant property of persons, with respect to justice, that their actions can affect others since justice is a response to people’s conflicting claims to resources, rights, opportunities, and powers. Darrel Moellendorf explains that “membership dependence” is the view that “the requirements of justice between persons are affected by associational membership either because the content of the duties is in some part membership dependent, or because the strength of the duties is.”152 In Part 2, I will argue against (Rawslian) “statism,” or what Moellendorf calls “non-compatriot nonegalitarianism,” that is, the view that only state membership requires egalitarianism; duties of justice to noncompatriots are not egalitarian. He says that “one strategy for rejecting non-compatriot nonegalitarianism is to reject membership dependence.”153 This is because all statist accounts are membership dependent. (As I mentioned above, there are also membership dependent cosmopolitanisms, including Moellendorf’s.) In Part 2, one of my aims is to use the 150 Quoted in Tan, Justice Without Borders: Cosmopolitanism, Nationalism, and Patriotism, 33. Thomas Pogge argues that Rawls’s focus on background justice also implies that distributive justice is global in scope. A just basic structure is necessary for establishing and maintaining background justice, which is necessary to make economic interaction free and fair. So the fact that there are transactions across borders and international institutions coordinating trade, implies that distributive justice is global in scope. Pogge, Realizing Rawls, chap. 6. However, making interaction free and fair does not require ongoing interaction. Two people in the circumstances of justice can affect each other’s interests, which could make any later interaction unfree or unfair. 152 Moellendorf, Global Inequality Matters, 40. 153 Moellendorf, 40. 151 110 Rawlsian constructivist ideas defended in Part 1 to argue that Rawlsian constructivists are not entitled to membership-dependence; they should accept instead humanity-centered egalitarianism. In the next three chapters, I examine the main broadly Rawlsian constructivist arguments for statism. I will in part use my Rawslian constructivist case for the institutional approach to the site of justice to respond to these membership dependence statist arguments. My general claim is that these doctrines, which aim to show how social justice is restricted to the state, conflict with Rawls’s own formulation of the problem of distributive justice defended in Part 1. I show instead that when the arguments are made consistent with the problem of conflicting claims, they support cosmopolitanism. The arguments rely on morally relevant properties of persons, and I argue that these are global in scope. Thus, Rawlsian constructivism supports the global scope of principles of distributive justice. 5.2 Chapter 5: Reciprocity and the Scope of Distributive Justice In this chapter, I examine reciprocity arguments for restricting the scope of distributive justice to the state. These arguments claim that a certain form of reciprocity is the distinct moral ideal that should regulate social cooperation among members of a state. Proponents of the argument hold that nondenizens of a state do not cooperate within a state’s basic structure and so have no distributive claim to our cooperative surplus based on reciprocity. International institutions constitute a different form of cooperation and reciprocity that requires less than distributive equality. For example, they might require principles of sufficiency or procedural fairness in trade. So state and 111 international cooperation are presented as separate problems requiring distinct principles of regulation. In other words, these arguments are “constructivist” in Korsgaard’s sense. The principles of regulation are constructed for the particular cooperation problem that they are meant to regulate. What is important, for my purposes, is that these arguments claim that the moral relation of reciprocal cooperation within state institutions must occur before distributive justice applies to an institutional structure. This is the distinguishing characteristic that separates the problem of interstate and intrastate cooperation. International cooperation involves a different form of reciprocity and a different cooperative product, which is usually said to be national gains on trade. This form of international reciprocity does not require distributive justice. I explain Samuel Freeman’s and Andrea Sangiovanni’s reciprocity arguments in section 3. In section 4, I explain a type of response, given by Abizadeh and others, to accounts similar to Freeman’s and Sangiovanni’s. These arguments claim that since reciprocity accounts use moral relationships as a precondition for the application of justice, they are pernicious. These critics claim that justice ought to be able to criticize relationships where norms of fairness do not yet occur. In section 5, I show that the use of moral relationships to define the scope of justice has a further problem besides being pernicious. This way of defining the scope of justice conflicts with how justice is social. I claim that the account unjustifiably uses ideas that operate within society to apply to society as a whole and the idea of social justice. It also attributes to people a fixed nature theoretically prior to socialization. These charges are roughly Hegel’s argument against the contract doctrines of his day, such as Locke’s. The charges are problematic for proponents of reciprocity accounts since they aim to show that social justice, with its 112 particular principles of distribution, only apply to state social and political institutions. That is, their account rests on certain social aspects of justice. These accounts are Rawlsian. I showed in Part 1 of the dissertation that Rawlsian constructivism is based on an understanding of the problem of distributive justice as a response to pluralism. On this account, distributive justice regulates social institutions; it does not directly regulate private associations and interactions within the institutional structure. In short, my case for the institutional approach to the site question rests on a Rawlsian conception of the problem of distributive justice, with its conception of the person and society. Reciprocity accounts hold that distributive justice is institutional, and they seem to also accept Rawls’s formulation of the problem of distributive justice. So I do not think that proponents of reciprocity accounts will want to reject the social aspects of the concept of social justice. I conclude that these arguments can only reasonably require the capacity to reciprocally cooperate as a condition for the scope of distributive justice. Since people share this morally relevant property, reciprocity arguments support the global scope of their distributive justice principles. That is, my claim that Rawlsian constructivism supports the idea that distributive justice is humanity-centered is supported. 5.3 Reciprocity Arguments for Statism Samuel Freeman and Andrea Sangiovanni offer arguments based on reciprocity to conclude that the scope of distributive justice is restricted to state cooperation while other economic principles apply globally.154 Sangiovanni is one of the main defenders of 154 James interprets Rawls as a giving a practice-dependent case where our shared understandings of the basic structure of society involves seeing it as social cooperation for the reciprocal benefit of co-members of society. James, “Constructing Justice for Existing Practice: Rawls and the Status Quo,” 304. 113 practice-dependence in his other works, which I examine in the next chapter. So I will primarily focus on Freeman’s argument in this chapter. Freeman offers an “institutional account” of distributive justice in that principles of distributive justice organize the institutions necessary for social and economic cooperation for the reciprocal advantage of its members.155 He argues, following Hume, that social and economic cooperation are normally made possible and productive by certain institutions, including the institutions of property law; law and conventions enabling transfers, inheritance, etc.; contract law; and government.156 Government, he claims, is necessary to resolve disputes and “to legislatively specify and judiciously interpret the many procedures and rights, powers, duties, etc., constituting property and other economic institutions.”157 These basic political and economic background institutions and sources of authority enforcing the institutional rules are what Rawls calls the “basic structure of society.”158 Again, following Hume, Freeman holds that these institutions, or “conventions of justice,” are not natural but rather human “artifices,” and therefore we must come up with principles of justice to regulate them.159 These institutions are organized and regulated by each state’s own terms of cooperation. Finally, global cooperative institutions, including the WTO and World Bank, depend on state cooperative institutions, which provide the “grounding and the matter for economic Freeman, “The Law of Peoples, Social Cooperation, Human Rights, and Distributive Justice”; Freeman, “The Social and Institutional Bases of Distributive Justice.” 156 Freeman, “The Social and Institutional Bases of Distributive Justice,” 198. 157 Freeman, 205. 158 The basic structure of society is its “major institutions,” i.e., the “political constitution and the principal economic and social arrangements.” Examples of the basic structure include “the legal protection of freedom of thought and liberty of conscience, competitive markets, private property in the means of production, and the monogamous family” (Rawls TJ, 7). 159 Freeman, “The Social and Institutional Bases of Distributive Justice,” 205. 155 114 treaties that coordinate and harmonize economic relations among different societies.”160 Freeman explains that cooperation is different from other coordinated activities in that it usually involves an idea of reciprocity. Based on some conception of fair terms, each cooperator assumes a share of the burdens of cooperation and is rewarded with a share of the benefits. The terms of cooperation define reciprocal sharing of burdens and benefits. Conceptions of justice assess these terms based on the type of cooperation in question to determine if they embody reciprocity and fairness. The distinct standards of reciprocity depend on the kind of cooperation being regulated, including the purpose of the cooperative institutions in question. Freeman distinguishes distributive justice from the wider topic of economic justice, which includes allocative justice and principles of redress. Economic justice applies to institutions that do not directly determine the distribution of wealth, opportunity, and economic powers among cooperators, such as laws regulating banking and finance and monetary policy.161 Principles of redress try to remedy past wrongs, and certain economic principles provide economic compensation for disability and misfortune. For Freeman, the problem of distributive justice is not the problem of how to organize institutions to allocate wealth. Instead, It is the problem of how to fairly organize and distribute economic powers and positions and divide up the resulting joint social product among productive socially cooperative agents. Economic production is a social 160 Freeman, 210. Aaron James sees international trade as mutual reliance by states to augment national income. He thinks that postwar trade was the result of states coming to an agreement for mutual advantage. They agree to liberalize their markets using international trade rules that provide long-term assurance of mutual benefit. James, Fairness in Practice: A Social Contract for a Global Economy. However, see Darrel Moellendorf, “Fairness in Practice: A Social Contract for a Global Economy, by Aaron James.,” MIND 122, no. 486 (2013): 551–52. He argues that global trade is a mixture of state interest and the interest of firms, investors, consumers, and workers which the states harness through trade policy. Before the postwar trade regime, global trade (and raiding and piracy) occurred and so states sought to stabilize property and economic relations through trade policies. 161 Freeman, “The Social and Institutional Bases of Distributive Justice,” 213. 115 activity resulting in a cooperative social product. The leading question of distributive justice is: How is the jointly produced social product, and the powers and positions behind it, to be fairly distributed among those who cooperate in productive activity by complying with a society’s institutional requirements and making their respective contributions?162 These goods and powers include consumer goods and the means of production. Distributive justice aims to fairly regulate both the distribution of and opportunities to attain powers and responsibilities in economic activity, which includes control over the means of production. To do this, principles of distributive justice seek to organize and assess the standards of reciprocity that distribute these benefits and burdens of socially productive economic activities. Because the social institutions described above are primary in determining the distribution of wealth, opportunity, and economic powers, Freeman calls them “distributive institutions.” Reciprocity, according to Freeman, requires being a member of cooperation. It requires doing your fair share in cooperation by complying with the terms of cooperative institutions and participating in and maintaining the institutions that make cooperation possible. To have a claim on socially cooperative benefits and responsibilities, one must be a member of the cooperative scheme and do their fair share “by complying with the rules” of the basic structure and “contributing to economic output”; that is, social reciprocity in economic cooperation requires membership in the basic structure of society.163 No matter the extent of their cooperation with citizens of another state, 162 Freeman, 199. Freeman, 199 and 211. Aaron James offers a global version of this claim. He says that nontrading countries do not have an equal claim to national gains on trade since the gains of trade are socially created by the cooperation of trading countries. See James, Fairness in Practice: A Social Contract for a Global Economy, 190. 163 116 nonmembers do not and cannot participate directly in the political and economic institutions of the state and so have no claim, based on reciprocity, to the cooperative benefits produced by the state in question.164 Freeman contends that international cooperative institutions do not require distributive justice because they are not distributive institutions or a “basic structure,” and so they involve a different idea of reciprocity. Since distributive institutions do not exist at the global level, the “subject matter of distributive justice does not exist.”165 Global institutions, such as the World Trade Organization, however, require principles of economic justice to fairly regulate trade relations, working conditions, and so on. In other words, a different idea of reciprocity organizes international institutions with their distinct cooperative product, which is mostly gains from international trade. Freeman agrees with Rawls that “advantaged peoples” have a duty to assist economically “burdened peoples” in political and economic development such that they become economically self-sufficient.166 He rejects the idea, however, that advantaged peoples have a positive duty to cooperate with others if this is not mutually beneficial. We have duties of assistance, redress, civility, among others to noncompatriots but not a duty to establish cooperative institutions with them. According to Freeman, just as we have duties to assist and be civil to strangers but do not have positive duties to befriend them, we have duties of assistance and fairness in trade with noncompatriots but we do not have a duty to establish complex cooperative economic institutions with them. People join 164 Further, global cooperation depends on intrastate cooperative institutions. So even if global cooperation requires distributive principles of justice, they only extend to the social product of international cooperation. In other words, the principles would fairly regulate the distribution of the burdens and benefits of the returns gained on trade (Freeman, “The Social and Institutional Bases of Distributive Justice,” 210). 165 Freeman, 214. 166 Freeman, 216. 117 together in societies for mutual benefit, and once this occurs principles of justice are needed to assess the standards of reciprocity involved. Freeman adds that his case against cosmopolitanism is added to the common Rawlsian argument that the basic structure of a state has an enormous impact on its members—it shapes their interests, it primarily determines their economic shares and positions, and it is primary in maintaining background justice. Freeman recognizes that this argument alone does not support the rejection of cosmopolitanism since, as many have argued, international institutions also have a big impact on life-prospects.167 However, he thinks that there is a difference in degree, and so a difference in the economic product produced, which helps to separate intrastate and interstate cooperation based on the type of cooperation involved with its distinct standard of reciprocity. Sangiovanni similarly claims that there are different types of cooperation with their own standards of reciprocity. He first argues for the empirical premise that state institutions are distinct from the international order because state institutions primarily provide us with collective goods, such as law and defense, that are necessary to have and act upon a plan of life.168 He grounds egalitarianism in the “luck egalitarian” claim that undeserved arbitrary natural and social inequalities should be eliminated or mitigated. However, he also holds that equality is a relational ideal of reciprocity. He says, “Reciprocity in the mutual provision of the basic collective goods necessary for acting on a plan of life conditions the content, scope, and justification of distributive equality.”169 For instance, see Cohen and Sabel, “Extra Rempublicam Nulla Justitia?”; Nicole Hassoun, Globalization and Global Justice: Shrinking Distance, Expanding Obligations (Cambridge: Cambridge University Press, 2014), chap. 5. 168 Sangiovanni, “Global Justice, Reciprocity, and the State,” 19–20. 169 Sangiovanni, 22. 167 118 We owe each other reciprocity in the provision of these goods, and this reciprocity takes the form of the luck egalitarian claim for the following reasons. Everyone’s ability to make use of their talents relies on the contributions of others in society, and developing one’s talents occurs within an institutional framework, which provides basic goods such as the legal system. This framework is provided and upheld by fellow denizens. Similarly, wealth is a function of society’s accumulated knowledge over generations. Thus, fair reciprocity—a fair return for one’s contribution—in the provision of necessary goods to sustain and act upon a plan of life triggers the presumption against arbitrary inequalities in the distribution of these social goods.170 In other words, your reciprocal return for fair cooperation should not be influenced by arbitrary inequalities. Only state institutions provide and distribute the collective goods necessary to act upon a plan of life, so the reciprocity grounding of the luck egalitarian presumption against inequality is not global in scope. Although Sangiovanni holds that morally equal persons are the ultimate units of moral concern and are owed equal respect, the luck egalitarian premise is triggered by relationships, specifically state based reciprocity, and does not apply to all arbitrary advantages and disadvantages such as place of birth. In short, “We owe obligations of egalitarian reciprocity to fellow citizens and residents in the state, who provide us with the basic conditions and guarantees necessary to develop and act on a plan of life, but not to noncitizens, who do not.”171 We get to a more substantive egalitarian conclusion, on his account, by combining reciprocity in the 170 Sangiovanni, 25–26. For Sangiovanni, upholding and participating in sate legal institutions, such as civil and criminal law, and political institutions that provide for defense and sustaining a system of property are more important than contributing to economic production for the mutual provision the necessary collective goods. This is partly because of what these institutional goods are and because of the luck egalitarian points above. 171 Sangiovanni, 20. 119 collective provision of social goods with the moral arbitrariness claim. Sangiovanni notes that we could, for example, use these premises to support Rawls’s original position justification for substantive egalitarian principles, such as the difference principle, and support restricting the contract argument to fellow denizens of a state. 5.4 Reciprocity Arguments and Perversity There are two main responses to reciprocity arguments. One is to resist the empirical claims made, such as the claims that a “basic structure” only exists at the state level. The other main criticism of reciprocity arguments, and ones like it, targets the use of moral relations as delimiting the scope of justice. That is, these accounts hold that one does not have a duty of distributive justice to nondenizens because they are not in the moral relationship that justice regulates. I will focus on the moral criticism. I explained that primarily the moral relation of “reciprocity,” but also “fairness” and “mutual benefit,” does the work of restricting the scope of distributive justice to the state. In this section, I explain a line of argument given by Abizadeh among others against accounts similar to reciprocity arguments. The arguments charge reciprocity accounts with holding that people must be under fair relationships before justice can apply to them, which is pernicious. I will then, in section 5, expand on the responses to show that reciprocity arguments are also unreasonable given the idea that justice is social. Abizadeh responds to the following Rawlsian argument for restricting distributive justice to the state. The argument is as follows: 1) distributive justice requires a basic structure, 2) basic structures are confined to state cooperation, 3) therefore distributive justice only applies to co-denizens of a state. Sangiovanni’s and Freeman’s arguments 120 can be seen as offering reasons based on reciprocity for why justice requires a state basic structure as the subject matter of distributive justice. In his response to this form of argument, Abizadeh distinguishes between three senses in which justice might be said to “require” a basic structure: as an “existence condition,” an “instrumental condition,” or a “constitutive condition” of justice.172 An “existence condition,” or triggering condition, is a state of affairs or relationship that must occur for the virtue of justice to arise or be applicable. An “instrumental condition” is something that is helpful or required for feasibly promoting or realizing justice.173 A “constitutive condition” is something that is necessarily part of a just society. For example, we might think that a just society requires equality and embodies reciprocity in the relationships among the people involved. He notes that only the basic structure as an existence condition could show that the site and scope of justice coincide.174 As I explained in the introduction, my case for institutionalism with respect to the site of justice does not support the basic structure as an existence condition. Because of this, I will not examine existence condition arguments, such as coercion and pervasive impact arguments for the scope of justice in this chapter. I focus instead on constructivist arguments as instrumental or constitutive conditions. Abizadeh argues that what he calls the “cooperation theory”—the argument that justice regulates terms of cooperation among members of a cooperative scheme–can only Abizadeh, “Cooperation, Pervasive Impact, and Coercion: On the Scope (Not Site) of Distributive Justice,” 324. For other criticisms of reciprocity accounts, or justice as mutual advantage, see Allen Buchanan, “Justice as Reciprocity Versus Subject-Centered Justice.,” Philosophy & Public Affairs 19, no. 3 (1990): 227–52; Nussbaum, Frontiers of Justice: Disability, Nationality, Species Membership, chap. 1; Tan, Justice Without Borders: Cosmopolitanism, Nationalism, and Patriotism, 58–60. 173 Liam Murphy thinks this is the main reason to be concerned with institutions. They are effective at promoting the aims of justice. “Institutions and the Demands of Justice.” 174 Abizadeh, “Cooperation, Pervasive Impact, and Coercion: On the Scope (Not Site) of Distributive Justice,” 321. 172 121 be a constitutive or instrumental condition of justice. An instrumental condition means that a basic structure is helpful for advancing justice, which does not restrict the scope of justice to the state since a global basic structure may effectively advance justice. Further, a constitutive condition cannot give us an answer to the scope question, according to Abizadeh, because it offers moral ideals for a just society or demands of justice.175 Rawls’s idea of social cooperation, which is regulated and defined by a basic structure, involves moralized elements such as “mutual benefit,” “fairness,” and “reciprocity.” According to Abizadeh, these are demands of justice—for example, a basic structure is not just if it is not for the reciprocal benefit of those involved. They therefore cannot restrict the scope of justice to existing basic structures since a basic structure regulated by principles of fairness and reciprocity is a requirement of justice. Constitutive conditions are not conditions to be met before justice can arise or be necessary and possible. Abizadeh holds that because of this, using a constitutive condition of justice to define the scope of justice—or conflating constitutive and existence conditions—arbitrarily favors the status quo and is “perverse.”176 He says, “By using the ideal of social cooperation to specify (and restrict) the scope of application of the ideal itself, [the cooperation argument for restricting the scope] perversely implies that demands of distributive justice arise only between persons whose social interactions are already conducted on fair terms.”177 This is because the idea of cooperation is a moralized idea that involves “fairness” or “reciprocity.” 175 He notes that instrumental and constitutive conditions may give us feasibility restrictions on the scope of justice for the purposes of advancing justice. However, they do not restrict the scope of justice. Abizadeh, 324. 176 Abizadeh, 313 & 330. 177 330. 122 Abizadeh’s criticism is similar to Darrel Moellendorf’s charge of “justice positivism.”178 Moellendorf claims that whether an advantage is fair depends both on whether it results from following established rules and whether the rules themselves are fair. To hold that an advantage is fair based solely on whether it is the result of following established rules is “justice positivism.” The fairness of social advantages, like advantages in a game, can be questioned both by appeal to the established rules and by questioning the fairness of the rules themselves. Justice positivism lacks a way of criticizing institutional social conventions since it only focuses on the fairness of established rules. With respect to reciprocity accounts, the critical claim would be that they lack a way of criticizing the convention of reciprocity that defines the scope of our distributive duties. For example, Rainer Forst uses the idea of justice positivism to respond to the Rawls inspired basic structure account explained above. He claims that holding that distributive justice only applies where norms of justice are already institutionalized in positive law or where mutually beneficial cooperation or mutually beneficial cooperative institutions exist amounts to a twofold justice positivism.179 Similar to Abizadeh, Forst claims that the justice-positivism claim is that the moralized ideals or institutional relationships are normative conclusions of arguments, not existence conditions or triggering conditions for the virtue of justice.180 Holding that an institutionalized norm of reciprocity must be established before justice applies means that 178 Moellendorf, Cosmopolitan Justice, 38–39; Moellendorf, Global Inequality Matters, 36. Rainer Forst, “Transnational Justice and Non-Domination: A Discourse-Theoretical Approach,” in Domination and Global Political Justice: Conceptual, Historical, and Institutional Perspectives, ed. Barbara C. J. Buckinx, Jonathan Trejo-Mathys, and Timothy Waligore (New York: Routledge, 2015), 99– 100. He calls it “practice-positivism” because he is responding to “practice-dependence accounts,” but he cites Moellendorf for the idea. 180 Forst, 99–100. See also Pogge, Realizing Rawls, 20. He refers to the basic structure as a “social system” rather than “society” since social systems are not always mutually beneficial. 179 123 we lack the means to criticize the form or scope of the institutional relation, which is perverse. Principles of justice ought to be able to criticize circumstances in which fairness norms do not prevail.181 Freeman and Sangiovanni both claim that people join together for mutually advantageous cooperation. It is only once this occurs that distributive justice becomes applicable because cooperation involves an ideal of reciprocity. They also hold that we do not have distributive duties based on reciprocity to noncooperators because they do not cooperate. Further, Freeman argues that we do not have a positive duty to establish cooperative institutions with noncompatriots unless it is mutually advantageous. The above arguments given by Abizadeh and others give us reason to doubt that these moral ideals or relationships can be preconditions or “existence conditions” to be met before social justice applies to people. Moral relationships, such as reciprocity, are conclusions of normative arguments for what kind of relationships should occur between people. It is perverse to hold that these ideals must regulate conduct before standards of justice apply. It implies that fairness or justice must occur among a set of people before fairness or justice can regulate or criticize their interactions. It prevents us from criticizing existing relationships and their scope.182 Also see Valentini, and Ronald Dworkin’s claim that “justice is our critic, not our mirror” quoted in Laura Valentini, “Global Justice and Practice-Dependence: Conventionalism, Institutionalism, Functionalism,” Journal of Political Philosophy 19, no. 4 (2011): 408. 182 Beitz makes a similar point with his example of slaves forced to contribute to a scheme of social coordination against their will without any advantage to themselves (Political Theory and International Relations, 1999, 131). 181 124 5.5 Against Reciprocity as Triggering Social Justice Besides being perverse, I think that reciprocity accounts fail to appreciate the concept of social justice. Since reciprocity accounts are Rawlsian, I will first argue that reciprocity can only be a constitutive condition of justice in Rawls’s theory. I will then argue that Rawls’s reasons for this, and his reasons for holding that social justice organizes the basic structure of society, are meant to capture the social aspects of justice. He is worried about Hegelian criticisms of other contract theories, such as Locke’s. I argue that reciprocity accounts are susceptible to the Hegelian criticisms. This is important since reciprocity accounts clearly try to capture the social aspects of justice since they hold, with Rawls, that the primary subject of social justice is social and political institutions. So I think that proponents of reciprocity accounts will not want to reject the social aspects of justice. Further, my Rawlsian constructivist case for institutionalism given in Part 1 relies on the social aspects of justice. That is, I argued that social justice regulates social institutions to fairly adjudicate between the conflicting claims of people who cannot be expected to share ideas of the good. What I called “the problem of conflicting claims” is a social problem rather than a problem among, say, members of a religious organization who freely join and share ends. Both reciprocity accounts follow Rawlsian ideas, so we can look at how reciprocity and mutual advantage function in Rawls’s theory for guidance. Cynthia Stark explains that Rawls’s contractarianism does not justify principles as mutually advantageous or reciprocally beneficial as measured against the “non-agreement point” or a state of noncooperation (what Rawls calls “general egoism”).183 Instead, justice is Cynthia Stark, “Contractarianism and Cooperation,” Politics, Philosophy & Economics 8, no. 1 (2009): 77–78. See also Rawls PL, Lecture 7. 183 125 reciprocity or mutual advantage with respect to an agreed upon baseline of equality.184 That is, the parties in his social contract start by considering equality and rationally decide to accept certain inequalities as justifiable departures because they are reciprocally beneficial compared to other possible arrangements. Rawls arrives at this conclusion in part by assuming that the parties represent people who are capable of full cooperation. According to Stark, this assumption is justified by the contractarian acceptance of the “anti-exploitation principle,” which says that those engaged in cooperation for benefits are due a portion of those benefits.185 The idea is that people are exploited by cooperation where they are not given a portion of the benefits that they help produce. So Rawls aims to derive his ideal of justice as mutual advantage or reciprocity among cooperators for the social product by assuming the capacity for full cooperation in his contract. Being incapable of cooperation is morally relevant for justice as reciprocity on this view because only those capable of cooperation are capable of reciprocating with respect to the social product.186 For Rawls, rational mutual advantage or reciprocity as judged against noncooperation does not have a role to play in justifying a conception of justice. In short, reciprocity and mutual advantage are constitutive conditions of justice among those capable of participating in social cooperation. As Rawls says, the difference principle “expresses a conception of reciprocity” (TJ, 102; my emphasis). Freeman claims elsewhere, however, that reciprocity is the main ground for the 184 Rawls notes that if we maximize the position of the worst off relative to other social arraignments, and everyone benefits by doing this, the benchmark is irrelevant “if not largely impossible to ascertain anyway” (1971: 80). According to Rawls, there is a further sense in which everyone benefits when the difference principle is satisfied because, given certain facts, inequalities are “chain-connected” (see 1971: 80-83). That is, “if an advantage has the effect of raising the expectations of the lowest position, it raises the expectations of all positions in between” (1971: 80). The relevant point is that the comparison class for reciprocity is other possible social arrangements that determine the systemic distribution of opportunities. 185 Stark, “Contractarianism and Cooperation,” 89–90. 186 Stark, 81. 126 difference principle for Rawls.187 He adds that the reciprocity ground explains why it is not appropriate for Rawls to apply the difference principle either globally (2006, 442) or to the disabled (2006, 106-108, 448). However, Henry Richardson responds that reciprocity in Rawls is expressed by the difference principle, so it does not and should not ground the principle. Instead, the arguments for it rely on a psychological principle of reciprocity (rather than a normative principle), which supports the idea that the difference principle would support a stable society better than the principle of utility, and the other ground for the difference principle relies on assumptions about the social bases of selfrespect.188 As Richardson notes, Freeman’s own reconstruction of Rawls’s case for the difference principle shows that other concerns, such as the social basis of self-respect, grounds the difference principle as a principle of reciprocity.189 Thus, Rawls does not (and should not) use reciprocity to ground his principle that expresses reciprocity. Reciprocity and mutual advantage are not compared to nonagreement for Rawls because his focus is social justice. He holds that social justice is the primary virtue of social institutions for fairly adjudicating between people who cannot be expected to share ideas of the good. The justice problem being addressed is social rather than a problem among members of a voluntary association. Given the concept, he offers at least three main reasons for why the basic structure is the first subject of social justice. First, it largely shapes the interests and aspirations of people. Second, its effects on inequalities 187 Rawls, 189. Henry Richardson, “Interpreting Rawls: An Essay on Audard, Freeman, and Pogge,” The Journal of Ethics 15, no. 3 (2011): 232 & 241–42. Freeman (Rawls, 345.) also explains that reciprocity adds an idea of fairness to cooperation which helps provide the baseline of equality for the argument. However, see my above comments on the baseline of equality. With Richardson, I think Freeman overemphasizes the role reciprocity plays in Rawls’s theory. 189 Freeman, Rawls, 193; Richardson, “Interpreting Rawls: An Essay on Audard, Freeman, and Pogge,” 232. Cf. Anderson, “Rawls’s Difference Principle.” 188 127 among persons’ rights and entitlements are profound and present from the start of life. Because of the first two points, he holds that inequalities in life-prospects are the most fundamental for social justice. Finally, when principles of social justice organize the institutional structure, ideas of the good are “framed” by just institutional rules and people are then left to effectively pursue their idea of the good consistent with justice. That is, the basic structure maintains “background justice” for society. Rawls stresses that the above reasons for focusing on the basic structure for social justice, and the reasons for holding that justice is the primary virtue of the basic structure, captures the fact that the idea of justice in question is social. In particular, Rawls argues that a social contract theory must allow for at least three facts given the subject of social justice and the function of a basic structure as maintaining background justice for a society. The facts are, [1] that membership in our society is given, [2] that we cannot know what we would have been like had we not belonged to it (perhaps the thought itself lacks sense), and [3] that society as a whole has no ends or ordering of ends in the way that associations and individuals [within a basic structure] do. [PL, 276] Rawls’s point with 1 and 2 is that one is born into one’s social institutions, so the parties to a social contract should not be allowed to compare how they would fare in other societies. You are born into and shaped by your basic background institutions, so the comparison of how you would fare or what you would be like in another society may even lack sense. His point with 3 is that there is no way to know one’s potential contribution to society, and it is morally irrelevant. Potential contribution is not relevant since, among other reasons, people are shaped by their institutional structure, and their level of contribution, and what counts as contribution, will be largely determined by their 128 social structure. Further, there are no shared ends by which to judge contributions in society to begin with. Social justice, recall, is social in part because it responds to conflict among those who do not share ends. For Rawls, the shared goal of a basic structure is to establish and maintain background justice; it is not to rank individuals’ contributions to society based on the goals of the association. Indeed, the point of background justice is to establish and maintain an institutional background within which associations and interactions are made to be free and fair. That is, social justice makes associations free and fair; it is not itself a private association where people are free to join and leave and where their potential contribution to the association can be measured. Rawls contends that other “social contract” theories, such as Locke’s, are not really social. Unlike his theory, these other theories are susceptible to the two following Hegelian criticisms. First, they unjustifiably extend notions of what Hegel called “civil society,” such as a contractual agreement among actual people, to the idea of social justice and so to the whole social structure. Second, the arguments fail to recognize the social nature of human beings and attributed to humans fixed natures and desires as theoretically prior to socialization.190 In short, these other forms of contract doctrines make justice a virtue for private associations, and they have no role for social institutions—the basic structure—because of this. The Hegelian argument applies to a general category of theories that Rawls terms “historical process doctrines,” which include contract theories such as Hobbes’s and Locke’s. These theories hold that the results of historical transactions that did not violate rights preserve justice. For example, Locke’s process doctrine holds that in a state of Rawls PL, 286. Also see Georg Wilhelm Friedrich Hegel, Hegel’s Philosophy of Right, trans. Thomas Malcolm Knox (Oxford: Oxford University Press, 1967), 156. 190 129 nature everyone has an equal claim to resources and can appropriate them as long as they leave enough and as good for others. This Lockean proviso may be lifted with universal consent, that is, only if everyone would be better off under a new arrangement. And Robert Nozick’s libertarian doctrine holds that minimal government associations are just if people freely joined the dominant protective agency or government. Further, current holdings are just if they are the result of a process of acquisition and transfer where no one’s rights were violated and satisfied the Lockean proviso. Rawls explains that these theories are susceptible to both of Hegel’s criticisms. First, these agreements result from “contingencies and accidents of the as-if just historical process which has no tendency to preserve or to move toward background justice” (PL, 287). Background justice’s role is to make transactions free and fair over time since individual transactions tend away from justice, that is, away from free and fair interaction. To know whether individual transactions are truly free and fair, we need to know about the background social institutions against which they took place. To hold that justice is iterative, where as long as individual rights were not violated at the time of the transaction, the result of each iteration of rights and transactions as a historical process is just, ignores these social aspects of interaction. In other words, by holding that agreements made in the past can make present conditions just, these doctrines unjustifiably extend notions of “civil society,” such as fairness of individual transactions and contracts among rational adults, to social justice.191 Second, these theories have no role for the basic structure as establishing a division of labor between social justice rules and interactions within the basic structure. It is the role of social justice to “frame” On this, also see Anderson, “Rawls’s Difference Principle”; Ronzoni, “What Makes a Basic Structure Just?” 191 130 permissible ideas of the good since the virtue is meant to adjudicate between the interests of people who cannot be expected to share ideas of the good. The process doctrines instead hold that justice is always a matter of actual acquisition and exchange between consenting adults. That is, these doctrines unjustifiably apply notions from civil society—exchange and acquisition between rational adults—to social justice and the basic structure. However, to determine whether exchange is free and fair for social justice, according to Rawls, we need to know about the background conditions that one was born into and which structured one’s life-time prospects. Finally, the process doctrines compare one’s situation in society to a state of nature or noncooperation. These types of comparisons, however, are only appropriate for private associations within a basic structure. The comparisons assume that humans have a fixed nature theoretically prior to being shaped by the background institutions that social justice regulates. However, as Rawls explains, this comparison of how you would fare in society versus a state of nature or another society may lack sense. That is, we cannot really know how “we” would fare in a state of nature versus our society since our society has shaped our interests and determined what talents are rewarded and what talents are nurtured. So the notion of mutual benefit versus other social arrangements misapplies a notion of “civil society,” the notion of a fully developed rational agent benefiting from one association over others, to social justice and society as a whole.192 Similar to historical process doctrines, Freeman’s and Sangiovanni’s response to noncompatriots’ claims on “our” social product misrepresents the social aspects of justice This is why Rawls’s parties in his contract are not really analogous to fully formed human beings. It would be a mistake for a contract argument to make the parties analogous to people. Instead, the parties represent the moral powers of people and general interests necessary for citizens to develop those powers. 192 131 and the nature of the social institutions justice regulates. Freeman claims that one gains a reciprocal claim to the distribution of the social product by doing one’s fair share in producing the social product. Sangiovanni requires reciprocity in supporting state institutions. In other words, both hold that one must be engaging in cooperation, with its ideal of reciprocity, to have a claim to the benefits and burdens of cooperation. Both authors also hold that distributive justice involves a large part of procedural justice; it regulates the background institutional rules from which a distribution results. Noncompatriots are not members of social cooperation within state institutions, so they have no distributive claim based on reciprocity to the benefits and responsibilities of cooperation. However, cooperative institutions—the procedure that distributive justice regulates—shape our interests from the start of life, determine our life-time prospects, and form the background of our interactions and associations. That is, the institutions are social, and social justice regulates these institutions. Consequently, the notion that reciprocity is part of cooperation and so noncooperators do not have a claim to distributive justice based on reciprocity misunderstands how justice is social in two ways. First, it unjustifiably applies notions for “civil society” to social justice. Private associations can justifiably require reciprocal cooperation as a precondition for receiving benefits or responsibilities. This is because, assuming relatively just background institutions, the association is free and fair. One can avoid joining the association or leave. So the association can use its benefits and responsibilities as a reward—mutual benefit—for cooperating. Social justice is different. One is born into and is unavoidably shaped by just or unjust social and political institutions. Because of this, one cannot be required to do anything to receive social justice—as if it were a reward. Indeed, Freeman 132 criticizes coercion arguments by claiming that their account makes distributive justice a type of compensation for being coerced.193 He claims that reciprocity accounts hold that justice is in large part procedural rather than compensatory. However, reciprocity accounts hold that cooperation and the scope of who is owed distributive justice is a type of compensation for being a member of cooperation. That is, being a member of cooperation and doing one’s fair share gives one a claim to benefits and responsibilities. This makes distributive justice a type of compensation for doing one’s fair share. However, distributive justice is part of social justice, so the reciprocity account unjustifiably applies notions for “civil society,” such as compensation, to social justice and society as a whole. These notions for civil society are only appropriate for civil society because we need to assume a just background structure of society to use them. For example, Nozick’s “natural liberties,” or "side constraints,” for property acquisition and exchange do not act to preserve fair background conditions for society. Instead, they focus on individual transactions between fully formed rational adults. However, to know whether these transactions are free and fair, we need to know under what background conditions they occurred. Thus, the notion of contractual agreements among rational adults is appropriate for civil society but not society as a whole since it ignores socialization among other things. Similarly, requiring that one do one’s fair share in cooperation to receive benefits based on reciprocity is a notion appropriate only for civil society. It makes distributive rules a sort of contracted compensation. However, we need to know about societal background conditions and socialization to determine whether such 193 “The Social and Institutional Bases of Distributive Justice,” 201. 133 contracted compensation is really free and fair. That is, reciprocity accounts imply that cooperating and doing one’s fair share is a choice that one makes that earns them a share of the social product, but society and social justice is not like this. Second, reciprocity accounts assume that individuals have a fixed nature theoretically prior to being shaped by the institutions that justice regulates. They compare being a member of reciprocal cooperation with not cooperating or a state of noncooperation. One gains a claim to the benefits and responsibilities of cooperation by being a member and doing one’s fair share. Proponents claim that everyone cooperating mutually benefits compared to not cooperating. So using reciprocal cooperation to determine the scope of justice attributes a fixed nature as theoretically prior to our socialization. However, we, as fully formed rational agents, do not decide whether we will benefit by cooperating over not cooperating or a state of nature. Instead, we are born into and shaped by social institutions. It is not relevant to social justice how fully formed rational individuals would fare in another society or in a state of nature. Social justice institutions shape who we are, our talents, and what counts as a talent. For example, it is not relevant to social justice to determine how a Maasai warrior, who is highly valued because of his skills relevant to protecting and herding cattle, might fare in another society or a state of nature. Social justice applies to the social institutions that shaped his interests and nurtured his talents relevant to herding cattle. I have argued that reciprocity accounts use notions for civil society and hold that individuals have a fixed nature prior to socialization. Consider how this affects the reasonableness of their argument. They hold that one gains a claim to distributive justice by being a member of cooperation and doing one’s fair share. However, if actually 134 reciprocating is a precondition of justice, the reciprocity justification to noncompatriots also applies to compatriot children and does not make sense of the fact that distributive institutions apply to us—in shaping us by education and so on—before we can reciprocate. The point is that reciprocity as a precondition or “existence condition” is unreasonable since distributive justice must govern life-time prospects and it shapes who we are—principles of distributive justice are principles of social justice. Once we adopt these features of the concept of social justice, we can no longer hold that noncompatriots are asking for our cooperative surplus that they have not helped to produce by asking for distribute justice. Instead, they are asking for a cooperative system of social rules governed by reciprocity to govern them and us. Their position is analogous to a compatriot child’s claim to a position in a distributively just society. The child has not yet contributed, but we do not say, “you have no claim on fair life-time prospects until you get a job, vote, etc.!”194 Reciprocity principles of distributive justice instead must expect reciprocal cooperation eventually, but then the response to noncompatriots is unpersuasive. Thus, Rawls’s reasons to focus on the basic structure as the site of justice, and the Hegelian criticisms of contract theories of social justice, should make us doubt that reciprocity is anything but a constitutive condition of a just society. Since we are born into institutional schemes that are just or unjust, we can only reasonably require the capacity to reciprocate as part of scope of justice. As Stark argues, the capacity is morally relevant to reciprocity conceptions of justice. However, Buchanan makes a similar point. He says, “If justice as reciprocity were understood as the thesis that actual (not just potential) contribution is a necessary condition for having rights, it would have even more startling implications-for example, that not just the severely handicapped but all normal children, prior to the age at which they make net contributions to the cooperative surplus, have no rights” “Justice as Reciprocity Versus Subject-Centered Justice,” 233. 194 135 this capacity extends to persons across borders. So to maintain statism, proponents of reciprocity accounts would have to hold that membership in a cooperative scheme in fact determines who will cooperate. Where a scheme of cooperation exists, reciprocity norms judge the cooperation. This modified account would not exclude compatriot children. The response to noncooperative members now is that they are not members of cooperation and so the reciprocity norms governing their cooperation does not apply to them. However, this is no longer a reciprocity defense of the scope of justice. It says instead that cooperation, with its membership rules, generates or triggers new reciprocity principles of justice—that is, as constitutive conditions of justice. So we are back to Blake’s apparent tension in liberal thought. Membership in a cooperative scheme as triggering social justice looks arbitrary. The notion that people “join together” for mutual advantage makes a similar mistake as reciprocity.195 The question of joining together for mutual benefit never arises in questions of social justice. One is born into and shaped by an institutional scheme that is just or unjust. Justice is not a relationship for one generation or a private association; one is born into it, and it occurs over time. So any question of “joining together” is simply a question of whether we owe justice to others. If we do, a just institutional scheme must be established and maintained. In short, as I explained above, mutual benefit as compared to noncooperation is an idea used by historical process doctrines, which is susceptible to the Hegelian criticisms. Finally, Freeman’s idea that we do not have a duty to cooperate with noncompatriots if it is not mutually beneficial as measured against the status quo is 195 See, for instance, Freeman, “The Social and Institutional Bases of Distributive Justice,” 216. 136 similarly unreasonable. He compares reciprocal cooperation to friendship. However, this unjustifiably applies notions of “civil society” to social justice. People freely decide who their friends are, whereas they are born into their background institutions. Again, reciprocity accounts assume that justice is a relation between fully formed rational adults, but this type of relation is only appropriate for private associations within background institutions. 5.6 Conclusion I examined reciprocity arguments for restricting the scope of distributive justice to cooperation among members of a state. These arguments hold that justice only applies to members of cooperation who do their part in cooperation by complying with cooperative institutions. The kind of cooperation and reciprocity that requires distributive justice only occurs at the state level. Since noncompatriots do not cooperate, distributive justice—based on reciprocity—does not apply to them. I explained the response from Abizadeh, among others, that this argument is pernicious because it holds that one must be cooperating by fair terms for distributive justice to apply to one’s interactions. I added to this criticism the claim that the reciprocity argument is incompatible with the concept of social justice. I argued that the Hegelian criticisms of social contract theories apply to reciprocity arguments with respect to the scope of justice. These arguments assume that people have a fixed nature theoretically prior to socialization, and they unjustifiably apply notions of “civil society,” such as agreements to join an association, to society and social justice. The upshot of my argument is that reciprocity arguments do not give a reason to accept statism. Indeed, I concluded that reciprocity must be a constitutive 137 condition of justice, and only the capacity to reciprocally cooperate is morally relevant for determining the scope of justice. Since persons share this capacity across borders, these “statist” Rawlsian arguments instead support humanity-centered egalitarianism. In response to these criticisms, proponents of reciprocity arguments can adopt a practice dependence interpretation. On this reading, reciprocity accounts do not require that cooperation is actually reciprocal or mutually beneficial for distributive justice to apply. They will instead hold that reciprocity is the socially interpreted goal of a practice, and that justice judges the different standards of reciprocity for different practices. This means they can consider what Forst calls “negative cooperation,” such as exploitative and coercive legal, political, cultural, and economic institutions.196 They will therefore avoid the charge of “positivism” or that their account is pernicious since they can judge circumstances where fairness norms do not prevail. Sangiovanni is a proponent of practice-dependence in other works, and Freeman sometimes seems amenable to the view.197 I will examine practice-dependence in the next chapter. I do this because it is a popular Rawlsian constructivist and institutionalist account supporting statism, and it seems to offer a plausible reply to some the arguments of this chapter. Forst, “Transnational Justice and Non-Domination: A Discourse-Theoretical Approach,” 100. For example, Freeman says that each system of social cooperation has its own standards of fairness and reciprocity by which it distributes, but that “Whether these distributions are fair is of course a substantive question of distributive justice,” Freeman, “The Social and Institutional Bases of Distributive Justice,” 208. Thus, at this point in his paper, it seems like he holds that reciprocity is a socially interpreted goal of social cooperation rather than a relation that is a precondition of the application of justice. 196 197 CHAPTER 6 AGAINST PRACTICE DEPENDENCE: SOCIAL JUSTICE AND CONSTRUCTIVISM In recent years, the somewhat stagnant debate about the scope of distributive justice has shifted to methodological issues inspired by Rawls, Michael Walzer, and Gerald Dworkin.198 In the previous chapters I explained that for Rawls the circumstances of justice, including moderate scarcity of resources and conflicting claims to those resources, make cooperation and fair terms of cooperation both necessary and possible. However, within the circumstances of justice, he claims that there will be different conceptions of justice appropriate to different kinds of cooperation depending on the nature of cooperation, its purpose, and so on.199 He mentions the family, voluntary associations like churches and universities, political and social cooperation, and international cooperation as different kinds of cooperation that will likely require different regulative principles. Similarly, Walzer holds that distinct distributive principles depend on the good being created by the common good association the principle is meant to regulate.200 I will not discuss Dworkin’s view. He uses it for different purposes and explicitly rejects it as a method for justice since “justice is our critic.” 199 For example, TJ, 457; PL, 261; JAF, 10-11. 200 Walzer, Spheres of Justice: A Defense of Pluralism and Equality. He also holds that we ought to justify these principles within the cultural “sphere” of the common good association by interpreting the “actually existing morality”—our cultural understandings of moral categories, relationships, and so on—because 198 139 Recently, many statists claim that we should use a constructivist methodology, termed “practice dependence,” to determine the justification and scope of distributive principles of justice. 201 They claim that this sort of methodology is what Rawls and Walzer had in mind when distinguishing between types of cooperation and the regulative principles for each type. Practice dependence accounts of justice began with an influential interpretation of Rawls by Aaron James.202 James explains that the practice dependence account that comes from his interpretation is a kind of “constructivist methodology,” and he notes that he is influenced by Korsgaard’s understanding of constructivism.203 Recall that Korsgaard holds that constructivism is the view that moral principles are responses to practical interaction problems. The principles are “constructed” by human reason to solve human problems in large part by examining the problems themselves. The practice dependence account adds a restriction regarding the construction of principles and their scope to Korsgaard’s account. Proponents claim that a social interpretation of the goals or function of a practice, and the function justice is meant to play within the practice, should be used in part to justify and determine the kind of principles of regulation for the practice. The goals or function of the practice also determine the scope of the principles by determining the scope of the practice. Thus, the they are constitutive of our identity (Walzer, 21.). Sangiovanni calls Walzer’s view a “cultural conventionalist” type of practice dependence theorist. Sangiovanni, “Justice and the Priority of Politics to Morality,” 8–10. I will only discuss what Sangiovanni calls “institutionalism,” which is the “Rawlsian version” of practice dependence. Sangiovanni, sec. II B. 201 The name comes from Sangiovanni. James just calls it a constructivist method. I examine the view as supporting statism. However, Ronzoni argues, using the method, that it can support a move toward cosmopolitanism. It can support the establishment of a new practice to make global interaction free and fair. She argues that global interaction can constitute, given certain empirical conditions, a case of background injustice. Ronzoni, “The Global Order: A Case of Background Injustice? A PracticeDependent Account.” 202 James, “Constructing Justice for Existing Practice: Rawls and the Status Quo.” 203 James, Fairness in Practice: A Social Contract for a Global Economy, xxii; James, “Constructing Justice for Existing Practice: Rawls and the Status Quo.” Fairness xxii. 140 scope of distributive justice, on this method, is determined by demarcating the scope of the practice it properly regulates. What is distinctive about practice dependence constructivism is how it determines the scope of the practice that justice addresses by using social interpretations of the moralized aims of practices, such as cooperation for reciprocal benefit.204 Rawlsian statists use this method to argue that distributive justice principles are the proper response to the practice of social cooperation within a state given the function of social cooperation. For example, practice dependent interpretations of the reciprocity accounts in the previous chapter hold that political and social cooperation among members of a state has the socially recognized goal of reciprocal benefit of necessary collective goods. Principles of distributive justice are justified using this social interpretation for the practice and the practice defines the bounds of the principles. International cooperation is a distinct practice with a distinct function, for example, reciprocally augmenting national income, which requires less than distributive equality— for example, it might require sufficiency or principles of procedural fairness in trade.205 In section 1, I explain practice dependence. I also show that this interpretation of reciprocity accounts seems to avoid the charges of status quo bias and perniciousness. In section 2, I explain that stressing the fact that practice dependence is a constructivist account of social justice helps it respond to its main criticisms. I show that the social Eva Erman and Niklas Moller, “What Distinguishes the Practice-Dependent Approach to Justice?” Philosophy & Social Criticism 42, no. 1 (2016): 3–23. argue that practice dependence is not a distinct method of theorizing about justice. I only examine the practice dependent answer to the scope question. This answer is indeed distinctive from other forms of constructivism. Practice dependence accounts use a social interpretation of the morally legitimate goals of a practice to define its bounds and claim that justice applies to each practice. 205 Sangiovanni’s view and James’s view, respectively. Sangiovanni, “Global Justice, Reciprocity, and the State”; James, Fairness in Practice: A Social Contract for a Global Economy. 204 141 aspects of justice help to rebut Laura Valentini’s and Saladin Meckled-Garcia’s arguments against practice dependence. In section 3, I show that the social aspects of justice—ones that proponents of practice dependence accept—imply that social justice is not like the regulative principles for other practices. For example, social justice applies to our lifetime prospects, whereas other practices are voluntary. Practice dependence uses shared understandings of the moralized goals of a practice, such as reciprocity, to delineate who and what justice regulates. Because of this, the practice dependence method for restricting the scope of justice is incompatible with the social aspects of justice. In other words, similar to reciprocity accounts and historical process doctrines, practice dependence is susceptible to the Hegelian criticism that it unjustifiably applies notions from “civil society” to social justice and society as a whole.206 If I am right about rejecting the practice dependence way of demarcating the scope of justice for Rawslian constructivism, then the method cannot save the reciprocity case for statism. Instead, my case for the institutional approach to the site of justice shows that reciprocity arguments, once practice dependence is rejected, imply that distributive justice is global in scope, that is, distributive justice is “humanity-centered.” The case begins with the problem of conflicting claims, which includes all those in the problem of conflicting claims. The attempts to further restrict the problem to state practices are rejected. I argued in the previous chapter that only the capacity to reciprocally cooperate can restrict the scope of justice. Thus, justice as reciprocity applies to all those with the capacity to reciprocally cooperate in the problem of 206 The other criticism is that it assumes people have a fixed nature theoretically prior to socialization. This criticism only applies to contract arguments that compare one’s situation in a practice to not being a member of the practice. It is not essential to practice dependence that it provide a contract argument comparing the state to the state of nature, for instance, so I will focus on the first criticism. 142 conflicting claims. 6.1 Practice Dependence The main proponents of practice dependence are Aaron James and Sangiovanni. Both James and Sangiovanni agree that practice dependence is a kind of constructivism. Recall that Korsgaard holds that constructivism is the view that justice is the solution to practical interaction problems. We construct principles of justice from human reason by examining the very problem being addressed along with other moral ideas. Practice dependence adds to constructivism a central role of “social interpretation” for constructing regulative principles for the problems or “practices” in question.207 According to James, the method starts with examining existing social practice.208 He quotes Rawls’s characterization of “social institutions” or “practices” as “any form of activity specified by a system of rules which define offices, roles, moves, penalties, defenses, and so on, and which gives the activity its structure,” for example, “games and rituals, trials and parliaments, markets and systems of property.”209 At the stage of interpretation, we must look to history and politics to determine the function or aim of the practice and the role justice is meant to play as governing the practice. After the interpretive stage, we use the social interpretation to in part determine the scope, content, and justification of our conception of justice—this is the critical or normative stage. As Sangiovanni’s “Practice-dependence Thesis” states, “The content, scope, and justification Sangiovanni, “Justice and the Priority of Politics to Morality,” 6. Also see James, Fairness in Practice: A Social Contract for a Global Economy, 27–31; James, “Constructing Justice for Existing Practice: Rawls and the Status Quo,” 282–86, 300. 208 James, “Constructing Justice for Existing Practice: Rawls and the Status Quo,” 282. 209 James, 283. 207 143 of a conception of justice depends on the structure and form of the practices that the conception is intended to govern.”210 This is only so for the first principles for certain practices.211 In particular, proponents hold that first principles of justice are needed for political and social institutions, and practice dependence methodology is most reasonable for justifying these principles. Institutions are formal or informal rules, procedures, and norms that regulate the political or social activity. The political and social activity determines the division of resources of which there is a competitive struggle.212 In other words, social and political institutions form the practices that justice is meant to govern. For example, Sangiovanni says, “in justifying any conception of justice, we first need an interpretation both of the point and purpose of the institutions that the conception is intended to govern, and of the role principles are intended to play within them.”213 Institutions that enforce their rules put people in special relationships, which gives rise to justice. How people understand their institutional relationships, their goals and function, shapes justice and its scope for their practice. Sangiovanni and James hold that there are three steps to practice dependence methodology.214 First is the “pre-interpretive stage” (Sangiovanni’s term) or “identification” (James’s term), which is where one assumes the object of interpretation, for example, the state. This is simply to fix the basic contours and agree on what practice one is interpreting. It should be done in uncontroversial sociological terms. Second is Sangiovanni, “Justice and the Priority of Politics to Morality,” 2. Sangiovanni, 2. 212 Sangiovanni, 6. 213 5. 214 James, “Constructing Justice for Existing Practice: Rawls and the Status Quo,” 282; James, Fairness in Practice: A Social Contract for a Global Economy, 26–28; Sangiovanni, “Justice and the Priority of Politics to Morality.” 210 211 144 what Sangiovanni calls the “interpretive stage,” which involves two steps. Step 1 is to determine the purpose of an institution by social interpretation. For example, this may involve looking for ideas in preambles and constitutions as data to individuate the practice. Step 2, “the interpreter assumes the point of view of the participants in order to reconstruct what reasons they might have for affirming its basic rules, procedures, and standards.”215 In the interpretive stage the goal is to interpret the entire system of institutions as a whole practice, such as the state with its many institutions, including law, defense, etc. At this stage, we ask how the concern for justice arose in its historical and political circumstances and what role justice is meant to play in the practice. James terms this step “moralized characterization.” He claims that the interpreter aims to characterize the “general purposes, aims, or nature of the practice as independently identified.”216 The characterization can use moral concepts but it is a social interpretation of the practice, not a normative argument. The interpretation of a practice comes from the shared understandings of its function and aims. However, we do not use particular shared understandings or majority opinion. Instead, the aim is an interpretation that “yields the most consistent, coherent, comprehensive, illuminating interpretation of its various elements as a whole.”217 This does not mean, however, that the interpretation is detached from shared understandings discovered by social research. For example, James’s aim in Fairness in Practice was to “understand the global economy [as a practice] in terms that hew closely to its recognized rationales—especially its economic rationales—as they came to be clarified, defended, and implemented over the two hundred or so year period Sangiovanni, “Justice and the Priority of Politics to Morality,” 12. James, Fairness in Practice: A Social Contract for a Global Economy, 28. 217 James, 29. 215 216 145 since Adam Smith’s Wealth of Nations.”218 In short, the second stage offers an interpretation of the moralized goals of a practice and how justice is meant to function within the practice. Only practices with morally legitimate purposes—practices that potentially can be justified to those affected—are given such an interpretation. For instance, slavery is a morally illegitimate practice, and so no matter how it is structured by regulative principles, it is unjustifiable. The final stage, the “post-interpretive stage” (Sangiovanni) or “moral assessment” (James), is critical or normative. At this stage one uses the social interpretation to “provide structure” for the justification or it “conditions the appropriate criteria” for justice.219 However, substantive moral reasoning is also used. For James, we ask what way of organizing the institutions of the practice are reasonably acceptable to all parties involved. In other words, we ask how existing practices would have to be reorganized to be justified. The justification is “framed by the underlying (perhaps moralized) conception of the practice and the interests that are morally relevant in the context in question.”220 Principles cannot be justified in the abstract, nor can they be derived purely from social interpretation. Instead, they must be justified through substantive moral reasoning “specifically for, and from, an independent conception of the practice in which the principles are to have a regulative, governing role.”221 Sangiovanni similarly claims that moral principles, such as the equality of persons, can be used at this stage, along with the social interpretation, to justify principles of justice.222 For Sangiovanni, political institutions are first introduced for Hobbesian goals of Aaron James, “Reply to Critics,” Canadian Journal of Philosophy 44, no. 2 (2014): 286–304. Sangiovanni, “Justice and the Priority of Politics to Morality,” 14–15. 220 James, Fairness in Practice: A Social Contract for a Global Economy, 28. 221 James, 27. 222 Sangiovanni, “Justice and the Priority of Politics to Morality,” 11–12. 218 219 146 mutual preservation or to facilitate trade. Once state or international institutions exist, they structure our lives and put us in special relationships—we are members of practices. The practice then structures how justice is supposed to regulate how we interact within the practice. In other words, people join together first in a practice and then justice concerns arise to regulate the specific practice. This is why, according to James, it would be a mistake to simply, for example, apply Rawls’s original position globally. Instead, the grounding of the original position argument, or constructivist one’s like it, depends on the independently identified context and social structure, which gives rise to fairness concerns.223 Similarly, for Sangiovanni, distributive justice concerns only arise within a state because co-member cooperation within a state requires reciprocity in the mutual provision of the goods necessary to act upon a plan of life. 6.1.1 Reciprocity, Status Quo Bias, and Perniciousness Reciprocity accounts, interpreted as practice dependent, can use moralized ideals as the social goals of certain types of cooperation. That is, they can claim that certain types of reciprocity are the function of practices. This is determined, not by moral argument, but by using informed social science and history. For Sangiovanni, a state’s purpose is to reciprocally provide the goods necessary to have and act upon a plan of life. For Freeman, social cooperation’s goal is to reciprocally benefit cooperators with its political and economic system. International cooperation has a distinct form of reciprocity with its own product, for example, national gains on trade. Similarly, according to James, international trade is a practice among nations using a shared market 223 James, Fairness in Practice: A Social Contract for a Global Economy, 25–26. 147 to augment national income. This shared market, the agreements to liberalize trade, has the aim of providing assurances to nations of long-term gain.224 We can justify principles of distributive justice for the practice, which also justifies the practice, by interpreting the type of reciprocity and cooperation involved in the practice and using practice independent moral principles, such as equality of persons and intersubjective justifiability. Once distributive principles are justified for the practice, the response to noncooperators is that they do not have a claim to the cooperative product based on reciprocity because a distinct standard of reciprocity applies to international cooperation, which is a distinct practice with its own type of reciprocity. This interpretation helps reciprocity accounts reply to the charge of status quo bias. The practice dependent view also might seem biased in favor of the status quo or against nonmembers since it starts with the moral ideals of existing practices. However, proponents claim instead that it is sensitive to existing practice and the status quo but not biased. This is because nonmembers can complain about exclusion, for example, but they cannot complain about participation-based claims such as exploitation.225 These claims also mean that practice accounts have a way of replying to the charge of perniciousness. One can criticize the function of a practice from the standpoint of another practice or from other moral ideals on the practice account. This is how they reject the practice of slavery, for example. The goal of the practice is morally illegitimate, and so there is no way to change the rules of the practice to make it a legitimate practice. 224 225 James, chap. 2. James, “Constructing Justice for Existing Practice: Rawls and the Status Quo,” 312. 148 6.2 Practice Dependence and Institutional Social Justice In this section, I show that the predominant criticisms of practice dependence are met by stressing that practice dependence statists are constructivists focused on social justice. I first explain Laura Valentini’s conceptual challenge to practice dependence. I show that any reasonable response to her will involve the claim that social justice regulates social institutions, which practice dependence proponents accept. I show next that since practice dependence is a constructivist account for social justice, it avoids Saladin Meckled-Garcia’s “justificatory” challenge. Stressing the social aspects of justice helps the practice account respond to its main critiques. However, I will argue in the next section that the practice account conflicts with the social nature of justice. Valentini charges practice dependence with a conceptual problem. The problem is that it is unclear how practice dependence proponents can explain the concept of justice as separate from other moral concepts and as unified across different practices. Valentini claims that, [B]y suggesting that the very function of justice changes across contexts, [practice dependence accounts] seem to end up robbing justice of its normative distinctiveness. They tell us that different conceptions of justice apply to different institutions in light of our different interpretive understandings of them, but do not tell us how such conceptions represent interpretations or instantiations of the same concept.226 For example, many practice dependence advocates hold that state institutions establish cooperation for reciprocal benefit among members of the practice, which requires distributive principles of justice. They hold that international institutions, on the other hand, only require principles of sufficiency or principles of structural fairness. What Valentini, “Global Justice and Practice-Dependence: Conventionalism, Institutionalism, Functionalism,” 412–13. 226 149 makes these distinct conceptions of justice instances of the same concept, “social justice”? What distinguishes the concept “justice” from other moral concepts, such as “beneficence”? The practice dependence method cannot rely on its substantive practice independent moral principles since these principles do not plausibly distinguish justice. For instance, Valentini points out that James’s principle of intersubjective justifiability does not distinguish justice from beneficence—both virtues are justifiable to others.227 Nor does Sangiovanni’s or James’s use of the principle of moral equality of persons help to distinguish the concept “justice” since it can be used to support many other ethical principles. Valentini concludes that justice looks like a completely different concept for each practice on this account since the function of each practice determines what justice is.228 Valentini argues that we should hold instead that justice has one function to unify the concept across distinct practices and to distinguish it from other moral concepts. She notes that cosmopolitans focus on social institutions and interactions that in fact shape life-prospects and opportunities from which people cannot escape instead of appealing to interpretive understandings of the goals of existing institutions. She holds that the function of justice is “respecting persons’ equal right to lead lives they deem worth living,” so cosmopolitans are right to think that how we in fact shape each other’s holdings and opportunities should be consistent with this equal right.229 She notes that 227 For example, James, Fairness in Practice: A Social Contract for a Global Economy, 134. The other option Valentini considers is that practice dependence accounts hold that co-members of states affect each other in the most important ways for cooperation and so social justice only applies to comembers. However, this narrow definition of justice “turns a contingent (and perhaps superseded) empirical fact—the fact that respecting other people’s spheres of agency requires distributive justice only within domestic political communities—into a conceptual truth about justice—the idea that justice is about the distribution of certain kinds of goods between citizens.” Valentini, “Global Justice and PracticeDependence: Conventionalism, Institutionalism, Functionalism,” 412. 229 Valentini, 414. 228 150 statism is much harder to support once we give justice a unified function because cosmopolitans are also right to point out that global institutions greatly affect our opportunities to act and our life-prospects. However, she adds that our beliefs about empirical facts will ultimately determine whether we are cosmopolitans regarding distributive justice, or what principles we think apply globally and domestically, on her “functionalist” account. A practice dependence response to Valentini must explain the concept of justice while still making conceptions of justice practice dependent. They might start by, for example, holding that justice is a secular virtue; it is enforceable; it impartially responds to conflict regarding resources, power, and opportunities; it organizes social and political institutions; and it applies to, and must be acceptable to, moral equals.230 They can then hold that how the concept is instantiated, or how separate conceptions are justified and their scope is defined, is shaped by the practices the virtue regulates. Proponents of practice dependence clearly accept that justice regulates social institutions. It is also clear that their reasons for accepting this claim in part rely on Rawls’s reasons for thinking that justice is the first virtue of institutions. Further, in Part 1, I argued that Rawlsian constructivism, with its conception of the person, supports relegating distributive justice to organizing socioeconomic institutions. Thus, since practice dependence accounts are Rawlsian, it is reasonable to think that the social aspects of justice can help distinguish the concept from other virtues. James offers a way of thinking about the social nature of justice by thinking about collective power. His reasons for why social justice applies to social and political 230 These features were given in Chapter 4. 151 institutional practices go some way toward answering Valentini.231 He claims that Rawlsians support the idea that certain principles of social justice apply only to the basic structure as a two-step process. First, they separate practices as the subject of justice as opposed to individual interactions. Second, they argue that certain principles of social justice apply to certain practices. James notes that Rawls’s reasons for focusing on the basic structure as the primary subject of social justice give some support to his claim. However, Rawls does not fully show why social practices require principles of a special moral kind. James aims to fill this gap by claiming that social institutions exercise collective forms of power that must be justifiable to those affected. Those affected also exercise collective responsibility for this power. Social justice does not directly apply to individual actions because no one person is responsible for the power of social institutions. In short, the special moral significance, requiring distinct principles of right, is that social practices have a collective social power that no one individual is responsible for.232 This state of affairs gives rise to a special occasion for justification since no one person is responsible for the power. And social justice appraises coordinated activity where the coordination exercises collective power.233 However, James notes that this explanation simply adds a general mark of importance that helps explain why practices need distinct principles of right from individual actions.234 He recognizes that this still would not completely unify the concept of social justice. Collective power is too broad a problem to distinguish justice. For example, principles that regulate football and other private voluntary associations would count as principles of justice, which does not James, “Power in Social Organization as the Subject of Justice.” James, 42. 233 James, 42. 234 James, 45. 231 232 152 reasonably capture our concept “social justice.” He suggests that to give a full account of justice, we would need to at least fill out the specific kind of power exercised. I take it that James means to further fill out the concept of social justice in part by explaining what kinds of collective power it is meant to regulate. That is, certain institutional practices have a certain power that requires that they be just as a first virtue rather than expedient, benevolent, and so on. This is important because practice dependence defenders also need to hold that there is a general problem or family of problems or practices that justice regulates to unify the concept. Korsgaard uses the example of a chair as the solution to a human problem to help explain constructivism. I think her example also shows why practice dependence constructivism must hold that there is a family of practices that justice regulates to unify the concept. Korsgaard says that the concept of a chair is the solution to a general human problem of needing a place to sit. However, there are different conceptions of a chair or different instantiations of a chair for the particular problems being addressed, such as a camping chair that folds up and is light weight or a reclining sofa. Thus, Korsgaard’s problem of needing a place to sit is really a family of problems that can be subsumed under the general problem of needing a place to sit. The concept of justice must also function like this to unify it and distinguish it from other moral concepts on constructivist accounts. It is not enough to claim that justice is institutional and involves collective power since this will include voluntary private associations. Korsgaard explains that constructivism is a long-standing method in political philosophy. The problems addressed by constructivists always include a conflict of interests, and in recent discussions of social justice, the conflict is focused on moderately scarce resources, opportunities, and powers. In other words, the 153 circumstances of justice are being responded to even when more specific relationships of conflict are addressed within the circumstances. So to unify the concept of justice, Rawlsian practice dependent proponents can at least hold that justice regulates social and political institutions to adjudicate between conflict among those who cannot be expected to share ideas of the good. Social and political institutions exercise power, in James’s framework, over people who cannot be expected to share ideas of the good. This helps to distinguish justice, which is traditionally meant to apply across ideas of the good and religious doctrines, from voluntary associations of people to which justice does not directly apply.235 In short, the general practice is the circumstances of justice and distinct conceptions of justice will apply to practices involving distinct types of conflict of interests within these circumstances. Practice dependence accounts can add the following features of social and political institutions, and social justice, to unify and distinguish the concept of social justice. I explained above that Rawls holds that social justice and social institutions are involuntary institutional relations that one is born into and is usually affected by over a life-time. “Social justice” is social in part because the institutions it regulates shape who we are, they maintain the background conditions to our interactions and associations, and they determine our life-time prospects starting from birth. These are some of the features that distinguish social justice from criminal justice and social justice from other related moral concepts, such as beneficence. These features begin to answer Valentini’s conceptual charge since the concept is 235 We can also use our historical concept of social justice to go back and forth to refine our concept of justice and what it is a response to. We do not have to have a specific order where we must first define the problem that justice addresses before we are clear on the concept. 154 unified as a social institutional virtue that responds to a social conflict between persons over resources, rights, powers, and opportunities. According to the practice view, there are different instantiations of justice based on what the virtue is an answer to or what problem within a practice it regulates. In short, practice dependence proponents will aim to distinguish practices, which require distinct conceptions or instantiations of justice, within the larger problem of the circumstances of justice. These features of practice dependence accounts also help to show that another prominent criticism of practice dependence, given by Meckled-Garcia, is off-base. He claims that practice dependence determines the scope of principles of distributive justice using methodological considerations. There are distinct conceptions of justice for national and international cooperation because they are distinct practices determined through non-normative social interpretation. The distinction between practices determines substantive moral standards—one for national and one for international practices. Meckled-Garcia argues that this fails his “justifiability constraint” for any account of moral principles. The constraint says, “Any argument supporting a substantive moral standard for a given group of people must be justifiable by reference to at least one moral value or independently derived moral principle.” Practice dependence offers substantive moral conclusions, including the claim that we do not owe distributive justice to noncompatriots. However, they rest these claims on methodological considerations based on the socially determined functional role of a practice. In response to Meckled-Garcia, practice dependence proponents can say that their view is a constructivist account of social justice. Justice is an answer to a human interaction problem or practice. The concept of social justice is determined in part by 155 determining what it is a response to—that is, the family of practices or problems I outlined above. This means that the scope of substantive moral requirements will be determined by social facts rather than normative moral principles on the constructivist method. Since constructivists think that moral principles are solutions to human interaction problems, Meckled-Garcia’s argument begs the question. For a constructivist, one’s rights and entitlements are determined by substantive moral principles, which are responses to the interaction problems people are in. If they are not in certain sorts of social interactions, then certain moral principles will not apply to them. 6.3 Social Justice is not Practice Dependent The common features of social justice explained in the previous section help the practice dependence account reply to critics. However, the features also, I claim, show that the practice dependence way of determining the scope of justice is unreasonable given the social nature of justice and the social institutions that it regulates. Recall that the problem with the moral relationships of mutual benefit and reciprocity as preconditions for the application of justice is that they are unreasonable given the concept of social justice. They unjustifiably apply notions from “civil society” to social justice. As such, reciprocity accounts make justice out to be a virtue for voluntary associations. The practice account similarly makes this mistake.236 Recall that social justice is social in that it organizes social institutions that one is In the next chapter I explain Anderson’s criticism of using local principles, especially merit, to govern entire socioeconomic systems. She explains that it is pernicious since it supports one class’s exalted virtues. It is premised on class warfare. It therefore cannot be an answer to the problem of conflicting claims. Pluralist peoples will not find these class-specific principles acceptable. This point strengthens my Hegelian arguments in this chapter and the previous one. 236 156 born into and that determine life-prospects, shapes interests, and maintains the background of their associations and interactions. It is also social in that it fairly adjudicates between the conflicting interests of people who cannot be expected to share ideas of the good. Private associations, by contrast, are voluntary (if they take place within just, or relatively just, background institutions), usually do not affect one’s prospects from birth, and can expect their members to share goals. Determining what practice or problem social justice responds to must at least take these social features of the concept of justice into account. That is, delineating the practice must consider how social justice is distinct from voluntary associations. The practice account uses a social interpretation of the moralized goals of a practice to delineate it. Practice dependence accounts do not require that social cooperation actually be mutually beneficial or reciprocally beneficial; they hold instead that moralized goals define the practice that justice regulates. That is, they separate justice problems not by actual relationships but by the socially interpreted aims of practices. Once this part of the account is filled in, they claim reciprocity with respect to the social product—or some other moral notion—is the aim of state social cooperation, and augmenting national income through a mutually reliant market system is the functional aim of international cooperative institutions. It is the socially interpreted goal of practices, which is always some moralized concept, that conflicts with social justice.237 The Rawlsian constructivist concept of social justice is the response to distinct agents who cannot be expected to share final ends because of their separate agency. 237 It is always a legitimate moral concept. For example, James excludes the black market as part of the global trade practice since it is illegitimate. It exists because nations cannot stop it. James, Fairness in Practice: A Social Contract for a Global Economy, 39. 157 These agents are also born into a society with certain background rules, which shape their interests, and so on. Because of this, Rawls, for instance, says that society will have no shared ends other than those that are part of the conception of justice (JAF, 20). Practice dependence accounts, by starting with socially interpreted ends of practices to both justify the scope of the practice and in part justify the principles of justice for it, attribute such a shared goal to members of the practice. Specifically, the justification of the scope and principles requires the acceptability of the conception. If people accept a practice dependence conception of justice, they also accept that their practice has a certain moralized aim, such as reciprocal cooperation. However, members of society cannot share goals other than justice. This is what makes social justice a social problem rather than a problem of voluntary association. The idea that members of a practice can share goals other than justice, or that a practice is delineated by its moralized aims, is a notion only appropriate for within civil society. In other words, the socially determined moralized aim of a practice unjustifiably applies a notion from within “civil society” to social justice. Recall that notions for civil society are only appropriate for civil society because we need to assume a just background structure of society to use them. For example, Nozick’s “natural liberties,” or "side constraints,” for property acquisition and exchange do not act to preserve fair background conditions for society. As Rawls says, agreements result from “contingencies and accidents of the as-if just historical process which has no tendency to preserve or to move toward background justice” (PL, 287). Instead, Nozick’s principles focus on individual transactions between fully formed rational adults. However, to know whether these transactions are free and fair, we need to know under what background 158 conditions they occurred. Thus, the notion of contractual agreements among rational adults is appropriate for civil society but not society as a whole since it ignores socialization, the involuntariness of society, among other things. By defining the scope of justice using socially interpreted moralized goals, practice dependence makes justice the solution to a problem of a voluntary association rather than a social virtue that organizes social institutions. Determining the scope of a virtue that regulates a private association or practice within just institutions can be determined by interpreting the association’s socially interpreted goals. This is because the background institutions make joining or leaving the association, for example, free and fair, so a voluntary association can require accepting shared goals. The voluntariness, established by fair background conditions, legitimizes the shared goal. However, using James’s terminology, social institutions and social justice exercise a different kind of “collective power.” Socially just or unjust institutions are not voluntary: you do not normally choose to join or leave, you cannot choose to not be shaped by them, you cannot escape how they maintain the background of your interactions. Thus, the practice, or problem, that social justice applies to cannot be delineated by the socially determined aim of the practice. These notions are unjustifiable for social justice since social justice is meant to organize institutional structures that people are born into and shaped by in order to adjudicate between conflicting claims. I conclude that we should reject practice dependence as a way of saving the statist reciprocity arguments. Thus, my argument from the previous chapter—that only the capacity to reciprocate can be reasonably required for distributive justice, and thus Rawlsian distributive justice is humanity-centered—still stands. In the next chapter I 159 offer further support for my claim that Rawlsian distributive justice is humanity-centered. As part of my argument, I will use Elizabeth Anderson’s distinction between local and systemic principles of justice. Her arguments will offer further support to my Hegelian points in Chapters 5 and 6. CHAPTER 7 COSMOPOLITAN CONSTRUCTIVISM: A HUMANITY-CENTERED APPROACH TO THE ECONOMIC DIVISION OF LABOR In Part 1 of the dissertation, I offer a constructivist case for the institutional approach to the site of distributive justice. I begin my case in Chapter 2 by defending the internal consistency of institutionalism. I then ask whether the institutional approach is justified. My arguments in Chapter 2 point to constructivism and value pluralism as potential grounds for the view. In Chapter 3, I show that the idea of a division of labor, as making space for value pluralism, is not a good reason to accept institutionalism. Tan’s and Rawls’s idea that justice must frame pluralistic ideas of the good is more reasonable, but I argue that act-egalitarianism might also be able to frame ideas of the good. I present my case for institutionalism, which is not susceptible to the actegalitarian reply, in Chapter 4. In that chapter, I argue that we should accept institutionalism not because it is less demanding than anti-institutionalist accounts but because the institutional approach to distributive justice is the only approach acceptable to people in a society with value pluralism. My case for institutionalism relies on an interpretation of constructivism, where the problem of distributive justice is interpreted as the problem of conflicting claims. (I call this interpretation of constructivism “Rawlsian 161 constructivism.”) Conflicting claims, recall, is framed by assuming a conception of the person as a moral agent. Since a society of moral agents will lead to value pluralism, their claims to rights and resources based on their incompatible values will conflict. I use this framework to compare Rawls’s institutionalism with the two main antiinstitutionalist accounts, utilitarianism and act-egalitarianism. I argue for two claims. First, if the concept of distributive justice, as per constructivism, refers to the solution to conflicting claims and the reasoning for conceptions of justice (again, as per constructivism) represents conflicting claims, then Rawls's institutionalism is justified, and the two anti-institutionalist conceptions are not justified. Second, if we examine the common justifications for utilitarian and act-egalitarian conceptions, we find that their natural justifications do not represent conflicting claims; a different conception of the person is represented instead. I conclude that if we accept conflicting claims as the problem of distributive justice, then principles of distributive justice should only organize the institutional rules of the game; they should not also apply to individual economic interactions. Proponents of Rawlsian constructivism, including Rawls himself, often support statism. However, the problem of conflicting claims uses a conception of the person that crosses borders. So Rawlsian constructivism seems to lead to humanity-centered egalitarianism, which is a cosmopolitan perspective. In Part 2, I examine and reject the main reasons Rawlsian constructivists give to restrict distributive justice to compatriots. My primary aim is to offer a positive argument for humanity-centered egalitarianism as an implication of the Rawslian constructivist account I used to support institutionalism in Part 1. I aim to show that each argument for restricting the scope of Rawlsian 162 constructivism to the state, when made consistent with the conception of the person and society used to frame Rawlsian constructivism, supports humanity-centered egalitarianism instead. Thus, if we accept Rawlsian constructivism as the most reasonable way of supporting institutionalism, we have good reason to be cosmopolitans. With this aim in mind, I argue, in Chapters 5 and 6, that reciprocity and practice dependence fail to support statism. At most, they show that the capacity to reciprocate determines the scope of distributive justice; but this capacity is shared across borders, so the arguments support humanity-centered egalitarianism. In this chapter, I conclude the dissertation by arguing that a premise that most anticosmopolitan Rawlsian accounts share in fact implies cosmopolitanism. Arash Abizadeh calls the main anticosmopolitan Rawlsian argument “the basic structure argument” about the scope of justice.238 Many readers of Rawls think his case against cosmopolitanism rests on two claims: that distributive justice only becomes relevant in the context of a basic structure, and that there is no global basic structure. According to Abizadeh, there are three main ways, which all have some basis in Rawls, that anticosmopolitans interpret a basic structure to claim that that structure makes distributive justice “relevant.”239 First, the basic structure is a set of basic coercive institutions. Second, the basic structure is a set of institutions that determine and regulate the terms of social cooperation. Third, the basic structure is a set of institutions that has a profound and pervasive influence on life chances. On these accounts, the scope of I say “for the scope of distributive justice” since I also discussed the basic structure argument for the site of justice in Chapter 2. Abizadeh disambiguates the Rawlsian site and scope arguments. Abizadeh, “Cooperation, Pervasive Impact, and Coercion: On the Scope (Not Site) of Distributive Justice.” 239 I explained in Chapter 5 that Abizadeh shows that the basic structure must make distributive justice “relevant” as an existence condition for the basic structure argument to be valid. 238 163 distributive justice extends to the set of people subject to the same basic structure, whichever of these three ways that structure is understood. Rawlsians often offer some combination of these basic structure interpretations to support the idea that distributive justice only becomes relevant within the confines of an existing state’s basic structure. There are three main challenges to this argument.240 First, Rawlsians need to show that the basic structure relation is not global in scope; that is, they must show that there is no global basic structure.241 Second, Rawlsians need to explain why the basic structure relation, such as coercion, makes distributive justice (rather than allocative, retributive, etc.) the relevant regulative concept.242 Third, Rawlsians must offer a reasonable concept of distributive justice, which captures our intuitions about the concept and is consistent with relevant empirical research. In this chapter, I claim that each version of these Rawlsian arguments shares a key premise, which I call the “division of labor thesis.” Briefly, the division of labor thesis holds that distributive justice becomes relevant only where members of a division of labor jointly produce some socioeconomic product. In section 1, I show how the most reasonable versions of the Rawlsian basic structure argument are constructivist, which is, I argue, what allows them to use the division of labor thesis. They hold that a state’s basic structure organizes a division of labor, and the concept of distributive justice 240 I explained and offered several other challenges in the two previous chapters. For arguments against the claim that the basic structure relation only applies to the state, see for instance Abizadeh, “Cooperation, Pervasive Impact, and Coercion: On the Scope (Not Site) of Distributive Justice”; Buchanan, “Rawls’s Law of Peoples: Rules for a Vanished Westphalian World”; Caney, Justice Beyond Borders: A Global Political Theory; Hassoun, Globalization and Global Justice: Shrinking Distance, Expanding Obligations; Moellendorf, Cosmopolitan Justice; Pogge, Realizing Rawls; Cohen and Sabel, “Extra Rempublicam Nulla Justitia?” 242 For example, see Freeman, “The Social and Institutional Bases of Distributive Justice,” 201; Moellendorf, Global Inequality Matters, chap. 2. Caney also argues that institutions are not relevant to the concept of distributive justice. Justice Beyond Borders: A Global Political Theory, 111. 241 164 becomes relevant with the existence of the social practice of a state. I explain how this interpretation of the basic structure argument helps proponents respond to the first two challenges identified above. In section 2, I examine what I take to be the most plausible response to the third challenge. The response is to accept what Elizabeth Anderson calls “the modern systemic concept of distributive justice.” The systemic concept is the idea that distributive justice organizes entire economic systems with respect to their distributive consequences or what relationships they maintain, rather than only regulating individual transactions.243 Anderson argues that this concept is Rawls’s idea that distributive justice applies only to the basic structure of society.244 This account of distributive justice meets the third challenge because it is consistent with systemic economic explanations of wealth and inequality and matches intuitions—or at least the intuitions supporting Rawlsian constructivism. In what follows, I will assume the modern concept as the correct way of conceiving of distributive justice and the correct interpretation of Rawls’s basic structure claim. In section 3, I show that if we accept the modern concept of distributive justice, we must reject the anticosmopolitan basic structure argument. The basic structure argument’s claim that the existing basic structure of a state makes distributive justice relevant among compatriots is incompatible with the systemic concept of distributive justice. My argument for this conclusion relies on the following three premises. 1) The systemic concept of distributive justice assesses, in part, the nature and scope of a division of labor. 2) According to the basic structure argument, an existing division of Anderson, “Thomas Paine’s ‘Agrarian Justice’ and the Origins of Social Insurance”; Anderson, “Rawls’s Difference Principle.” 244 Anderson, “Rawls’s Difference Principle.” 243 165 labor makes relevant the concept of distributive justice as what properly regulates it. This concept of distributive justice therefore cannot assess the nature or scope of the existing division of labor that occasions it. 3) Existing basic structures, with their existing divisions of labor, cannot make the systemic concept of distributive justice the relevant regulative concept for the basic structure in question. The systemic concept assesses the nature and scope of divisions of labor, so it cannot be the relevant regulative principle for existing divisions of labor. If Rawlsians accept the modern systemic concept, they can hold, at most, that the capacity to participate in a division of labor is relevant for the scope of distributive justice. However, this capacity is almost universal, so it cannot restrict distributive justice to the state. I accept the systemic concept of distributive justice. I therefore conclude my Part 2 argument: the main Rawlsian constructivist arguments for distributive justice imply that distributive justice is global in scope. Rawlsian constructivism supports humanity-centered egalitarianism. In section 4, I conclude the dissertation by briefly discussing whether the difference principle specifically should apply globally. The arguments that I claim apply globally are either arguments for the difference principle or some unspecified principle of reciprocity. I provide reasons to think that the difference principle should be the principle of reciprocity we advocate, but I caution that there are further obstacles for humanity-centered egalitarianism. 166 7.1 Anticosmopolitan Rawlsian Constructivism and the Division of Labor The Rawlsian argument for restricting the scope of distributive justice to the basic structure of domestic society is as follows. 1. The scope of distributive justice extends over the set of persons a) who are subject to their society’s coercive basic structure, and/or b) whose basic structure organizes their cooperation for a social product, and/or c) whose life chances are profoundly and pervasively impacted by their society’s basic structure. 2. The set of persons subject to an existing basic structure, defined by any or all of a-c, does not include all persons. Rather it includes persons as occupants of states. Only states have a basic structure in the relevant sense. 3. The scope of distributive justice is not global. It extends over individual states. In this section, I show how each version of the argument (using a-c) relies on what I call the “division of labor thesis” in order to conclude that distributive justice is only relevant for regulating a state’s basic structure. I show, in other words, that these arguments all maintain that it is only when socioeconomic institutions organize a complex division of labor that distributive justice becomes relevant among the members. Most proponents of the argument type I just outlined are Rawlsian constructivists. According to Christine Korsgaard, “constructivism” is the view that ethics and justice attempt to solve practical problems. She explains that Rawls’s concept/conception distinction in A Theory of Justice illustrates how constructivism with respect to justice works. She says, [T]he concept of justice refers to the solution to a problem. The problem is what we might call the distribution problem: people join together in a cooperative scheme because it will be better for all of them, but they must decide how its benefits and burdens are to be distributed. A conception of justice is a principle that is proposed as a solution to the distribution problem, arrived at by reflecting on the nature of the problem itself.245 245 Korsgaard, The Constitution of Agency: Essays on Practical Reason and Moral Psychology, 322. 167 Principles of justice are constructed from reason in part by examining the very problem being addressed. Rawlsians use this distinction between the concept of distributive justice and conceptions of justice to restrict the scope of distributive justice. It allows them to use Rawls’s idea that “the correct regulative principle for anything depends on the nature of that thing” (TJ, 29). They interpret Rawls’s claim in two different ways, corresponding to two ways of interpreting constructivism. On the first interpretation, the concept of distributive justice refers to the solution to the problem of distribution understood as a practical interaction problem. On the second interpretation, often called “practice dependence constructivism,” a central role for “social interpretation” is added to Korsgaard’s claim that we arrive at a solution to the problem of distribution by reflecting on it. In other words, on this account, we construct regulative principles for the problems or “practices” in question in part by interpreting those practices.246 According to Aaron James, the method starts with examining existing social practice.247 Principles of justice are justified, in part, from and for the social practices they are supposed to regulate. According to Andrea Sangiovanni’s “Practice-dependence Thesis,” “The content, scope, and justification of a conception of justice depends on the structure and form of the practices that the conception is intended to govern.”248 Constructivism, on either interpretation, allows the division of labor thesis to be added to the Rawlsian basic structure argument and so helps it meet the first two challenges. Most Rawlsians admit that their favored basic structure definition, for Sangiovanni, “Justice and the Priority of Politics to Morality,” 6. Also see James, Fairness in Practice: A Social Contract for a Global Economy, 27–31; James, “Constructing Justice for Existing Practice: Rawls and the Status Quo,” 282–86, 300. 247 James, “Constructing Justice for Existing Practice: Rawls and the Status Quo,” 282. 248 Sangiovanni, “Justice and the Priority of Politics to Morality,” 2. 246 168 example, coercive institutions, is at least partly global in scope, which means that at least some form of justice is global in scope. Their claim is not that justice only applies within states but that distributive justice only applies within states.249 Distributive justice becomes relevant as the correct regulative concept for the social practice of a state. A state’s basic structure, as defined by a-c, organizes a division of labor for creating a social product; and distributive justice is the correct regulative principle for this practice. Distributive justice is the relevant regulative principle because the social practice organizes labor, capital, and land to produce a social product; and distributive justice organizes divisions of labor for a social product. The division of labor within a state’s basic structure, Rawlsians claim, is a distinct practice from the division of labor among states established through international institutions. Thus, the practice that requires distributive justice is not global in scope. 7.1.1 Coercion There are several versions of the coercion premise.250 I will focus on Michael Blake’s account since it is explicitly constructivist and uses the division of labor thesis, which means it can better reply to the first two challenges discussed above. Blake holds that coercion is in prima facie conflict with personal autonomy and so must be justified or eliminated. He thinks that only the coercive laws of a state require distributive justice—or what he calls “principles of relative deprivation,” as opposed to principles of Nagel makes the stronger claim that justice only applies within states. Nagel, “The Problem of Global Justice.” The stronger claim makes his view especially susceptible to the three challenges. For a response, see Julius, “Nagel’s Atlas.” 250 Blake, “Distributive Justice, State Coercion, and Autonomy”; Miller, “Cosmopolitan Respect and Patriotic Concern”; Nagel, “The Problem of Global Justice”; Risse, On Global Justice. 249 169 “absolute deprivation” or sufficiency—as their justification. Globally, we only have duties of sufficiency—in particular, a duty to provide others with an absolute amount of resources required for minimal autonomy. Coercion arguments are often criticized by pointing out that state borders and some international institutions are coercive.251 Blake admits that there are coercive international institutions; he maintains however that they do not require distributive justice as their justification. He holds that the justification required for coercion differs depending on the type of coercion involved. In other words, Blake is a constructivist who is offering an interpretation of the “correct regulative principle” claim. He thinks that state institutions include a system of coercive laws, such as tax laws, tort laws, and so on, whose “job” it is to define material holdings. Since these laws define what counts as property, what resources count as private or public, and so on, they collectively define “how we may hold, transfer, and enjoy our property and our entitlements.”252 Only state institutions are directly coercive of individuals and are meant to define holdings. It follows that the justification of state institutions, owed to those coerced, will be a principle of distributive justice, such as the difference principle. Blake outlines a situation of consent to introduce Rawls’s original position as a way of justifying state coercion. He claims that its resulting principles could provide justification for the coercive laws of the state system. In other words, the coercion account partly relies on the claim that distributive justice only becomes relevant in a system that maintains a complex division of labor among its Abizadeh, “Cooperation, Pervasive Impact, and Coercion: On the Scope (Not Site) of Distributive Justice,” 348–49; Cohen and Sabel, “Extra Rempublicam Nulla Justitia?”; Julius, “Nagel’s Atlas”; Moellendorf, Global Inequality Matters, chap. 2; Laura Valentini, “Coercion and (Global) Justice,” The American Political Science Review 105, no. 1 (2011): 205–20; Van Parijs, “International Distributive Justice.” 252 Blake, “Distributive Justice, State Coercion, and Autonomy,” 281; also see 265 and 280; Michael Blake, “Coercion and Egalitarian Justice,” The Monist 94, no. 4 (2011): 563–64. 251 170 members for a social product. Interstate divisions of labor are separable from the international economic system. The international system of borders, trade agreements, and so on, is coercive but will have a distinct type of justification (if the current international practices can be justified at all). Borders are not meant to be “productive” or to define holdings, and thus the justification for them will not require principles of relative deprivation. Similarly, statists often claim that international trade agreements are meant to augment national income,253 and do not directly coerce individuals. Thus, the division of labor thesis helps distinguish a state’s coercive basic structure from international coercion; and the division of labor thesis helps to explain why the coercive institutions of a state require distributive justice, while international institutions require distinct forms of justice. The coercion premise seems to meet the first two challenges. 7.1.2 Reciprocal Cooperation Several readers of Rawls interpret him as offering a cooperation account of the scope of justice. On such an account, a theory of justice is worked out from the ideas of each cooperative practice, and social cooperation includes the idea that members of a division of labor cooperate to jointly produce a social product.254 I explained Freeman’s, Sangiovanni’s, and James’s version of the reciprocal cooperation argument in the two previous chapters. They hold that reciprocity is a virtue of cooperative schemes for joint production. Each scheme will involve its own type of reciprocity depending on the type 253 Aaron James argues that this is the purpose of international institutions. James, Fairness in Practice: A Social Contract for a Global Economy. 254 Freeman, “The Law of Peoples, Social Cooperation, Human Rights, and Distributive Justice”; James, “Constructing Justice for Existing Practice: Rawls and the Status Quo.” Rawls is explicit that “peoples” are partly politically autonomous and so economically distinct (LOP, 117-118), but there is debate on what his anticosmopolitan argument ultimately rests on. 171 of cooperation involved. For example, borders might require reciprocal respect. The relevant type of reciprocity for the cooperation involved in an economic division of labor for a joint social product is distributive justice. The reciprocal cooperation within a state’s economic institutions produces a socioeconomic product, so distributive justice is needed to fairly organize this productive process, with its rights, opportunities, and powers. For example, Freeman says that distributive justice is distinct from economic or allocative justice in that it is “how to fairly organize and distribute economic powers and positions and divide up the resulting joint social product among productive socially cooperative agents.”255 He claims that if distributive reciprocity applies to global institutions, it only applies to the joint product of the productive social activity, mainly gains on trade. We do not have to share the social product we produced with those who did not help produce it. As practice dependence accounts, reciprocity arguments use the social interpretation of state economic practice as a division of labor among members for joint production of necessary economic resources. Reciprocity is taken to be the relevant moral ideal for cooperative social practices. The relevant problem that distributive justice is constructed for is the conflicting claims of members of a division of labor cooperating to produce a joint social product. That is, the division of labor thesis makes distributive justice relevant as the type of reciprocity needed for compatriot economic cooperation. Thus, reciprocity accounts seem to meet the first two challenges. They explain why distributive justice is only relevant among compatriots within their cooperative division of labor, and they distinguish between a state’s basic structure, with its division of labor for a social product, and international institutions with their distinct 255 “The Social and Institutional Bases of Distributive Justice,” 213. 172 production of goods. 7.1.3 Pervasive Impact Pervasive impact accounts of the basic structure hold that distributive justice becomes relevant when people are subject to the same set of basic institutions that profoundly and pervasively impacts their life chances from birth.256 This definition is usually combined with one of the other basic structure definitions.257 The argument’s empirical premise—that only states have a pervasive impact—is often criticized. G. A. Cohen claims that individual actions and not simply institutions have profound impacts on life chances.258 Further, many cosmopolitans dispute the claim that domestic institutions primarily have a profound impact.259 The Rawlsian constructivist interpretation of the argument, with its division of labor thesis, helps offer a reply to these criticisms. Basic structure institutions organize a division of labor for a social product. How the division of labor is organized profoundly and pervasively affects life chances among the members of the division of labor. The fact that the basic structure of a state organizes a division of labor for a social product makes principles of distributive justice the relevant regulative principle for the practice. Although international institutions will have a profound impact on life chances, they do not do so by organizing a division of labor for production. It follows that the basic structure of a 256 For example, see Rawls TJ, 7 & 96; PL, 269; JAF, 52-57. For example, see Freeman, “The Social and Institutional Bases of Distributive Justice.” 258 Cohen, Rescuing Justice and Equality, chap. 3. 259 Abizadeh, “Cooperation, Pervasive Impact, and Coercion: On the Scope (Not Site) of Distributive Justice”; Brock, Global Justice: A Cosmopolitan Account; Buchanan, “Rawls’s Law of Peoples: Rules for a Vanished Westphalian World”; Cohen and Sabel, “Extra Rempublicam Nulla Justitia?”; Hassoun, Globalization and Global Justice: Shrinking Distance, Expanding Obligations; Moellendorf, Global Inequality Matters; Pogge, Realizing Rawls; Tan, Justice Without Borders: Cosmopolitanism, Nationalism, and Patriotism. 257 173 state is distinct from the international realm. Thus, the division of labor thesis helps the pervasive impact premise reply to the first two challenges. In the next section, I explain the modern concept of distributive justice as the most plausible response to the third challenge. 7.2 The Modern Concept of Distributive Justice In this section, my main objective is to outline the systemic concept of distributive justice and what it assesses. I will also explain how the systemic concept is Rawlsian and is a plausible response to the third challenge. According to Samuel Fleischacker, the modern concept of distributive justice does not get a clear and systematic representation until Rawls articulates it. For an example of the novelty of Rawls’s concept of distributive justice, Fleischacker notes that “When Rawls tells us that retributive justice must be concerned with a person’s character but distributive justice should not be so concerned (TJ, 311–315), he almost reverses the view Aristotle had proposed of these two types of justice.”260 Samuel Scheffler explains that Rawls is not concerned with individual characters because he focuses on the basic structure for distributive justice.261 It is Rawls’s view that distributive justice only applies to the basic structure that is truly novel. Elizabeth Anderson calls this view the “modern systemic concept of distributive justice.” It holds that distributive justice assesses overall distributive outcomes of socioeconomic institutions, or the background conditions of interaction, rather than a person’s character or individual transactions. Anderson and Fleischacker trace the 260 261 Fleischacker, A Short History of Distributive Justice, 12. Scheffler, “Cosmopolitanism, Justice & Institutions,” 72. 174 origins of the concept to the late 18th century.262 Anderson claims that it starts with Thomas Paine’s Agrarian Justice263 where Paine argues that poverty is systemic and neither deserved nor inevitable. She says, Until Paine, justice was primarily conceived in a transactional sense—in terms of duties to render particular things to particular people, and to avoid force, fraud, theft, and other wrongs in person-to-person interactions. Paine’s essay made a major advance toward the modern conception of justice as a virtue of entire systems of property, to be assessed in terms of its consequences for everyone’s interests.264 Paine made a significant step towards the modern concept by rejecting the claims that poverty was natural or the result of individual vice. He also held that poverty could be prevented and that people have a right to a property system that does not produce poverty. To explain the modern concept of distributive justice, Anderson distinguishes between “systemic” and “local” principles. Local principles are meant to assess the conduct of individual people or groups in acquiring goods and distributing them to particular people. For example, agents enter into contracts with one another, which should be free of fraud, theft, and so on. Systems of production and distribution consist of local principles applying to particular agents and transactions. Systemic principles set standards for the total distributive consequences of these systems, for example, whether the systems realize equality of opportunity. She says that the modern concept of Anderson, “Thomas Paine’s ‘Agrarian Justice’ and the Origins of Social Insurance”; Anderson, “Rawls’s Difference Principle”; Fleischacker, A Short History of Distributive Justice. 263 Anderson, “Rawls’s Difference Principle”; Anderson, “Thomas Paine’s ‘Agrarian Justice’ and the Origins of Social Insurance.” However, she notes that there is often a tension between systemic and local concerns in many theories including Paine’s. Also see Sean Monahan https://www.jacobinmag.com/2015/03/thomas-paine-american-revolution-common-sense/ 264 Anderson, “Thomas Paine’s ‘Agrarian Justice’ and the Origins of Social Insurance,” 56. Fleischacker contrasts “commutative justice” with the modern concept. See for example Fleischacker, A Short History of Distributive Justice, chap. 1. 262 175 distributive justice assesses the justice of a system’s overall consequences. The justice of local principles is judged systemically. That is, the entire system made up of local principles is judged by its overall consequences or the relationships it maintains. “Local rules are subordinate to systemic principles and may be altered to satisfy them.”265 Recently, Anderson uses her distinction to explain and defend Rawls’s principles.266 She claims that Rawls’s idea of distributive justice uses local principles, such as entitlement and desert, but they are always judged systemically. His goal is to offer systemic principles that relate people as free and equal persons. Anderson says, “The inequalities permitted by Rawls’s principles are only those that avoid consolidating socioeconomic inequalities into distinct class identities.”267 Proponents of thinking of distributive justice systemically often cite the following related ideas as supporting the systemic concept. They observe that local principles of justice alone cannot explain financial crises and market shocks since they have systemic causes, such as opaque markets.268 Further, the cumulative effects of local principles alone cannot result in a fair property system over time. As Rawls says, “fair background conditions may exist at one time and be gradually undermined even though no one acts unfairly when their conduct is judged by the rules that apply to transactions within the appropriately circumscribed local situation” (PL, 267). The systemic concept of distributive justice is part of what Rawls calls “background justice,” or the set of social Anderson, “Thomas Paine’s ‘Agrarian Justice’ and the Origins of Social Insurance,” 72. Cf. Anderson, “Rawls’s Difference Principle”; Ronzoni, “What Makes a Basic Structure Just?” 266 Anderson, “Rawls’s Difference Principle.” 267 Anderson. 268 Our interconnected fates put pressure on moralized conceptions of poverty and local principles of justice. For example, it is obvious to those whose poverty was caused by unemployment due to recession or disability due to pollution that their poverty was not their fault. Anderson, “Thomas Paine’s ‘Agrarian Justice’ and the Origins of Social Insurance,” 73. 265 176 conditions necessary for individual transactions to be free and fair.269 I explained above Paine’s idea that poverty and wealth have systemic causes. Fleischacker’s explanation of the modern concept of distributive justice adds that Hume, Smith, and Rousseau helped to bring about the modern concept by helping to change the perception that the poor were different in skill and virtue than the rich. Each of the above thinkers helped to show that wealth and inequality cannot be explained by individual vices and virtues. Instead, proponents of the systemic concept hold that levels of wealth, talents, opportunities, and social relations are primarily determined systemically, by property systems, rather than by individual actions and characters. Because these roles are primarily determined from birth, systemic distributive justice assesses the entire opportunity structure of the economic system. According to Anderson, a key factor leading to the systemic concept is the fact that economists began to understand the economy as a “system of interconnected mechanisms that led to aggregate outcomes that were not intended by any individual but that could be predictably affected by state politics.”270 269 For example, PL pp. 265-269, and TJ pp. 548, 563, and 31-32. Anderson, “Thomas Paine’s ‘Agrarian Justice’ and the Origins of Social Insurance,” 72. Thomas Piketty explains that inequality has recently increased in the United States because of changes in pay norms allowing for the pay of top managers to greatly increase relative to workers (Capital in the Twenty-First Century, trans. Arthur Goldhammer (Cambridge: Harvard University Press, 2017), chap. 9.). Joseph Stiglitz agrees and adds that many forms of rent seeking are to blame. Competitive markets drive prices down, so businesses create barriers to entry in their market (e.g., through patents), exploit market power or market imperfections, gain monopoly control over a market through buying or selling to governments at above or below market price, and so on. In short, rent-seeking is how the political process helps the well-off exploit the poor, and how people gain in a market without laboring or producing. For example, Carlos Slim acquired a large share of Mexico’s telecommunications when it privatized at well below market price (52), banks exploit the less educated through predatory lending (50), and banks use the creation of the opaque derivatives market to gain an advantage over buyers since it is difficult for consumers to get a good deal because sellers know the market much better and the sellers are constantly selling so they have more information (44). The last example led to instability and ultimately the crash of 2008. Piketty also argues that when the rate of return on investments outpaces the rate of growth (r>g), inequality increases as it has done in Europe. He (6–8) also thinks that Ricardo’s principle of scarcity—the prices for scarce goods will grow ever higher as population and wages increase—makes r>g even worse since it can increase the rate of return on investment in scarce goods, such as land. Branko Milanovic proposes the idea of “Kuznets waves” to explain inequality. Kuznets argued that inequality is low at low levels of income, it then rises as the economy develops, and it falls at high income levels. Kuznets waves are alternating increases and 270 177 I think there are at least three parts to the systemic concept. First, it is concerned with fairly organizing socioeconomic institutions to organize a division of labor. This includes what counts as labor, what counts as property, what type of property rights is allowed including use or ownership rights of land and the means of production, rules of acquisition and transfer, and so on. These establish what I will call “functional roles” in a socioeconomic scheme—for example, doctor, land-owner, parent, or student. Occupying a role comes with rights, powers, and opportunities. Second, systemic distributive justice is concerned with dividing the social product for each functional role, that is, the procedures whereby people in functional roles earn wages, gain property rights, and so on. There may be minimum wage laws; markets of supply and demand; a basic income; employment and wages determined by bargaining power in the market; income, wealth, sales, property, and inheritance tax; or other procedures of attaching wealth, property, powers, and opportunities to functional roles. The first two parts of systemic distributive justice can be contrasted with Robert Nozick’s way of seeing production as coming from self-sufficient “Robinson Crusoes.” He sees production as between self-sufficient people who produce things themselves before the point of trade, and trade provides everyone with the marginal product of labor. Functional roles are not assessed on his account. According to the systemic concept, on the other hand, every product produced in an economy relies on everyone working decreases in inequality driven both by malign (e.g., wars) and benign (e.g., increasing levels of education) forces that are both economic (e.g., forces of supply and demand) and political (creating opaque markets and tax havens). Milanovic says, for example, “It is the interplay between economic and political factors that drives Kuznets waves…Income inequality is, almost by definition, an outcome of social and political struggles, sometimes violent ones” (p. 86). The wave must replace the Kuznets curve because the curve does not explain the recent rise in inequality. Branko Milanovic, Global Inequality: A New Approach for the Age of Globalization (Cambridge: Harvard University Press, 2016), chap. 2. He thinks that often the causes of increasing inequality are overdetermined (e.g., p. 109). 178 together. Anderson says, From the point of view of justice, the attempt, independent of moral principles, to credit specific bits of output to specific bits of input by specific individuals represents an arbitrary cut in the causal web that in fact makes everyone’s productive contribution dependent on what everyone else is doing. Each worker’s capacity to labor depends on a vast array of inputs produced by other people—food, schooling, parenting, and the like … In addition, the productivity of a worker in a specific role depends not only on her own efforts, but on other people performing their roles in the division of labor. Michael Jordon could not make so many baskets if no one kept the basketball court swept clean. Millions of people could not even work if public transportation workers went on strike.271 Anderson adds that non-wage-earning dependent caretakers, usually women, also contribute to production—including household production that would otherwise have to be hired out, raising future workers and rehabilitating injured or sick workers, and by relieving others of the human responsibility of care so they can participate in the market economy.272 Systemic distributive justice assesses entire systems of production, so it is concerned with organizing functional roles and how the social product is divided amongst roles. We cannot take functional roles as “given” and assess only local transactions. Third, systemic distributive justice is concerned with the distribution of lifetime opportunities of the persons subject to justice to attain functional roles with their associated opportunities, rights, and powers. Recall that rather than vice or virtue determining opportunities for wealth, economic powers, and rights, these are primarily determined from birth by the entire system. For example, education, inheritance, and other systemic causes primarily explain opportunities for functional roles, so the systemic concept assesses the entire economic system’s opportunity structure. The systemic concept of distributive justice is how at least these three factors, and the local principles 271 272 “What Is the Point of Equality?” Ethics 109, no. 2 (1999): 321–22. 323. 179 regulating each transaction within them, produce overall consequences or social relationships. For example, do they produce poverty and oppression or relate people as equals and provide fair equality of opportunity? Do people reciprocally benefit from the system? 7.3 Organizing Property and Functional Roles, the Fruits of Functional Roles, and Opportunities for Functional Roles In this section, I will use my division of the aspects of the systemic concept of distributive justice to show that the concept assesses the nature and scope of a division of labor—that is, its functional roles and opportunities and therefore who is capable and permitted to participate in the division of labor. Then, in the next subsection, I use this claim to argue that accepting the systemic concept requires rejecting the basic structure argument. Recall my division of the systemic concept into three parts. First, it is concerned with fairly organizing a division of labor or socioeconomic institutions, which includes what counts as property and labor. These rules establish what I call “functional roles” in a socioeconomic scheme, for example, land-owner. Second, systemic distributive justice is concerned with dividing the social product for each functional role, that is, the procedures whereby people in functional roles earn wages, wealth, and powers. Third, systemic distributive justice is concerned with the distribution of lifetime opportunities of the persons subject to justice to attain functional roles with their associated opportunities, rights, and powers. I will use this division of the aspects of the modern concept of distributive justice 180 to argue for the following. First, I argue that the first two parts of the concept—what functional roles exist and what rights, powers, and opportunities they afford—largely determines the third part. That is, what rights, powers, etc., are afforded to the various functional roles determines who is eligible to attain those functional roles; and so those rights and roles determine de facto membership in a division of labor. For instance, if all previously available roles to a person become automated, she or he lacks opportunity to fill a functional role—no one, in this case, is eligible to attain that role. I argue, second, that the third part of the systemic concept of distributive justice—the distribution of life time opportunities—also aids in determining whether one is a de facto member of the division of labor. The opportunities provided for the system of production in place determine whether one has real opportunities to attain functional roles in an economy. For instance, the talents developed for the system of production in place will effectively exclude some from functional roles if they cannot develop the talents needed to attain a role. Thus, how systemic principles of justice organize local principles constituting a socioeconomic system determines who is permitted and capable of being a member of the division of labor. In other words, the systemic concept of distributive justice assesses the nature of a division of labor—its functional roles and opportunities—and therefore who is capable and permitted to participate in the division of labor. Consider my first claim, namely that parts 1 and 2 of the concept influence the overall opportunity structure to determine who can be a member of a division of labor— or who can fill what I called a “functional role”—and who is effectively excluded. Allen Buchanan argues that being able to contribute to a scheme of social cooperation depends 181 on what cooperative framework is in place.273 Our society’s cooperative rules are complex, so a large number of people are deemed “incompetent,” that is, they are deemed unable to fill a cooperative role at all—for instance, they cannot enter into contracts, labor, or acquire property. In a different cooperative arrangement, such as a traditional peasant village, these “incompetents” would likely be full members of cooperation. I will add that people without the full use of their body but with great intelligence, such as Stephen Hawking, are considered valuable members of our current division of labor, whereas these people would likely be excluded from functional roles in a peasant village. Buchanan claims further that a cooperative scheme does not simply choose for excellence from a given set of skills; it chooses the skills required to cooperate. He asks us to imagine someone with a mild cognitive impairment and dyslexia. In our society, being literate is required for full participation since one must be able to read and understand contracts, but these skills are not necessary in a hunter-gatherer society. The systemic concept of distributive justice organizes functional roles, in part by determining what counts as “productive labor” and the “social product,” and so it determines who is a member of the division of labor. Buchanan concludes from his examples that “justice as reciprocity”—the view that distributive justice is reciprocity among cooperators for a social product—is a seriously impoverished view. His point is that cooperative frameworks are socially determined, so these should be assessed from the standpoint of justice.274 In other words, the ability to contribute is determined by what cooperative framework is in place, which is socially determined. So questions of justice do not simply arise among contributors; 273 274 Buchanan, “Justice as Reciprocity Versus Subject-Centered Justice,” 237. Buchanan, 237. 182 justice is needed to determine or assess the sort of cooperative institutions we choose, which determines who has the opportunity to contribute. Justice as reciprocity is “incomplete” and thus “defective” since it cannot tell us if our scheme of cooperation unjustly excludes some people from contributing.275 “Justice as reciprocity” can respond to Buchanan’s worry by showing that the systemic concept of distributive justice holds that organizing the division of labor and the opportunities for these functional roles is part of what distributive justice assesses. Buchanan is right to think that distributive justice ought to be able to say when people are being exploited or being marginalized because they do not count as members of a division of labor. Contrary to Buchanan’s claims, reciprocity conceptions of distributive justice, when applied systemically to the entire property system, can do this.276 If a principle is said to express a conception of reciprocity, we can evaluate the principle based on whether it excludes or marginalizes people. Reciprocity conceptions of justice can satisfy the three parts of systemic distributive justice outlined above, and we can assess them on this basis. They can fairly assess what functional roles exist and the opportunities for these roles while maintaining that cooperation is not possible for all people. For instance, recall (in Ch. 5) that Stark interprets Rawls as holding that reciprocity applies to those fully capable of cooperating, even though justice should apply to all. She claims that those who are incapable of reciprocally cooperating, on Rawls’s view, include both nonpersons in Rawls’s sense (e.g., those who lack the capacities necessary for rational agency) and persons who are 275 Buchanan, 238. Buchanan offers an interpretation of reciprocity conceptions as commutative principles of justice (e.g., 229.). He thinks reciprocity is owed from one worker to other workers. It is hopefully clear from the two previous chapters why this is a mistake. Reciprocity must apply throughout the course of our lives and so is not what I as a worker owe to other current workers. It is a systemic virtue that organizes functional roles, the fruits of functional roles, and life-time opportunities. 276 183 impaired such that despite technological help and political will, they cannot help produce benefits or shoulder burdens of cooperation.277 We can generalize Stark’s interpretation to include any socioeconomic scheme of cooperation to make sure that systemic conceptions of justice are not unjustly excluding the disabled. Buchanan’s worries are avoided since the people in his examples are clearly capable of reciprocally cooperating; they are simply left out of our current schemes. Buchanan’s examples, nevertheless, show that functional roles and the property rules attached to them determine who participates in social cooperation. Thus, systemic conceptions of justice determine de facto membership in a division of labor. Consider the following examples as further support for my claim. Many socioeconomic systems equate functional roles with the competitive market and thus do not count care work as part of the productive division of labor.278 Consider how Marxist explanations leave out women and care work. Paula Casal criticizes G. A. Cohen’s defense of Marx’s historical materialism—the idea that history is moved by stages of economic development with each stage’s class conflict helping to lead to the next stage.279 Casal notes that Cohen fails to consider women in his defense of historical materialism. For example, women are often legally prohibited from inheriting productive means, which would make them “serfs” on Cohen’s definition, in time periods that do not correspond to feudalism—the right was only recently won, for example, in Nepal. Further, Cohen attempts to consider all forms of subordination but does not consider common sexist ones, such as traditional marriage. These factors lead “to a gross Stark, “Contractarianism and Cooperation,” 81. For example, Rawls does this in several places. He assumes, for instance, that cooperation means full participation in the competitive market. Cf. JAF, 179. 279 Casal, “Marx, Rawls, Cohen, and Feminism.” 277 278 184 miscalculation of the numbers in each position (slave, serf, independent producer, and proletarian) and a mismatch of descriptions and periods.”280 Casal concludes that historical materialism’s core predictions are wrong because of this. Contrary to Cohen, slaves, serfs, and proletarians have co-existed together in large numbers, there are more than four categories of class, and each category’s presence cannot simply be explained by productivity levels; instead, cultural phenomena such as sexism are needed to explain why women are in these roles. For instance, the move from feudalism to capitalism did not really occur for half the population. Because of how the theory defines and applies its categories, it often leaves out women as nonmembers of the socioeconomic system. Thus, women are mostly left out of the claims of exploitation and class-conflict. For my purposes, it is clear that the theory’s definitions and applications of its categories determines de facto membership in a socioeconomic system. An obvious example is chattel slavery in the United States. Not only did the American institution of slavery not count slave work as “labor,” slaves were instead considered property in this system. Often, this was justified by a Lockean “just acquisition” of the slave as property.281 For example, the institution of slavery started with indentured servants and slowly moved to including the descendants of servants to get to chattel slavery of Africans.282 By determining what functional roles exist and what (or who) counts as property, systemic rules governed by some principles of justice determine who participates and who is excluded. 280 Casal, 813. For example, John Quiggin, “John Locke’s Road to Serfdom” https://www.jacobinmag.com/2015/10/locke-classical-liberalism-treatise-nozick-constitution/ 282 W. E. B. Du Bois, “The Suppression of the African Slave-Trade,” in Du Bois: Writings (New York: Literary Classics of the United States, 2007). 281 185 Consider finally the Lockean labor theory of value for justly acquiring property, so long as there are enough and as good resources left over. Jeffersonian Democrats implemented these ideas following the Louisiana Purchase. Expropriation of land from Native Americans was justified by claiming that the natives did not improve upon the land with productive labor and that uncultivated land was almost worthless. Natives, slaves, and indentured servants are excluded from membership in the division of labor or socioeconomic system given Lockean principles of property rights. These people are either incapable of acquiring property, given that they have a hunter-gather lifestyle, or their labor can only acquire property for their master if they themselves, or their labor for a given amount of time, were “justly acquired” as property.283 Again, my first main claim is supported: principles organizing the system of property and functional roles determine membership in a socioeconomic system. Recall that my second main claim is that the third part of the systemic concept of distributive justice—the distribution of lifetime opportunities—also aids in determining whether one is a de facto member of the division of labor. The opportunities provided for the system of production in place determines whether one has real opportunities to attain functional roles in an economy. Opportunities begin to develop from the early stages of life through education, property, and other means. If talents are not fostered, if literacy is not developed, if capital cannot be inherited, and so on, one is effectively excluded from attaining a role in a division of labor. There are many historical examples of people being excluded from functional roles in a division of labor because they were not John Quiggin “Locke’s Folly” https://www.jacobinmag.com/2016/08/locke-property-manifest-destinyjefferson-slavery-indigenous/ “John Locke’s Road to Serfdom” https://www.jacobinmag.com/2015/10/locke-classical-liberalism-treatise-nozick-constitution/ “John Locke Against Freedom” https://www.jacobinmag.com/2015/06/locke-treatise-slavery-private-property/ 283 186 afforded opportunities to attain a role. These include classes of people deemed unfit to be given opportunities by their culture. For example, it was a struggle in the years following the civil war in the United States to afford African Americans opportunities for functional roles. Black colleges were founded to train teachers that would educate freed slaves because white southerners refused to teach them, and not enough white northerners volunteered.284 This was an institutional correction to an economic system that did not provide opportunities to freed slaves. Consider also health as a determinant of opportunity for functional roles. Norman Daniels explains that health, or normal human functioning,285 has special moral importance for justice because it protects opportunity.286 The socially controllable factors that affect health or “social determinants of health”—such as personal medical care and nonpersonal medical measures including environmental protection, housing, heating, and nutrition—promote (or fail to promote) health and therefore opportunity. Justice requires the protection of opportunity, according to Daniels, so it requires the protection of health through health policies that affect socially controllable health factors. To show that justice requires the protection of opportunity and thus health, Daniels uses Rawls’s conception of justice, which supports the idea of fair equality of opportunity (FEO) as part of distributive justice. Daniels claims that FEO aims to protect W. E. B. Du Bois, “The Souls of Black Folks,” in Du Bois: Writings (New York: Literary Classics of the United States, 2007), chap. 6. 285 Daniels defines health needs as those necessary for “normal species functioning” Norman Daniels, Just Health: Meeting Health Needs Fairly (Cambridge: Cambridge University Press, 2012), 34. Following Boorse, Daniels defines “health” as the absence of pathology, or “any deviation from the natural functional organization of a typical member of a species” Daniels, 37. This need not be completely naturalistic, but Daniels rejects the strongly normative conceptions of health. All Daniels requires is that normal and abnormal function be ascertainable by public methods including biomedical methods, Daniels, 42. 286 For this argument, see Daniels, Just Health: Meeting Health Needs Fairly, chap. 2. 284 187 people’s “fair share of the normal opportunity range.”287 The normal opportunity range is the normal set of life-plans available to someone given one’s talents and skills. If positions are simply open to talents, the opportunity for them is arbitrarily affected by disparate education, race, family background, and other disadvantages that affect talents and opportunity. FEO attempts to correct for these arbitrary disadvantages (JAF, 43). The intuition behind the principle is that the opportunity range should not be determined by arbitrary social factors such as race, class, and gender. This way everyone has a fair share of the normal opportunity range—the range of opportunity people would have if their life prospects were not affected by morally arbitrary social factors such as race. Everyone’s share is equal in that each of these impediments to opportunity is eliminated.288 Daniels argues that health, or normal functioning, should count as part of our fair share of the normal opportunity range. The distribution of pathology is largely a function of social factors such as class, race, and gender. Our opportunities instead should be determined by talents and skills given normal functioning, which means that many diseases and pathologies must be prevented or addressed to ensure normal functioning.289 Daniels then draws on social epidemiological work to show that the social determinants of the level and distribution of health are those that Rawls’s principles govern.290 Given 287 Daniels, 43–44. Daniels, 60. 289 According to Daniels, Rawls takes some natural distribution of talents as a baseline. His difference principle “mitigates” the effects of natural talents on the distribution, but it does not seek to eliminate the effects of natural talents on the distribution. Daniels also holds that FEO should structure the institutions that determine health. Health care is like education in that FEO must structure these institutions to fairly correct for special social circumstances, such as effects of race and family background on opportunity. Opportunity in these areas cannot be equalized by fairly distributing “primary goods,” such as wealth. See Daniels, 53–61. 290 Daniels, chap. 3. 288 188 Daniels’s arguments, the members of a division of labor or socioeconomic system organized by Rawls’s principles would be much different from the members of a division of labor in a Lockean society where equality of opportunity is not required. I conclude that the concept of systemic distributive justice determines the nature of functional roles and opportunities for these roles, which determine de facto membership in a division of labor for joint production. The systemic concept therefore assesses the nature and scope of a division of labor. 7.3.1 The Systemic Concept of Distributive Justice Versus Anticosmopolitans In the previous sections, I aimed to the show the plausibility of the systemic concept. Furthermore, I explained that Rawls and Rawlsians seem to accept the systemic concept of distributive justice, which means that they may use the concept to respond to the third challenge to the basic structure argument, which is to provide a reasonable concept of distributive justice. However, if we accept the systemic concept of distributive justice, then we must reject the Rawlsian basic structure argument with respect to the scope of justice. Recall that for anticosmopolitan Rawlsians, the existing basic structure of a state organizes a division of labor among compatriots, which makes the concept of distributive justice relevant for the social practice of a state. The social practice or interaction problem exists among the members of an existing division of labor with the functional roles and opportunities organized by their state’s basic structure. It follows that the concept of distributive justice for the state social practice cannot assess the nature and scope of the state’s division of labor. The nature and scope of the division 189 of labor are determined empirically or by interpreting state practice. Principles of distributive justice for the state practice can assess the fairness of the productive process among the members of the division of labor; they cannot assess the nature or scope of the division of labor since the concept of distributive justice is made relevant by the practice or interaction problem of an existing division of labor. Thus, existing divisions of labor cannot make the systemic concept of distributive justice relevant for regulating them. As I argued in the previous section, the concept of systemic distributive justice assesses the nature and scope of divisions of labor, so distributive justice understood in the systemic way cannot be made relevant by existing divisions of labor. If we accept the systemic concept, we can at most require the capacity to be a member of the socioeconomic system—including currently unpaid roles that support the system—as restricting the scope of systemic distributive justice. Distributive justice is concerned with organizing functional roles; the rights, powers, and opportunities attached to functional roles; and opportunities for functional roles. All those with the capacity to be a member of a division of labor have an interest in how this is done, in how property systems will have power over them. I reached a similar conclusion in Chapter 5. There I claimed that reciprocity accounts of distributive justice can only require the capacity to cooperate as limiting the scope of justice. The only neutral definition of “capacity” to cooperate or be in a division of labor—one that is not itself the result of a theory of systemic justice—is personhood, which all the arguments already invoke, or the generalized version of Rawls’s definition of “capacity to cooperate” according to Stark’s reading of him. Thus, Rawlsian constructivism supports humanity-centered 190 egalitarianism.291 In our currently partially globalized world where state economic systems are partly distinct, country of birth is still the number one indicator of wealth and lifetime opportunity.292 Cosmopolitans often claim that country of origin is morally arbitrary much like race or class.293 This is often said to be a “luck egalitarian claim.” However, with respect to the concept of systemic distributive justice, it can also be interpreted as a claim about the existence of property systems in the circumstances of justice as explaining distributions of wealth and class. I aimed to show that the nature of functional roles and opportunities is an important distributive question since these factors, as part of entire property systems, largely explain one’s wealth and opportunities. This power must be justified to be just. So according to the systemic concept, being excluded from the prosperity of a property system—whether it is by lack of wheelchair ramps in public places or by state borders—is just as relevant to distributive justice as being an exploited member of the system in question. It is the power that property systems wield over persons that is relevant to justice. The systemic concept of distributive justice is concerned with this power—it assesses the functional roles and opportunities of a system—and therefore lends support to cosmopolitanism rather than statism as proponents of the basic structure argument think. Given Stark’s definition, distributive justice as reciprocity would not quite be “humanity-centered” since some persons are incapable of reciprocating. Recall that my argument for cosmopolitanism uses morally relevant properties of persons. My claim is that if the capacity to reciprocate is morally relevant, then it applies to all persons with this capacity. So the argument is still a “humanity-centered” one. 292 Milanovic, Global Inequality: A New Approach for the Age of Globalization, 131. 293 For example, Moellendorf, Cosmopolitan Justice, 55 & 79; Caney, Justice Beyond Borders: A Global Political Theory, 112. 291 191 7.4 A Global Difference Principle? I started the dissertation by defending Rawls’s difference principle as a systemic principle. I then offered a more general justification of institutionalism with respect to the site of distributive justice. However, the argument also specifically defended Rawls’s original position, especially its mutual disinterest condition, which justifies the FEO and difference principle. I have now completed my defense of cosmopolitan constructivism. I have shown that the main Rawlsian constructivist arguments support the global scope of the principles. Does this mean that the difference principle should apply globally? The arguments above all point to a reciprocity principle applying globally, but some of the arguments are not specific as to whether the difference principle should be the principle of reciprocity supported. Briefly, I think the difference principle is the most reasonable principle of reciprocity for two reasons. First, I argued in Part I that mutual disinterest is an important criterion for representing the problem of distributive justice that supports institutionalism. I showed that when this criterion is represented, the difference principle is justified for institutions. Second, I agree with Thomas Nagel that any distributive principle justifiable to all will be the difference principle. He claims that if each person’s point of view is considered separately, as I argued the mutual disinterest condition does, and there is a conflict of interests, no result will be completely acceptable to all. We can only achieve unanimity with the distribution that is most acceptable for the least well-off person. This is the option that is least objectionable to them and so the only option that can achieve unanimity.294 294 Nagel, Mortal Questions, 123. 192 There are still a host of questions, however, about the existence of separate political institutions and how this affects distributive justice.295 Richard Miller argues, for example, that even a global original position with representatives of persons would not justify a global difference principle. This is because the parties will protect the fundamental interests of freedom of association and self-reliance. Self-reliance is part of self-respect according to Miller. He thinks that when the two interests are combined, as collective self-reliance and an interest in civic friendship among compatriots, the fact of distinct political societies will block global egalitarianism.296 He compares patriotism to special concern for one’s family, which results in inequality. Families pool risks, devote special attention to one another, and so on. The parties will apply these notions to states and so would not choose a global difference principle. There is also a question of whether political institutions require greater equality or simply sufficiency.297 Finally, Aaron James argues that cosmopolitanism unjustifiably idealizes away the international system. He claims that political principles must be normative in that they give rise to political action. To do this, political philosophy can only require social cooperation that is “epistemically available” to us. Action requires reasonable assurance that others will likewise act, which is what social cooperative institutions in part supply. We are much more uncertain about what others will do at the global scale without a system of cooperation in place for assurances. With respect to global justice, James concludes that we “can only require cooperative arrangements that we can know—with reasonable 295 For other potential problems with cosmopolitanism, and responses, see Moellendorf, Global Inequality Matters, chap. 7. 296 Miller, “Rawls and Global Justice: A Dispute over Kantian Legacy,” 300. 297 Miller, “Rawls and Global Justice: A Dispute over Kantian Legacy,” thinks political institutions require greater equality of resources. Moellendorf thinks political institutions require sufficient resources to ensure equal political influence. Moellendorf, Global Inequality Matters, 55–56. 193 confidence—that we can jointly establish and maintain, starting from our current agential situation.”298 We do not know what a global institutional revolution might mean or how the structure might work, so we must work with the international system, even for global collective action problems. Implementation of global egalitarianism or of a global difference principle is said to require institutions that comprehensively affect lifechances, an institutional scheme that is currently epistemically unavailable to us. I cannot adequately respond to every criticism here. However, I will note that Miller’s argument implies that the difference principle cannot apply within states. If states are treated as free associations in the way that families are, then Rawls’s principles will restrict their form, but they will not apply directly to their internal workings.299 As for James’s challenge, global egalitarianism would eventually require great institutional change, but I am not sure why it requires an end to the nation state anytime soon. Egalitarianism can use representative persons to measure the worst-off. Many people agree that poverty can be eliminated. And James thinks trade can be made fairer. Both goals require identifying representative persons from classes and what institutional changes, including coordination of state policies, can mitigate poverty and make the consequences of trade fairer to those it affects. So I think it is an important conclusion for liberalism that its principles of justice are universal. Many liberals agree that we can target poverty and make trade fairer, but we owe more than poverty relief and principles of fair trade to noncompatriots. Everyone is entitled to an economic system that benefits 298 299 James, Fairness in Practice: A Social Contract for a Global Economy, 116; 112–22. For instance, see Rawls JAF, §50. 194 all.300 Further, the goals of eliminating poverty and decreasing inequality often overlap. Policy-wise, the goals may include actions that simultaneously move us toward stronger international institutions, greater global economic equality, and poverty relief. These sorts of policies include making markets and wealth more transparent such as Piketty’s proposed global wealth tax or making sure housing markets are not opaque, as they were leading to the 2008 crash due in part to the practice of mortgage-backed securities; different ways of preventing privatization of public resources or monopolization of a newly privatized resources or the prevention of other forms of rent-seeking through fairer trade, national policy, and political and social pressure; and supporting democratization of the workplace or unions both nationally and internationally. These policies can be worked towards from our current institutional scheme. Indeed, many of them are national policies that also affect people abroad. I said in Chapter 5 that if establishing the institutions required by justice is feasible and can be done without too much cost, we may have a duty to establish new institutions. Since the policy goals of ending poverty, making trade fairer, and increasing economic equality tend to overlap, the changes required by a global difference principle seem to be feasible and to not require too much cost. For example, according to Milanovic, in Marx’s day class was most important for explaining inequality since countries were closer in wealth. Today country of birth is most important for explaining inequality.301 Development moves countries closer to convergence (in GDP per capita). So increasing development increases global equality. Further, capital is highly mobile while labor is not. Milanovic thinks that liberalizing 300 Caney offers more ways that humanity-centered egalitarians can respond to criticisms and accommodate some of the intuitions of membership dependent accounts. Caney, “Humanity, Associations, and Global Justice: In Defence of Humanity-Centred Cosmopolitan Egalitarianism,” 525–29. 301 Milanovic, Global Inequality: A New Approach for the Age of Globalization, 131–37. 195 migration, allowing labor to move across borders, will do more than simply development to equalize global wealth.302 Thus, it is unclear why we need a global basic structure anytime soon to move closer to satisfying a global difference principle. Instead, accepting a global difference principle requires removing barriers to convergence between nations. Our primary focus for distributive justice, then, can again move to class differences. It is often said in the global justice debate that as the world becomes more interconnected, our duties of justice grow wider.303 On my account, it is justice that requires greater connections. For instance, the European Union was created in the aftermath of World War II to foster cooperation and prevent further conflict.304 Similarly, I think that to prevent exploitation, domination, restrictions of freedom of movement, and being cut off from resources, everyone is entitled to a global economy that can be justified to them. Most humans have long shared the circumstances of justice with one another, which I have emphasized with my historical examples. I have made the case that it is our greater understanding of wealth, inequality, and opportunity that expands the bounds of justice. Our ideals of liberal justice, shaped by modern explanations of wealth, inequality, and society, apply to all. It is not our duties of justice that are expanding on my account. As Milanovic contends, it is political tensions that are growing with our greater interconnectedness and levels of development. He quotes Kuznets’ claim that “Since it is only by contact that recognition and tension are created…the reduction of 302 For example, Milanovic, 143. Brock, Global Justice: A Cosmopolitan Account; Cohen and Sabel, “Extra Rempublicam Nulla Justitia?”; Hassoun, Globalization and Global Justice: Shrinking Distance, Expanding Obligations; Moellendorf, Global Inequality Matters. 304 Cf. Rawls LOP, 48. 303 196 physical misery [in underdeveloped countries]…permit[s] an increase rather than a diminution of political tensions.”305 As our understanding of socioeconomic systems increases, and our connections to those in underdeveloped countries allow comparisons, injustice becomes more visible and understood, especially by those suffering injustice. 305 Milanovic, Global Inequality: A New Approach for the Age of Globalization, 148. REFERENCES Abizadeh, Arash. “Cooperation, Pervasive Impact, and Coercion: On the Scope (Not Site) of Distributive Justice.” Philosophy & Public Affairs 35, no. 4 (2007): 318– 58. Anderson, Elizabeth. “Rawls’s Difference Principle.” In The Cambridge Companion to A Theory of Justice, edited by Matthew Clayton and Andrew Williams, Forthcoming. ———. “Thomas Paine’s ‘Agrarian Justice’ and the Origins of Social Insurance.” In Ten Neglected Classics of Philosophy, edited by Eric Schliesser. Oxford University Press, 2017. ———. “What Is the Point of Equality?” Ethics 109, no. 2 (1999): 287–337. Arneson, Richard. “Rawls Versus Utilitarianism in the Light of Political Liberalism.” In The Idea of a Political Liberalism: Essays on Rawls, edited by Victoria Davion and Clark Wolf. Lanham: Rowman & Littlefield, 2000. Barry, Brian. “Humanity and Justice in Global Perspective.” Nomos 24 (1982): 219–52. Baynes, Kenneth. “Ethos and Institution: On the Site of Distributive Justice.” Journal of Social Philosophy 37, no. 2 (2006): 182–96. Beitz, Charles R. “Cosmopolitan Ideals and National Sentiment.” The Journal of Philosophy 80, no. 10 (1983): 591–600. ———. Political Theory and International Relations. Princeton: Princeton University Press, 1979. ———. Political Theory and International Relations. Princeton: Princeton University Press, 1999. Berkey, Brian. “Against Rawlsian Institutionalism about Justice.” Social Theory and Practice 42, no. 4 (2016): 706–32. ———. “Double Counting, Moral Rigorism, and Cohens Critique of Rawls: A Response to Alan Thomas.” Mind 124, no. 495 (2015): 849–74. 198 Blake, Michael. “Coercion and Egalitarian Justice.” The Monist 94, no. 4 (2011): 555–70. ———. “Distributive Justice, State Coercion, and Autonomy.” Philosophy & Public Affairs 30, no. 3 (2001): 257–96. Brock, Gillian. Global Justice: A Cosmopolitan Account. Oxford: Oxford University Press, 2009. Buchanan, Allen. “Justice as Reciprocity Versus Subject-Centered Justice.” Philosophy & Public Affairs 19, no. 3 (1990): 227–52. ———. “Rawls’s Law of Peoples: Rules for a Vanished Westphalian World.” Ethics 110, no. 4 (2000): 697–721. Caney, Simon. “Humanity, Associations, and Global Justice: In Defence of HumanityCentred Cosmopolitan Egalitarianism.” The Monist 94, no. 4 (2011): 506–34. ———. Justice Beyond Borders: A Global Political Theory. Oxford: Oxford University Press, 2005. Casal, Paula. “Marx, Rawls, Cohen, and Feminism.” Hypatia 30, no. 4 (2015): 811–28. ———. “Occupational Choice and the Egalitarian Ethos.” Economics and Philosophy 29, no. 01 (2013): 3–20. Christiano, Thomas. The Constitution of Equality: Democratic Authority and Its Limits. Oxford: Oxford University Press, 2012. Cohen, G. A. Rescuing Justice and Equality. Cambridge: Harvard University Press, 2008. Cohen, Joshua. “Taking People as They Are?” Philosophy & Public Affairs 30, no. 4 (2002): 363–86. Cohen, Joshua, and Charles Sabel. “Extra Rempublicam Nulla Justitia?” Philosophy & Public Affairs 34, no. 2 (2006): 147–75. Daniels, Norman. “Democratic Equality: Rawls’s Complex Egalitarianism.” In The Cambridge Companion to Rawls, by Samuel Freeman, 241–76. Cambridge: Cambridge University Press, 2006. ———. Just Health: Meeting Health Needs Fairly. Cambridge: Cambridge University Press, 2012. Du Bois, W. E. B. “The Souls of Black Folks.” In Du Bois: Writings. New York: Literary Classics of the United States, 2007. 199 ———. “The Suppression of the African Slave-Trade.” In Du Bois: Writings. New York: Literary Classics of the United States, 2007. Erman, Eva, and Niklas Moller. “What Distinguishes the Practice-Dependent Approach to Justice?” Philosophy & Social Criticism 42, no. 1 (2016): 3–23. Estlund, David. “Liberalism, Equality, and Fraternity in Cohen’s Critique of Rawls.” Journal of Political Philosophy 6, no. 1 (1998): 99–112. Fleischacker, Samuel. A Short History of Distributive Justice. Cambridge: Harvard University Press, 2004. Forst, Rainer. “Transnational Justice and Non-Domination: A Discourse-Theoretical Approach.” In Domination and Global Political Justice: Conceptual, Historical, and Institutional Perspectives, edited by Barbara C. J. Buckinx, Jonathan TrejoMathys, and Timothy Waligore. New York: Routledge, 2015. Freeman, Samuel. “Congruence and the Good of Justice.” In The Cambridge Companion to Rawls, 277–315. Cambridge: Cambridge University Press, 2003. ———. Rawls. New York: Routledge, 2007. ———. “The Law of Peoples, Social Cooperation, Human Rights, and Distributive Justice.” Social Philosophy and Policy 23, no. 1 (2006): 29–68. ———. “The Social and Institutional Bases of Distributive Justice.” In Cosmopolitanism Versus Non-Cosmopolitanism Critiques, Defenses, Reconceptualizations, edited by Gillian. Brock. Oxford: Oxford University Press, 2013. Goodin, Robert E. “What Is So Special about Our Fellow Countrymen?” Ethics 98, no. 4 (1988): 663–86. Hassoun, Nicole. Globalization and Global Justice: Shrinking Distance, Expanding Obligations. Cambridge: Cambridge University Press, 2014. Hegel, Georg Wilhelm Friedrich. Hegel’s Philosophy of Right. Translated by Thomas Malcolm Knox. Oxford: Oxford University Press, 1967. Herzog, Lisa. “Distributive Justice, Feasibility Gridlocks, and the Harmfulness of Economic Ideology.” Ethical Theory and Moral Practice 18, no. 5 (2015): 957– 69. Hirose, Iwao. “Aggregation and the Separateness of Persons.” Utilitas 25, no. 02 (2013): 182–205. 200 Hodgson, Louis-Philippe. “Why the Basic Structure?” Canadian Journal of Philosophy 42, no. 3–4 (2013): 303–34. Hope, Simon. “The Circumstances of Justice.” Hume Studies 36, no. 2 (2010): 125–48. James, Aaron. “Constructing Justice for Existing Practice: Rawls and the Status Quo.” Philosophy & Public Affairs, 2005, 69–104. ———. Fairness in Practice: A Social Contract for a Global Economy. New York: Oxford University Press, 2012. ———. “Power in Social Organization as the Subject of Justice.” Pacific Philosophical Quarterly 86, no. 1 (2005): 25–49. ———. “Reply to Critics.” Canadian Journal of Philosophy 44, no. 2 (2014): 286–304. Julius, A. J. “Basic Structure and the Value of Equality.” Philosophy & Public Affairs 31, no. 4 (2003): 321–55. ———. “Nagel’s Atlas.” Philosophy & Public Affairs 34, no. 2 (2006): 176–92. Kant, Immanuel. Kant Political Writings. Edited by Hans Reiss. Translated by H. B. Nisbet. New York: Cambridge University Press, 1996. Korsgaard, Christine. “Realism and Constructivism in Twentieth-Century Moral Philosophy.” In The Constitution of Agency: Essays on Practical Reason and Moral Psychology, 302–28. Oxford: Oxford University. Press, 2010. ———. The Constitution of Agency: Essays on Practical Reason and Moral Psychology. Oxford: Oxford University Press, 2010. Kymlicka, Will. Contemporary Political Philosophy: An Introduction. Oxford: Oxford University Press, 2002. Lippert-Rasmussen, Kasper. “Inequality, Incentives and the Interpersonal Test.” Ratio 21, no. 4 (2008): 421–39. Meckled-Garcia, Saladin. “On the Very Idea of Cosmopolitan Justice: Constructivism and International Agency.” Journal of Political Philosophy 16, no. 3 (2008): 245– 71. Melenovsky, C. M. “The Basic Structure as a System of Social Practices.” Social Theory and Practice 39, no. 4 (2013): 599–624. Milanovic, Branko. Global Inequality: A New Approach for the Age of Globalization. Cambridge: Harvard University Press, 2016. 201 Miller, David. National Responsibility and Global Justice. Oxford: Oxford University Press, 2012. Miller, Richard. “Cosmopolitan Respect and Patriotic Concern.” Philosophy & Public Affairs 27, no. 3 (1998): 202–24. ———. “Rawls and Global Justice: A Dispute over Kantian Legacy.” The Philosophical Forum 43, no. 3 (2012): 297–309. Moellendorf, Darrel. Cosmopolitan Justice. Boulder, CO: Westview Press, 2002. ———. “Fairness in Practice: A Social Contract for a Global Economy, by Aaron James.” MIND 122, no. 486 (2013): 548–53. ———. Global Inequality Matters. New York: Palgrave Macmillan, 2009. Murphy, Liam. “Institutions and the Demands of Justice.” Philosophy & Public Affairs 27 (1999): 251–91. Nagel, Thomas. Equality and Partiality. New York: Oxford University Press, 1991. ———. Mortal Questions. Cambridge: Cambridge University Press, 1979. ———. “The Problem of Global Justice.” Philosophy & Public Affairs Philosophy Public Affairs 33, no. 2 (2005): 113–47. Norcross Alastair. “Two Dogmas of Deontology: Aggregation, Rights, and the Separateness of Persons.” Social Philosophy and Policy 26, no. 1 (2009): 76–95. Nussbaum, Martha. Frontiers of Justice: Disability, Nationality, Species Membership. Cambridge: Harvard University Press, 2006. Olsaretti, Serena. “The Inseparability of the Personal and the Political: Review of G.A. Cohen’s Rescuing Justice and Equality.” Analysis 72, no. 1 (2012): 145–56. O’Neill, Onora. Bounds of Justice. Cambridge: Cambridge University Press, 2000. Parfit, Derek. Equality or Priority? Lawrence: University of Kansas, 1995. Persson, Ingmar. “In Defense of Extreme Egalitarianism.” In Egalitarianism: New Essays on the Nature and Value of Equality, edited by Nils Holtug and Kasper LippertRasmussen. Oxford: Clarendon Press, 2011. Piketty, Thomas. Capital in the Twenty-First Century. Translated by Arthur Goldhammer. Cambridge: Harvard University Press, 2017. 202 Pogge, Thomas. “On the Site of Distributive Justice: Reflections on Cohen and Murphy.” Philosophy & Public Affairs 29, no. 2 (2000): 137–69. ———. Realizing Rawls. Ithaca: Cornell University Press, 1989. ———. World Poverty and Human Rights: Cosmopolitan Responsibilities and Reforms. Cambridge: Polity Press, 2016. Rawls, John. A Theory of Justice. Cambridge: Harvard University Press, 1971. ———. Collected Papers. Edited by Samuel Freeman. Cambridge: Harvard University Press, 1999. ———. Justice as Fairness: A Restatement. Edited by Erin. Kelly. Cambridge: Harvard University Press, 2001. ———. Lectures on the History of Political Philosophy. Edited by Samuel Freeman. Cambridge: Harvard University Press, 2008. ———. Political Liberalism Expanded Edition. New York: Columbia University Press, 1993. ———. The Law of Peoples, Cambridge: Harvard University Press, 1999. Richardson, Henry. “Interpreting Rawls: An Essay on Audard, Freeman, and Pogge.” The Journal of Ethics 15, no. 3 (2011): 227–51. Risse, Mathias. On Global Justice. Princeton: Princeton University Press, 2012. Ronzoni, Miriam. “The Global Order: A Case of Background Injustice? A PracticeDependent Account.” Philosophy & Public Affairs 37, no. 3 (2009): 229–56. ———. “What Makes a Basic Structure Just?” Res Publica: A Journal of Moral, Legal and Social Philosophy 14, no. 3 (2008): 203–18. Rousseau, Jean-Jacques. The First and Second Discourses. Edited by Roger D. Masters. Translated by Roger D. Masters and Judith R. Masers. New York: St. Martin’s Press, 1964. Sangiovanni, Andrea. “Global Justice, Reciprocity, and the State.” Philosophy & Public Affairs 35, no. 1 (2007): 3–39. ———. “Justice and the Priority of Politics to Morality.” Journal of Political Philosophy 16, no. 2 (2008): 137–64. 203 Scanlon, Thomas. What We Owe to Each Other. Cambridge: Harvard University Press, 2000. Scheffler, Samuel. Boundaries and Allegiances: Problems of Justice and Responsibility in Liberal Thought. Oxford: Oxford University Press, 2001. ———. “Cosmopolitanism, Justice & Institutions.” Daedalus 137, no. 3 (2008): 68–77. ———. “Is the Basic Structure Basic?” In The Egalitarian Conscience: Essays in Honor of G. A. Cohen, Christine Sypnowich, 102–29. Oxford: Oxford University Press, 2006. ———. The Rejection of Consequentialism. Oxford: Oxford University Press. Sen, Amartya. The Idea of Justice. Cambridge: Harvard University Press, 2009. Shue, Henry. Basic Rights: Subsistence, Affluence, and U.S. Foreign Policy. Princeton: Princeton University Press, 1980. Simmons, John. “Ideal and Nonideal Theory.” Philosophy & Public Affairs 38, no. 1 (2010): 5–36. Stark, Cynthia. “Contractarianism and Cooperation.” Politics, Philosophy & Economics 8, no. 1 (2009): 73–99. Stemplowska, Zofia. “What's Ideal About Ideal Theory?” Social Theory and Practice 34, no. 3 (2008): 319–40. Tan, Kok-Chor. Justice, Institutions, and Luck: The Site, Ground, and Scope of Equality. Oxford: Oxford University Press, 2012. ———. Justice Without Borders: Cosmopolitanism, Nationalism, and Patriotism. New York: Cambridge University Press, 2004. Thomas, Alan. “Cohen’s Critique of Rawls: A Double Counting Objection.” Mind 120, no. 480 (2011): 1099–1141. Valentini, Laura. “Coercion and (Global) Justice.” The American Political Science Review 105, no. 1 (2011): 205–20. ———. “Global Justice and Practice-Dependence: Conventionalism, Institutionalism, Functionalism.” Journal of Political Philosophy 19, no. 4 (2011): 399–418. ———. Justice in a Globalized World: A Normative Framework. Oxford: Oxford University Press, 2011. 204 ———. “On the Apparent Paradox of Ideal Theory.” Journal of Political Philosophy 17, no. 3 (2009): 332–55. Van Parijs, Philippe. “International Distributive Justice.” International Economic Law 1 (2011): 62–79. Walzer, Michael. Spheres of Justice: A Defense of Pluralism and Equality. New York: Basic Books, 1983. Weithman, Paul. Why Political Liberalism?: On John Rawls’s Political Turn. New York; Oxford: Oxford University Press, 2013. Williams, Andrew. “Incentives, Inequality, and Publicity.” Philosophy & Public Affairs 27, no. 3 (1998): 225–47. ———. “Justice, Incentives and Constructivism.” Ratio 21, no. 4 (2008): 476–93. Young, Iris Marion. Justice and the Politics of Difference. Princeton: Princeton University Press, 1990. Ypi, Lea. Global Justice and Avant-Garde Political Agency. Oxford: Oxford University Press, 2012.