Rawls's A Theory of Justice at 50 1009214691, 9781009214698

Citation preview

Rawls’s A Theory of Justice at 50

John Rawls’s A Theory of Justice transformed twentieth-century political philosophy and it ranks among the most influential works in the history of the subject. This volume of new essays marks the fiftieth anniversary of its publication with a multifaceted exploration of Rawls’s most important book. A team of distinguished contributors reflects on Rawls’s achievement in essays on his relationship to modern political philosophy and twentieth-century economic theory, on his Kantianism, on his transition to political liberalism, on his account of public reason and contemporary challenges to it, on his theory’s implications for problems of racial justice, on democracy and its fragility, and on Rawls’s enduring legacy. The volume will be valuable for students and scholars working in moral and political philosophy, political theory, legal theory, and religious ethics. Paul Weithman is the Glynn Family Honors Professor of Philosophy at the University of Notre Dame. He is the author of Religion and the Obligations of Citizenship (Cambridge, 2002), Rawls, Political Liberalism and Reasonable Faith (Cambridge, 2016), and Why Political Liberalism? (2010).

Cambridge Philosophical Anniversaries

The volumes in this series reflect on classic philosophy books from the second half of the twentieth century, assessing their achievements, their influence on the field, and their lasting significance.

Titles Published in This Series Cavell’s Must We Mean What We Say? at 50 Edited by Greg Chase, Juliet Floyd, and Sandra Laugier Rawls’s A Theory of Justice at 50 Edited by Paul Weithman This list can also be seen at www.cambridge.org/cpa

Rawls’s A Theory of Justice at 50 Edited by

Paul Weithman University of Notre Dame

Shaftesbury Road, Cambridge CB2 8EA, United Kingdom One Liberty Plaza, 20th Floor, New York, NY 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467 Cambridge University Press is part of Cambridge University Press & Assessment, a department of the University of Cambridge. We share the University’s mission to contribute to society through the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781009214698 DOI: 10.1017/9781009214704 © Cambridge University Press & Assessment 2023 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press & Assessment. First published 2023 A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in-Publication Data Names: Weithman, Paul J., 1959– editor. Title: Rawls’s ‘A theory of justice’ at 50 / edited by Paul Weithman, University of Notre Dame, Indiana. Other titles: Rawls’s ‘A theory of justice’ at fifty Description: 1st edition. | New York : Cambridge University Press, 2023. | Series: Cambridge philosophical anniversaries | Includes bibliographical references and index. Identifiers: LCCN 2023000469 (print) | LCCN 2023000470 (ebook) | ISBN 9781009214698 (hardback) | ISBN 9781009214667 (paperback) | ISBN 9781009214704 (epub) Subjects: LCSH: Rawls, John 1921-2002. Theory of justice. Classification: LCC JC578.R383 R38 2023 (print) | LCC JC578.R383 (ebook) | DDC 320.01/1–dc23/eng/20230215 LC record available at https://lccn.loc.gov/2023000469 LC ebook record available at https://lccn.loc.gov/2023000470 ISBN 978-1-009-21469-8 Hardback Cambridge University Press & Assessment has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

To the memory of John Rawls

Contents

List of Figures List of Contributors Acknowledgments Abbreviations of and Bibliographic Information for Rawls’s Works Introduction paul weithman Part I

page x xi xiii xiv 1

Rawls and History

1 Taillight Illumination: How Rawlsian Concepts May Improve Understanding of Hobbes’s Political Philosophy s. a. lloyd 2 The Theory Rawls, the 1844 Marx, and the Market daniel brudney

17 36

3 Rawls, Lerner, and the Tax-and-Spend Booby Trap: What Happened to Monetary Policy? aaron james

60

4 Rawls’s Principles of Justice as a Transcendence of Class Warfare elizabeth anderson

78

5 The Significance of Injustice peter de marneffe

95

Part II Developments between A Theory of Justice and Political Liberalism 6 On Being a “Self-Originating Source of Valid Claims” stephen darwall

109 vii

viii

Contents

7

Moral Independence Revisited: A Note on the Development of Rawls’s Thought from 1977–1980 and Beyond 121 samuel scheffler

8

The Method of Insulation: On the Development of Rawls’s Thought after A Theory of Justice rainer forst

9

The Stability or Fragility of Justice japa pallikkathayil

140 151

Part III Rawls, Ideal Theory, and the Persistence of Injustice 10 The Circumstances of Justice erin i. kelly 11 Why Rawls’s Ideal Theory Leaves the Well-Ordered Society Vulnerable to Structural Oppression henry s. richardson 12 Race, Reparations, and Justice as Fairness tommie shelby

169

181 199

13 On the Role of the Original Position in Rawls’s Theory: Reassessing the “Idealization” and “Fact-sensitivity” Critiques 220 laura valentini Part IV Pluralism, Democracy, and the Future of Justice as Fairness 14 Public Reason at Fifty kevin vallier 15 Reasonable Political Conceptions and the Well-Ordered Liberal Society samuel freeman 16 Religious Pluralism and Social Unions paul weithman

239

257 277

17 One Person, at Least One Vote? Rawls on Political Equality … within Limits 296 david estlund

Contents

ix

18 Reflections on Democracy’s Fragility joshua cohen

313

19 A Society of Self-Respect leif wenar

336

References Index

356 372

Figures

14.1 14.2 14.3 14.4 14.5 14.6 14.7 14.8 14.9 19.1 19.2

x

Model 1: A Theory of Justice Model 2: Political Liberalism Model 3: The paperback edition Model 4: Symmetry with wide legitimacy Model 5: The Order of Public Reason Condorcet Paradox preferences Increasing return dynamics Model 6: The Tyranny of the Ideal Model 7: The Open Society and Its Complexities The OP (in JF) The OP (in Law of Peoples)

page 241 243 244 246 248 250 251 252 253 337 339

Contributors

elizabeth anderson is the Max Shaye Professor of Public Philosophy and John Dewey Distinguished University Professor of Philosophy and Gender and Women’s Studies at University of Michigan. daniel brudney is the Florin Harrison Pugh Professor of Philosophy at the University of Chicago. joshua cohen is a member of the faculty at Apple University; Distinguished Senior Fellow in Law, Philosophy, and Political Science, University of California, Berkeley; coeditor, Boston Review. stephen darwall is the Andrew Downey Orrick Professor of Philosophy at Yale University. david estlund is the Lombardo Family Professor of Philosophy at Brown University. rainer forst is Professor of Political Theory and Philosophy, and Director of the Research Center Normative Orders, at Johann Wolfgang Goethe University, Frankfurt/Main. samuel freeman is the Avalon Professor Emeritus in the Humanities and Professor of Philosophy and Law at the University of Pennsylvania. aaron james is Professor of Philosophy at the University of California, Irvine. erin i. kelly is Professor of Philosophy at Tufts University. s. a. lloyd is Professor of Philosophy and Law at the University of Southern California. peter de marneffe State University.

is Professor

of

Philosophy at Arizona

japa pallikkathayil is Associate Professor of Philosophy at the University of Pittsburgh. xi

xii

List of Contributors

henry s. richardson is Professor of Philosophy at Georgetown University in Washington, DC, and is a Senior Scholar at the Kennedy Institute of Ethics. samuel scheffler is University Professor and Professor of Philosophy and Law at New York University. tommie shelby is the Caldwell Titcomb Professor of African and African American Studies and of Philosophy at Harvard University. laura valentini is Professor of Philosophy and Political Theory at the Ludwig Maximilian University of Munich. kevin vallier is Associate Professor of Philosophy at Bowling Green State University. paul weithman is the Glynn Family Honors Professor of Philosophy at the University of Notre Dame. leif wenar is the Olive H. Palmer Professor in Humanities, Professor of Philosophy and, by courtesy, Professor of Political Science and Professor of Law at Stanford University, and a Senior Fellow of the Stanford Woods Institute for the Environment.

Acknowledgments

It is a great pleasure to acknowledge the foundations and institutions whose generosity made possible the conference whose proceedings are published in this volume. Conference planning would never have gotten off the ground had the Sekyra Foundation, and its founder Ludek Sekyra, not put the wind at the organizer’s back with a very generous initial grant. The National Endowment for the Humanities helped the conference take flight by providing a generous grant under its program for advancing collaborative research. The Fritz Thyssen Stiftung für Wissenschaftsförderung in Cologne provided funding that made Rainer Forst’s presence at the conference possible. The conference was also supported by Notre Dame’s Nanovic Institute for European Studies, directed by Clemens Sedmak, and a Henkels Large Conference Grant, administered by Notre Dame’s Institute for Scholarship in the Liberal Arts (ISLA). Special thanks are due to Ken Garcia of ISLA for his tireless assistance with grant applications. Conference logistics were adroitly managed by ISLA’s conference staff, headed at that time by Lauri Roberts. That the conference went off so smoothly is due largely to Lauri’s efforts. The labor shortage that the COVID-19 pandemic brought in its wake did not spare Notre Dame. Staffing and catering a large international conference in the fall of 2021 proved a challenge. That challenge was met by the good people of Notre Dame’s Morris Inn and of Venue ND. Special mention goes to Dawn Howard, who put in long hours to show conference guests the hospitality for which the Inn and Venue ND are justly well known. Finally, I am grateful to Rev. John Jenkins, CSC – professor of philosophy and the seventeenth president of the University of Notre Dame – for taking time out of his busy schedule to open the conference with some words of appreciation and welcome. Hilary Gaskin has been, as always, a helpful and encouraging editor. Isabel Canfield helped compile the index.

xiii

Abbreviations of and Bibliographic Information for Rawls’s Works

SGEK

TJ PL PLR LP CP LHMP LHPP JF BI

“A Study in the Grounds of Ethical Knowledge: Considered with Reference to Judgments on the Moral Worth of Character,” doctoral dissertation submitted to the Princeton University Department of Philosophy, February 1, 1950. A Theory of Justice. Cambridge, MA: Harvard University Press. 1971 and 1999. Political Liberalism. New York: Columbia University Press. 1993; paperback 1995; expanded edition 2005. Political Liberalism, revised draft. Unpublished, 1998. Law of Peoples. Cambridge, MA: Harvard University Press. 1999. Collected Papers, ed. Samuel Freeman. Cambridge, MA: Harvard University Press. 1999. Lectures on the History of Moral Philosophy, ed. Barbara Herman. Cambridge, MA: Harvard University Press. 2000. Lectures on the History of Political Philosophy, ed. Samuel Freeman. Cambridge, MA: Harvard University Press. 2000. Justice as Fairness: A Restatement, ed. Erin Kelly. Cambridge, MA: Harvard University Press. 2001. Brief Inquiry into the Meaning of Sin and Faith, ed. Thomas Nagel. Cambridge, MA: Harvard University Press, 2010.

Three of Rawls’s articles are sometimes cited in their original venues of publication rather than in the versions published in his CP: “Kantian Constructivism in Moral Theory.” The Journal of Philosophy 77 (1980): 515–72. “Justice as Fairness: Political Not Metaphysical.” Philosophy & Public Affairs 14 (1985): 223–51. “The Idea of an Overlapping Consensus.” Oxford Journal of Legal Studies 7(1987): 1–25.

xiv

Introduction Paul Weithman

The year 2021 saw the fiftieth anniversary of the publication of John Rawls’s A Theory of Justice, a work that was said to revive political philosophy when it was published in 1971. To mark the anniversary, and the centenary of Rawls’s birth, the University of Notre Dame hosted a two-day international conference: John Rawls’s A Theory of Justice at Fifty. Planning for the conference began with the issuance of invitations in 2018. While the conference itself was held in the midst of a global pandemic that could not have been foreseen when planning began, it took place in the window between the spread of the Delta and Omicron variants of the coronavirus. Some who hoped to attend were unable to do so because of public health protocols. But the conference drew 140 registrants from fifteen states and a half dozen countries. For almost everyone in attendance, the conference was their first since the onset of the pandemic. Their shared sense that the academic world was opening up again made the occasion all the more celebratory. Because planning for the conference began so far in advance, it was possible to secure commitments from many of the best political philosophers now working. The conversation within and between sessions was deep and probing. While it is impossible for a volume of conference essays to recreate the excitement of a conference, it is to be hoped that this volume will give readers some sense of the very high quality of work done at the conference and will advance both Rawls scholarship and political philosophy. It is unfortunate that the volume does not include essays by two philosophers who were to have taken part: Jerry Gaus and Charles Mills. Their deaths are great losses to the field and their absence from the conference was sorely felt. But Jerry’s position was ably represented by his student and friend Kevin Vallier, whose essay appears here. Charles Mills’s illness would have prevented him from traveling to Notre Dame, but he had hoped to deliver his remarks by zoom. Sadly, he passed away days before the conference and we missed the benefit of even his virtual presence. Tommie Shelby and Henry Richardson, who 1

2

Paul Weithman

were to have shared the podium with him, paid tribute to him in the session in which all three of them were to have spoken, as did the session chair James Sterba. In the course of developing and revising his theory, Rawls contributed to – and sometimes set the agenda for – a number of sub-areas of political and moral philosophy. He was also an accomplished historian of philosophy whose readings of his predecessors have proven immensely influential. Because Rawls worked on so many questions and had such wide influence, it is understandable that there were no lines of inquiry that ran through the conference in its entirety and that unite all of the essays in this volume. There were, however, threads that bound together groups of essays, sometimes in unexpected ways. The essays in this volume have been grouped so as to reflect those thematic continuities, but there are also interesting intersectional dialogues among the essays that will be noted along the way. I.1

Rawls and History

Rawls lectured regularly on the history of political and moral philosophy. Many of his students have gone on to do important and creative historical work. It was therefore natural to have a section on Rawls and the history of philosophy at the conference and to devote Part I of the conference volume to it. In Chapter 1, S. A. Lloyd argues that Rawls’s theory provides what her title refers to as “taillight illumination.” By that she means that Rawls’s theory casts light backwards in the history of political philosophy, illuminating previously neglected features of his predecessors’ work. The predecessor whom she takes Rawls to illuminate is Thomas Hobbes, on whom Lloyd has done groundbreaking work, beginning with Loyd (1992). Lloyd argues that in light of Rawls’s development of political liberalism, certain features of Hobbes’s work can be seen in sharper relief. The feature on which she focuses is Hobbes’s political conception of the person. Seeing that conception at work in Hobbes, she thinks, both raises the possibility that Hobbes pioneered political liberalism and helps to break the hold of the game-theoretic interpretations of Hobbes that have been so prominent in the literature on him. Daniel Brudney’s “The Theory Rawls, the 1844 Marx, and the Market” (Chapter 2) juxtaposes the Rawls of 1971 with a thinker with whom he is not often compared: the Marx of 1844. If Lloyd uses Rawls to provide taillight illumination of his predecessors, Brudney uses Marx to provide what we might call “headlight illumination” of Rawls. For he uses Marx to highlight an important but underappreciated element of

Introduction

3

A Theory of Justice and to bring to light an unappreciated and worrisome tension in that work. More specifically, Brudney argues that Marx’s true communist society would be characterized by a “give/receive” relationship in which citizens care for one another’s well-being. Having isolated the defining features of that relationship, Brudney then argues that such a relationship would also characterize the well-ordered society of A Theory of Justice. That is the underappreciated element of Rawls’s theory and the one that gives rise to the unappreciated tension. Brudney argues that the market activity Rawls allows in a well-ordered society would threaten the relationships of mutual care to which he is committed. But Brudney also argues that restricting the role of markets so as to preserve those relationships may well run afoul of what members of a market society may take as a tenet of common sense, since in a market society citizens may believe that indifference rather than mutual care is natural to us. In that case, the market restrictions necessary to preserve the give/receive relationship among citizens might not be publicly justifiable by appeal to the common sense of citizens and the deliverances of social science, as Rawls’s commitment to public reason requires. It seems that Rawls can have mutual care or mutual justifiability but not both. In raising the possibility that mutual care cannot be had, Brudney’s is the first essay in this collection – but not the last – to raise the possibility that the society Rawls envisioned may be vulnerable to, rather than stabilized by, the internal dynamics its institutions generate. Aaron James’s “Rawls, Lerner, and the Tax-and-Spend Booby Trap: What Happened to Monetary Policy?” (Chapter 3) locates Rawls, not in the history of philosophy, but in the history of economic thought – specifically in the economic thought of the mid-twentieth century. James argues that Rawls, like much of political philosophy, accepts the axioms of what he calls “sound finance.” According to sound finance, budgets should be balanced and expenditures – including transfers and social programs needed to satisfy the principles of justice – should be paid for out of tax revenues. Accepting the second of these axioms leaves Rawls vulnerable to Nozick’s famous critique of redistributive views: that the taxation needed to finance them is tantamount to forced labor and is therefore morally unacceptable. James notes that as early as the 1930s and ’40s some economists – notably Abba Lerner, whom James says Rawls read with care – had developed an alternative to sound finance: “functional finance.” These economists argued that the view of taxation as a revenue-raising device is obsolete in a society that has moved off the gold standard. Once the gold standard has been abandoned, central banks can provide all the resources societies need – including all that they need to satisfy the difference

4

Paul Weithman

principle – by increasing the money supply. The question James raises is why Rawls ignored functional finance. This is not just a historical question, since the embrace of sound finance left Rawls open to Nozick’s objection and since functional finance has recently been championed in an important new work of economics (Marglin 2021). Moreover, James observes in passing that Rawls’s acceptance of sound finance led him to rely on markets, suitably regulated, to achieve distributive justice. If Brudney is right about the tension in Rawls’s view induced by his reliance on markets, then the question James raises about Rawls’s neglect of functional finance is even more pressing than James says. In “Rawls’s Principles of Justice as a Transcendence of Class Warfare” (Chapter 4), Elizabeth Anderson locates Rawls in the history of social democratic thought. She begins by observing that in the 1980s, corporate elites in the United States initiated a series of assaults on the New Deal consensus. They justified their largely successful assaults and their economic gains by meritocratic arguments. Those arguments, Anderson says, are similar to the desert-based arguments wielded by elites in the class warfare of the late nineteenth century. But, she observes, it was not only nineteenth-century elites who appealed to desert. Representatives of all different class interests appealed to desert – albeit different conceptions of desert – to ground their claims. Where all such arguments misfire, Anderson says, is in the confusion of local and systemic principles. Local principles “directly guide agents in distributing goods to specific individuals.” Systemic principles “govern the choice of systems of local principles by constraining their overall distributive consequences, defined in terms of opportunities for various goods.” Desert may be appropriate for local application, but it is an inappropriate basis for systemic principles of distributive justice. One side may gain the upper hand in class warfare if it successfully enforces its desert claims. But Anderson argues that class warfare can be transcended only with general acceptance of appropriate systemic principles grounded on shared democratic citizenship. This approach was pioneered early in the twentieth century by the Social Democratic Party of Sweden. “Rawls’s principles of justice,” Andersons says, are also grounded on common citizenship and so “aim to … end[] a class-based society” and thereby transcend class warfare. She warns that the elimination of class distinctions may not lead to the elimination of other forms of hierarchy. But by attempting to transcend class warfare, and to envision a society in which all share the status of free equals, Anderson thinks Rawlsians fall squarely within the tradition of social democracy. Peter de Marneffe’s essay “The Significance of Injustice” (Chapter 5) is perhaps the most provocative in the collection. De Marneffe begins

Introduction

5

with what he calls “the traditional view of justice.” That is a view of justice he finds in Plato, Aristotle, Aquinas, Hobbes, Locke, Adam Smith, Hume, and Kant. According to that view, injustice entails that a person suffering an injustice has been wronged. “Injustice in the traditional sense,” de Marneffe says, “has a special kind of significance for our relations to each other.” But, he continues “the failure of our social institutions to conform to Rawls’s two principles of justice does not have [that significance].” The reason it does not, de Marneffe argues, is that institutions can fail to conform with Rawls’s principles without anyone having performed actions that would warrant the reactive attitudes the tradition thinks are natural responses to injustice. It follows that Rawls’s claim to the contrary notwithstanding (TJ 1999, 9–10), his theory of justice does not “tally with” the traditional view of justice. De Marneffe concludes that what Rawls has given us is not a theory of right and wrong or part of a theory of right and wrong. What he has given us is “a vision of the ideal democratic society” – perhaps, though de Marneffe does not says so, a vision of an ideal social democratic society à la Anderson. That society may be “something worthy of aspiration” but, de Marneffe concludes, realizing that society is “not necessary for mutual respect.” I.2

Developments between A Theory of Justice and Political Liberalism

Part II of the volume is made up of four essays on the development of Rawls’s views in the crucial two decades between the publication of A Theory of Justice and the publication of Political Liberalism. It opens with an essay – Stephen Darwall’s “On Being a Self-Originating Source of Valid ‘Claims’” (Chapter 6) – that dovetails nicely with de Marneffe’s. Darwall notes that Rawls introduces his description of persons as “selforiginating sources of valid claims” in his Dewey Lectures, published in 1980. The Dewey Lectures are the writings in which Rawls’s Kantianism was at its apogee. Darwall argues that by adding the description of persons as self-originating sources of claims to his theory, Rawls filled a lacuna or remedied a defect in the Kantian interpretation of the theory he had laid out in section 40 of A Theory of Justice. In section 40, Rawls says that we express our nature when we act from principles chosen in the original position. A society in which everyone realizes her nature may be what de Marneffe called “something worthy of aspiration.” But, Darwall argues, the fact that we express our nature when we act from the principles does not imply that those principles are principles of right. For Darwall, like de Marneffe, thinks that

6

Paul Weithman

principles of right are principles whose violation properly elicits certain reactive attitudes, such as guilt from oneself and blame from others. A failure to express our nature – or, as de Marneffe might have it, to realize something worthy of aspiration – elicits attitudes of an entirely different kind. Once persons are conceived of as self-originating sources of valid claims, however, they are conceived of as having the authority to hold one another accountable, and so to blame one another, for their failures. Parties in the original position, who represent persons so conceived, then choose principles that can properly be called principles of right. Darwall does not address the question of whether violations of Rawls’s two principles, which apply to the basic structure, properly occasion reactive attitudes with respect to other persons. His essay does not, therefore, have implications for the most fundamental worry de Marneffe raised. But it does show how Rawls supplemented his view after the publication of Theory so as to incorporate what de Marneffe calls “the traditional view of justice.” The essay also shows, in a very satisfying way, how Darwall’s own pioneering work on the second-person standpoint bears on the interpretation and viability of Rawls’s theory. Samuel Scheffler’s “Moral Independence Revisited” (Chapter 7) also concerns the development of Rawls’s thought after publication of A Theory of Justice – in Scheffler’s case, the development that eventually led Rawls to recast his theory as a political liberalism. In the introduction to Political Liberalism, Rawls credits Scheffler (1979) with having played an important role in that development. Scheffler begins his contribution to this volume by reviewing the argument of his essay and by asking what it was about that essay that led to such significant changes in Rawls’s thought. To answer that question, he turns not only to Rawls’s published work – prominently including “The Independence of Moral Theory” and the Dewey Lectures – but also to unpublished correspondence and to unpublished lectures that Rawls delivered in the late 1970s and 1980s. The result is an informed and nuanced discussion of Rawls’s attempt simultaneously to establish political philosophy’s independence from metaphysical questions about personal identity and philosophy of mind, and to give his conception of the person the central place that constructivism demands. Scheffler concludes by observing that Rawls thought he finally resolved the tension between those goals only in Political Liberalism, where he developed the political conception of the person and the political version of constructivism. Rainer Forst’s “The Method of Insulation” (Chapter 8) – like Darwall’s and Scheffler’s essays – treats of the development of Rawls’s

Introduction

7

thought between A Theory of Justice and Political Liberalism. His contribution, like Scheffler’s, attends closely to Rawls’s “Independence of Moral Theory” and his Dewey Lectures. Like Scheffler, Forst zeroes in on Rawls’s attempt to insulate political philosophy from other areas of philosophy. And like Scheffler, Forst discerns a tension in Rawls’s view, albeit a different one than Scheffler identified. The tension Forst discerns is between (i) the autonomy or insularity of a political conception of justice from what Rawls famously called “comprehensive doctrines” and (ii) the political conception’s dependence on those doctrines for the moral force they need to trump unreasonable views. Forst is far less sanguine than Scheffler about Rawls’s ability to resolve the tension he discerns. He concludes that “the struggle for ever more independence led to the danger of philosophical and normative dependence, and the island that Rawls tried to create is in danger of being washed away.” Japa Pallikkathayuil opens “The Stability or Fragility of Justice” (Chapter 9) by laying out the concerns about Theory’s treatment of stability that, Rawls says, led him to recast justice as fairness as a political liberalism. That Rawls should have been moved by these concerns shows that he regarded stability as an important feature of a conception of justice. Pallikkathayuil asks why he accords it such importance, a question she pursues by addressing Gerald Cohen’s claim that stability is alien to justice (Cohen 2008, 328). She argues that Rawls takes stability to be an important feature of justice as fairness because he thinks the role of that conception is to establish fair terms of cooperation and that any such terms are bound to be stable. But she also contends, in sympathy with another of Cohen’s criticisms, that the concept of justice has application even in circumstances where mutually advantageous cooperation is not possible – hence not only in what Rawls identifies as the circumstances of justice. “What matters” for the applicability of justice, she says, is precisely the claim that Darwall said Rawls needed to make if the parties in the original position are to choose principles of justice properly so called: “that we are ‘self-authenticating sources of valid claims.’” Pallikkathayil insists that broadening the circumstances of justice beyond the circumstances of cooperation does not require giving up what she regards as the core Rawlsian insight: that the point of justice is to establish an inherently stable relationship of mutual justifiability among citizens. But she concludes by suggesting that a relationship of mutual justifiability might not be secured by the Rawlsian ideal of public reason. Stability and mutual justifiability may be achievable, she thinks, but they may require a non-Rawlsian conception of how citizens reason together.

8

Paul Weithman

I.3

Rawls, Ideal Theory, and the Persistence of Injustice

Essays in Part III of the volume ask whether Rawls’s theory is capable of handling cases of injustice, prominently including racial injustice. As would be expected, Charles Mills figures prominently in all of the essays here. It is regrettable beyond measure that he was not able to contribute an essay of his own to this part of the volume. The first essay in Part III, like the final essay in Part II, argues for modifying Rawls’s account of the circumstances of justice. In “The Circumstances of Justice” (Chapter 10), Erin Kelly argues that the circumstances should be taken to include contingent facts about a society’s history of injustice and exploitation. Kelly is especially interested in the US history of racial injustice. Such historical circumstances make justice necessary and, Kelly argues, parties in Rawls’s original position should take account of them when adopting principles of justice. Thus does she think the resources of Rawlsian political philosophy can be brought to bear on legacies of domination and structural injustice. Moreover, Kelly thinks that once we see Rawlsian principles as remedies to historic injustice, we can add to the reasons Rawls offers for thinking that justice is a good. Justice is a good because it helps to effect moral repair in response to collective wrongdoing. Kelly contrasts her approach with the way it is usually thought Rawls would handle problems of historical injustice, by relegating them to the realm of non-ideal theory. She concludes by sketching some affinities between her approach to the grave injustice of racism and the approach of Charles Mills. Henry Richardson begins “Why Rawls’s Ideal Theory Leaves the Well-Ordered Society Vulnerable to Structural Oppression” (Chapter 11) with the frank acknowledgment that Mills’s work reveals a flaw in Rawlsian ideal theory. The well-ordered society of justice as fairness does not contain adequate bulwarks against racism, so even a just society could eventually come to be marred by bigoted oppression. Richardson’s conclusion recalls and promises to vindicate Elizabeth Anderson’s closing warning that elimination of economic class hierarchy does not mean the elimination of hierarchies of other kinds. And it anticipates arguments by Kevin Vallier and Joshua Cohen that the justice of Rawls’s well-ordered society may be more vulnerable than Rawls acknowledged. The vulnerability of a well-ordered society is due, Richardson argues, to Rawls’s overly juridical understanding of the basic structure and his overly moralized conception of power. The basic structure as Rawls conceives it consists only of some legally constituted institutions and of the legally constrained aspects of others. Power as Rawls conceives it is

Introduction

9

overly moralized because it excludes various kinds of de facto power and the power exercised by informally constituted in-groups. Members of these groups may gain power through access to cultural and social capital that others do not enjoy. Even in a well-ordered society, the elite snobbism that can develop among those with such access can eventually – through stages Richardson hypothesizes – lead to exclusion and domination that are unjustifiable but beyond the reach of law. But while Charles Mills criticized Rawls for pursuing ideal theory, Richardson believes the flaws he identifies in Rawlsian ideal theory can be repaired; how to repair them is the one of the subjects of Richardson’s ongoing work. In “Race, Reparations and Justice as Fairness” (Chapter 12), Tommie Shelby acknowledges that Rawlsian ideal theory would be gravely flawed if it could not condemn racial injustice. He has argued in other work that it can. In this essay, he argues – contra Charles Mills – that reparations for past racial injustice are consistent with the essentials of Rawls’s theory. Thus he argues that parties in the original position would acknowledge a natural duty of reparation. He also argues, in the spirit of the priority Rawls accords the basic structure as a subject of justice, that attaining a just basic structure should have priority over the duty to make reparations for past injustices and that reparations made for recent injustices must be consistent with attaining and maintaining a just basic structure. In “On the Role of the Original Position in Rawls’s Theory” (Chapter 13), Laura Valentini asks what a normative theory is and insists on the importance of distinguishing the desiderata of normative theories from the evidence supporting them. This distinction is, she says, crucial to rebutting two quite different critiques of Rawls. One is Charles Mills’s critique that Rawls’s theory is excessively idealized. The other is a criticism touched on by Pallikkathayil’s essay: Gerald Cohen’s critique of Rawls’s constructivism as fact-sensitive. Both Mill’s and Cohen’s criticisms of Rawls zero in on the role of the original position. But, Valentini argues, these criticisms are not criticisms of Rawls’s theory properly speaking because the original position is not, properly speaking, part of Rawls’s theory. Arguments from the original position are evidence for, rather than part of, that theory. Rawls’s aims justify his reliance on evidence provided by an idealized choice situation in which principles are adopted for idealized conditions. Mill’s critique of ideal theory therefore misfires. Moreover, given those aims – which include formulating an account of justice that can enduringly well-order human societies in the circumstances of justice – a theory that has those aims must be fact-sensitive. So Cohen’s critique of Rawls’s principles as

10

Paul Weithman

fact-sensitive misfires as well. Of course one can reject Rawls’s aims, as Mills and Cohen seem to do. But, Valentini insists, to reject the aims of a theory is not to reject the theory itself. I.4

Pluralism, Democracy, and the Future of Justice as Fairness

Kevin Vallier’s “Public Reason at Fifty” (Chapter 14) concerns what he calls “the public reason project” as it developed through Rawls’s writings and beyond. The public reason project is the project of determining what basic social arrangements, if any, can be justified to everyone who is subject to them. Rawls clearly pursued that project. Indeed, we saw that the argument of Daniel Brudney’s essay turned on Brudney’s claim that mutual justifiability by public reason is one of Rawls’s central commitments. The late Gerald Gaus also pursued it. Vallier defends the superiority of Gaus’s approach to the project because, he thinks, it effectively responds to the failure of Rawls’s. To see the failure that Vallier thinks undermines the Rawlsian approach, recall a point Pallikkathayil discussed in her essay: Rawls thinks the best conception of justice is the one that is most likely to be enduringly or stably adhered to once it is instituted. The Rawls of A Theory of Justice defended justice as fairness on the ground that it satisfied that desideratum. Vallier presents the development of Rawls’s thought after Theory as a series of attempts to remedy what he found unsatisfactory in that defense. Vallier contends that the free society Rawls’s principles of justice require is bound to be characterized by reasonable pluralism about justice. The three models of public reason Vallier says Rawls developed in A Theory of Justice, and then in Political Liberalism and its paperback edition, attempt first to deny and then to contain such justice pluralism. Rawls’s hope was to show that the desideratum he asserted in Theory would be satisfied by justice as fairness or by a small family of liberal conceptions of justice. But, Vallier thinks, none of the Rawlsian models succeeds. Gaus’s response to this failure developed significantly over many years; Vallier identifies four models of public reason in Gaus’s writings. Though Vallier does not say so here, it seems that the momentum of Gaus’s thought was carrying him toward the denial that stability is a desideratum. If that is right, then a non-Rawlsian conception of public reasoning leads to a very different place than Pallikkathayil suggests at the end of her essay. Samuel Freeman (Chapter 15), like Vallier, is interested in the ways Rawls revised his view to accommodate the possibility of what Vallier

Introduction

11

called “justice pluralism.” Freeman notes that in his published writings of the middle 1990s, Rawls introduced the ideas of a well-ordered liberal society and a well-ordered constitutional democratic society. These are societies governed by one or another reasonable liberal political conception of justice. Much of Freeman’s essay is devoted to identifying the formal and substantive features of such societies. One important formal feature, he says, is that citizens accept the criterion of reciprocity. Reciprocity is shown by citizens’ willingness to offer one another shared public reasons, considerations whose reason-giving force all citizens recognize in virtue of their shared status as free equals. This requirement differs from the conception of public reason that Pallikkathayil recommends at the end of her essay and from the view of public reason endorsed by Vallier in other work (Vallier and Muldoon, 2021). These two alternative views, Freeman might say, violate the condition of reciprocity. Conceptions of justice satisfy the criterion of reciprocity only if they require fair equality of opportunity, and include limits on economic inequalities and a commitment to full employment. It follows, Freeman thinks, that “neo-liberal and libertarian views including the classical and libertarian-liberal positions of Hayek, Friedman, Buchanan, Epstein, Gaus, Tomasi, Schmidtz, Brennan, and others” are not reasonable conceptions of justice. This conclusion enables us to see how Freeman would respond to Vallier. Vallier’s objection that Rawls cannot accommodate justice pluralism is premised on the claim that the range of reasonable conceptions of justice includes conceptions whose reasonability Freeman would deny. My own contribution to the volume, “Religious Pluralism and Social Unions” (Chapter 16), takes up the role and fate of the concept of a social union in Rawls’s work. Rawls’s claim that a well-ordered society would be a social union of social unions does important work in A Theory of Justice, but the idea of a social union virtually disappears from Rawls’s work after “The Basic Liberties and Their Priority.” In “Reply to Habermas,” Rawls says that the idea is no longer viable once we appreciate the fact of reasonable pluralism. I argue that it is no longer viable because Rawls’s use of the idea depended on an assumption I call “the security assumption.” According to that assumption, the security of the basic liberties in a social union of social unions makes it psychologically possible for citizens in the well-ordered society to enjoy the diverse public culture of a free society. I canvas reasons Rawls might have rejected the assumption and suggest that the security assumption – or its violation – helps to explain the degraded state of our own public life. The next two essays in Part IV are about Rawls on democracy. David Estlund has long-standing interests in the question of what justifies democratic procedures for political decision-making, and

12

Paul Weithman

whether those procedures can be justified because of their epistemic merits (Estlund 2008). In “One Person, At Least One Vote? Rawls on Political Equality … within Limits” (Chapter 17), Estlund argues on textual and doctrinal grounds that Rawls is not in-principle opposed to John Stuart Mill’s proposal of plural voting. According to Mill’s proposal, citizens who are most likely to make just and wise political decisions would be entitled to more than one vote or to cast votes weighted more heavily than those of other citizens. According to Estlund’s reading of Rawls, Rawls would allow such plural voting so long as that measure would do better than the alternatives at protecting and promoting the other liberties, especially the other liberties of those who have the least liberty. What Estlund calls Rawls’s “lift all boats exception” – the element of Rawls’s theory that justifies inequalities in income, wealth and opportunity – could justify inequalities in the right to vote as well. Joshua Cohen’s “Reflections on Democratic Fragility” (Chapter 18) explores Rawls’s concern that democracy might be fragile even under favorable conditions. Cohen assumes for the sake of argument that citizens all endorse reasonable comprehensive doctrines and reasonable political conceptions of justice, but that those conceptions will differ because of the burdens of judgment. Because all reasonable political conceptions endorse some form of democracy, all citizens will endorse it. But, Cohen observes, because of citizens’ different histories, experiences, and interests, they may support different conceptions of democracy. Indeed, Cohen observes “we may see deep disagreement among [citizens’] conceptions of justice, the forms of democracy they recommend, and the interests and identities of citizens who endorse those competing conceptions.” These disagreements may be so deep that each side may understandably lose confidence in the democratic convictions of the other side. Invocations of values of trust and civility may seem like invitations to a mug’s game. And [citizens] may then think it is permissible to break the democratic rules both because of their substantive commitments and because they are concerned that the other side is breaking the rules.

In A Theory of Justice and Political Liberalism, Rawls was concerned that even in a well-ordered society, citizens’ interest in pursuing their conception of the good – together with their uncertainly about others’ commitments – might move them to defect from the agreement that would be reached in the original position. That is, he was concerned that citizens might find it individually rational to defect from the choice that is collectively rational and that justice would be undone by a collective action problem. In both books, he offered intricate and powerful arguments for the claim that goodness and rightness are congruent.

Introduction

13

Summing up those arguments, he said that “the hazards of the generalized prisoner’s dilemma are removed by the match between the right and the good” (TJ 1999, 505). Cohen, in effect, raises the possibility that because citizens embrace different conceptions of justice, democracy may be threatened by a generalized prisoner’s dilemma even if Rawls’s congruence arguments succeed. Thus Cohen returns to the justice pluralism that figured prominently in Vallier’s and Freeman’s essays to argue, as Richardson had, that a well-ordered society might not be as stably just as Rawls hoped. Cohen recognizes that removing the hazard to which he calls attention requires that “an appreciation of reasonable disagreement on the fundamentals of justice – of the fact of reasonable political pluralism – itself have a genuine practical grip on the public, political culture.” This, he grants, “imposes great expectations on public reason and the associated duty of civility.” But he closes by expressing the hope that those great expectations can be met. The final essay in the collection is Leif Wenar’s (Chapter 19). Where the essays in Part I looked back to the history of political philosophy, Wenar looks fifty years forward, asking “What parts of Justice as Fairness might spread beyond the academy into American public culture by the 100th anniversary [of Theory], in 2071?”. He argues that the original position, the object of so much attention in the copious literature on Rawls, is too cognitively and culturally inaccessible to become part of public political culture. What Wenar finds in Rawls, and what he thinks might become part of that culture, is the model of social relations that gives his essay its title: “A Society of Self-Respect.” Wenar’s explication of the model depends upon a contrast that was drawn by Anderson in her essay, a contrast between Rawlsian and meritocratic justifications of distribution. When claims of merit and desert are used to justify large inequalities such as those that prevail in the contemporary United States, Wenar says, the result is a politics of envy and resentment in which some citizens derive their sense of selfrespect from class or ethnic identity. By contrast, the Rawlisan justification of equal basic rights, of fair equality of opportunity, and of whatever inequalities the difference principle allows, all bolster citizens’ selfrespect. Wenar’s focus is on the respect citizens have for themselves. But he observes that when citizens respect themselves, they are likely to respect one another. Wenar does not argue, against de Marneffe’s closing claim, that realizing the Rawlsian model is necessary for mutual respect. But he does make the case that realizing it is sufficient. In my own essay, I suggested that Rawls’s theory might help us understand some of the alienation characteristic of contemporary American politics. Wenar, like

14

Paul Weithman

Anderson, suggests that his theory might help us not only to understand the divisiveness of our politics, but also to transcend it. It would seem inappropriate to close this introduction without paying homage to the thinker whose powerful and compelling vision of a just society has been a source of inspiration and inquiry for half a century. But tribute is sometimes better paid in work than in words. It is sometimes better shown than said. The willingness of so many outstanding philosophers to lay aside their other projects, and to produce such superb essays, is a far greater tribute to John Rawls than anything that could be said here. The collection of those essays is dedicated to his memory. *. *. *. *. * Readers may be helped by a couple of notes about citation practices. Rawls’s works are cited parenthetically in the text according to abbreviations given in the front matter. Authors of the essays collected here have usually cited only the 1999 edition of A Theory of Justice. Sometimes reference to the 1971 edition or to both the 1971 and 1999 editions was necessary for comparison’s sake or to analyze the development of Rawls’s thought. In all cases, footnotes and parenthetical citations will make clear which edition is being cited. In the case of works published before 1900, every effort has been made to cite using the standard scholarly apparatus and the pagination of a contemporary edition.

Part I

Rawls and History

1

Taillight Illumination How Rawlsian Concepts May Improve Understanding of Hobbes’s Political Philosophy

S. A. Lloyd* In a short piece from 1993 entitled “Some Remarks About My Teaching,” Rawls asserted “we learn moral and political philosophy, and indeed any other part of philosophy by studying the exemplars – those noted figures who have made cherished attempts – and we try to learn from them, and if we are lucky to find a way to go beyond them” (quoted in LHPP, xiv). By now, Rawls himself has become one of the exemplars, and we fully expect that future political philosophies will continue to draw lessons from his. The question I wish to explore is whether Rawls’s philosophy can provide backward, or what we might call “taillight,” illumination of past philosophical systems. Analogously to the way that the taillights of an automobile let us see features of objects physically behind, can a philosophical system enable us to appreciate features of systems temporally behind? On the one hand, it wouldn’t be so surprising if Rawls’s work shed light on ideas in those prior philosophical systems from which Rawls drew to develop his own ideas – say Kant’s moral philosophy or Rousseau’s social contract – especially where Rawls’s formulation of an idea sharpens an idea that was indeed present but only latent, seminal, or inchoate in the earlier exemplars. What may be surprising is that Rawls’s work should illuminate our understanding of the political philosophy of a figure like Thomas Hobbes, which provided little of use to Rawls in his construction of “justice as fairness,” except by way of contrast.1 It would * I thank Paul Weithman for his many helpful comments on a draft of this essay, and for the magnificent conference that occasioned its composition. 1 For instance, the goal of Hobbes’s philosophy is to provide all with some or another sufficient reason to submit to a government possessing undivided and legally unlimited powers, rather than to identify principles of justice appropriate for a modern, liberal, democratic, pluralistic society; the “no agreement point” of Rawls’s modeling device is general egoism, whereas Hobbes’s is the state of nature; Rawls’s original position imposes a veil of ignorance whereas Hobbes’s state of nature assumes partial knowledge; Hobbes assumes partial compliance in varying degrees rather than strict compliance, as with Rawls; and whereas Rawls characterizes parties as mutually disinterested, Rawls saw Hobbes as characterizing persons as predominantly self-interested and also concerned

17

18

S. A. Lloyd

be a mistake to suppose that Hobbes could not have influenced Rawls because Rawls thought Hobbes’s philosophy unsophisticated or unimportant. To the contrary, Rawls’s teaching lectures on Hobbes begin by declaring In my own view and that of many others, Hobbes’s Leviathan is the greatest single work of political thought in the English language … its scope and its acuteness and interesting vividness of observation, its intricate structure of analysis and principles, and its presentation of what I think is a dreaded way of thinking about society which almost might be true and which is quite a frightening possibility … can have a very overwhelming and dramatic effect on our thought and feeling. (LHPP, 23)

And also, Rawls taught, on our philosophizing. He deemed it useful to think of modern British moral and political philosophy as beginning with Hobbes, and with critical reactions to Hobbes, from orthodox Christian moralists including Cudworth, Clarke, and Butler and from Utilitarians such as Hutcheson, Hume, Smith, and Bentham. Rawls said Hobbes’s original, formidable, and influential system of thought was for the next century and a half “something in regard to which one had to decide where one stood” (LHPP, 26). But if Rawls did not develop ideas from Hobbes in his own theory, how could Rawls’s theory illuminate Hobbes’s philosophy, except again, perhaps by way of contrast? Part of the answer lies in Rawls’s evolving understanding of his own theory of justice between Theory and Political Liberalism and in his revised understanding of Hobbes’s political philosophy from the mid-1980s onward. When Political Liberalism came out in 1993, Rawls sent me a copy inscribed “for Sharon, with appreciation and gratitude for your Hobbes, the First political liberal?” (question mark, emphasis on “First”). In his inscription, Rawls was asking me, was Hobbes, supposing your interpretation, the first political liberal?2 His question is at first puzzling, simply because Hobbes was so clearly not a liberal. Rawls located the historical origins of liberalism in (1) acceptance of the principles of toleration and liberty of conscience, (2) establishment of

2

with positional standing (TJ 1999, 126–27, 211). Rawls was doubtful even that Hobbes’s system actually belongs to the social contract tradition Rawls sought to advance (TJ 1999, 10, note 4), and I have no doubt that it does not. Rawls was familiar with my interpretation of Hobbes because he, along with T. M. Scanlon, had supervised the doctoral dissertation in which I began to develop it; and he had read my 1992 book on Hobbes (with evident attention, judging by his copious marginal notes in the copy I had sent him, which Erin Kelly kindly returned to me after his death). Rawls’s inscription was dated April 1993; on the 28th of that same month, Rawls wrote in a separate letter “I want to ask you: what might be the predecessors of political liberalism? How far is Hobbes one?.”

Taillight Illumination

19

constitutional regimes of limited monarchy, and (3) support for democracy and majority rule (LHPP, 11). Although Hobbes inclines toward (1) if, but only if, the population is willing peacefully to accept religious toleration, he explicitly rejects (2) and sees democracy as a legitimate but nevertheless undesirable form of regime because of its relative instability. Rawls understood liberal regimes to institutionalize priority for a list of basic rights and liberties, and to guarantee adequate material means to make use of those, in a system that limits executive power and balances it against legislative power. Hobbes, in contrast, criticizes constitutionally limited regimes, especially those with separation of powers, as fragile, prone to paralyzing stalemate and resorting to civil war to overcome it. Hobbes agrees that free exercise of religion would be best if it could “be without contention,” but religious strife was rampant at the time, and Hobbes was not optimistic about the future. Instead, he recommends a church establishment headed by the national civil sovereign, who enjoys ultimate authority to interpret religion’s requirements and to determine external profession and practice. Although Hobbes’s system recommends affording citizens the widest possible scope of liberty compatible with security, equal treatment under the law, and public support of the population, especially the poor, all as duties government owes to subjects under natural law, his absolutism allows for no legal guarantee of these things nor any moral right of rebellion against a regime that fails to provide them. Subjects are released from their duty of obedience only when their sovereign fails to protect them. As for support for democracy, and rights to vote and to run for office, although Hobbes’s system allows for a democratic form of government, so long as its powers are not limited, it deems democracy positively undesirable because it is more prone than other forms to destabilizing factionalization, to more widely spread corruption and influence peddling (because so many more legislators have favorites to enrich and must be lobbied), and because it elevates demagogues and demagoguery over reasoned political deliberation. This last consideration, that demagoguery eclipses reason when politics must persuade the masses as efficiently as possible, particularly concerned Hobbes. Those are reasons to hope not to live under a democratic regime, although, Hobbes insists, if one already does, it would be both immoral and imprudent to try to replace it with a more stable form of governance. Hobbes then, was no liberal, which, of course, Rawls well knew. If we instead focus on specifically “political liberalism,” Rawls’s question whether my Hobbes was the first political liberal makes much better sense. Political liberalism is characterized by a family of interrelated ideas including reasonable pluralism, a political conception of the person as free and

20

S. A. Lloyd

equal and as possessing capacities for rationality and reasonableness or reciprocity, a freestanding justification of political principles, overlapping consensus, stability for the right reasons, and public reason. If, as I shall argue, viewing Hobbes’s system in light of key features of political liberalism – particularly its deployment of a political conception of the person as both rational and reasonable – enables us to appreciate previously unrecognized features of it, Rawls will have provided taillight illumination of a system radically different from justice as fairness.3 Further, Rawls names and explains some ideas belonging to his political liberalism that we now can see were doing important work in Hobbes’s theory on my interpretation of it, lending support to Rawls’s suspicion that Hobbes developed a distinctly political justification, in addition to a comprehensive justification, for adhering to his favored principle of political obligation. What might have surprised Rawls though, is that the chief feature of Hobbes’s view on Rawls’s 1983 interpretation of that view4 – namely, that it allowed no room for any notion of the reasonable and so no possibility of understanding society as a system of social cooperation rather than of mere coordination (LHPP, 87) – turns out to be mistaken. Hobbes’s compendious concept of right reason dictates both rational norms of instrumental reasoning and reasonable reciprocity constraints on action. This discovery creates space in Hobbes’s system for principled accommodation of a degree of pluralism “in accordance with reason,” compatible with each citizen’s committing, for reasons of her own, to a common political principle that enjoys both support in public reason and something like an overlapping consensus of comprehensive doctrines.

1.1

Hobbes’s Central Problem and Strategy for Its Solution

The central problem Hobbes addresses is political instability, of states collapsing from subjects’ withholding cooperation or imploding into civil 3

4

Although I believe all six of these ideas are present in Hobbes in at least rudimentary form, I focus here primarily on showing that his conception of the person includes the capacity for reasonableness, both because orthodox interpretations deny this, assuming Hobbesian persons possess only instrumental rationality, and because establishing this capacity is necessary to make room for other aspects of a political liberalism. I briefly indicate how we might understand Hobbes to have offered a freestanding justification for his principle of political obligation, and attempted to demonstrate that it could receive the support of an overlapping consensus, thereby allowing for stability for the right reasons. I discuss Hobbes’s conception of public reason elsewhere (Lloyd 2018). This is the account of Hobbes’s theory Sam Freeman transcribed from his tape recordings of Rawls’s lectures in Harvard’s Philosophy 171 course in the spring term of 1983, supplemented with Rawls’s class handouts and handwritten lecture notes, and published in LHPP, 23–99.

Taillight Illumination

21

war as groups compete to harness state authority to promote their own and the society’s good as they see it. He was vividly impressed by the “miseries and horrible calamities that accompany a civil war,” (Leviathan XVIII.20) having lived through one and its chaotic aftermath of rapid regime changes. He attributed that war largely, though not exclusively, to the efforts of diverse religious factions to relocate political power into the hands of those who would use it to institute the requirements of their sectarian religious doctrines. Hobbes comments in his Six Lessons to the Savilian Professors of the Mathematics on his writing of Leviathan: the cause of my writing that book was the consideration of what the ministers before and in the beginning of the civil war, by their preaching and writing did contribute thereunto (Hobbes 1839, VII, 335)

and tells us in Seven Philosophical Problems that it was written in a time when the pretense of Christ’s kingdom was made use of for the most horrid actions that can be imagined; and it was in just indignation of that, that I desired to see the bottom of that doctrine … which divers ministers then preached for a pretense to their rebellion.” (Hobbes 1839, VII, 5)

Hobbes saw clearly that political stability cannot be reliably secured by force or threat of force. It is not merely that Hobbes deems coercion “all the way down” to be impossible, as he indicates in Behemoth: If men know not their duty, what is there that can force them to obey the laws? An army you will say. But what shall force the army? (Hobbes 1990, 59)

It is primarily that people undertaking political insurrection are often moved by transcendent interests – interests that override narrowly selfinterested concerns, and for the advancement of which they are willing to die if need be – in causes like religious reformation or in procuring salvation. Those willing even to sacrifice their lives in the service of their larger interest are not likely to be deterred by the sorts of punishments governments have at their disposal. Governments can offer to keep you safe, and can threaten you with a death penalty, but “Eternal life is greater reward, than the life present; and Eternal torment greater punishment than the death of Nature.” (Leviathan XXXVIII.1) If the command [of the civil sovereign] be such as cannot be obeyed, without being damned to eternal death; then it were madness to obey it.” (Leviathan XLIII.1)

In Philosophical Rudiments (De Cive) Hobbes had observed: “Neither is any man so mad, as not to choose to yield obedience rather to them who can remit and retain their sins, than to the powerfulest kings” (Hobbes 1839, II.17.25). “For every man, if he be in his wits, will in all things yield that man an absolute obedience, by virtue of whose sentence he

22

S. A. Lloyd

believes himself to be either saved or damned” (Hobbes 1839, II. 18.14). Government sanctions can’t compete with divine sanctions; and even sanctions aside, “it is manifest enough that when a man receiveth two contrary commands, and knows that one of them is God’s, he ought to obey that, and not the other, though it be the command even of his lawfull Sovereign” (Leviathan XLIII.1, emphasis added). Political regimes cannot hope to remain stable without buy-in from adherents of the main religious or moral doctrines present in the society, which requires that society’s members see political obedience to the existing regime as, at bare minimum, compatible with those commitments, and better, as positively supported by them. Hobbes’s proposed solution was to identify a principle of political obligation – a principle specifying the conditions under which a citizen is to obey the political authority under which she lives – that could, if widely followed, reliably secure domestic peace and then provide each citizen with what she could regard as a sufficient reason to adhere to that principle.5 However, different people have different values and interests, reflecting differences in their bodily constitutions, upbringings, experiences, habits, education, and self-conceptions, to such a degree that it is “impossible that … all men consent in the desire of almost any one and the same object” (Leviathan VI.6), and the “objects of the passions, which are the things desired … the constitution individual and particular education do so vary… as they are legible only” to God (Leviathan Introduction). Given this inevitable pluralism, Hobbes saw that providing each person a reason not overridden by contrary reasons required him to show that his proposed principle of political obligation advanced a multitude of standard types of interest, including: (i) prudential interests in safety and commodious living, (ii) moral interests in fulfilling one’s natural duties and voluntary obligations, (iii) the religious interest in fulfilling one’s duties to God, and (iv) “special-prudential” interests in 5

The principle he aimed to establish was that one should obey an existing sovereign so long as it is effective in protecting one, in all of its commands excepting only those that would require violation of one’s duty to God. As Hobbes summarizes his principle (omitting the effectiveness condition on political obligation) “subjects owe to sovereigns simple obedience, in all things wherein their obedience is not repugnant to the laws of God” (Leviathan XXXI.1). In De Cive, Hobbes says of the principle “it has been shown, both by natural reason and from holy scripture, that citizens should obey Princes and rulers of the commonwealth in all things, except in what is contrary to God’s commandments” (Hobbes 1998, 18.13). Hobbes carried out his attempt to reconcile religious and special-prudential interests with his principle of political obligation in the second half of Leviathan. There he argued that God’s natural law requires obedience to the commands of an effective sovereign and that revealed religion as contained in Scripture makes obedience to the sovereign’s (even erroneous) commands concerning religion one of two necessary conditions for salvation (the other being belief that Jesus is the Christ).

Taillight Illumination

23

receiving salvation or avoiding damnation. Hobbes’s effort to provide a confluence of distinct reasons converging in support of his proposed principle of political obligation looks to be an effort to show that there can be something like an overlapping consensus on his principle, allowing it to enjoy the principled commitment of diverse individuals embracing various moral or religious doctrines, thereby making their allegiance to it more stable and the society ordered by it more than a mere modus vivendi. This may be one of the features of Hobbes’s view on my interpretation of it that set Rawls to wondering whether Hobbes may have been the first political liberal.6 I have emphasized Hobbes’s insistence on the pluralism of human ends – a fact not often appreciated by standard interpretations – because it limits the sort of “political” conception of the person and of society available to Hobbes. He asserted in an early version of the theory that the desire to avoid bodily death is the strongest of desires “from nature” (meaning, presumably, biologically hardwired) in every (healthy) human being. However, because Hobbes recognizes that this natural desire is often overridden by competing desires both natural and “from culture,” including the desire to escape an unacceptable quality of life, desires for glory, honor or reputation, for divine rewards or to avoid divine punishments, and desires to advance valued ends including the good of the nation and the interests of loved ones, Hobbes does not implausibly

6

Although in Political Liberalism Rawls leaves it to adherents to various comprehensive doctrines to develop a supportive connection between their doctrine and the object of overlapping consensus, in Theory he offers an example of a possible such connection in his “Kantian interpretation” of justice as fairness. Of course, that interpretation may be contested by other Kantians, just as we might expect different Utilitarians to disagree as to how best to relate utilitarianism to justice as fairness. Reasonable disagreement is possible. Hobbes offers his own interpretation of Judeo-Christian religion in support of his principle of political obligation, as this is necessary in order to challenge the interpretations of those who would vest authority to interpret religious requirements in someone other than the civil sovereign, e.g., Independents (who privileged the individual’s interpretation of Scripture) and Roman Catholics (who privileged the Pope’s judgment). He needs to show in detail how his principle of political obligation is both compatible with and supported by Scripture, if he is to reassure his Christian audience that they may defer to their sovereign without compromising their eternal prospects or failing in their duties to God. A. P. Martinich has persuasively argued that Hobbes’s own interpretation is fairly typical of an English Calvinist of the period, once Hobbes corrected a couple of eccentric positions driven by misapplications of his philosophical innovations (Martinich 2021). For Hobbes, reasonable disagreement about the requirements of religion is possible because natural religion based on reason alone cannot settle all questions of interpretation of revealed religions’ authoritative texts. However, it can settle some questions and rule out some interpretations; thus, not all interpretations of a revealed religion are equally reasonable, and some may be unreasonable. Such a view is consistent with a political liberalism.

24

S. A. Lloyd

stipulate the universality of an overriding desire for self-preservation.7 Desire for a pleasant, commodious life is common; yet many are prepared to forgo such a life in pursuit of other ends. “All men agree on this, that peace is good” (Leviathan IV.40), but peace is often deemed less good than other ends by “needy and hardy men” seeking military advancement or by those who prefer victory to peace – or those who do not want peace on just any terms (comprising most of those prepared to fight and many of their civilian supporters). Hobbes does stress the strength and ubiquity of the desire to be highly esteemed by others, but insofar as high interpersonal standing is a positional good, this desire seems unpromising as the centerpiece of politically liberal conceptions of citizens and society. What Hobbes needs to ground such conceptions is a desire no person can fail to have, and which is never overridden. 1.2

A Political Conception of the Person?

In his teaching lectures, Rawls suggested that Hobbes may not have intended the conception of human nature he articulated to be strictly true but rather as appropriate for the limited purpose of devising a theory of political obligation (LHPP, 46). Although acknowledging that individuals are capable of benevolence (and so rejecting psychological egoism), Hobbes thought that for political purposes we ought not to assume that people will sacrifice their interests for the sake of strangers or fellow subjects but should instead assume them to be predominantly selfinterested. Rawls read Hobbes as attributing to all men interests in selfpreservation, conjugal affections, and commodious living – in that order – and as possessing the instrumental rationality to enable them to appreciate that submission to government could be expected better to advance those interests than could life in a state of nature. Rawls interpreted Hobbes’s laws of nature as articles of a shared “secular morality” meant 7

Indeed, if Hobbesian men really cared most for their self-preservation, we should not expect to see them engaging in the life-threatening activity of active rebellion – war being so obviously hazardous to one’s health – except perhaps against extraordinarily violent regimes attacking themselves or extraordinarily weak regimes exposing them to violence from others. The fact that Hobbes views seditions and rebellions against powerful coercive states as frequent happenings and an ever-present possibility suggests that he did not see fear of death as the main motivator of action. The prominent role of fear of death in Hobbes’s political philosophy is not to provide a necessary end of action but is rather to explain why we would judge others unreasonable to fault us for trying to defend our life and relying on our own judgment as to how to do so – “it is therefore neither absurd nor reprehensible, neither against the dictates of true reason, for a man to use all his endeavors to preserve and defend his body” (De Cive 1.7). Combined with the reciprocity requirement, this judgment yields the universal right of nature. See Lloyd (2009, 63–73).

Taillight Illumination

25

as a political doctrine (LHPP, 51, 53), thinly justified as instrumentally advancing those interests. He saw the content of those articles as enjoining reasonable constraints on conduct; but Rawls understood Hobbes to have offered a strictly rational justification for the law of nature’s reasonable precepts. Because understanding society as a system of social cooperation rather than of mere coordination requires attributing to participants a capacity for mutuality or reciprocity (LHPP, 56, 62), which those lacking the capacity for reasonableness do not have, Hobbes’s system could be at most one of social coordination. Rawls concluded “if moral right and obligation involves grounds different from the Rational, as I believe it does, Hobbes has no place for it in his official view” (LHPP, 66) and that “Hobbes has no place for a sense of fairness” (LHPP, 87). Let me now offer new support for Rawls’s characterization of Hobbes as having advanced a political conception of the person, while arguing for a quite different characterization of the interests and capacities comprising that conception. Hobbes’s method of demonstration requires that his premises be either conceptual truths settled by definition or empirical truths confirmable by introspection or observation – premises “as passion not mistrusting, may not seek to displace” (Hobbes 1839, IV, dedicatory epistle). On my account, Hobbes assumes: (1) People have ends they desire to advance by their actions. (2) People desire that the conditions necessary for their actions to be effective in achieving their desired ends obtain. These premises are entailed by Hobbes’s definition of man as a “rational animal,” which definition he believes is widely accepted. Qua animals, people have internal appetites and aversions suggesting ends of their voluntary actions; qua rational, people seek to secure the conditions necessary for their actions to achieve their ends.8 Putting this second 8

Cases in which we might be tempted to say that we desire an end but do not desire that the conditions needed to achieve that end obtain are best understood as cases in which we have not adequately characterized our desired end. When I desire to win the presidency but do not desire the election fraud that would be needed for me to win the presidency, what I actually desire is to win the presidency “fair and square”; and of course, I would want the conditions necessary for achieving that end to obtain. If I desire to inherit my parent’s wealth, but not the requisite death of my parent, my first-order desire is better characterized as the desire to inherit when the time comes. Hobbes explicitly affirms as a principle of rationality that he who wills the end must will what he believes to be the necessary means to that end; that principle suggests a tight connection between the necessary desire and every first-order desire, so if the necessary desire is a second-order desire, it will be a special kind of secondorder desire. Unlike ordinary second-order desires – e.g., “I desire that I shall desire to eat healthy foods rather than the fast food I actually desire – which are not entailed by the firstorder desire they govern – the necessary desire does seem to be implied by any first-order

26

S. A. Lloyd

point negatively, insofar as a person is rational, she desires to avoid a condition that can be expected to undermine her ability to act effectively. Hobbes’s subsequent argument explains why a state of nature is just such a condition. It is a condition in which mutual interference and insecure control of resources threaten achievement of any of our ends, no matter what those ends are. This explains why in each presentation of his state of nature argument, Hobbes adduces a long list of desirable ends unlikely to be realizable – achievements in arts, letters, sciences, geographical knowledge, society, elegancy, comfort, safety, riches, freedom from fear, etc. Hobbes views the desire to achieve one’s particular ends, and the desire to secure the conditions required for doing so –which I term “the necessary desire” – to be inescapable for any rational human. Because he intends his state of nature argument to demonstrate to everyone that they have at least one very good reason in common to submit to an effective political authority, it is crucial that that argument not depend on premises that are false of some people because it attributes to them desires they lack, or which in them are overridden by other desires. The “necessary desire,” by not presupposing possession of desire for any particular object, is as thin and as unladen with controversial values as Hobbes can make it.9 One advantage of this understanding of the foundation of Hobbes’s political philosophy over Rawls’s 1983 interpretation is that it provides a more clearly political (because less contentious) conception of the person upon which Hobbes can build his argument for political obligation. Rawls saw that Hobbes was “attempting to identify fundamental interests by which everyone is moved” (LHPP, 67). Hobbes’s actual identification – that we have an interest in securing the conditions necessary for our actions to be effective in achieving whichever ends we actually have, is far less controversial than assuming that in everyone, “our interest [is] first in preserving ourself” (LHPP, 46). Hobbes’s assumption of a desire

9

desire I really do have. Thanks to Susanne Sreedhar for encouraging me to think about the status of the necessary desire. Hobbes indirectly expounds this desire in his discussion of the desire for “powers,” understood as one’s present means to obtain some “future apparent good” (Leviathan X.1). People with desires must want powers (among which Hobbes includes good luck in external circumstances and cooperative allies such as friends, servants, and admirers, who help to advance one’s ends). Hobbes’s claim that people seek “power after power” (Leviathan XI.2) in order to “assure forever” the satisfaction of their future desires is best understood as the quite plausible claim that people sequentially seek various powers over time as they come to foresee the ends they are likely to have, rather than as attributing to each an effort to gain maximum power at every moment. An orderly social environment, by reducing the impediments posed by other people, reduces the types and magnitudes of powers needed to achieve one’s ends.

Taillight Illumination

27

to act effectively is also less contested than an assumption of merely predominant egoism; and the fact that it is it is analytically derivable from our nature as a rational animal, (in contrast to the primacy of selfpreservation), conforms to Hobbes’s stated “scientific” method.10 What Hobbes terms “the condition of mere nature” is a condition in which it is morally permissible for each individual to act on their own private judgment in every matter. In characterizing his imagined prepolitical condition this way, Hobbes alludes to two further features of his conception of the person for the purpose of establishing political obligation: persons are free and equal. Were persons born under an obligation to defer to the judgment of others, including to political authorities, action on their own divergent private judgments would be impermissible. Although Hobbes personally believes that children in fact owe obligations of obedience to the person who preserved them alive – always in the first instance to their mother – for political purposes, people are to be understood to be each and equally at liberty to make and to act on their own decisions. In the condition of mere nature, there is a universal right of each to act on private judgment in every matter. Hobbes does not offer as the basis of equal liberty some theory-laden claim about humans’ possession of innate dignity, or status qua God’s children. Instead, he calls our attention to the observable fact that as adult individuals, we are sufficiently similar in the sorts of capacities needed to impose our will on others or to resist imposition of theirs on us – bodily strength, intelligence, ingenuity, wiles – none of us can reasonably expect to get our way against everyone else. Even if any one of us were marginally superior to others in all the qualities mentioned, others can combine their talents to outdo us. People who are operationally equal will have similar hope to achieve their ends and will see no reason why they should defer to others’ judgments, rather than act on their own. Hobbes writes: In the faculties of body and mind … the difference between man and man is not so considerable as that one man can thereupon claim to himself any benefit to which another may not pretend as well as he. (Leviathan XIII.1) If nature therefore have made men equal, that equality is to be acknowledged; or if nature have made men unequal, yet because men that think themselves equal will not enter into conditions of peace but on equal terms, such equality must be admitted. (Leviathan XV.21)

This basis of equality is thinly political. The reason for assuming natural freedom is similarly thin: natural equals have no cause to accept 10

If Rawls’s idea of a political conception also requires that it be worked up from the public political culture, then Hobbes embraced a more minimal condition.

28

S. A. Lloyd

purportedly natural normative relations of domination and subordination. Hobbes argues that if it is permissible for you to judge all matters, including matters pertaining to me – such as the correctness of my judgments about how to act – it must be equally permissible for me to judge matters pertaining to you – including the correctness of your judgment about the correctness of my judgment as to how to act.11 The final element of Hobbes’s conception of the person is, like the necessary desire, entailed by his expansive notion of reason: qua reasoning creatures, humans adduce reasons, or justifications, for their actions (and beliefs), which reasons are subject to a consistency constraint. Not only does reason forbid simultaneously affirming contradictory beliefs, it also prohibits affirming and acting on contradictory practical principles. The ability to hold oneself to the same standard of practical action one applies to others – the capacity for reciprocity – is a central feature of Hobbes’s conception of the person; this ability makes it possible for Hobbesian persons to comply with the most basic natural law requirement – reciprocity – the common core from which all other laws of nature are derived. Hobbes refers to the reciprocity requirement variously as the “sum” of the laws of nature, as “containing” all the laws of nature, or says that it “just is” the law of nature. He formulates it more than a dozen times, both positively and negatively as dictating Whatsoever you require that others should do to you, that do ye to them. (14.5) Do not that to another which thou thinkest unreasonable to be done by another to thyself. (Leviathan XXVI.1)

In its political application, reciprocity demands That no man require to reserve to himself any right which he is not content should be reserved to all the rest. (Leviathan XV.22)

Notice that this is a “weak” reciprocity requirement in the sense that it never requires you to adopt the behavioral standards of others; someone else’s choice to treat you according to some general principle they propose does not impose any duty on you to conform to that principle in your treatment of them. Nor does it require you to adopt universalizable standards that could be adopted by all from some impersonal point of view. Rather, your own demands or evaluative judgments or practical attitudes set a standard for action that reason then requires you to apply 11

Hobbes writes, “say that another man is judge. Why now, because he judgeth of what concerns me, by the same reason, because we are equal by nature, will I judge also of things which do belong to him. Therefore it agrees with right reason, that is, it is the right of nature that I judge of his opinion.” (De Cive I.9)

Taillight Illumination

29

across the board – including to your own actions – and to accept from others as justifying their like actions, even in circumstances when doing so does not advance your self-interest. I mentioned earlier that Rawls viewed Hobbes’s laws of nature as a “secular morality” offered as part of a political conception. Some further evidence for that view, which Rawls does not mention, is that the moral virtues expressed in the laws of nature do not include any traditional moral virtues that redound to the personal benefit of the agent who possesses them without necessarily benefiting her political community. Hobbes excludes temperance, prudence, and courage from among the dispositions required by the laws of nature precisely because those virtues in an enemy may prove detrimental to the flourishing of our community. These “are not virtues of citizens as citizens, but as men,” whereas “good manners (that is, moral virtues)” are dispositions that facilitate and preserve civil societies (De Homine XIII.9). Hobbes’s laws of nature pick out only virtues or vices of persons engaged in social relations12 and not of humans per se. Hobbes’s reciprocity requirement is not itself justified instrumentally as necessary for securing self-interest and is thus not reducible to “the rational.” In fact, violating the reciprocity requirement is not always irrational, because, as Hobbes concedes, some people, in particular powerful people, can correctly expect to profit by doing so; Hobbes acknowledges that in this life (the only one we can observe) the wicked do prosper, many a time. Reciprocity is an “eternal and immutable” requirement binding at all times on everyone with the use of reason: Yet if its justification depended on its unfailingly serving self-interest, it could not have that status. Compare Hobbesian reciprocity with Rawls’s standard of reasonableness – which is that we be ready to propose and to abide by terms of cooperation we think others could, as free and equal persons, reasonably accept, so long as others also honor those terms. Hobbesian reciprocity does require willingness to abide by the standards we propose, and, because we believe ourselves (who are proposing them) to be reasonable, and we accept those standards, we obviously think a reasonable person could accept them. Rawls’s worry is that differences in bargaining position may enter here, and it is true that Hobbes makes no provision to abstract from those; but recall that Rawls was devising a theory of justice, whereas Hobbes was pursuing the different project of justifying submission to an absolutist government. If we agree that even imperfectly just 12

The law of nature requiring allowing safe passage to mediators would apply to relations among political communities and between hostile factions within a political community.

30

S. A. Lloyd

governments may still be better for securing human interests than no government at all, we should not define the reasonableness requirement for social cooperation so narrowly that no arrangements short of perfectly just ones will count as any form of social cooperation. Hobbesian reciprocity enables principled, internally motivated cooperation, as opposed to mere coordination imposed by external coercion. Reason’s reciprocity requirement has a deep hook in human psychology as Hobbes understands our psychology: Humans care very much to be and to be seen by others as being justified by reason in their actions, as each can introspectively confirm.13 Hobbes’s persistent insistence on the influence on our social interactions of our pride, self-conception, and concern for status (which he discusses under the headings of glory and vain glory as well as pride) explains why we feel it degrading to be exposed as deviating from a requirement of reason. Our higher ability to reason is what distinguishes adult humans from nonhuman animals. Actions “that proceed from error, ignorance, or folly [are] dishonorable” (Leviathan X.42); and “craft, shifting [cheating], neglect of equity is dishonorable” (Leviathan X.46). We take offense when others disagree with us on multiple matters because we see such disagreement as impugning our status as intelligent reasoners. Hobbes makes clear that the main way in which people depart from reason’s requirements is by hypocritically holding others to different practical standards of conduct than they apply to themselves. Rawls rightly observed that Hobbes’s laws of nature, both the reciprocity requirement itself and the specific secondary laws dependent on it for their derivation, “define a family of reasonable principles so far as their content and role discern” (LHPP, 64). But we can now see that reciprocity “accords with reason,” not because, or only when, it instrumentally serves personal desires, but rather because reason imposes a consistency constraint on our judgments and attitudes as well as on our beliefs. Reason dictates in more than one way – forbidding both logical contradiction and failure to fit perceived means to ends. Hobbes criticizes the reputedly wise Cato on the ground that with him “animosity should so prevail instead of judgment, and partiality instead of reason, that the very same thing which he thought just in his popular state, he should censure as unjust in a monarchical.” And the fact that one “gives a different judgment of an action when he does it than when someone else does the very same thing … [is among] the obvious signs that what moral 13

For discussion, and an argument that Hobbes views the desire for self-justification as a potentially powerful motive for conformity with the moral norms articulated in his laws of nature, see Lloyd (2020).

Taillight Illumination

31

Philosophers have written up to now has contributed nothing to the knowledge of truth” (De Cive Dedicatory Epistle). It is contrary to reason to hold contradictory attitudes toward the same justifying consideration, or reason, for a fixed action-type (an action under a general description). When we are acting, not on instincts or whims, but on proffered reasons, we are committed to the consistency requirement articulated by the reciprocity rule. Hobbes illustrates violation of the reciprocity requirement with the biblical story of the prophet Nathan’s rebuke of King David over his appropriation of Uriah’s wife, Bathsheba. Nathan poses David’s own action to David by way of an analogous case to be judged of a rich man who had many lambs but chose to sacrifice the only lamb of a poor man: When David judges “the man that hath done this thing shall surely die” the prophet answers, “Thou art the man.” David is rebuked – on Hobbes’s reading – not for coveting another man’s wife, nor even for causing Uriah’s death in order to gain her, but rather for inconsistently approving in himself the very same type of action he condemns in another (the rich man of the story). Another striking example is Hobbes’s application of the reciprocity requirement to defend his position that whatever a subject like the Christian convert Naaman, who bowed before an idol, does, not because he approves it but because the civil law requires it, is the action of his sovereign, and not of himself, and so is permissible. Hobbes reasons I ask [any objector], in case there should be a subject in any Christian Commonwealth that should be inwardly in his heart of the Mahomedan Religion, whether if his sovereign command him to be present at the divine service of the Christian Church, and that on pain of death, he think that Mahomedan obliged in conscience to suffer death for that cause, rather than to obey the command …. If he say he ought rather to suffer death, then he authorizeth all private men to disobey their Princes, in maintenance of their religion, true or false; if he say, he ought to be obedient, then he alloweth to himself, that which he denyeth to another, contrary to the words of our Saviour “Whatsoever you would that men should do unto you, that do ye unto them,” and contrary to the Law of Nature, (which is the indubitable and everlasting Law of God) “Do not to another that which thou wouldest not he should do unto thee.” ((Leviathan XLII.11), emphasis added)

More generally, Hobbes appeals to the reciprocity requirement in his condemnation of proselytizing efforts to convert foreign populations to our religion: The missionary “does that which he would not approve in another, namely, that coming from hence, he should endeavor to alter the religion there” (Leviathan XXVII.4). Notice that none of these condemnations of violation of the reciprocity requirement appeals to instrumental irrationality. The criticism is not

32

S. A. Lloyd

that one is failing to take necessary means to one’s own ends. The criticism is of inconsistency in one’s practical principles – a type of hypocrisy – in affirming moral standards by applying them to others, while not conforming one’s own behavior to those standards (not merely as a matter of weakness of will), but denying their application to oneself. What goes on in these violations of reason’s reciprocity requirement? Hobbes characterizes reasoning as calculative, as adding or subtracting “names in our affirmations,” in a syllogistic process.14 We can make sense of Hobbes’s condemnation of practical inconsistency if we understand him as holding that when a person offers a reason or justification for some action, evaluative attitude, or practical judgment – let’s call this an “item” – she is committing herself to a general claim – no indexicals or definite descriptions allowed – from which her item is supposed to follow by syllogistic argument. For example, taking “because” to mark a proffered reason or justification (a) I fault your behavior because it is harmful commits me to the general claim. (A) harmful behavior is faultworthy. By Hobbes’s account, if the “because” clause is to provide a candidate reason, it must follow deductively from a general claim/principle that properly interrelates its component terms. Thus, only if “is faultworthy” contains everything “named” by “is harmful” has (a) expressed a reason for faulting your behavior. This creates a tight link between Hobbes’s conceptions of reasoning and of having/being/offering a reason. However, this link operates in both directions between general principle and specific conclusions, so the general principle that justifies my particular “item” will also justify anyone else’s like item. And my tacit appeal to that principle in claiming reason for my own item commits me to acknowledging equal reason for anyone else’s like item. So, if I offer (a), I then behave contrary to reason when I refuse to fault my own harmful behavior or refuse to accept from you that the fact that my behavior is harmful is also a reason for you to fault it. This is because those practical stances, when converted into propositional form – (a0) my harmful behavior is not faultworthy, (a00) that my behavior is harmful is no reason for you to fault it – directly contradict the general claim (A) upon which my proffered reason depended. All right reasoning in any domain depends on consistency. The reciprocity rule expresses reason’s requirement of consistency in the domain of 14

He writes “that making of syllogisms is that we call RATIOCINATION or reasoning.” (Elements I.5.11)

Taillight Illumination

33

the practical. That is the claim it makes on humans, for whom, qua “rational animals”, reason is normative. The reciprocity requirement is not justified instrumentally, as contributory to the satisfaction of any desire, and so is not reducible to a tenet of means-ends rationality. It expresses a requirement of reasonableness, which Hobbesian persons do accept and which they have the capacity to meet, completing Hobbes’s conception of the person as free and equal, rational and reasonable, and having the desire to secure the conditions necessary for acting effectively in pursuit of the ends she has. This does look like an appropriately political conception of the person. 1.3

A Better Argument from the State of Nature for Submission to Government

Assembling these elements, we can now state Hobbes’s argument: In the state of nature, (1) each individual has moral liberty to act on her own private judgment in every matter, with no duty to defer to anyone else. In this sense, persons are free. (2) Persons are also equal, meaning that none can rationally expect to impose his will over others or “claim any benefit they cannot claim as well.” (3) Persons have ends they desire to achieve by their actions, and, as rational, (4) they desire that conditions for their actions to be effective in achieving their ends obtain. (5) Persons are capable of reciprocity – of holding themselves to the same standards they demand others meet – and insofar as they act in conformity with reason, abide by the reciprocity requirement. Hobbes then argues: Because individuals not infrequently disagree in their practical judgments and pursue incompatible ends, (as experience confirms), the condition of universal private judgment can be expected to result in mutual interference, and insecure access to needed resources (including the cooperative aid of others), that seriously compromises any given person’s prospects for acting effectively to achieve their ends. Rational persons must therefore demand that others give up their right to act according to their own private judgment in everything. But, by reciprocity, what we demand of others we too must do. Thus, reason dictates “that a man be willing, when others are too, to give up his right to all things, as far as he thinks necessary, and be contented with so much liberty against other men as he would allow other men against himself” (Leviathan XIV.5). This is Hobbes’s second law of nature, operationalized by undertaking political obligation, because the only reciprocal alternative to governance by universal private judgment is universal deference in some range of matters to a public judgment that can arbitrate disputes and enforce rules to create a navigable social environment in which we can expect to effect our ends.

34

S. A. Lloyd

This certainly looks like an argument “freestanding” from controversial comprehensive doctrines.

1.4

Conclusion

Rawls’s query whether Hobbes may have been the first political liberal is tough to answer partly because it runs so contrary to the understanding of the basic contours of Hobbes’s political philosophy that has been dominant for a good three hundred years and reinforced in the latter half of the twentieth century. There was an influential school of interpretation, beginning now fifty years ago with David Gauthier in the late 1960s and developed by Greg Kavka and Jean Hampton in the 1980s, that applied a thick layer of alien rational choice theory and game theory over Hobbes’s original argument that assumed Hobbesian persons to be narrowly rational, wholly or predominantly self-interested actors.15 My argument here has indicated how those interpretations distort Hobbes’s political philosophy beyond recognition. They are false to the complex psychology of Hobbesian persons (and of us); they completely obscure the moral basis of Hobbes’s system; they oversimplify his conception of the operations of reason; and they would render ineffective Hobbes’s primary recommendation for preserving a well-designed commonwealth, which is, not to change the payoff matrix by instilling terror in the population, but to reform university education and pulpit preaching to acculturate citizens in their moral, political, and religious duties (Leviathan XXX.4, 7–14; Hobbes [1990] [Behemoth] 39–40, 56, 58–59). I was reminded not too long ago of misunderstood Hobbes by a New York Times report that Johannes Vermeer’s famous painting, “Girl Reading a Letter at an Open Window” turns out to contain a naked Cupid hidden under a rectangle of paint that makes the wall to the letterreader’s right appear totally bare, empty.16 This fact had been discovered forty years ago by X-ray analysis of the painting, but it was only recently, when museum restorers applied a solvent to the paint rectangle, that they realized it must have been applied later by a foreign hand and that the 15

16

Even Rawls, in his 1983 teaching lectures on Hobbes, described Hobbes’s state of nature as having a structure analogous to the prisoner’s dilemma game (LHPP, 73–78, 88–90), which explained why the advantages of adhering to agreements and observing the requirements of natural law are unavailable to individuals living without a sovereign enforcer. Less guardedly, Rawls (1999, 238), states that “Hobbes’s state of nature is the classical example” of “the general case of the prisoner’s dilemma.” See also Gauthier (1969, 79–85); Kavka (1986, 109–13) and passim and earlier articles; Hampton (1986); Curley (1994, XXIV–XXV). www.nytimes.com/2021/09/09/arts/design/vermeer-cupid-restoration.html?smid=url-share.

Taillight Illumination

35

Cupid was in fact Vermeer’s, a realization that altered and enhanced scholars’ understanding of the painting’s intended meaning. (For instance, that the girl was more likely reading a love letter than a vender’s demand for payment). I am hopeful that a similar sort of lost-and-found is happening in Hobbes studies. However well-intentioned the application of the shiny new tools of game theory to Hobbes’s political philosophy may have been, the result was to obscure some of its most important elements, including its appeal to citizens’ reasonableness as a central and irreducible presupposition of the theory and its recognition that citizens’ transcendent religious or moral interests can be significant sources of social disorder.17 Thinking now in the Rawlsian terms of political liberalism about Hobbes is not exactly like applying solvents to paint overlays on an Old Master, but it does aid us in washing away the innovative additions of Hobbes’s later interpreters. Rawls’s political liberalism provides “taillight illumination” of Hobbes by enabling us to notice those thick and controversial assumptions about human nature and motivation both Hobbes’s admirers and detractors coated him in and by supplying tools for understanding what Hobbes was doing that reveal him to have been more original and interesting than we had imagined.

17

One might wonder whether such prisoner’s dilemma arguments could helpfully supplement Hobbes’s system, not in their advocates’ intended role of providing the state of nature argument for submission to government but by assuring us that reasonability and rationality align for the most part, so that we can see our commitment to reciprocity as a part of our good, in turn enhancing social stability by limiting defections from cooperative norms. When Rawls addressed the question in A Theory of Justice whether having a sense of justice is a part of our good according to the thin theory, he asked this about members of a well-ordered society who have already acquired a sense of justice. Hobbes, in his famous reply to the Foole, engages the more ambitious project of showing that even an atheist without a sense of justice who rebels unjustly acts “against reason” and in discordance “with his own good” (Leviathan XV.5, 91 note 5, Latin variant) because she relies on a faulty rule of inference (viz., if an action turns out well, it cannot have been against reason to perform it), and because she incorrectly extrapolates from experience (expecting to go undetected even though most such deceptions are discovered). This reply depends in no way on game theory modeling of strategic choice. The Foole simply lacks both sapience and prudence and is, as such, a defective person. See Lloyd (2009, chapter 7). In contrast, ordinary people accept the natural law requirement of reciprocity as a dictate of reason and would be embarrassed, ashamed, or offended to seen by others as incapable of reasoning, or as too weak of will to act on their reason, or as hypocrites who endorse for others requirements they are unwilling to accept themselves. For Hobbes, part of our good is to be, and to be acknowledged as being, not overall inferior to others; this provides a natural motive for trying to live up to reason’s requirements. For discussion of how the desires for selfadmiration and the respect of others may motivate compliance with the moral requirements articulated in the laws of nature, see Lloyd (2020).

2

The Theory Rawls, the 1844 Marx, and the Market Daniel Brudney*

2.1

Introduction

In this chapter, I connect A Theory of Justice to certain strands in the history of political philosophy. Rather than do the obvious, namely, to talk of Kant or of Sidgwick’s Methods of Ethics, I will be more speculative. I will place Theory mostly with respect to the young Marx, but themes from Aristotle and other writers will emerge as well. Here is a preview of my argument. In his work of 1844, Marx claims that “the history of industry and the established objective existence of industry are the open book of human essential powers” (Marx 1975b, 302, emphasis in original, translation amended). Toward the end of A Theory of Justice, Rawls asserts that the collective maintenance of a just society is “the preeminent form of human flourishing” and that “persons best express their nature” by maintaining just institutions (TJ 1971/1999, 529/463). Two similarities here are obvious. First, with each writer, a specific joint activity – the production of material goods, the achievement and maintenance of just institutions – is said to be a central component in the way human beings realize their nature. Second, these are basic framing activities. They create the ongoing framework, the specific form of social reproduction, within which other activities, in particular, the pursuit of individual ends, can proceed. In discussing Marx and Rawls, I will note other similarities. Among them are these: what will turn out to make these joint activities central to the human good is in significant part the relationships they establish and maintain among human beings, specifically, among individuals who, for the most part, do not know of one another’s existence, relationships among distant unknowns.

* I thank Paul Weithman and Jan Kandiyali for extremely helpful comments on earlier versions of this chapter.

36

The Theory Rawls, the 1844 Marx, and the Market

37

Moreover, a necessary condition for these relationships to obtain in the right way, that is, for agents actually to be in these relationships, will turn out to be that these relationships are regarded and believed to be regarded in a certain way, that is, under a certain description. There will be a crucial role for the social ethos of a society. This will eventually point to an issue for Theory’s account of justice as fairness. If a standard leftwing critique of the market is cogent, and if the well-ordered society of Theory involves a widespread market, then it might be that the several elements in the desired social ethos of justice as fairness are in tension with one another and cannot all be satisfied simultaneously. I will briefly present the parallels between the 1844 Marx and the Theory Rawls, and then move to the left-wing critique of the market and the possible problem it raises for Rawls. I should note four limitations to my presentation. First, the possible problem is not clearly actual. The issue will depend in large part on whether decisions about the basic structure of society may use certain controversial social scientific considerations. Second, I limit myself to A Theory of Justice, indeed, to the first edition. I invoke bits from Rawls’s later writings but only to make clear what I take to be the account in Theory. Similarly, I mostly limit myself to the 1844 Marx. With each writer, some later texts are at odds with the writer’s earlier work. Third, I will be pushing Rawls’s text hard. Doing so yields what I take to be an appealing account of justice as fairness, but I am pushing the text hard. Finally, the tale I want to tell is a long one. Unfortunately, here, it has to be severely compressed.

2.2

The 1844 Marx

I begin with the 1844 Marx, who holds that the following obtain in a true communist society (TCS).1 (i) Agents engage in the production of goods and services as their mode of individual self-realization.2

1 2

My reading of Marx draws on Brudney (1998). See also Brudney (1997, 2010, 2014). I say “goods and services” because that is the apt expression for the goal the 1844 Marx has in mind. However, Marx’s texts, certainly the early texts, are inadequate in their account of the service activities (as distinct from the factory, fishing, farming activities) that are essential to any developed economy.

38

Daniel Brudney

(ii) Agents engage in the production of goods and services as part of the human species’ joint transformation of nature in order to survive and to live at an increasingly high material and spiritual level. (iii) Agents engage in the production of goods and services in order to provide other agents with the means for the pursuit of their individual self-realization. (iv) Agents recognize and appreciate that others are producing goods and services for them. (v) Agents have concern for (distant, unknown) others’ well-being and agents believe that (distant, unknown) others have concern for their well-being. There is much to say about Marx’s view. I limit myself to what makes for links to justice as fairness. Agents in TCS see their collective activity as what I have called “a basic framing activity.” By producing goods and services and maintaining the relationships needed to do so, agents know they are keeping in place the framework within which social life as a whole is reproduced and individuals are able to pursue their individual ends (Marx 1975a, 227). The relationships in the basic framing activity have a give to / receive from form. The 1844 picture tracks the from each / to each formula of the 1875 “Critique of the Gotha Program” (Marx 1989, 87). That formula specifies not a duty/entitlement pairing but a description of the lives of agents who are concerned for one another’s well-being. They see themselves both as pursuing their individual projects and as giving to and recognizing that they are receiving from one another.3 Like Aristotle, the 1844 Marx finds human flourishing in a particular kind of activity but, for him, the site of such flourishing is neither politics nor philosophy but the ordinary productive activities of ordinary life. Now, TCS has no role for conventional political activity. Nevertheless, ordinary productive activity in TCS is oriented to the common good. Charles Taylor remarks on the modern praise of ordinary life, “My work in my calling ought to be for the general good …. With the affirmation of ordinary life, agapē is integrated in a new way into an ethic of everyday existence” (Taylor 1989, 258). The 1844 Marx accepts this thought. TCS involves a form of connectedness that has two elements. First, there is the confirming element. As in capitalism, in TCS agents make products that others use. However, in contrast to capitalism, in TCS 3

Rawls is among the commentators who see the “Gotha Program” formula as merely descriptive of social life in a true communist society. See Rawls LHPP, 370. For an excellent discussion of this way of understanding relationships in TCS, see Kandiyali (2020).

The Theory Rawls, the 1844 Marx, and the Market

39

agents regard productive activity as simultaneously expressing my individuality and doing something for others. And others’ – that is, consumers’ – recognition and appreciation of what I am doing for them confirms for me that I am doing what counts as realizing my nature (Marx 1975a, 227–28). Your recognition and appreciation of what I have produced reflects back to me a confirmation of my own selfunderstanding. Marx puts it that “our products would be so many mirrors in which we saw reflected our essential nature” (Marx 1975a, 228). Second, there is the completing element. For Marx, this itself has two components. First, I am completed qua seeing myself as human in the sense of a strong identification with the species as a whole. This links Marx to the line of nineteenth-century humanism that focuses on identification with a collective entity, the human species. This focus is found in a number of writers, including Ludwig Feuerbach, Auguste Comte, and John Stuart Mill. Marx is in this line of thought, but one should not overstate his commitment. He does occasionally talk as if it is important to realize the destiny of the human species and, at one point, he echoes the usual rhetoric that the individual is mortal, but the species is not (Marx 1975b, 299). Still, the bulk of the 1844 presentation of TCS stresses that the realization of our species nature involves the right kind of reciprocal relations to other (distant, unknown) human beings. More significant for the comparison with Rawls, identification with others also enables us partially to transcend the division of labor, namely, with regard to our varied outputs. In TCS, we would see others’ products as, in a sense, our own. The 1844 Marx stresses that in capitalism the worker is alienated not just from her own but also from others’ labor and output (Marx 1975b, 277). But then, in TCS, one will be properly related not only to one’s own output but also to others’ output. The importance of this issue varies across Marx’s texts. At times, especially in The German Ideology, his stress is elsewhere, on the “allaround realization of the individual,” that is, on defeating the division of labor’s inhibition of the individual’s development of her varied capacities (Marx and Engels 1976, 292; see also 263). That theme does exist in the work of 1844; but in that work, there is also a strong emphasis on seeing our overall output as our output. In TCS, I relate to your output as also my own in such a way that we live in a social world that we see as something we have collectively built. Both the confirming and completing elements of community are connected to the fact that agents regard themselves as in a give to / receive from relationship. Their sense of themselves as in such a relationship is

40

Daniel Brudney

what ties agents together. Of course, agents in TCS do not stand back, reflect on, and then deliberatively affirm the elements noted above. Rather, Marx thinks that such matters are part of the taken-for-granted background of practical life in TCS, part of what is self-evident. They are simply how the world appears (Marx 1975b, 300 and 305). 2.3

The Theory Rawls

I turn now to A Theory of Justice and specifically its “Kantian interpretation” (TJ 1971/1999, §40). With the Kantian interpretation, the features of human beings that Rawls focuses on are what in “Kantian Constructivism in Moral Theory” he calls our “two moral powers … the capacity to understand, to apply and to act from (and not merely in accordance with) the principles of justice … [and] the capacity to form, to revise and rationally to pursue a conception of the good” (CP 312; see also TJ 1971/1999, 505/442). In the well-ordered society of justice as fairness, he says in Theory, “moral personality” – in effect, the two moral powers – is regarded “as the fundamental aspect of the self” (TJ 1971/ 1999, 563/493). The well-ordered society provides the conditions to realize the fundamental aspect of the self. Here, I can highlight only those features of Theory’s form of justice as fairness that resonate with what I have taken from the 1844 Marx. To begin with, there is the instantiation of a basic framing activity. Just institutions maintain the proper conditions for individuals to pursue their individual ends. There is also a parallel to some of the Marxian forms of connectedness. Rawls stresses that justice as fairness involves what he calls “the values of community” (TJ 1971/1999, 520/456). I want to note three elements of Rawlsian community. First, individuals follow their own life plans but “our person and deeds [are] appreciated and confirmed by others” (TJ 1971/1999, 440/386). This obtains both within individual social unions and within the larger social union of social unions where, by hypothesis, my contribution to our common good is recognized and appreciated (TJ 1971/1999, §79). This is the confirming element of community. Second, Rawls refers to individuals’ life plans as complementary and says that each takes pleasure in what others do. “We appreciate what others do as things we might have done but which they do for us, and what we do is similarly done for them” (TJ 1971/1999, 565/495; see also 529/464). We identify “with the good of others,” and “the successes and enjoyments of others are necessary for and complementary to our own good” (TJ 1971/1999, 501/438 and 523/458). Our interdependence is

The Theory Rawls, the 1844 Marx, and the Market

41

thus crucial to individual flourishing. “Only in a social union is the individual complete” (TJ 1971/1999, 525ft/460ft). I need others not only to provide opportunities for me to exercise my own talents. I also need others so that I can participate in their exercise of their talents. This is the completing element of community. I will get to the third element of Rawlsian community at the close of this section; but before doing so, some further development of the Theory picture is needed. My remarks above suggest that what I have called the give to / receive from form of citizens’ relations obtains in Theory’s well-ordered society. With the 1844 Marx, the primary focus is on giving others physical objects, things that have been produced, but one could also take a broader view. In TCS, individuals jointly give one another not only things but opportunities, specifically the opportunities to pursue their individual life projects and to participate in the process of maintaining proper social arrangements. This also obtains in justice as fairness. Citizens participate in the process of maintaining the social arrangements within which each citizen has the opportunity to pursue her life projects and to participate in maintaining those very social arrangements. Of course, some sort of participation in maintaining one’s social arrangements might obtain in almost any society. With both Marx and Rawls, this is understood to be an activity whose content is collectively determined (with Rawls, it is determined through democratic activity). Importantly, in each case participation in the basic framing activity is regarded as a part of one’s good (TJ 1971/1999, 528/463). There might seem to be a central difference between the well-ordered society and TCS. Suppose justice as fairness involves a widespread market in goods and services. Citizens might see their individual market activities as entirely self-regarding. Of course, if they believe their society is just, they believe that having a market ultimately benefits the worst off. Still, it seems that, merely qua just citizens, they are likely to describe their own activities as giving their fellow citizens not, as in TCS, the means that enable their fellow citizens to pursue their individual good but, rather, justice. (In Political Liberalism Rawls writes that a wellordered society “has the aim of giving justice to all its citizens”; PL, 276.) I will return to this point – the issue of the market in justice as fairness – more than once. What, then, is involved in giving one another justice? I think of it in three ways: as compliance with the laws, as application of the laws, and as political participation. With the first, citizens give one another peace and stability. With the second, they engage in the good faith interpretation that is needed to handle ambiguous situations. Indeed, I take Rawls to

42

Daniel Brudney

imagine citizens in the well-ordered society as resolving most disputes without litigation.4 As for the last thing that citizens give one another, political participation, this involves the many elements of well-functioning democratic governance, for example, voting, providing one another with arguments and ideas about how to maintain and adapt shared social arrangements, an understood commitment to sustain democratic processes and to support their results. Let’s note two opportunities for political participation that a just democratic polity provides. There is the opportunity to attain positions of power, to rule as well as to be ruled, to be “in the room where it happens.”5 Call this Opportunity (1). Opportunity (2) is more mundane. It is the opportunity to act with others for the common good. Every citizen has this opportunity, one that can be instantiated in larger and smaller ways. It is among the opportunities of ordinary life. The ordinary can be seen as the political and be understood to be important as such. In Political Liberalism, Rawls rejects the view that a human being’s “essential nature is most fully realized in a democratic society in which there is widespread and vigorous participation in political life” (PL, 206). Here, I take him to have Opportunity (1) in mind. However, I take both the 1844 Marx and the Theory Rawls to highlight Opportunity (2), not (1). The shared thought is that, within proper institutional arrangements, our reciprocal political and economic dependence makes Opportunity (2) pervasive. Indeed, availing ourselves of it is simply what we do. If one regards what one is doing in a certain way, then one can see one’s ordinary activities – work, discussing the issues of the day, voting, abiding by the results of the vote, etc. – as part of maintaining a just society and so as part of human flourishing. And one can be appreciated by and appreciate others for such activities.6 Here, again, one could invoke Taylor’s remarks on the modern prizing of the ordinary. Let’s push the Marx/Rawls parallels a bit further. Their views are in contrast to views that see the attainment of the proper conditions of social 4

5 6

Rawls writes, “One conception of justice is more stable than another if the sense of justice that it tends to generate is stronger and more likely to override disruptive inclinations.” See TJ 1971/1999, 454/398. I take a strong sense of justice to include a strong desire to resolve disagreements in a way that all will agree is just, and I take that desire, when broadly shared, to make recourse to litigation less frequent. See “The Room Where It Happens,” Miranda (2015). The Rawls of Political Liberalism would probably reject not only (1) but also (2) as something to be fostered for all citizens. He would not want to build any view of human flourishing into a purely political view. I thank Jan Kandiyali for pushing me to note this issue.

The Theory Rawls, the 1844 Marx, and the Market

43

life as, ideally, a once-and-for-all achieved state of affairs. Rather, Marx and Rawls see the basic framing activity as something that, even ideally, should not come to an end. If the good life involves engaging in certain activities, then the good life requires that the conditions for these activities not be transcended.7 Other writers reject this picture. Consider the eternal vigilance that is said to be the “price” of liberty. The metaphor suggests a wish to relinquish vigilance, to have liberty for free. That our vigilance must be eternal is seen as a misfortune, the political form of our regrettable need to get our bread with the sweat of our brow.8 Many writers see political institutions as merely instrumentally valuable, a view Rawls explicitly rejects.9 Such writers seem to see our reciprocal dependence as only a danger and not also a basic opportunity. On this view, dependence simply exposes us to domination. It has no upside. Devices to forestall domination, to insulate us from danger, are, then, pure costs, the price of liberty. The Theory Rawls and the 1844 Marx are certainly aware of the dangers of dependence. Rawls’s two principles are supposed to mitigate those dangers in a justifiable way. Marx is wrong to think that all such dangers will disappear with the abolition of capitalism, but making them disappear is, for him, a major reason for that abolition. Both writers see free individual development as central to a good human life, and that clearly requires protection from domination. Nevertheless, both see the process of overcoming the dangers of dependence as also part of a good life. The alternative tradition, the dependence as mere danger tradition, does not. I take Marx and Rawls to go yet a step further. They do not regret that reciprocal dependence is a condition of human life. One could believe that, given the circumstances of justice, there is value to the activities that respond to those circumstances but also believe that it would be even better to transcend them, for example, to have manna rain down from heaven. I take Marx and Rawls to reject any such aspiration. For the 1844 Marx, the circumstances of justice never really disappear. Rather, in TCS we 7

8

9

Some writers term ends that have this structure, “infinite ends.” See Rödl (2007, 34–43); Pinkard (2017, 41–45 and 140–68). Ben Laurence argues that justice is an infinite end. See Laurence (2021, 73–74 and 99). Phillip Pettit notes, “The writers at whom we have been looking, the writers who identify with the broad republican tradition of thinking, take liberty to be defined by a status in which the evils associated with interference are avoided rather than by access to the instruments of democratic control, participatory or representative.” See Pettit (1997, 30). See also Skinner (1984). Rawls rejects the view that “institutions are not thought to have any value in themselves, the activity of engaging in them not being counted as a good but if anything as a burden.” See TJ 1971/1999, 521/457.

44

Daniel Brudney

continually overcome them in the sense of continually producing the material plenty that obviates the need for rules of distributive justice. And that we must do so – continually produce both material abundance and a certain kind of relationship to others – is believed to be a condition of our good.10 As for Rawls, in discussing the circumstances of justice in Theory and the way they condition what counts as the achievement of autonomy, he writes, “this simply means that the conception of autonomy is that fitting for human beings; the notion suited to superior or inferior natures is most likely different” (TJ 1971/1999, 516/452). Here, he does not quite affirm that it is good that we are the kind of beings we are, but there is no intimation of regret. I suspect he believes it would not be conducive to the human good to transcend the basic conditions of human life.11 He is more explicit in the Lectures on the History of Political Philosophy: “The evanescence of justice, even of distributive justice, is not possible, nor, it seems, is it desirable” (LHPP, 322). And: “The absence of concern with justice is undesirable as such, because having a sense of justice, and all that it involves, is part of human life” (LHPP, 372).12 A final way to put the point is to note that what, in Theory’s last sentence, Rawls calls “purity of heart” is a virtue that requires a particular background if it is to count as a virtue, namely, the conditions of human life (TJ 1971/1999,

10

Rawls does not appreciate this feature of the 1844 Marx. In the Grundrisse (1857–58) Marx stresses the self-realization involved in overcoming necessity. Certainly, labour obtains its measure from the outside, through the aim to be attained and the obstacles to be overcome in attaining it. But … this overcoming of obstacles is in itself a liberating activity … further, the external aims become stripped of the semblance of merely external natural urgencies, and become posited as aims which the individual himself posits – hence as self-realization [Selbstverwirklichung], objectification of the subject, hence real freedom [reale Freiheit], whose action is, precisely, labour.

11 12

See Marx (1973, 611). This passage fits well with the work of 1844 but seems at odds with Marx’s later assertion, in Capital, volume 3, that “the realm of freedom [Reich der Freiheit] actually begins only where labor which is determined by necessity and external expediency ceases; thus in the very nature of things it lies beyond the sphere of actual material production.” See Marx (1967, vol. 3, 820). For Rawls’s view of Marx and the circumstances of justice, see LHPP, 371. I thank Paul Weithman for the formulation in this last sentence. Rawls says further at LHPP, 372: “To act always as we have a mind to act without worrying about or being aware of others’ claims, would be a life lived without an awareness of the essential conditions of a decent human society.” Here, Rawls is criticizing Marx for aspiring to a society beyond the need for distributive justice. However, the 1844 Marx thinks it good for people to provide for one another. He would substitute “needs” for “claims” and accept the following modified Rawlsian thesis: “To act always as we have a mind to act without worrying about or being aware of others’ needs, would be a life lived without an awareness of the essential conditions of a decent human society.”

The Theory Rawls, the 1844 Marx, and the Market

45

587/514). The Theory Rawls thinks that continuing properly to respond to those conditions is a central part of the human good.13 Here is yet one more parallel. It is clear that the agents in the 1844 Marx’s TCS have concern for one another’s good. There, to provide others with means to advance their good is among one’s goals in producing.14 As noted above in my remarks about a market in the well-ordered society, this might be thought to make for a contrast with Theory, since that text seems to focus on giving others justice, not goods and services, on respect rather than concern for others. The 1844 Marx’s focus is on concern; it is often said that Rawls’s focus is on respect.15 There is surely something to this contrast. However, it omits an important part of the story. Rawls presses that there is a way in which, at least in Theory, justice as fairness does involve a concern for others’ good. Consider his assertion that we can see the difference principle as an interpretation of fraternity (TJ 1971/1999, 105–6/90–91). Rawls denies that this involves “ties of sentiment and feeling” across a large society (TJ 1971/1999, 106/ 90), yet merely modeling fraternity in distributive terms does not actually yield fraternity. Fraternity is constituted by particular relations, specifically those involving a concern for one another’s good that agents in those relations recognize and appreciate. If there is to be fraternity in the wellordered society, such reciprocal concern must somehow obtain. In Theory’s well-ordered society, citizens respect one another as what, in “Kantian Constructivism,” Rawls calls “self-originating sources of valid claims” (CP, 330). Still, to see others as sources of valid claims could merely constrain one’s conduct. One might refrain from violating others’ valid claims yet not care how they fare. One would still not do things for others. There would not be fraternity. I take what Rawls is after to go further. My thought is that, without there being implausibly strong ties of sentiment and feeling, citizens can have concern for one another. It does seem possible for me to have concern for distant strangers’ well-being – say, the well-being of those who are, and take this as the full available description, the victims of an earthquake in Siberia. Rawls could and should see justice as fairness as involving concern for the well-being of fellow citizens. How strong must such reciprocal concern be to amount to something in the realm of “fraternity”? I am not sure. At least strong enough to provide significant support to sustaining commitment to the two principles of

13

14 15

There is a difference here between Marx and Rawls. Marx sees the conditions of human life as always involving the need to provide for material wants of many kinds. Rawls, but not Marx, adds the need to deal with human disagreement about the good. See Brudney (2010). This is most explicit with political liberalism. See Larmore (2008, chapter 6).

46

Daniel Brudney

justice.16 Below I offer some textual evidence to think that Theory’s wellordered society is supposed to involve that degree of reciprocal concern. Before doing so, though, it will be useful to consider two non-virtuous, or, rather, two vicious attitudes. Call the first indifference. In The Condition of the Working Class in England, after explaining to a Manchester bourgeois all the ills of working-class life in that city, Friedrich Engels reports: “The man listened quietly to the end, and said at the corner where we parted: ‘And yet there is a great deal of money made here; good morning, sir’” (Engels 1975, 563). Engels’s bourgeois shows no contempt for the workers and takes no satisfaction in their misery. Their condition simply does not matter to him. In his Discourse on Inequality, Rousseau presents another attitude: If one sees a handful of powerful and rich men at the pinnacle of greatness and fortune while the masses grovel in obscurity and misery, it is because the former value the things they enjoy only to the extent that the others are deprived of them, and, without any change in their own state, they would cease to be happy if the People ceased to be miserable. (Rousseau 2019, 189)

Engels’s bourgeois is indifferent to whether the people rise from misery. Rousseau’s rich men are not indifferent: they want the masses to remain in misery. Rousseau’s rich men are rivalrous. They aspire to positional happiness. With each attitude, there is a failure to relate properly to others. One could gloss the failure in different ways: as a failure to treat others as selforiginating sources of valid claims or as ends in themselves; alternatively, as a failure to have concern for others’ well-being. Marx does not deal with positional happiness. Rawls argues that, in the well-ordered society, the aspiration to it would be much reduced (TJ 1971/1999, §§80–82). However, I think that each sees proper citizens as not indifferent to one another. As I say, this is obvious with Marx. With the Theory Rawls, different categories are used at different places. For instance, Rawls writes that “the unconditional concern of other persons and institutions for our good is far stronger on the contract view [than with average utility]” (TJ 1971/1999, 499/437; see also 486/426). And, “The concern which is expressed for all persons by counting each as one (by weighing 16

Rawls’s worry that ties of feeling might be “unrealistic” is a misguided echo of an unfortunate nineteenth-century polemic against Mill where the critics, especially James Fitzjames Stephen, mistakenly assumed that concern for distant others must be an extension of what it is to love an individual – and such an emotional extension would be implausible. For discussion of this issue, see Brudney (2010). For Stephen’s criticism of Mill on this issue, see Stephen (1992, 241).

The Theory Rawls, the 1844 Marx, and the Market

47

everyone’s utility equally) is weak compared to that conveyed by the principles of justice” (TJ 1971/1999, 500/437). Moreover, the two principles publicly affirm that “everyone’s good is included” in the overall scheme of justice as fairness (TJ 1971/1999, 179/156).17 In the wellordered society of justice as fairness, then, citizens would see themselves as regarded in a certain way (as concerned for their good) by their fellow citizens and their shared institutions. I think such concern would in fact be seen as part of one’s commitment to, as well as being expressed by, the two principles. On the other hand, in a more Kantian vein Rawls writes: “Another way of putting this is to say that the principles of justice manifest in the basic structure of society men’s desire to treat one another not as a means only but as ends in themselves” (TJ 1971/1999, 179/156). The Rawlsian desire to treat others as ends is, I assume, not the desire merely to conform to the requirement “Act in such a way that …” but, rather, the desire to respond to others as ends, to regard them, these individuals, as beings whom one takes to be ends, as beings who matter.18 At this point, I indulge a speculative claim. I have just talked of agents mattering to one another. In doing so, my hope is to finesse the usual philosophical categories and to register a similarity behind standard philosophical differences. I take mattering to fit with both a sentimentalist and a Kantian view. I take it to be the everyday thought about our proper relations to others from which philosophers often begin. B could matter to A for many reasons – perhaps because A has sympathy for B, perhaps because A sees in B a rational being, perhaps because B is human. The common thread is the horizontal relation: B matters to A, not merely through a covering principle (“B is a rational being/B is a human being with needs; I should treat rational beings as ends/try to satisfy others’ needs; therefore, I should deal with B a certain way.”) Rather, A responds to B, herself. My claim is that in both TCS and Theory’s well-ordered society agents matter to one another, although Theory expresses the thought in more than one category. I don’t deny that, if required to provide deeper philosophical underpinnings to their views, Rawls and Marx would differ. My claim is merely that, above or beneath such a difference, both

17

18

Rawls says that to have everyone’s good included “supports men’s self-esteem.” See TJ 1971/1999, 179/156. But surely it does so only within a horizontal relation and not if institutions and my fellow citizens are oriented merely toward “justice” and not also toward me. Ronald Dworkin long ago insisted that Theory’s justice as fairness involves both respect and concern. See Dworkin (1977, 180–183, especially 181).

48

Daniel Brudney

the Theory Rawls and the 1844 Marx instantiate a basic positive orientation to others. “Mattering” is intended as a label for this.19 (I am using a concept with great present-day importance. I take part of the point of insisting that “Black Lives Matter” to be to acknowledge and overcome white America’s history, with regard to Black Americans, of somehow instantiating both indifference and a demand for superiority. The history of white America is a history of profoundly failing to respond properly to Black Americans.) In the first edition of Theory, Rawls writes, “in justice as fairness men agree to share one another’s fate” (TJ 1971/1999, 102).20 Such sharing could be merely parallel play, as when we agree to share the bus going downtown. Alternatively, it could be construed as agreeing to become people who matter to one another. There might even be an echo of the final line of the Declaration of Independence, where the signers “mutually pledge to each other our Lives, our Fortunes and our sacred Honor.” Justice as fairness involves hypothetical, not actual, agreement, yet there is a way in which the citizens of the well-ordered society see themselves as having signed on to an enterprise in which their lives are reciprocally entangled. I do not attribute to Rawls the claim that this actually entails reciprocal mattering (it does not), but one might hope – consistent with Theory’s text, Rawls could hope – that such a thing would obtain. Of course, in addition to these parallels, there are also significant Marx/Rawls contrasts. Most obviously, justice is Rawls’s topic. Marx has no use for that concept. And as noted, individuals seem unlikely to be oriented to one another’s good within a market sphere of exchange in the well-ordered society. There is also a third element in Rawlsian community, an element altogether lacking in Marx. This is the shared justification element. We agree to share one another’s fate on the basis of reasons that we share. These must be reasons that we all do, or at least could, see as acceptable. This is clearly the case from the standpoint of the original position (there, a good reason for one is a good reason for all), but it should also be the

19

20

I think many writers have gestured at something along these lines. See, for instance, Diamond (1991). And see Iris Murdoch’s claim that a “just and loving gaze directed upon an individual reality” is the “proper mark of the active moral agent.” Murdoch (1985, 34.) NB: both “just” and “loving.” I do not claim that either Diamond or Murdoch is talking exactly of what I call “mattering.” I do claim that each wants to start from a description of a proper shared way of living with one another and not then transition to some allegedly “deeper” property that is supposed to be the real thing with moral relevance. This remark is absent in the revised version.

The Theory Rawls, the 1844 Marx, and the Market

49

case from the standpoint of a citizen in the well-ordered society.21 Any well-ordered society inevitably restricts some citizens’ liberty and/or lifeprospects in order to facilitate others’ liberty and/or life-prospects. No citizen should see the exercise of such power as arbitrary. Indeed, it is in significant part because citizens believe they share such institutional justifying reasons that, to invoke Aristotle, they can regard their fellow citizens as other selves, that is, can have a form of civic friendship. Marx is not worried about justification to others. He thinks the basic facts about human nature, and so the reasons to achieve and maintain TCS will seem obvious in TCS, as self-evident as any unhindered sensory perception. In TCS, he says, “the senses have … become theorists directly in their practice … the essential reality of human beings … has become evident in practice, sensuously perceptible” (Marx 1975b, 300 and 305, emphases in original, translation amended). The giving of justifications will be unnecessary.22 2.4

Theory’s Social Ethos and the Market

A couple of times I have mentioned that the market is a sphere in which the give to/receive from orientation might not obtain in justice as fairness. It is time to focus on this issue. So I now shift gears and examine the implications of a standard left-wing critique of a widespread market in goods and services for the social ethos of Theory’s justice as fairness. This critique says that the desiderata of that ethos would be significantly undermined. Since TCS excludes the market, there is no analogous problem for Marx. Therefore, from here on, and although his work is a major source for the left-wing critique, I leave Marx aside. Rather, I explore the impact of the market as seeming to point to a problem internal to Rawls’s view. The concept of a social ethos descends from such earlier concepts as the eighteenth century’s “opinion” and Hegel’s Sittlichkeit.23 Recent writers spell it out in a wide range of ways.24 For our purposes, two things are important. First, a social ethos captures how citizens regard 21 22 23

24

Rawls stresses this in 1957 in “Justice as Fairness.” See CP, 59. T. M. Scanlon mentions the relevant passage in Scanlon (2003a, 161). See Brudney (1998, chapter 6). See also Brudney (2001). For “opinion,” see David Hume, “Of the First Principles of Government,” in Hume (1987) and see Rousseau (1997, book II, chapter 12). For Sittlichkeit, see Hegel (1991, Part III). The idea of a social ethos goes back to Plato’s discussions in The Republic of the different personalities that correspond to the different kinds of state. For G. A. Cohen, a social ethos is “the set of sentiments and attitudes in virtue of which [a society’s] normal practices and informal pressures are what they are.” See Cohen (2008, 144). For another account of a social ethos, see Wolff (1998, 105).

50

Daniel Brudney

both one another and the goal or meaning of their shared social practices. Seemingly similar practices can have different ethoi, as when a musical group stops playing for fun and becomes professional (see TJ 1971/1999, 523–25/459–60).25 And second, while a social ethos involves beliefs, sentiments, and attitudes, these need not be articulated or reflected upon. They might simply be taken for granted. As Judith Shklar puts it, they might express what seems to be “common sense” (Shklar 1990, 17). That how we regard one another and our shared practices is part and parcel of how we live with one another is not news. (Aristotle insists on the point.)26 Still, it is important to specify the form this takes in justice as fairness. In Theory, Rawls writes that “the difference principle represents, in effect, an agreement to regard the distribution of natural talents as a common asset” (TJ 1971/1999, 101/87, emphasis added; see also 179/ 156).27 And in the well-ordered society, citizens will act not merely in accordance with the two principles but from a sense of justice, and this will be publicly understood; that is, citizens will regard one another as so acting. Moreover, a society can be a social union of social unions only if citizens regard it as such, regard it as a shared activity with a certain point. To agree to share one another’s fate is to regard the well-ordered society in a certain way. Rawls also insists that the two principles “publicly express men’s respect for one another” (TJ 1971/1999, 179/156). And (invoking concern rather than respect), he says, “there is a sense in which acting in accordance with [the two principles] can be interpreted as acting from a concern for mankind or for the good of other persons” (TJ 1971/1999, 486/426). The expressive role of the two principles can operate only if citizens interpret – that is, regard – the two principles in a particular way and so regard one another in a certain way. Now, a standard critique of the market is that it makes human relationships instrumental. In the market, each of us is oriented toward her own interests and ignores others’ interests. Moreover, such instrumentalization spills beyond the market. At the limit, it pervades human life. We become as mutually indifferent as the Manchester bourgeois. This

25

26 27

See also PL, 320–21. An orchestra in which musicians want to play for the sake of playing and want to do so with other musicians (and not, say, with wonderfully programed robots) seems best to capture Rawls’s idea of a social union. Rawls uses the analogy of the shared enjoyment in the good play of a game. See TJ 1971/1999, 525–26/461. Here is Terence Irwin on relations among Aristotle’s ideal citizens: “All this is true if and only if each friend regards the other as another self.” See Irwin (1990, 91). The revised addition says, “in some respects a common asset.” See also 179/156.

The Theory Rawls, the 1844 Marx, and the Market

51

critique is often associated with Georg Lukács’s concept of reification (see Lukács 1972). With a nod to Engels, I will call it the indifference thesis. A second standard critique is that the market induces us to see human life in competitive terms. It generates a rivalrous society, where we are envious and grudging, where our happiness requires believing we are ranked above others. We become like Rousseau’s rich person. (We seek positional happiness.) Call this the rivalry thesis.28 These vicious attitudes are different from one another. They might be mutually incompatible. Still, either singly or jointly, they are a threat to the desired ethos of justice as fairness.29 Either singly or jointly, they would undermine the give to/receive from relation among citizens. That relation expresses the fact that citizens matter to one another, but in an indifference society, they do not matter to one another, and in a rivalry society they matter in the wrong way. If citizens assume that other citizens are indifferent or rivalrous, it would not be rational to interpret their compliance with just principles “as acting from a concern … for the good of other persons” (TJ 1971/1999, 486/426). Nor, I think, would it be rational to regard the distribution of talents as a common asset, as a collective possession. Indifference and rivalry threaten the confirming and completing elements of community in justice as fairness. If citizens’ relations are substantially indifferent or rivalrous (or, somehow, both), citizens will not properly recognize and appreciate or identify with one another, and these two elements of community are unlikely to obtain.30 Thus far I have not discussed the social bases of what Rawls thinks of as “perhaps the most important primary good,” namely, self-respect (TJ 28 29

30

For an excellent discussion of “Rivalry-Defining Arrangements,” see Hussain (2020). For a more exhaustive list of the alleged ills of the market, see Lukes (2018). In a very perceptive essay, Stuart White raises a variant of this worry. His proposed solution is for the parties, at the constitutional stage, to opt for what he calls a “republican” form of citizenship. White generates the worry from within propertyowning democracy. I am agnostic about how far indifference and rivalry problems would arise in such a context (see Section 2.5). However, if White is correct and they would arise there, then the problem goes even further than in my presentation. See White (2014). Paul Weithman has been thinking about why the market ought not to count as a form of social union: after all, some social unions, such as competitive games, also have winners and losers. Weithman notes that market exchanges involve no analogue to the idea of a good and fair play of the game. (Personal communication.) It might be that Weithman’s thought is this: When done right, fair play in a game is not merely adherence to the rules. If a player on the other side is injured and has to leave, my own team might remove a player to even the sides again. This resonates to the difference principle – it expresses our unwillingness to take advantage of mere natural happenstance in our relations to others. By contrast, taking advantage of others’ situation is what market participants do.

52

Daniel Brudney

1971/1999, 440/386). It seems likely that indifference and rivalry would undermine these bases as well. In the well-ordered society, self-respect is supposed to be supported in several ways. Citizens would see one another as equal bearers of rights, and this would be more important to self-respect than comparative wealth and income; moreover, each citizen would belong to one or more social unions in which her contributions would be recognized and appreciated; finally, citizens would see themselves as recognized and appreciated as equal members of a just social union of social unions (TJ 1971/1999, 544–46/477–78; see also 440/386). Unfortunately, in a modern rivalry society, wealth and income are often the central metric of superiority. If, despite what Rawls hopes, self-respect becomes tied to relative wealth and income, and, as would then seem probable, requires that one be in a top group, it will become a scarce good. And if people are either indifferent or rivalrous (or some nasty amalgam), it is hard to see how social unions can flourish to the degree that Rawls’s view requires. Finally, a social union of social unions needs citizens to identify with one another and to regard their social world as a collective enterprise in which all participants matter equally. In an indifference and/or rivalry society, such identification and reciprocal regarding seem unlikely to obtain. Overall, if the indifference and rivalry theses are true, a market society endangers crucial elements of Rawlsian community as well as the social bases of self-respect. Is there a Rawlsian solution to this problem? Would market restrictions whose goal would be to forestall or remedy indifferent and/or rivalrous relations be an acceptable solution? Rawls thinks that at least one thing, liberty, has priority over greater economic output. At some not forbiddingly high level of output, the importance of additional wealth and income diminishes relative to the importance of activities whose protection requires the liberty principle (TJ 1971/1999, 541–45/ 474–78).31 Maybe at some not forbiddingly high level of output Rawls would accept that a proper social ethos is more important than greater material plenty. And fostering the give to/receive from relation should be important to Rawls. A society in which citizens are mostly indifferent or rivalrous is quite unappealing. Also, Rawls insists on preventing “the social conditions that undermine self-respect” (TJ 1971/1999, 440/386). Buttressing the social bases of self-respect might provide a further reason to limit the market.

31

The argument in the revised edition is somewhat different than in the first edition.

The Theory Rawls, the 1844 Marx, and the Market

53

My point has been that, if we grant a common left-wing critique, Rawls might have important reasons internal to justice as fairness – call them Rawls reasons – to severely limit the market. Of course, objections could be made to the left-wing critique. Most importantly, the indifference and rivalry theses each relies on a causal claim about the market’s impact on human psychology. Why think that either’s claim is true? Why not think instead that human nature simply has certain vices? Indifference and rivalry are hardly unique to capitalist societies. They have existed forever. For utter indifference to others’ wellbeing, read Thucydides; for rivalry, read Homer (and that is just the ancient Greeks!). Isn’t it mere table-thumping to say that such features of human life are dramatically more widespread in market societies? This brings us to a third thesis in the standard left-wing critique. In The Myth of Ownership, Liam Murphy and Thomas Nagel remark that features of a society that are “sufficiently pervasive can come to seem like a law of nature – a baseline for evaluation rather than something to be evaluated” (Murphy and Nagel 2002, 8). The idea is that social institutions might generate the false belief that the status quo reveals our hard-wiring, our basic possibilities. The limits of the present falsely seem to be the limits of our nature. Any claim that social life could be otherwise seems to go against common sense. Call this the distorted baseline thesis. It is a very old thesis, going back to Plato. The left-wing version says that a widespread market generates the deep belief, the common sense, that pervasive indifference and rivalry are simply how things are and will inevitably be with human beings. They seem to be our nature, the natural human baseline. This version of the distorted baseline thesis (hereafter, simply “the distorted baseline thesis”) says that such skepticism about the market’s impact – skepticism that the market significantly expands the scope of indifference and rivalry – is itself part of the pernicious impact of the market. In effect, it becomes part of the reigning social ethos. Rawls thinks that, once past a certain level of material plenty, citizens in the well-ordered society might find participation in a variety of nonmarket activities (e.g., family, culture, sports, religion, politics) sufficiently important that they would opt out of maximizing income and wealth. The market’s impact on individuals’ lives might abate. However, this would result from individual choices. I have claimed that in a similar way sustaining ties of community and/or a give to/receive from relation and/or the social bases of self-respect could come to be more important than greater income and wealth. But if the left-wing critique is cogent, sustaining these things would probably require state action severely restricting the market, that is, state action restricting individual choices.

54

Daniel Brudney

One would then need to justify such restrictions to those whose choices would be restricted. 2.5

The Apparent Dilemma

In this final section of the chapter, I raise the worry that severe market restrictions, if justified by the theses sketched above, might undermine the third element of Rawlsian community, the shared justification element. Legal restrictions aimed at fostering a Rawlsian social ethos might themselves undermine an aspect of that ethos. Jointly, the indifference, rivalry, and distorted baseline theses claim that a particular element of social life perniciously affects important beliefs, sentiments, and attitudes and makes having these perniciously affected beliefs, sentiments, and attitudes seem natural, common sense, truth-tracking. To claim they are not is to shoulder a heavy epistemic burden. As noted, one might point out that these beliefs, etc. have existed in pre-market societies. One might also claim that if they are more extensive in modern societies, the culprit is modernity, not the market. The leftwing critic would respond that, as a matter of the best empirical arguments, the market is the culprit. She would draw a parallel to Theory’s willingness to let social scientists determine whether the market is likely to be more economically productive than central planning (TJ 1971/ 1999, §42). She would try to meet her epistemic burden by saying, “let the best social scientists win.” Can things be that simple? On the one hand, the parties behind the veil of ignorance are supposed to have “knowledge of the general facts about society” (TJ 1971/1999, 547/479). They have “true general beliefs” (TJ 1971/1999, 547/480) and “must consider the general facts of moral psychology” (TJ 1971/1999, 181/157). These are conditions for making a rational choice among candidate principles of justice. So far, so good for the appeal to social science. But there is a problem. Rawls wants the citizens in the well-ordered society to be able to justify their fundamental institutions to one another – this is the shared justification element. To do this, they must agree on what counts as a legitimate reason to justify state coercion. In Theory, Rawls writes: “The reasoning leading up to the initial agreement is to be accessible to public understanding … we must rely upon current knowledge as recognized by common sense and the existing scientific consensus” (TJ 1971/1999, 548/480). Keep in mind that the common sense and scientific consensus in question are those of the well-ordered society. Political Liberalism makes this clear.

The Theory Rawls, the 1844 Marx, and the Market

55

A liberal political conception include[s] … guidelines of inquiry that specify ways of reasoning and criteria for the kinds of information relevant for political questions … those guidelines and rules must be specified by reference to forms of reasoning and argument available to citizens generally [i.e., in the well-ordered society], and so in terms of common sense, and by the procedures and conclusions of science when not controversial. (PL, 223 and 162)

Of course, at the legislative stage, decisions must sometimes rely on controversial scientific claims. Decisions might have to be made, say, about which type of vaccine research to fund, where there is not yet a settled scientific consensus. Still, I take Rawls to hope that there will be substantial agreement on “the procedures of science” and that ordinary citizens will sufficiently understand and accept those procedures. More important, although policy decisions might have large consequences, they are fundamentally different from the decision to severely restrict a central element of the basic structure of society (the market). About the latter decision, I take the Theory Rawls to believe that citizens ought to be able to rely upon “common sense and the existing scientific consensus.” It is true that Theory’s specific invocation of common sense, etc. is for “the initial agreement” (TJ 1971/1999, 548/480). However, the Theory Rawls seems not to have taken into account the possibility of fundamental controversy about social theory. For example, he says that at the constitutional stage agents understand “the principles of social theory” (TJ 1971/1999, 197/172) – but the left-wing critique suggests that some important principles are in dispute. My thought is that Theory implicitly extends the reliance on common sense and the scientific consensus to the choice of such things as central parts of the basic structure but that it is overly sanguine about how this will work. Here is the possible problem. If we grant the distorted baseline thesis, the indifference and rivalry theses are unlikely to reflect common sense in a well-ordered society with a widespread market. In such a society, common sense will see the relevant tendencies as hardwired into us, not as consequences of market relations. And the indifference and rivalry theses are probably not within “the existing [social] scientific consensus.” Certainly, they are difficult to test by procedures thought to be uncontroversially compelling. Moreover, there is a long history not only of market critique but also of market encomium. The market has been said to promote desirable personal development, eliciting talent, effort, selfreliance.32 In a well-ordered society with a widespread market, there is

32

There are also writers who deny that the market must generate indifference and/or rivalry. See Bruni and Sugden (2008).

56

Daniel Brudney

likely to be significant controversy both about what the market actually does to us as well as about whether what it does warrants restriction. Yet, as noted, maybe a Rawlsian legislature would be the proper place to respond to claims about the market’s impact. Dealing with some disagreements about values (say, mass transit versus highways) seems within its purview and, as noted, sometimes legislators must decide between competing empirical claims.33 In addition, Rawls countenances the possibility that in the well-ordered society there would remain an urge for greater status based on income share (the desire for positional happiness), and he suggests a legislative fix to prevent it from undermining the social bases of self-respect (TJ 1971/1999, 546/479). A Rawlsian legislature seems to have scope to address some competition-generated social ills. However, Rawls’s specific proposal here is merely to adapt the application of the difference principle, not to severely restrict market activity. I think an issue remains, even at the legislative stage, about justifying severe market restrictions if neither common sense nor social science can sufficiently uncontroversially endorse the indifference and rivalry theses. Assessing the distorted baseline thesis is even more challenging. It says that current common sense – and specifically, common sense skepticism about the indifference and rivalry theses – is epistemically suspect. But that is unlikely to be a deliverance of common sense. Is the distorted baseline thesis redeemable “by reference to forms of reasoning and argument available to citizens generally”? That seems very unlikely. It amounts to a theory of social epistemology; in effect, it amounts to a theory of false consciousness. Such things will surely be controversial. The worry is this. To severely limit the market is profoundly to affect the basic structure of society. If we accept the indifference, rivalry, and distorted baseline theses, there might be good reasons internal to justice as fairness – Rawls reasons – to severely limit the market. However, there is also a reason internal to justice as fairness – another Rawls reason – not to make a central decision about the basic structure by invoking reasons that fail to rely upon “common sense and the existing scientific consensus.” It might be very difficult to justify the indifference and rivalry theses via such reliance – even more so with the distorted baseline thesis. If the indifference and rivalry theses are true, the market threatens Rawlsian values, but if we rely on theses that are currently controversial (i.e., in the well-ordered society) to justify market restrictions intended to promote

33

T. M. Scanlon makes this point with a similar example in Scanlon (2003a, 163).

The Theory Rawls, the 1844 Marx, and the Market

57

community and self-respect, then we threaten a further Rawlsian value, shared justification. Let’s look at things from a different angle. According to the indifference and rivalry theses, the market causes significantly bad things. Yet severely restricting (not to mention completely eliminating) the market involves restricting liberty. To raise this as an issue does not presume that market liberties are basic liberties (see JF, 44). It merely notes that considerations must be balanced in making a central decision about the basic structure of society. Suppose this balancing may involve only considerations that citizens generally consider legitimate. If I reject the indifference, rivalry, and distorted baseline theses, and deny that their institutional acceptance could be a matter of relatively uncontroversial social science, then I will not see any restriction of liberty based on them as a matter of legitimate balancing. It will seem arbitrary. Alternatively, if I accept the three theses, but their relevance to the content of the basic structure is excluded on epistemic procedural grounds, I will be profoundly frustrated by what I take to be the existence of fundamental bads – by reference to Rawls reasons – for which no remedy seems possible. Either way, citizen disaffection seems likely.34 Call this the apparent dilemma. Let me note four things about the apparent dilemma. (1) The issue differs from a related issue for the 1844 Marx. That Marx rests justification of social relations in TCS on perceptions in TCS, but our present perceptions (in our capitalist society) contradict those in TCS, so we do not have access to the reasons (their perceptions) that make people in TCS believe Marx’s claims to be (selfevidently) true.35 By contrast, in the well-ordered society, the limit on the considerations that are permitted to be invoked to justify the controversial theses is not built into that society’s epistemic situation. Rather, it is self-imposed. (2) The problem is not a problem within reflective equilibrium. In reflective equilibrium, all reasons, including controversial reasons, are available. There, any reason to believe that the three left-wing theses are true is available. 34

35

In talking of justificatory community G. A. Cohen’s claim is that such community is best promoted if the instantiation of the difference principle is not left solely to institutions; instead, individual citizens also act, and see others as acting, from the difference principle in their individual choices. For Cohen, at stake is the generation/maintenance of a certain social ethos via widespread individual decision-making (See Cohen [2008]). My focus is different, namely, on the legitimate reasons to justify liberty-restricting state action in order to generate/maintain a certain social ethos. See Brudney (1998, chapter 6). See also Brudney (2001).

58

Daniel Brudney

(3) The issue I am raising about Theory anticipates the issue that Political Liberalism explores as the constraints of “public reason” (see PL, Lecture VI). Properly to address whether appeal to the three leftwing theses would violate those constraints requires more space than I have here. However, I should note that the later book arguably rejects or at least minimizes aspects of Theory’s justice as fairness that I have highlighted (namely, reciprocal mattering). Therefore, there is less scope for conflict among what I have called “Rawls reasons.” It seems to me an open question whether this shows that, on balance, Political Liberalism is an advance over or a decline from Theory.36 (4) At the beginning, I said that I wanted to raise, not to resolve an issue. I do think the issue is quite real, but I should acknowledge some reasons to demur, to think it merely apparent. First, in various texts Rawls, himself, relies on speculative social psychological claims, claims that might not be amenable to justification by either common sense or uncontroversial social science.37 Yet in doing so he seems unworried that he has overstepped his own boundaries. Perhaps, one could interpret Theory’s justice as fairness so that it permits the legislature to make central decisions about the basic structure by reference to controversial claims, or at least by reference to the indifference and rivalry theses – although I find it hard to see how this would square with Rawls’s view of proper justification, with his view of the way citizens should be able to justify their shared institutions to one another. Second, citizens in the well-ordered society have, and know that others have, an effective sense of justice. Moreover, they believe that the reason to permit a market is to benefit all citizens, especially the worst off. Despite the way that markets impose self-interested incentives on us all, maybe citizens could also see their market activities as indirectly tied 36

37

I should note that what I call “mattering” to one another need involve neither sharing a conception of the good nor sharing what Political Liberalism calls “a comprehensive doctrine” (see PL, XVIIIFF.). It need not require what Political Liberalism thinks of as “community” (PL, 146). For instance, the discussions of envy in TJ 1971/1999, §§80 and 81 seem quite speculative. As another example, see the argument against restricted utility in Justice as Fairness: A Restatement. It relies in part on the badness of the feeling of alienation that the worse off would have about not being the object of concern from institutions and from those who are better off, that is, about the unpleasantness as well as the self-respect undermining impact of the indifference relation. For this to be a relevant consideration, the parties in the original position must accept the causal thesis that such alienation would result from restricted utility. Would Rawls want to rest his argument here on that causal thesis being redeemable by common sense or the current scientific consensus? See JF, §38.

The Theory Rawls, the 1844 Marx, and the Market

59

to a concern for others’ well-being. (This might be akin to the way an attorney could act zealously for her client but also believe the adversarial legal process is, as a whole, the most effective route to justice.) Maybe these and other elements of the well-ordered society, say, its form of moral education (see TJ 1971/1999, §70–72), would diminish the market’s undermining of the desired social ethos. Third, by the time of Justice as Fairness: A Restatement, Rawls favored property-owning democracy for the well-ordered society. Perhaps property-owning democracy, even if not eliminating the market, would so alter it as to significantly mitigate indifference and rivalry.38 None of these considerations seems to me, either singly or together, to put the worry to rest. Of course, one could simply reject my reading of Theory as committed to citizen relations that are inimical to indifference and rivalry. One could see Theory as committed only to a “let’s obey just rules for the right reason” social ethos, where citizens’ mattering to one another need not obtain. There would then be no Rawls reasons to restrict the market. It would take a further essay to make the needed more exhaustive textual argument for my reading of Theory. * In too compressed a way I have tried to connect the Theory Rawls to strains in the history of philosophy with which he is not usually associated. I have also tried to show the richness of the desired relationships among citizens in Theory’s well-ordered society as well as a possible problem, derived from a standard left-wing critique, that a widespread market might pose for those relationships. It should not be surprising that A Theory of Justice resonates to many strains in the tradition of Western political philosophy. That is what it is for a work to be a landmark.

38

For Rawls’s discussion of property-owning democracy, see especially JF, §§41–44.

3

Rawls, Lerner, and the Tax-and-Spend Booby Trap What Happened to Monetary Policy?

Aaron James

[We need] the superstition that the budget must be balanced at all times …. One of the functions of old-fashioned religion was to scare people by … what might be regarded as myths into behaving in a way that the long-run civilized life requires. Paul Samuelson, Nobelist and chief architect of neo-classical economics (Samuelson 2014)

Certain elementary parts of economic theory are brought in solely to illustrate the content of the principles of justice. If … the received doctrine is itself mistaken, I hope that for the purpose of the theory of justice no harm is done. John Rawls (TJ 1971/1999 section 42)

Rawls barely mentioned money, finance, and banking in A Theory of Justice.1 What he does say in tentative and scattered comments (in secs. 42–43) has all the marks of a neo-classical economy: the prominence of markets, the limited significance of money and banks, limited and “independent” central banking, along with “sound” tax-and-spend public finance, as if on a soft gold standard. Looking back fifty years later, this is puzzling and unfortunate. It’s puzzling because by the 1940s economist Abba Lerner, among others, had already explained that “sound” finance was an anachronism of the gold standard era. In a fiat money system, finance is “functional” or to be judged simply by its results (Lerner 1941, 1943, and 1944; Roml 1946). As neo-classical economist Paul Samuelson eventually explained (in the interview quoted in the epigraph), a “sound” balanced budget requirement – with spending offset dollar for dollar by taxes or borrowing (repaid in taxes) – was a “superstition” akin to “old-fashioned religion,” a 1

Given my concern with the development of Rawls thought across both editions of A Theory of Justice, citations hereafter will be to both the 1971 and 1999 editions, with page numbers given to each respectively, except where context indicates reference to one or the other.

60

Rawls, Lerner, and the Tax-and-Spend Booby Trap

61

“myth” needed to “scare” people (Samuelson 2014). Noble Lies run afoul of Rawls’s publicity requirement. And yet to this day, in part due to Rawls’s influence, that old time monetary religion continues to confuse political philosophers about such basic matters as liberty, equity, and the relation between them, even in the cathedrals of high theory. In this discussion, I examine Rawls’s uncritical acceptance of “sound” finance, how it reflects his use of his method, his influences in neoclassical economics, and what I regard as the regrettable consequences. I’ll also suggest how we can unlearn our practiced deference to neoclassical thinking, using Rawls’s method, in view of an egalitarian form of central banking presented more fully elsewhere (Hockett and James 2020).

3.1

Equity, Liberty, and Money from Nothing

In Robert Nozick’s famous liberty objection to Rawls’s difference principle, taxes were construed as robbing Peter to pay Paul (Nozick 1974, 169–70, 172–73, and 264). Involuntary contributions had to be taken from productive earners in order to give money to the disadvantaged that would not otherwise be forthcoming. But this is simply not true under fiat money. Money can be given to the least well off without “raising revenue” by either taxes or borrowing, by simply issuing it afresh, “from nothing.” The central bank would have to manage any risks of deflation and unhealthy inflation, to be sure. But since that may or may not require taxation, what is the grave threat to liberty again? Rawlsians who responded to Nozick often went along with the misconception. There is an equity/liberty trade off, they admitted. But, they replied, equity did not require wrongful, rather than fair, taxation. By granting that taxes were necessary to pay for equitable benefits, they stepped right into the tax-and-spend booby trap. Eventually, philosophers did learn to side-step, noting the role of ex ante “pre-distribution” rather than tax-and-spend “re-distribution.”2 And yet political philosophy has only begun to pay attention to monetary policy, one of the chief ways pre-distribution might steer money in a more egalitarian direction (Hockett and James 2020; Dietsch 2021). Think of what Milton Friedman disparaged as “helicopter money,” but instead of wind and dust and bills flying everywhere, imagine a more discriminate delivery (perhaps via distributed ledger technology). As 2

Murphy and Nagel (2002) did reframe the tax system as part of the larger system of property. For an early appeal to “pre-distribution” specifically, see O’Neill and Williamson (2012a). See also Thomas (2017) and Scanlon (2018, 102).

62

Aaron James

financial lawyer Robert Hockett and I explain in Money from Nothing, the US Federal Reserve could automatically open an account for every citizen and make regular dollar deposits (e.g., of $1,000–2,000 per month) (Hockett and James 2020, chapter 10) Much as the Fed now pays interest on reserves held by chartered banks, it could pay a rate of interest on each citizen account, which can be raised or lowered to influence saving and spending decisions. If allowed, further inflation management tools could be deployed as well: open-market asset purchases that “collar” systemically significant prices; dollar credits with a must-spend expiration date; or, when all else fails, temporary impounds of a portion of credited funds, with interest accruing. Over time, the free, sure money for every citizen, along with full and free inclusion in the public banking system (and perhaps “baby bond” payments as well), would reduce inequality in money holdings, shifting the distribution of purchasing power in an equitable direction.3 This is indeed a sort of “redistribution” but without taxation. If the new money issued were allowed to inflate prices slightly (e.g., by 2–4 percent), the affluent – especially net creditors – may see a reduction in their money’s purchasing power over time. But rest assured, despite the inevitability of righteous howling about “debasing the currency,” a modest, controlled “rebalancing” of purchasing power is hardly tantamount to “default” on money’s promise, formally or informally. It may well boost the economy’s prospects and the currency’s credibility, which is its promise of future redemption. Inflation is not “taxation without legislation,” as Friedman reportedly said in a moment of bombast. A tax is a legal debt and not a mere cost, and a mere decline in your money’s purchasing power won’t settle your (nominal) debt to the tax authority. Nor is a modest “devaluation” comparable to the tax authority or its armed agents coming after the real fruit of your labor. Suppose, arguendo, that you enjoy Nozick’s Lockean property rights over certain apples and oranges, grown putting spade to garden and hand to plow. Since fruits spoil, you trade them for some money, to be used as a “store of value.” While you could have bartered for pigs, beef jerky, gold, or bitcoin – which all store better than apples – you instead freely choose to store your new wealth in a money or money-denominated asset. But then have you not voluntarily assumed the risk of a change in those assets “real” purchasing power – meaning that you now lack any continuing claim to their previous “real” value? Asset price fluctuation is inevitable and entirely 3

Rawls resisted an unconditional basic income but did propose a “social minimum” paid, e.g., as a Friedman-style negative income tax (TJ 1971/1999, 275/243).

Rawls, Lerner, and the Tax-and-Spend Booby Trap

63

foreseeable, whether due to daily market forces or by central bank actions (perhaps acting as an asset buying or selling “whale” in the market). And surely, having taken a voluntary gamble, you lack a personal entitlement to the larger social environment needed to preserve your assets’ original “real” value. (Estimated how and when? At time of trade and in perpetuity? No matter how predicable the variation? No matter how modest the devaluation? No matter how well equity is served by one?) Perhaps you had “no real choice” in trading your fruits for money. Fair enough. In “tacitly” joining the social contract, you then have equity claims against that money’s devaluation but balanced against the stronger equity claims of the less well-off who’d benefit from it. On the other hand, Rawls does frankly endorse confiscatory taxation. He says expressly that taxes are needed “not to raise revenue (release resources to government)” but simply to “correct the distribution of wealth and prevent concentrations of power” (TJ 1971/1999, 277/245). So Nozick still has a target. Still, what carried his critique was always the image of the state taking one’s stuff in order to give it the poor, as forced charity. And that core objection simply misfires if equity can be advanced “from below” with money from nothing and little or no risk to liberty. And if taxing the rich to “prevent concentrations of power” is itself necessary to uphold liberty, its effective worth, or equal political liberty, it’s far less plausible to object to the infringement of liberty done for the sake of liberty. Nozick’s critique is, to a large extent, hollowed out.4

3.2

The Money Finance Taboo

That this line of reply was rarely if ever mooted in decades of discussion and teaching of Rawls vs. Nozick reflects a striking fact of recent intellectual history. Political philosophy has long operated within the confines of “sound” finance, marked by at least two related assumptions. First, balanced public budgets: any government expenditure should be offset, dollar for dollar, with either tax revenue or borrowing at a rate of interest, itself later repaid by future tax receipts, at least eventually. Second, the money finance taboo: taxes or borrowing, repaid in taxes, are our only options; the further, technically feasible option of “money financing” or “printing money” – that is, the government paying for things with money 4

On my reading of Nozick (1974, chapter 7), without this core argument, Nozick’s prior statement of an entitlement theory is mainly stipulative. His entitlement-based objection to taxation in the service of any “end-state” or “patterned” theory (as “on a par with forced labor”) is what gives the entitlement theory independent force. And without that, his subsequent burden-shifting probe of Rawls’s theory is mainly a stand-off.

64

Aaron James

created by fiat without new taxes or borrowing – is verboten, except in war or other crises. And since we must live in Victorian comportment to sound budgetary virtue, as if we still lived on a gold standard, we unfortunately cannot issue money to help the worst off among us, as Rawls and Nozick and decades of philosophers assumed. Monetary religion has weakened lately. Large public debt outlays have become a new normal. It’s now widely appreciated that governments that issue their own money and borrow in that money without a fixed exchange rate – enjoying “monetary sovereignty”5 – cannot involuntarily default on public debt; they always can and do pay by issuing new money. At the same time, the taboo against “money financing” remains in force. “Printing money,” or rather issuing it without new taxes or borrowing, is still treated as permissible during wars and other crises but improper in “normal times.” Rawls, however, quite expressly stipulated the need for a “scheme of taxation to raise the revenues that justice requires” (TJ 1971/1999, 278/ 246, emphasis added). As noted, he did say taxes should be used to take money from the wealthy simply in order “correct the distribution of wealth and prevent concentrations of power,” and to that end he wisely favored “a proportionate expenditure tax” over income taxes (TJ 1971/ 1999, 278/246). Yet he also says they are necessary for “transfer payments” to the least well off: Social resources must be released to the government so that it can provide for the public goods and make transfer payments necessary to satisfy the difference principle. (TJ 1971/1999, 278/246, emphasis added)

Likewise, he writes: Proportional expenditure (or income) taxes are to provide revenue for public goods, the transfer branch and the establishment of fair equality of opportunity in education, and the like so as to carry out the second principle. (TJ 1971/1999, 279/247, emphasis added)6

This was true under the late nineteenth-century gold standard. The gold commitment in the United States was precisely a promise to not “print money” (as the United States did with the paper “Greenbacks” 5 6

See Hockett and James (2020, Appendix I). As for public goods, Rawls illustrates with the case of national defense: tanks could be built, soldiers paid, and so forth, only if citizens pay their share of taxes for it (TJ 1971/ 1999, 267–68/237). And in connection with Knut Wicksell’s “unanimity criterion” for which public goods to have, he grants that, at least assuming a just background distribution, there is no justification for “using the apparatus of state to compel some citizens to pay for unwanted benefits” (TJ 1971/1999, 283/250), again assuming taxes would be needed to “pay for” benefits.

Rawls, Lerner, and the Tax-and-Spend Booby Trap

65

used to pay Civil War soldiers). This was the very promise FDR wisely called off in 1933 for domestic dollar-to-gold convertability (the international gold standard, a promise to foreign central banks, was abandoned in 1971 by Richard Nixon). In a 1946 article entitled “Taxes for Revenue are Obsolete,” New York Federal Reserve Chairman Beardsly Roml explained the new financial reality in plain terms. Citing the Fed’s establishment in 1913, and FDR’s 1933 decision, along with a list of the proper functions of taxes, he wrote, since “our Federal Government has final freedom from the money market in meeting its financial requirements,” taxes only matter for their “social and economic consequences,” not for “raising revenue” (Roml 1946, 35–36). In the years prior, Lerner had systematically explained this “functional” view of public finance. He admitted that “printing money” was a “subordinate” financial instrument, since without other, primary instruments of government, it “has no effect at all on the economy” for merely sitting in bank accounts (Lerner 1944, 314). But, he argued, money issuance is a perfectly useful way to normally achieve desired social outcomes, quite aside from wars and other emergencies, whether or not we decide to also tax or borrow. Taxes might be apt for various purposes, for instance, to “destroy” money and “leave less in the hands of the taxpayer,” as Lerner puts it. They can also, we might add, tamp down inflated asset prices, deter unwanted activities, or create a sense of vested interest, whether as good symbolism or to ensure the political durability of social insurance. Yet, for Lerner, taxes should “never be imposed merely as a means of raising money for the government on the grounds that government needs the money. The government can raise all the money it needs by printing it if the raising of the money is the only consideration” (Lerner 1944, 307, emphasis added). Must the funds then be “borrowed” and later repaid, perhaps from future tax collection? No, because the United States doesn’t need to get its dollars from anyone. Indeed it can’t get them from anyone unless they are first spent or lent into existence by the government or its authorized franchisors, the chartered banks (Hockett and Omura 2017, 1143). There is no other legal source for dollars, since counterfeiting is illegal. Why “borrow” at a rate of interest at all, then, as the US Treasury does? Because issuing Treasury securities has uses other than that of “raising funds,” such as pulling liquid dollars out of the banking system, offering a safe asset for investors, and so on. Indeed, Lerner went so far as to say that “borrowing by the government should never be undertaken merely as a means of raising money,” because if it wants its own money, it can simply issue it (Lerner 1944, 308, emphasis added).

66

Aaron James

There are limits on money issuance, to be sure, including an economy’s real productive capacity, beyond which extra money, if poorly directed, merely inflates prices. Lerner’s basic point is about how the matter should be judged.7 As Lerner explains, The central idea is that government fiscal policy, its spending and taxing, its borrowing and repayment of loans, its issue of new money and its withdrawal of money, shall all be undertaken with an eye only to the results of these actions on the economy and not to any established traditional doctrine about what is sound or unsound.

The question is simply “the way [financial instruments] work or function in the economy,” quite apart from the “principle of trying to balance the budget over the solar year” (Lerner 1943, 39). In that case, Rawls was simply wrong to say that “social resources must be released to the government so that it can provide for the public goods and make transfer payments necessary to satisfy the difference principle” (TJ 1971/1999, 278–79/246–47, emphases added). Since the United States issues dollars, it has no need to “release” dollars from anyone to finance security services or other public goods, public education, retirement or unemployment insurance, or basic income and other payments that raise prospects for the least well off. Spending without offsetting taxation may amount to a budget deficit, but this is of no significance in and of itself, as Lerner explained. More money is thereby left in circulation, but how much money is too much money and who has it is to be judged by its consequences (e.g., for deflation or inflation or the distribution of wealth), given the management tools available (whether taxes, borrowing, interest rate adjustment, raising bank reserve requirements, public investment in productive activity, the novel tools noted earlier, and so on). As economics, this should not be especially controversial. Heterodox economists have confirmed, in rigorous study of government accounting and sectoral balances, that functional finance best describes how fiat money systems actually operate (See Wray 2000; Goodley and Lavoie 2006). Key claims of functional finance have also been developed using mainstream economics methods by Harvard’s Stephen Marglin (Marglin 2021, part V, chapters 14–16). Even the early Milton 7

A case in point is the post-Covid 2022 surge in inflation: real productive capacity was limited by post-lockdown supply chain bottlenecks and then Russia’s invasion of Ukraine. The increase in prices might have been far less significant had central banks been better equipped to use the inflation management tools noted earlier, for example, by “collaring” key commodity prices or by raising interest rates on citizen accounts with the central bank.

Rawls, Lerner, and the Tax-and-Spend Booby Trap

67

Friedman, in a 1947 review of Lerner’s 1944 book The Economics of Control – EC hereafter – largely agreed with Lerner (Friedman, 1947) Friedman himself offered a version of functional finance in 1948. (Friedman 1948; Wray 2002). And when he subsequently revived “sound” finance as monetarism, it was for administrative rather that purely economic or financial reasons. Here it is important to appreciate how far a basic question of administrative bookkeeping has often been elevated to a matter of quasi-sacred principle. As Hockett and I explain in MFN, the alchemy of public finance is just the ordinary reality of promising. Ordinary acts of promising create debt-obligations and their corresponding claims “from nothing” on a promisor’s say so (e.g., I tell you, “yup, see you there!”). Likewise, the dollar promises that are money are issued “from nothing” as public debt-obligations on behalf of the United States by the Federal Reserve. In computer keystrokes that at once create liabilities on its balance sheet and new dollar credits in bank accounts, Fed officials simply type “high-powered money” credits into the reserve accounts of chartered banks, who then make the credits available to their intended recipients as “demand deposits” in their own accounts, who in turn save them or spend them into the economy. Crucially, the liabilities and the assets travel together by analytical necessity: a dollar just is, by its very nature, at once a liability, debt, and obligation for the Federal Reserve (and the United States) and an asset, credit, and claim for the party who holds it.8 Once created, if nothing else is done, the central bank liabilities are simply carried on the central bank balance sheet, which amounts to interest-free public promissory debt, that is, credit-money, in someone’s bank account or pocket. Such is the origin of a dollar, but this is yet to address a large further question: precisely who should be authorized to issue, account for, and adjust and allocate promissory debts and credits on behalf of the polity? History has answered by evolved law and convention, which first took shape in the United States in the 1950 Fed/Treasury Accord. The Fed thus came to be seen as the technocratic handmaiden of Treasury and Congress, which are themselves directly subject to democratic politics. While Fed officials have narrow scope for “independent” policy discretion within their legal mandate (e.g., over interest rates), the public liabilities that formally begin their existence on the central bank balance sheet at the request of Treasury were to be shifted over onto the Treasury 8

As Lerner puts the point (Lerner [1944] 1970, 304), speaking of national debt generally, “Every debt has a corresponding credit because there must be someone to whom the debt is owed.”

68

Aaron James

balance sheet, in the form of interest-bearing Treasury bonds (instead of liquid dollars).9 Fed assets and liabilities would thus move up or down by the billions or trillions daily but net to zero by the daily close of business. And in keeping a clean Fed balance sheet, “money financing” was off the table. The rest of the government would have to finance its spending much as under a gold standard: by raising revenue, if not from taxes, then with Treasury borrowing, the liabilities being carried on the Treasury balance sheet with interest payments. Whatever its merits, this was always an administrative convention, which could have been different. Public liabilities can also be – and increasingly are – created and left on the central bank balance sheet indefinitely, with no interest payments promised (except whatever interest the Fed happens to pay on bank reserves). Either the Fed or the Treasury balance sheet can work; both consolidate to one national government balance sheet in any case. And here we should ask: does it really matter, ultimately, which US balance sub-sheet the figures get parked on?10 Should the worse off be less well off than they might because use of the Fed balance sheet is verboten? Indeed, the Fed has already weakened the taboo for purposes of crisis management, with nice advantages. There’s no “borrowing” with interest steadily accruing, no debt “left for the grandchildren” to repay, and less public confusion. Why not do this in normal practice? The taboo on “money financing” took hold largely due to the perceived need for central bank “independence.” Friedman’s “sound” finance was an austere version, which requires a clean Fed balance sheet and no or very limited allocative authority. By comparison, Hockett and I propose a relaxed political “independence” from short-term political pressure, akin to an independent judiciary, with the central bank allowed to allocate key resources during crises and in normal times (Hockett and James 2020, chapter 14). Lerner, for his part, urged us not to remove our hands from “the economic steering wheel,” as under laissez-faire capitalism, but never 9

10

US Treasury “borrowing” merely changes the composition of government debt (along with the extra liabilities in promised interest payments, later to be redeemed in new money). As Lerner puts the point, “The primary effect of borrowing is to decrease the amount of money held by the public and increase the amount of government bonds held by the public” (Lerner [1944] 1970, 309). The Fed now pays interest on reserve accounts directly to help manage reserve bank account balances. The Treasury issues securities to provide a safe asset to financial markets; but that favor can be equally performed, for instance, by time-deposit accounts (perhaps with a promise of interest) at the Federal Reserve, or by Fed or climate bank-issued “green bonds,” which anyone can hold as a safe, productive, interest-bearing asset.

Rawls, Lerner, and the Tax-and-Spend Booby Trap

69

was clear about what specific administrative regime should do the steering. His account of the underlying economics is nevertheless clarifying: it highlights the fact that Fed “independence” coupled with a taboo on monetary financing is but one possible administrative regime, a matter of political economy, not mere economics. And that helps us see that Rawls, the great theorist of justice, somehow took a taboo for granted.

3.3

Rawls Read Lerner

We know from Rawls’s archives (thanks to David Reidy) that Rawls carefully read Lerner’s 1944 EC, which includes a clear formulation of functional finance (in chapter 24). Yet Lerner is not cited in A Theory of Justice. Perhaps Rawls read Lerner’s book only after A Theory of Justice’s publication in 1971. But no discussion or reference to Lerner or functional finance appears in either the 1999 second edition of A Theory of Justice or his 2001 Justice as Fairness: A Restatement. A charitable explanation is that Rawls read EC only late in his career, after Rawls made the revisions for German publication that were subsequently incorporated into the 1999 edition but before his 2001 Justice as Fairness. In Justice as Fairness, there’s at least no mention of needing taxes to “release” funds for public goods or transfers to the least well off. This is consistent with functional finance, though the reader is given no indication of a change in thinking or its importance.11 If Rawls did read Lerner before either edition of A Theory of Justice, the question is why he knowingly didn’t follow Lerner’s lead. But even if we assume Rawls read Lerner late, it remains an interesting question how “sound” finance held sway over Rawls’s thought. Lerner’s functional finance was well-known by the time Rawls began intensive study of economics in 1949–50.12 Why wasn’t he exposed to it? What seems fair 11

12

The key passage comes where he reiterates that income tax may not be necessary at all, adding that in any case the “the principle of progressive taxation might not be applied to wealth and income for the purposes of raising funds (releasing resources to government), but solely to prevent accumulations of wealth …” (JF, 161). According to Daniel Little’s helpful “Rawls and Economics,” Rawls studied with economist Jacob Viner at Princeton on fellowship in 1949–50. In Fall of 1950, he attended a seminar taught by economist William Baumol, which carefully studied Paul Samuelson’s Foundations of Economic Analysis along with J. R. Hick’s Value and Capital (Little 2014, 508). A further study group of his focused on Leon Walras’s Elements of Pure Economics, John von Neumann and Oskar Morgestern’s Theory of Games and Economic Behavior, and Frank Knight’s Ethics of Competition. He would have been exposed to Joseph Schumpeter, whose Ten Great Economists appeared in 1951. Rawls’s references to economists in his essays from 1955–71, before A Theory of Justice, drew

70

Aaron James

to say is that Rawls’s assumption of “sound” finance in both editions of Theory was influenced by the neo-classical economics he studied in both his formative years and the decades following. The question is by whom and in what ways.

3.4

Abstraction and Omission

When Samuelson called a balanced budget a “superstition” comparable to old time religion, he meant that it was useful for disciplining legislators, noting approvingly the views of James Buchanan. To Buchanan, self-interested legislators are prone to carry public debt, spending without raising unpopular taxes, in order to curry favor with the masses who elected them. Emergencies aside, a balanced budget norm was thus needed to discipline public spending in normal times.13 If Rawls could hardly endorse monetary religion, why assume “sound” money at all? The answer, it seems, is simply that it was prominent in the public goods literature Rawls responded to in Theory, which grew out of Wicksell’s, Samuleson’s, and Buchanan’s treatments. When Rawls is quite explicit that “financing public goods must be taken over by the state and some binding rule requiring payment [of taxes] must be enforced,” he cites not only Buchanan but also his one-time economics professor at Princeton, W. J. Baumol, suggesting à la Baumol that “many of the traditional activities of government, insofar as they can be justified, can be accounted for in this way” (TJ, 268/238). Perhaps Rawls saw this as the “received doctrine” and felt a need to respond to it specifically. Rawls does note that his political economy assumptions could turn out to be mistaken (TJ, 264/234, quoted in the epigraph). If we distinguish Rawls’s fundamental theory of justice from what might best implement it, the matter may seem at worst an administrative technicality: he made factual assumptions for illustrative purposes that he could simply drop, which anyway were never part of his ideal theory. Yet even if offered “solely for illustration,” as he says, I submit that Rawls’s political economy assumptions were not just expositional color or part of “non-ideal theory” he could just as well drop. He could indeed have adopted different assumptions; but he could not have entirely divorced his ideal theory account of justice from its practical role in a realistic economy. His own constructive method required him to unite the two.

13

heavily from the 1950s and 1960s, and, in A Theory of Justice, are disproportionately drawn from those decades. On Buchanan’s shift from empirical argument about the reality of strategic political selfinterest to Victorian moralism, see Alvey (2011).

Rawls, Lerner, and the Tax-and-Spend Booby Trap

71

By his own method, if principles of justice are to be for the regulation of certain social institutions and practices, they must be justified in the first instance in light of some (perhaps constructive) interpretation of what those institutions and practices are like, given how their participants might themselves understand them (James 2005). At least some basic version of them would have to be recognizable to the participants if they are to be addressed with conclusive requirements for their association. Otherwise, the proposed ideals lose normative purchase. So Rawls had to offer a realistic, credible if idealized vision of how his proposed principles might operate in practice. Principles “depend on general facts,” as he put it, and had to be tested for their “practicable bearing,” judged from the point of view of the citizen (TJ 1971/1999, 265/234). Rawls’s moral argument therefore was not that, given ideal theory, we can just as well be skeptical that a neo-classical, “neo-liberal” economy could ever engineer justice. Quite the contrary, Rawls’s conjecture, in effect, was that it could. We can therefore keep an economy much like the one the neo-classicals envisaged and demand justice from it, instead of concluding that justice requires us to start over or move to radically different arrangements. And yet, if ideal theory principles are also supposed to prescribe deep revision in going practice, they can’t be too embedded in prevailing social understandings or existing policy regimes. They then lose critical depth, at risk of being insensitive to relevant underlying injustices. Rawls no doubt saw Buchanan’s own version of liberal contractarianism and constitutionalism as unduly biased toward the status quo. What’s needed, instead, is moral reasoning that abstracts away from and neutralizes the existing distribution of advantage, with the dark political realities of perverse legislative incentives left for “non-ideal theory.” So ideal theory brought conflicting pressures, upward toward abstraction and downwards toward concreteness. In the upward direction, it’s useful to abstract away from the peculiar and more detailed workings of prevailing institutions, along with assuming favorable conditions. Rawls was thus not very specific about the design of political institutions and economic administration, given all their hazards and incentives. In turning to political economy in Theory, he could follow Richard Musgrave and merely sketch operating “branches” of government, defined by their general functions, in terms that feel recognizable enough. He could also allow refinement of their “division of labor” as justice required it – a crucial point I return to. Yet in the downward direction, toward concreteness, a justification of a particular set of principles – at least those specific enough to be of any interest – also required an examination of their meaning and potential

72

Aaron James

implications in practice – their “practicable bearing.” Otherwise, how could we decide they are finally acceptable and live by them? The Goldilocks finesse (“not too high, not too low, but just the right altitude”) is to characterize institutions in fairly abstract terms but also to offer more detailed suggestions about how policies could credibly play out, for illustrative purposes. If those suggestions prove mistaken, one hopes “no harm is done” (TJ 1971/1999, 264/234). But in that case, Rawls’s neglect of Lerner is only more curious. Functional finance offers a descriptive interpretive foundation for prescriptive moral principles, describing the various levers for their implementation. And to the extent “sound” finance goes beyond economics as a set of administrative conventions, often functioning as monetary religion, it seems particularly important to note this fact, and indeed elaborate feasible political economy alternatives, perhaps with the clarity provided by contrast with functional finance. But then why didn’t Rawls even so much as footnote Lerner, not even in Justice as Fairness? He presumably wished to stave off potential “sound” finance objections. But he could have done that while also showing the consistency of his principles with credible functional finance alternatives. The question is why he didn’t. 3.5

Friedman’s Influence?

A possible clue to Rawls’s thinking is that (so Reidy informs me) Rawls’s copy of Lerner’s EC includes a note to himself “see M. Friedman’s review in Journal of Political Economy (1947).” As noted earlier, there is a route through functional finance to “sound” finance – the route Friedman himself took. Assuming Rawls followed Friedman’s writings from the 1940s–60s, Rawls may have read Lerner but deferred to Friedman’s evolving views, or at least felt reluctant to disagree with them, perhaps just in time to put the final touches on A Theory of Justice before its appearance in 1971. Assuming Rawls read Friedman’s 1947 review of Lerner’s EC, he would have found Friedman objecting, not to functional finance itself but to Lerner’s neglect of “administrative problems of economic institutions” as well as non-economic threats to liberty (Friedman 1947, 416). Friedman says Lerner had nicely clarified the “factors involved in governmental fiscal operations” (to secure aggregate demand). He simply regarded the result as platitudinous (“a brilliant exercise in logic”), for lacking in “prescriptive” use (an “exhortation to do the right thing with no advice about how to know what is the right thing to do”) (Friedman 1947, 413–15).

Rawls, Lerner, and the Tax-and-Spend Booby Trap

73

Although Friedman went on to propose a version of functional finance himself in 1948, as noted earlier, by 1968 he endorsed what is in effect a soft gold standard, for administrative reasons of social engineering (Friedman 1969). If the money supply cannot be fixed to the slowly rising and troublingly variable supply of gold, central bankers could instead ensure a steadily rising but limited supply of fiat money. And while they can also ameliorate “major disturbances” not arising from money, they should not try to do “fine tuning” for changing economic conditions but merely try to “avoid wide swings.” “We simply do not know enough,” Friedman counsels, to effectively address “minor disturbances,” not without making things worse. Such was the hard lesson of the Fed’s early history according to Friedman. “In the first flush of enthusiasm about the newly created Federal Reserve System, many observers attributed the relative stability of the 1920s to the System’s capacity for fine tuning – to apply an apt modern term.” But looking back over the Fed’s blunders, he concluded, “‘fine tuning’ is a marvelously evocative phrase in this electronic age, but it has little resemblance to what is possible in practice …” (Friedman 1969, 1–3). If Keynesian stimulus or a reallocation of resources was needed, that could, and should, be left to Congress and the Treasury. The Fed should itself stick to broad monetary modulation, avoiding wide swings, and first and foremost ensure price stability, even at the price of higher unemployment. In one of the few places where A Theory of Justice touches upon monetary policy explicitly, Rawls gives an impression of this sort of arrangement: In conformity with political decisions reached democratically, the government regulates the economic climate by adjusting certain elements under its control, such as the overall amount of investment, the rate of interest, and the quantity of money, and so on. There is no necessity for comprehensive direct planning. (TJ 1971/1999, 273/241)

Aside from rejecting comprehensive central planning, Freidman had argued for targeting what Rawls calls “the quantity of money” (though, having tried it, the Fed learned to target interest rates instead). Presumably with a central bank in mind, Rawls notes “adjusting certain elements under its control,” perhaps in view of Friedman’s concern about what is not in our control. Rawls’s proviso, “in conformity with political decisions reached democratically,” might also be said to reflect the emerging view – which Friedman, Lerner, and Samuelson all expressed in 1960s testimony before Congress – that Fed decisions and appointments should not cater to private banker interests, in rejection of then-traditional central bank “independence” from public accountability.

74

Aaron James

Eventually, the public norm of “independence” came to mean much more, in part due to Friedman’s influence. Appointed central bank technocrats were seen as doing putatively value-free monetary policy, leaving politically significant distributive matters entirely to the legislature and its democratically elected representatives. When, following Richard Musgrave, Rawls explains how public administration might work in functionally distinct “branches” (for “allocation,” “stabilization,” “transfer,” “distribution,” and “exchange”), he suggests something like this institutional segregation. The “stabilization branch,” he says, “strives to bring about reasonably full employment in the sense that those who want work can find it and the free choice of occupation and the deployment of finance are supported by strong effective demand” (TJ 1971/1999, 276/244). This sounds akin to the US Fed, given its full employment mandate (except that the price stability mandate – which Friedman prioritized – is not mentioned). Yet, crucially, Rawls says the stabilization branch is at least functionally distinct from the “transfer branch,” which itself “guarantees a certain level of well-being and honors the claims of need” (TJ 1971/1999, 276/244). Rawls mentions the negative income tax, which Friedman championed, as a means of paying a social minimum and invites the functional separation of “transfers” from the central bank’s independent task in regulating the general “economic climate,” and here again Friedman fully agreed. So Rawls seems to have deferred to a then-emerging view of central bank “independence,” despite the fact that it precludes a direct means of promoting the difference principle.14 On the other hand, Rawls did leave room for a more qualified form of “independence.” He is careful to note that the default functions of different “branches” may need to be adjusted in a working division of labor among administrative bodies (TJ 1971/1999, 275–76/243). A transfer branch in particular will have to account for and weigh relevant claims that a “competitive price system gives no consideration to,” and he generally suggests “a division of labor between the parts of the social system in answering the common sense precepts of justice” (TJ 1971/1999, 276–77/244). Though the idea is not mooted, that leaves a crack in the door for a central bank, otherwise tasked with stabilization of the general economic climate, to also allocate basic income payments via accounts held directly with it, as Hockett and I propose. Indeed, it was arguably always a mistake to suppose that “modulation” of the overall supply of money could be separated entirely from “allocation” or “distribution” across an economy in any case; 14

But see T. Klooster (2019), who contrasts Rawls’s “democratic” position with Friedman’s “independence.”

Rawls, Lerner, and the Tax-and-Spend Booby Trap

75

equity and price stability must go hand in hand.15 And so, Hockett and I suggest, central banks rightly have “allocative” or “distributive” functions, so long as their importance is relatively uncontroversial and decisions are generally subject to vigilant democratic accountability (Hockett and James 2020, chapter 14). 3.6

Rawls’s Neo-classical Conjecture

But then if Rawls did not entirely defer to Friedman, why ultimately did he assume “sound” rather than functional finance or at least fail to note both? Here I suggest we look further back to Rawls’s early study of neoclassical economics in 1949–50, around the time he says the basic outlines of his theory of justice occurred to him, when he read Samuelson, Hicks, and others in seminars at Princeton with Baumol and Viner.16 Here it’s important to emphasize that the neo-classicals’ putative “synthesis” of classical theory with Keynes’s insights arguably did not share or even fully appreciate Keynes’s views about the significance of money and central banking. Following his economics teachers, Rawls may have paid less attention to what was far more visible and contested during the monetary policy battles of late nineteenth and early twentieth century: money and banking’s destabilizing and thus deep political, economic, and distributional importance. It was precisely Keynes’s view of money – as credit, along with its analytical correlate, debt – that separated him from the classicals of his day (Keynes 1930 and 1933). For Keynes, it was the dynamic instability of the “credit money” economy that animated booms and crashes, unemployment, and persistent disequilibrium, even in the long run. Keynes thus saw the dire need for central banks to perpetually manage, stabilize, and fine tune their respective domestic economies, along with a global central bank and international money – the “bancor” – that makes such domestic fine tuning possible (See Hockett 2013). To classical theorists such as Marshall and Pigou, by comparison, money was at bottom a “neutral” transaction technology, a “medium of exchange” needed simply to overcome the inconveniences of barter, as Adam Smith said. Banks were seen as “neutral” overlay, merely allocating existing capital between savers and borrowers for a fee. Much as in Walras’s early general equilibrium model, as carried through from Arrow/Debreu to more recent dynamic stochastic general equilibrium models, money and banking are inessential, and, if somehow added, 15 16

See also Hockett (2021). See Freeman (2007, 13); Little (2014, 508); and Reidy (2014).

76

Aaron James

make no difference. In Samuelson’s influential equilibrium model, fiat money is in effect just a government issued transaction technology – a sort of “paper gold” – another commodity subject to the usual laws of supply and demand.17 Thus the neo-classical “Keynesian” story of depressions is not a story of money but of “sticky wages,” “frictions,” “rigidities,” and other failures of perfect competition that prevent automatic self-correction to the longer run full employment equilibrium.18 Rawls catered to the neo-classical vision of the economy. He carefully presents the theory of general equilibrium and, again citing his Princeton economics professor Baumol, suggests that “perfect competition is a perfect procedure with respect to efficiency” (TJ 1971/1999, 271–72/ 240). Rawls notes that of course the necessary idealizations are “seldom if ever fully satisfied in the real world” and that government will have to correct “market failures and imperfections,” as well as provide public goods, where “the market fails altogether.” He nevertheless recommends the idealization to both “clarify the related notion of pure procedural justice” and show how it might credibly operate in practice. Much as in general equilibrium theory, this “ideal conception” of justice is supposed to bear in turn on the real world, being “used to appraise existing arrangements and as a framework identifying the changes that should be undertaken” (TJ 1971/1999, 272/240). The intended analogy is not just instructive. Rawls notes the “considerable use of market arrangements” in his own account, adding that “it is only in this way, I believe, that the problem of distribution can be handled as a case of pure procedural justice” (TJ 1971/1999, 274/242, emphasis added). In so relying on general equilibrium theory, Rawls in effect granted that economic models could abstract away from money, credit, banks, and finance and still retain practical relevance – enough relevance, at least, to show how pure procedural justice could credibly work in practice. We see this in Rawls’s broadly neo-classical conjecture about justice, which, it must be noted, appears only in A Theory of Justice’s first edition: Suppose that law and government … keep markets competitive, resources fully employed, property and wealth widely distributed … by the appropriate forms of taxation, or whatever, and … guarantee a reasonable social minimum. Assume also that there is fair equality of opportunity underwritten by education for all; and that the other equal liberties are secured. Then it would appear that the resulting distribution of income and wealth … will tend to satisfy the difference principle. (TJ 1971, 87, emphasis added)19

17 18 19

Samuelson ([1967] 1972, 557). For discussion see Mehrling (2014). On John Hicks’s influence and evolution on money, see Mehrling (2019). The passage would have appeared on p. 76 of the second edition.

Rawls, Lerner, and the Tax-and-Spend Booby Trap

77

The difference principle, in short, can be “left to take care of itself” (albeit “within a restricted range”).20 This can sound incredible to us now after fifty years of stagnant wages and yawning inequality. Perhaps Rawls dropped the passage from the 1999 edition in view of the emerging trend. The “whatever” (italicized in the quoted passage) does leave room for further measures, including progressive central banking. What’s noteworthy is that Rawls never saw the need to mention it. Rawls later emphasized the need for a high social minimum and wide distribution of property for a “property-owning democracy,” in contrast with both laissez-faire and welfare state capitalism, and he explained that this would require “pre-distribution” measures outside of the tax and redistribute model relied upon in A Theory of Justice (JF, 139). But if those measures should and arguably must include something like functional finance and progressive central banking, Rawls never noted their role or importance. 3.7

Conclusion

Rawls clearly saw the limits of the neo-classical’s sparse reaction to the horrors of the Great Depression and the World War that followed; his great ambition was precisely to develop a deeper, value-infused conception of a lasting social contract.21 But like the neo-classicals, he seems to have missed the full need for central banking in view of the importance of money and of financial market instability – even as that itself, among other causes, set the stage for war and, for Keynes, had to be part of any plan for post-war peace. This is entirely forgivable given Rawls’s many gifts to us. But fifty years later, having seen neo-classical, neo-liberal economics ravage justice in real time, now would be a good time for circumspection about what political philosophy, post-Rawls, has lost from its genuflection to neoclassical economics and its curious inattention to money.22

20 21 22

In the same passage, Rawls highlights the importance of fair equality of opportunity, which is arguably very demanding and at best modestly approximated. See Reidy (2014, 10). For a mea culpa in the theory of fairness in trade, see James (forthcoming).

4

Rawls’s Principles of Justice as a Transcendence of Class Warfare Elizabeth Anderson

4.1

The Return of Class Warfare

Fifty years after the publication of A Theory of Justice, class warfare is back. As billionaire Warren Buffett acknowledged, “there’s class warfare, all right, but it’s my class, the rich class, that’s making war, and we’re winning.” (Stein 2006) One could argue that the rich have always been fighting it. Yet in the postwar era, America’s corporate elite was mostly reconciled to the New Deal. The 1980s marked a decisive turn of a significant part of the corporate elite toward class warfare, by way of concerted assaults on progressive taxation, the welfare state, labor unions, antitrust, and other regulatory constraints on business models founded on strategies of redistributing income and wealth from workers to capital owners (Mizruchi 2013). They have been wildly successful: 44 percent of the increase in shareholder wealth from 1989–2017 was due to such redistribution (Greenwald et al. 2021). Yet in the United States, workers had barely mobilized against this class war until the financial crisis of 2007–8. Political philosophers have mostly focused on the libertarian rationales advanced to justify the upward redistribution of income, wealth, and opportunity. Yet in popular political discourse, meritocratic rationales for it have loomed larger. Meritocracy is the view that great inequalities in income, wealth, position, and social status are justified by great inequalities in skill, industry, initiative, and productivity: everyone deserves their place in the socioeconomic hierarchy. Meritocracy has the ideological advantage of grounding distributions in an intuitive notion of desert. (By contrast, the libertarian idea of historical entitlement looks less and less legitimate as the influence of inheritance on the distribution of wealth and opportunity increases.) Meritocracy also has the political advantage of allying the 1 percent with the top 20 percent of earners in the business, professional, and managerial classes, and of offering at least symbolic status to the middle 60 percent. That middle has suffered the most from neoliberal globalization and deindustrialization, relative to their position 78

Rawls’s Principles of Justice and Class Warfare

79

in the postwar decades (Milanovic 2016). Meritocracy offers the middle the consolation prize of salving their wounded pride by focusing their contempt and resentment on the bottom 20 percent. It legitimizes these feelings by stigmatizing the bottom as lazy and welfare-dependent. In the post-Theory era of egalitarian theorizing, luck egalitarians joined the ideological battle by fighting on meritocracy’s turf. They conceded the legitimacy of desert-based inequalities, while contesting the degree to which existing inequalities could be justified on these terms. On their view, undeserved inequalities are unjust. I stand with Rawls in rejecting any notion of desert as a basis for a theory of distributive justice at the systemic or structural level. In prior works, I have criticized luck egalitarianism for offering relief to the disadvantaged on stigmatizing terms and for failing to appreciate the allocative functions of market distributions (Anderson 1999, 2008). In this chapter, I explain why desert-based principles of justice are inherently unsuited for a systemic theory of justice, and hence why the complaint that Theory ignores desert is misguided. This is a prelude to my central argument, which points out that we have been here before. Desert-based theories of distributive justice similarly dominated the previous era of hyper-capitalist inequality in the late nineteenth and early twentieth centuries and similarly defined the ideological terms of class warfare. But no desert-based theory of distributive justice could supply an acceptable and analytically coherent basis for settling this war. Social democracy rather proposed to transcend class conflict by replacing desertbased justice with a conception of justice based on the idea of democratic cooperation among free and equal citizens. A Theory of Justice developed this social democratic thought more rigorously – and also carried it further. Where social democracy aimed to transcend class conflict while preserving socioeconomic classes, a rigorous realization of Rawls’s principles of justice would transcend class society itself by eliminating the possibility of concentrating socioeconomic advantages into heritable class identities. Given the resurgence of class warfare today, it is high time that we recover the ideals of social democracy that Rawls encapsulated in his theory of justice and consider how the difference principle in particular may offer a useful tool for constructing a society of equals. 4.2

Why Systemic Theories of Justice Cannot Be Desert-Based

A Theory of Justice has been criticized for ignoring claims to justice based on desert. Some critics have supposed that Rawls objected to desertbased principles of justice on the ground that people can’t deserve

80

Elizabeth Anderson

anything if they don’t deserve the things (such as genetic endowments and good parents) that enable them to make a good effort.1 Although Rawls seems to suggest this (TJ 1999, 64, 89), nothing in his argument hangs on this erroneous thought. Rather, he rejects desert as a principle of systemic justice because considerations of desert can only apply to individuals’ claims within a given structure of opportunities, not to the structure itself. Rawls’s difference principle (DP) appears to cut most sharply against the grain of desert-based claims. In its general form, it requires that “social and economic inequalities are to be arranged so that they are … reasonably expected to be to everyone’s advantage.” Specifically, they must be arranged so that they are “to the greatest expected benefit of the least advantaged.” (TJ 1999, 53, 72) Intuitively, the least advantaged have contributed considerably less to production or social welfare than successful entrepreneurs and professionals. So it seems that the DP perversely resists rewards in accordance with desert. This objection confuses local and systemic principles of distributive justice. Local principles directly guide agents in distributing goods to specific individuals. There are two types of local principle: individually allocating and procedural. Individually allocating principles focus on features of the individual and take the form “to each, according to their x.” The most common values for x are need and desert. Procedural principles focus on features of the process by which someone acquires or transfers something. The principle of contract, that rights to a good can be legitimately transferred by the owner to another by voluntary agreement in return for consideration, states a procedural rule for a just transfer of goods. Systemic principles govern the choice of systems of local principles by constraining their overall distributive consequences, defined in terms of opportunities for various goods. Opportunities are attached to coarsegrained classifications of social positions – for example, the expectations of a so-called unskilled laborer, an executive, or a citizen. Individuals attain particular outcomes by taking advantage of their opportunities, which deliver goods in accordance with local principles. Systemic principles thus constrain the basic structure of opportunities in society. They do not determine specific outcomes for particular individuals (TJ 1999, 48–49). The DP and the principle of fair equality of opportunity – that inequalities must be attached to positions fairly open to all – are systemic principles. 1

See, for example, Sher (1979) and, in an otherwise commendable essay, Schmidtz (2002).

Rawls’s Principles of Justice and Class Warfare

81

Systemic principles do not displace local principles in determining what particular individuals get. They are satisfied in the course of following local principles. If the local rules of property, contracts, and so forth supply a structure of opportunity that satisfies Rawls’s principles of justice, and people comply with them, then whatever particular distributive outcomes result from individuals freely making choices among the opportunities available to them are just. Distributive justice is “left to take care of itself,” in the sense that there is no need to track individual outcomes and alter them in accordance with a procedure-independent standard of justice (TJ 1999, 76). Rawls argued that his principles of justice could be realized either in a liberal socialist regime, in which the means of production are owned and managed by the state, or in a property-owning democracy, in which the means of production are privately held but ownership is widely dispersed. Either way, a just system largely relies on competitive markets to determine prices and thereby the allocation of resources (TJ 1999, 240–42). Reliance on markets and other impersonal rules (such as of corporate law, labor law, bankruptcy, etc.) enables the difference principle to be realized through a system of pure procedural justice. In a property-owning democracy, the specification of local rules of the legal infrastructure of markets and property will be guided by the DP so as to minimize non-functional pre-tax inequalities, rather than relying on post-tax redistribution (TJ 1999, XV). This leaves room for considerations of desert to play local roles within a just basic structure. Rawls calls such considerations “precepts” of justice. There are three bases of desert: virtue, productive contribution, and personal cost (effort or sacrifice). While a competitive price system has little to do with moral virtue, it offers some room for contribution and sacrifice. At the local level in perfectly competitive markets, wages reflect workers’ marginal productivity. This crudely corresponds to the precept “to each, according to their contribution.” To elicit a supply of workers to jobs requiring skills that are costly to acquire, or that involve significant risks to health or other noxious features, firms typically offer wages that will cover the costs of education and compensate workers for unpleasant aspects of the job, compared to other jobs they could get. This crudely corresponds to the precept “to each, according to their effort or sacrifice.” Since the DP relies on pure procedural justice as realized in competitive markets to determine factor prices, the precepts of reward according to contribution and personal cost thereby get some weight in an economy systemically governed by Rawls’s principles of justice. Yet they are of local significance only. The personal cost precept cannot explain systemic differences in wages across jobs that don’t compete for the same workers.

82

Elizabeth Anderson

Neither precept accurately compensates inframarginal producers for their contributions or personal costs. Neither precept scales up to the whole economy. Moreover, the systemic principles governing the basic structure of opportunities affect the weights given to the precepts. Under fair equality of opportunity, many more people will be well-educated compared to a system that limits educational access to the rich. This will increase the supply of highly skilled workers and reduce the wage premium on skill, thereby reducing the weight given to the precept of effort or sacrifice (TJ, 267–73). The distinction between local and systemic principles of distributive justice falls under Rawls’s more general distinction between justifying individual actions within a practice and justifying the rules of the practice (CP, 20–46). The umpire may justify his individual action of calling a batter out by appealing to the rule of practice that a batter who swings at three pitches without hitting the ball is out. But to justify that rule, we must appeal to considerations outside the rules of the game, such as getting a balance between offense and defense that keeps the game exciting. It might be thought that the principle of desert could simultaneously justify particular actions within a practice and justify the rules of the practice itself. Further reflection shows that desert cannot supply systemic principles of distributive justice. Consider the prize structure of an athletic competition. Should the winner take all, or should there be prizes for second, third, and lower places? How should the prize money be divided among all prize winners? Desert, as measured by excellent performance, may determine which place individuals will win (particularly in judged competitions such as figure skating, where winners are not determined by objective rules of pure procedural justice). But desert cannot determine the prize structure, which is determined before the performances are tallied.2 This point applies equally well to the basic structure (TJ 1999, 275). Consider the structure of job opportunities in a developed market economy: there will be particular numbers of professionals, managers and executives, skilled workers of various kinds, and so-called unskilled workers. The number of each type of position depends on the demand for the products and services each type of worker supplies. Positions are justified by the functions (needs, demands) they serve, not by desert. It would be absurd to hire more doctors beyond what the patient population needs or demands, on the ground that more people “deserve” to be 2

For a discussion of the kinds of principles that bear on the structure of athletic prizes, see Brown (2014).

Rawls’s Principles of Justice and Class Warfare

83

doctors. Only after the market or public policy determines how many openings for each job are demanded – only after the structure of opportunities is determined – does desert operate as a local principle used by managers for awarding particular jobs to particular applicants. The same point applies to the structure of incomes attached to positions. Factor prices tell people where resources should be directed according to other people’s interests and demands, so as to generate mutual gains from the overall system of production. Desert claims, by contrast, are oriented toward an individual’s internal merits or sacrifices. For this reason, efficient market prices do not track individual desert (Anderson 2008). However deserving slide rule makers were, the electronic calculator made their product obsolete. The collapse of their incomes properly directed them to better uses of their labor. Such uses of market prices to reallocate people’s labor to more productive uses are in everyone’s interest, including the least well off, and can be made in the long-term interest even of workers in obsolete jobs, with the institution of unemployment insurance and retraining opportunities. Hence, they are consistent with the difference principle. Attempts to make up the entire income shortfall of otherwise deserving people in obsolete jobs, even outside the market, would defeat the allocative function of market prices and leave everyone worse off. Allocation of workers to positions would then have to be centrally directed, defeating one of the great advantages of the market – its consistency with freedom of occupational choice (TJ 1999, 241). 4.3

From Desert Theory to Democratic Equality: Moving beyond Class Conflict

From the perspective of ideal moral theory, all principles of desert are parochial in reflecting contested conceptions of the good that lack authority for citizens generally in a pluralistic society. This is most obvious for the principle of reward in accordance with virtue. There are too many conflicting conceptions of virtue for this to serve as an impartial, shared, public standard of justice. Moreover, the three intuitive bases of desert – reward in proportion to virtue, contribution, and personal cost – conflict with each other. Rawls follows John Stuart Mill in arguing that intuitive moral argument is incapable of resolving these conflicts (TJ 1999, 268). While Rawls and Mill are correct about the limits of intuitive moral reasoning, their framing of the issue in terms of contradictory moral intuitions elides the political dimensions of conflicts over the bases of desert. Throughout the nineteenth century, arguments over desert

84

Elizabeth Anderson

supplied the terms in which class conflict was waged in ideological terms. Different interpretations of desert were tied to particular class positions. Rawls’s conception of justice as fairness, from which the difference principle is justified, is an attempt to supply a normative perspective that transcends class conflict. Within a democratic society, it expresses a universal rather than a parochial, class-based perspective. The history of attempts to raise desert-based claims to the level of systemic theories of distributive justice reveals their class-bound origins. The idea of a systemic theory of distributive justice – one concerned with justifying the structure of distributive outcomes generated by the whole system of local principles – did not arise until the late eighteenth century.3 However, most of the original arguments concerning distributive justice failed to clearly distinguish local from systemic principles. Instead, they attempted to scale up local intuitive principles to a systemic level. For a century, most of these arguments appealed to a productive contribution conception of desert. They distinguished different sources of income – and thereby classes of people who drew their income from these sources – in terms of their contribution. The arguments divided into three broad camps. The non-Marxist radical tradition from Paine and Ricardian socialists to Henry George advanced a labor theory of value that designated workers, managers, and entrepreneurs as the sole productive classes, who were exploited by idle landlords, financial speculators and manipulators, and recipients of government largess (state bondholders, occupants of state sinecures, and sometimes those on poor relief ). Since the latter classes produced nothing, but merely manipulated the political and financial systems for unilateral gain, their income was undeserved and could be justly taxed and redistributed to the productive classes.4 Some in the Marxist tradition applied the labor theory of value more narrowly. They identified workers as the sole productive class entitled to the entire social product and treated landlords, shareholders, and even capitalists managing their own firms as exploiters.5 Advocates 3

4 5

Fleischacker (2004) identifies the idea of distributive justice with the idea that the distributive system should be arranged so as to guarantee a basic level of material wellbeing to all. While I agree with his chronology and accept that the concern to abolish poverty animated the original theorists of distributive justice, I resist his identification of the idea of distributive justice with a particular material outcome. Advocates of capitalism in the nineteenth century also advanced systemic claims about distributive justice while insisting that it was compatible with mass poverty. I prefer a formal definition of the idea of distributive justice in terms of systemic principles of justice. See Paine (1792 and 1796); Hodgskin (1825); George (1879 [1912]). Marx himself (1875), while he used the labor theory of value as an analytical tool, criticized the German Social Democratic Party for treating it as a normative foundation for a theory of distributive justice.

Rawls’s Principles of Justice and Class Warfare

85

of laissez-faire capitalism argued that in competitive market equilibrium, each factor of production – land, labor, and capital – received its marginal product, and hence that the market system delivered to the owners of each factor precisely what they deserved – even if that meant the workers were impoverished.6 On this view, the unproductive exploiters were recipients of poor relief, trade protections, and the state administrators of these schemes (Sumner 1911; Spencer 1981; Bastiat 1995). Each of these views expressed the position of one side in society riven by class warfare. It is evident that neither radical tradition could be acceptable to the classes designated as pure exploiters. Nor was marginal productivity theory acceptable to the workers. This was not only because it rationalized their misery and denied the unemployed any recourse. It also undermined the justification for treating productive contribution as a basis of desert. The ostensibly non-moral bases of desert – productive contribution and personal sacrifice – got much of their normative force from the fact that they were assumed to be measures of the virtues extolled by the Protestant work ethic. They thereby served as interpretations of the idea of reward according to moral desert. Productive contribution, understood in terms of labor, was normatively compelling because workers exercise the virtues of industriousness, diligence, perseverance, and the active cultivation and exercise of talents. This rationale could not be extended to the productive contributions of land and capital. Even if land and capital contributed value, their owners played no real causal role in production and exercised no virtue. Their contributions were merely artifacts of the legal system that entitled owners to exercise veto power over others’ productive use of their property. To refrain from exercising the veto did not make one productive. Land would produce its bounty even if it had no owners. The same argument could be made for pure (non-entrepreneurial) returns to capital.7 One way around this critique was to shift the desert basis from productive contribution to personal cost. On this view, those who contribute to production at some cost to themselves deserve compensation for their troubles. Nassau Senior led the way, justifying interest income as the reward for abstinence from consumption (Senior 1836, 153). This restored a moral basis for passive income by tying it to work ethic virtues 6 7

Clark (1899); see also Burke (1999) for an early statement of a similar view prior to the advent of marginalist theory. Painites and Georgists acknowledged that managers and entrepreneurs were entitled to profits, but these were seen as returns to organizational and creative labor. Their critique applied most forcefully to absentee landlords, passive investors, and financial manipulators.

86

Elizabeth Anderson

of self-restraint, self-denial, and foresight. The theory could be extended to justify extraordinary profits as compensation for extraordinary risktaking, albeit at the cost of representing recklessness as even more virtuous than prudence. By parallel reasoning, returns to labor are compensation for foregone leisure. Personal cost was a perilous desert basis for defenders of returns to capital, however. It was a stretch to suppose that the wealthiest rentiers sacrificed anything of significance in investing their money.8 The difficulty was not due merely to the declining marginal utility of money. It was that the wealthiest cannot personally consume all their wealth: there is not enough time. It was even harder to imagine how landowners bore personal costs in abstaining from “consuming” their wheat fields and coal mines. Moreover, workers sacrificed far more in laboring than capitalists did in abstaining from consumption, for far lower returns. Personal cost theory, like productive contribution theory, created new terrain for ideological class warfare. In the late nineteenth and early twentieth centuries, many people – including Fabian socialists, New Liberals, progressives, Christian socialists, and social democrats – were seeking ways to think about distributive justice that would transcend class conflict. In material terms, this required a systemic theory that would secure a decent standard of life for everyone, regardless of class position. In ideological terms, this required construction of a normative point of view that members of all classes could share. Theorists pioneered two paths. One, advanced by Fabians, New Liberals, and some American progressives, continued to try to raise desert theory from a local to a systemic level. The other, advanced by social democrats, constructed a conception of distributive justice founded on the reciprocal rights and duties of citizens to each other in a democratic society. The Fabian/New Liberal path was to divide every person’s income, regardless of source, into deserved and undeserved (unearned) parts. Fabians and New Liberals recognized that marginal productivity theory had fatally undermined the labor theory of value. So they embraced it but replaced contribution with personal cost as the basis of desert. At market equilibrium, the marginal product of each factor of production is equal to its marginal cost. Because the marginal unit of each factor obtains a return equal to the cost to the owner of providing it, its entire return is 8

Ferdinand Lassalle, founder of the Allgemeiner Deutscher Arbeiterverein, the predecessor to Germany’s Social Democratic Party, ridiculed the idea that in investing their wealth, the rich in effect practiced asceticism. See the discussion of his and others’ criticisms of Senior’s theory in Fried (1998, 137–42).

Rawls’s Principles of Justice and Class Warfare

87

deserved. However, inframarginal units reap some rent in excess of personal cost, due to the good fortune of embodying scarcer talent, more fertile soil, favorable location, or some other quasi-monopolistic market position. Since these rents are undeserved, individual recipients have no property right to them. These unearned rents could therefore be justly taxed and redistributed on an egalitarian basis. The Fabian/New Liberal strategy ingeniously combined three advantages. First, it abandoned a parochial class perspective in acknowledging that all three factors of production deserved some return and received taxable rents. It thereby put the recipients of rents on wages, profits, and land on an analytical par. Second, it affirmed what was then seen as a Lockean theory of private property, based on sacrifice as measured by personal costs of production. Third, it preserved market efficiency by restricting just taxation to rents – that is, to returns that make no difference to individuals’ decisions to devote factors they own to production.9 Fabian/New Liberal desert theory won these advantages at significant analytic and normative cost, however. Analytically, rent (like marginal productivity) is meaningful only locally, holding for a given transaction in a given market, taking all else as given. It cannot be aggregated across all factor owners in all markets to generate a systemic measure of undeserved income available for taxation.10 Normatively, the theory replaces a substantive normative conception of personal cost in terms of the burdens of work and foregone consumption with the concept of reservation price. Factors command a reservation price because of competing demands for them in the market, even if their owners sacrifice no substantive good in dedicating them to production, as in the cases of landowners who cannot consume their fields and mines, rentiers who own more than they can consume, and workers who enjoy their work. The concept of cost needed to analyze an efficient allocation of resources is therefore distinct from the concept needed for a sacrifice theory of desert. Even worse from an egalitarian point of view, Fabian/New Liberal desert theorists had a hard time justifying their favored policy of progressive income taxes. Wage taxation proved the most embarrassing case.

9

10

For the Fabian and New Liberal economic theory of rents, see Webb (1888) and Hobson (1891). For its normative uses to justify redistributive taxation, see Hobson (1909). Barbara Fried explains how some American progressives adopted rent theory and justified it in terms of Lockean property theory in (1998, chapter 4). My discussion of Fabian/New Liberal desert theory is indebted to her analysis. The reservation price of a factor – the minimum price needed to induce the factor owner to dedicate it to a particular productive use – depends on the owner’s wealth. Taxing rent (income above the reservation price) in one market will reduce producers’ wealth, which will change the owners’ reservation prices for all other factors they own.

88

Elizabeth Anderson

The theory entails that all wages above a worker’s reservation price constitute an unearned rent properly subject to taxation. This could result in regressive wage taxes, because the reservation wage of poor workers would be low (rendering the income above that wage undeserved, hence taxable), while that of the better off would be relatively high.11 Fabian/New Liberal desert theory also supplies no normative basis for a systemic principle for distributing rents, since rents are not deserved by anyone. We do no better if we attempt to return to a productive contribution theory rather than a sacrifice theory of deserved compensation. As Ricardian socialist William Thompson argued, in any economy with a division of labor, it is in principle impossible to attribute specific bits of production to specific increments of work and other factors. The whole product is due to the ways different factors are organized to work together (Thompson 1827, 37). Asking how much of the total output was caused by this or that factor unit makes as much sense as asking how much the hands of a watch have advanced due to the motions of any single gear in the watch, apart from its engaging the activities of the other gears. Even if attributions to current factor contributions were possible, they could not account for the entire product. For current productivity is due in large part to common knowledge and other public goods, as well as unpaid parental and community investments in child development. The Social Democratic Party of Sweden (SAP) set aside desert theory and advanced a conception of distributive justice based on democratic citizenship (Berman 2006, chapter 7). It was the first social democratic party in Europe to abandon the Marxist ideology of class conflict and revolution, fully embrace parliamentary democracy as an essential part of socialism, welcome cross-class coalitions, and construct an identity not as a party for workers alone but for all the citizens of a state. Small farmers, small business owners, intellectuals, and others suffered during the Great Depression as well as workers. The SAP advanced measures to help them, too, as it represented its goal as helping all who suffered, regardless of class identity. Party chair Per Albin Hansson justified SAP policies in terms of an ideal of Sweden as “a society of free and equal individuals in democratic cooperation, where common resources are used to ensure security and wellbeing for all” (Berman 2006, 176). Where Fabian/New Liberal desert theory side-lined class conflict by treating members of different classes on a par with respect to receiving undeserved income, the SAP transcended it by treating them as sharing a 11

See the illuminating discussion of the difficulties of justifying progressive taxation under this type of desert theory in Fried (1991, 203–4, 152–57).

Rawls’s Principles of Justice and Class Warfare

89

common identity as Swedish citizens. It regarded society as composed of citizens joined in a cooperative enterprise for mutual benefit, where the role of democracy is to shape social institutions so as to “ensure security and wellbeing for all,” particularly those who are suffering – the least advantaged. In that context, it recognized that private property and its attendant inequalities could be justified to the extent that they result in economic growth that benefits everyone. In keeping with its identification with all citizens, the SAP focused on universal benefits rather than benefits that were means-tested or exclusively targeted to the working class (Esping-Anderson 1990). The two major schools of egalitarianism in contemporary philosophical discourse – luck egalitarianism and Rawlsian democratic egalitarianism – descended from these the two earlier traditions. Luck egalitarians follow the Fabian/New Liberal path of attempting to raise desert theory to a systemic principle, dividing everyone’s income into a deserved and an undeserved portion, respecting individual property rights and resulting inequalities in the deserved portion, and appropriating the rest for egalitarian redistribution. Rawlsians follow the social democratic path of deriving principles of justice from the point of view of free and equal democratic citizens who view their society as a cooperative enterprise for mutual benefit and aim to structure the basic institutions of society so that all, particularly the least advantaged, benefit from instituted inequalities. Regarding distributions, the key disagreement between the two schools focuses on which inequalities are just. Luck egalitarians say they are just if and only if deserved.12 The difference principle says they are just if and only if they redound to the maximum benefit of the least advantaged. However, the inequalities in question lie at different levels of analysis. For luck egalitarians, the relevant inequalities are actual distributions among individuals. For the difference principle, they are inequalities built into the structure of opportunities. Actual inequalities among individuals in a society ordered by Rawls’s principles are just if and only if they arise through just local procedures defined by rules of property, 12

Here I elide the distinction between desert and “personal responsibility,” which claims that individuals should receive the benefits and bear the costs of their voluntary choices. The latter standard, embraced by luck egalitarians such as Ronald Dworkin, was advanced by laissez-faire advocate Herbert Spencer. Compare Dworkin (1981) with Spencer (1891, Part IV, §35). The difficulty again is that the specific benefits and costs caused by one’s actions are due to the overall organization of institutions and cannot be attributed to specific actions independently of that organization. In practice, Spencer’s standard was used to justify things such as making workers bear the costs of workplace injuries (e.g., because they accidentally moved their hands into the saw) rather than holding factory owners responsible for installing unsafe machines or imposing a relentless pace of production on exhausted workers.

90

Elizabeth Anderson

contract, market regulations, taxation, public goods provision, and so forth, which have been chosen so as to realize the difference principle within the constraints of fair equality of opportunity and equality of basic liberties. Although both schools can credibly claim to have left class conflict behind, luck egalitarians’ attempts to raise desert (or personal responsibility) from local to systemic status continue to raise intractable analytical, normative, and technical difficulties.13 A major advantage of the difference principle is that it avoids many of these difficulties by leaving distributions to individuals up to pure procedural justice, instead of assuming the burdens on privacy, freedom, and dignity of adjusting each individual’s outcomes in line with controversial judgments of desert that require keeping track of “the endless variety of circumstances and the changing relative positions of particular persons” (TJ 1999, 76).

4.4

Property-Owning Democracy: Beyond Class Society

It is worth investigating in more detail how Rawls’s principles of justice transcend class conflict. Consider two ways to do so. One is to retain distinct classes (with accompanying large inequalities in the distribution of wealth) but reconcile everyone to their place in the class system by means of a democratic settlement that ensures a decent standard of living to all. This is what a welfare state attempts to do (TJ 1999, XVI). The other is to end a class-based society. A class-based society is one in which (1) ownership of wealth is highly concentrated, (2) effective access to superior positions is mostly limited to those with more wealth (perhaps by their effective monopolization of superior education), and (3) people transmit their class status to their descendants, so that the wealthy effectively shut out the lower classes from access to superior positions. The current US system of neoliberal meritocracy entrenches class-based inequality.14 Rawls’s principles of justice attempt to end class society by preventing inequalities in income, wealth, education, and occupation from consolidating into distinct and heritable class identities. Rather than focus on the details of Rawls’s argument for his principles of justice, we can see the point more clearly by pulling back to the big picture. The principles of justice are the principles that persons conceiving of themselves as free and equal citizens of a democratic society would 13 14

As I have argued in Anderson (1999, 2008, and 2012). See Sandel (2020, 23–27, 75, 95–104, 165–68) for pithy documentation of these claims and of their devastating impact on the social bases of self-respect and political participation of the less educated.

Rawls’s Principles of Justice and Class Warfare

91

choose to regulate the structure of opportunities of their socioeconomic system. A central aim guiding their choice is to ensure that life in accordance with the principles will sustain their relations to one another as free and equal citizens. The ideal of social equality has historically been worked out by identifying objectionable forms of social hierarchy and experimenting with strategies to dismantle them. Egalitarians have focused on three types of objectionable hierarchy: relations of domination and subordination, of adulation and stigmatization, and of considerability and inconsiderability (in which individuals’ interests count heavily or slightly in the deliberations of others). Rawls objects to a class-based society because it sustains all three types of hierarchy. When wealth is highly concentrated, the rich dominate the political process and thereby subordinate other citizens (TJ 1999, XIV–XV, 198–99). They are liable to influence lawmakers to weight their interests more heavily than others, rendering the less wealthy inconsiderable in access to publicly funded goods, including education. Excessive inequality of expectations for income, wealth, education, and positions also undermines the social bases of self-respect of the less advantaged (TJ 1999, 468). When the less wealthy have little prospect of gaining access to the education that could develop estimable skills and qualify them for superior positions, they suffer stigmatization. Under these conditions, merely formal equality of opportunity leads to a meritocratic society, in which individuals strive to “leave the less fortunate behind in the personal quest for influence and social position” (TJ 1999, 91).15 Given the heritability of class status, the lower classes see themselves shut out in this quest and so feel an excusable envy of the upper classes, who jealously guard their access to privilege against encroachment from below (TJ 1999, 467–69). A welfare state, in tolerating extreme wealth inequality, fails to counter these forms of hierarchy and therefore cannot sustain relations of equality among citizens (TJ 1999, XV). Rawls’s principles of justice aim to realize equal social relations by ending a class-based society. He envisions that his principles can be

15

If the concentration of wealth gets too extreme, then the highest class may expect an income from capital that greatly exceeds the highest salaries that can be earned in the most prestigious occupations. In that case, the economy of esteem is liable to resemble that of nineteenth-century France and England, in which the scramble for status led people to value dynastic inheritance, strive to marry into great wealth, and disdain all who had to work for a living. See Piketty (2014, 238–41). All but the top 1 percent lack the social bases of self-respect in such a society. As Piketty demonstrates, if the rate of return on capital consistently exceeds the rate of economic growth, wealth inequality will increase until it leads to this point.

92

Elizabeth Anderson

realized either in a liberal socialist regime or a property-owning democracy. A liberal socialist regime institutes public ownership of the means of production, which abolishes incomes from profits, interest, and rent, and thereby the economic classes defined by receipt of such forms of income (TJ 1999, 241). All the able-bodied must therefore work for a living. Fair equality of opportunity and limits on wage inequality imposed by the difference principle eliminate the remaining bases for class society. Without extreme wage inequality, no class can arise that controls politics or access to superior education. A property-owning democracy replaces the abolition of income from profit, interest, and rent with progressive taxes on inheritance and gifts to break up large concentrations of wealth. These policies are justified by the principle of equal basic liberties and fair equality of opportunity (TJ 1999, 245). The principle of equal basic liberties includes a guarantee of the fair value of political liberty, which requires broadly equal chances to influence the political process. The principle of fair equality of opportunity requires broadly equal access to education for those of equal innate ability. Both justify policies to block inheritance of hugely unequal fortunes. The fair value of political liberties may even demand a cap on inequalities more stringent than what the difference principle would permit (TJ 1999, 70). Fair equality of opportunity, by spreading education to all and barring any group from monopolizing education or access to better positions, also tends to reduce top wage incomes by increasing the supply of workers qualified for top positions (TJ 1999, 136–37) and blocks the heritability of unequal positions. If excessive inequality is already limited by the other principles, why bother with the difference principle? The DP is a very useful tool for identifying and prohibiting the myriad local (firm level) rules that constitute the redistributive business strategies that have shifted so much income from wages to profits and from less to more advantaged workers. These include such devices as noncompete agreements, mandatory arbitration, outsourcing of work to purportedly independent contractors who are nevertheless intensively managed, and the microdivision of labor into tasks for which workers are hired by the piece on ever shifting and unpredictable schedules, making it difficult for them to assemble enough separate jobs for a stable and secure life. From the point of view of the DP, the dissolution of jobs into routine microtasks, undertaken to reduce workers to interchangeable parts lacking bargaining power, amounts to an unjust deprivation of opportunities for the exercise of skill.16 16

Thus we should not assume that managers’ decisions on how to divide labor are driven only by efficiency considerations, as opposed to the purely redistributive desire to

Rawls’s Principles of Justice and Class Warfare

93

The DP also plays several distinctive roles in orienting the culture of a just society. First, it limits the justification of inequality to a strongly egalitarian conception of reciprocity. The more advantaged can only justify their higher expectations by showing that they are necessary to maximize the prospects of the least advantaged. This gives substantial weight to the interests of the least advantaged and thereby publicly affirms their considerability. The more advantaged refrain from seeking further advantages when this can come only at the expense of the least advantaged. This replaces a meritocratic culture with one founded on fraternity (TJ, 1999, 90–91). Second, the least advantaged enjoy a respectable considerability under the difference principle, because their shares are justified by a principle of reciprocity rather than pity for their misfortune (JF, §42.3). In a system in which incomes are largely obtained from market-determined wages (TJ 1999, 245–46), and wages reflect workers’ marginal productivity, reciprocity is a two-way street. Under the difference principle, the more advantaged recognize that they owe their higher productivity in part to the greater contributions of the less advantaged. Executives become more productive with more productive and hence more highly paid subordinates, who spare them from additional tasks that do not take maximum advantage of executives’ special talents. Public recognition of this fact raises the esteem enjoyed by the less advantaged. Third, unlike the principle of fair equality of opportunity, which only guarantees equal education to the equally capable, the difference principle guarantees the least advantaged greater educational investments, so that they can enjoy a richer culture and the esteem that comes from cultivated abilities (TJ 1999, 87, 91–92). Finally, in a property-owning democracy, the difference principle justifies not only breaking up concentrations of financial wealth but ensuring that each citizen enjoys a substantial share of it. Where a welfare state allows highly concentrated ownership of the means of production, and uses the tax-and-transfer system to redistribute income after the fact, a property-owning democracy focuses on ensuring that ownership of productive assets is widely dispersed from the start. This “put[s] all citizens in a position to manage their own affairs” (TJ 1999, XV). Rawls’s conception of a property-owning democracy culturally shaped by public affirmation of the DP offers a way out of recent worries about the re-emergence of an economy in which the share of income derived capture a greater share of producers’ surplus. Even more importantly, the content of jobs – the skills and responsibilities exercised in them – are subject to the DP and not just their external benefits. For further discussion, see Freeman, (2013).

94

Elizabeth Anderson

from ownership of capital overshadows labor’s share (Piketty 2014). If capital and labor become more substitutable with the rise of robots and intelligent computers, wages may fall along with the demand for labor. This could lead to the emergence of a large underclass of unemployed. If capital ownership is relatively evenly dispersed, however, then there is little cause to worry about the relative shares of income accruing to capital and labor. A property-owning democracy would lead instead to the rise of a leisure society, in which the share of time each person must devote to paid labor is modest.17 Once we interpret the DP in causal and cultural terms – that is, in terms of the kind of social relations it sustains – rather than only as regulating quantitative distributions, a way is open to envisioning how it might be extended to domains of inequality not essentially defined by inequalities in income, wealth, and education. These include inequalities of power between managers and workers, and gender inequality as manifested in dependent care and other domestic labor arrangements. We should not suppose that once a certain type of social hierarchy, defined in terms of a particular conception of class, is left behind, that other unjust hierarchies do not remain. Ultimately, extensions of the cultural logic of the difference principle may lead us beyond primary goods as a metric of distributive justice and beyond property-owning democracy as an embodiment of it.18

17

18

James Meade articulated his conception of property-owning democracy, from which Rawls borrows, in response to earlier worries about unemployment due to automation. See Meade (1965, 25–26, 45–62). For reflections along these lines, see Hsieh (2012); Robyns (2012); and Schweickart (2012).

5

The Significance of Injustice Peter de Marneffe

The concept of justice that A Theory of Justice theorizes is, as Rawls puts it early on, “the proper distribution of the benefits and burdens of social cooperation” (TJ 1999, 5). Shortly thereafter, he comments, “now this approach may not seem to tally with tradition” (TJ 1999, 9), and it is the difference between Rawls’s concept of justice and the traditional one that is my topic here. My main observation is that the significance of injustice in Rawls’s sense is very different from the significance of injustice in the traditional sense. Traditional injustice entails that someone has been wronged in a way that warrants resentment, guilt, and indignation. Injustice in Rawls’s sense does not entail this. Before going on, I will quote the entire paragraph where Rawls explicitly recognizes the difference between his concept of justice and the traditional one. He writes: Now this approach may not seem to tally with tradition. I believe, though, that it does. The more specific sense that Aristotle gives to justice, and from which the most familiar formulations derive, is that of refraining from pleonexia, that is, from gaining some advantage for oneself by seizing what belongs to another, his property, his reward, his office, and the like, or by denying a person that which is due to him, the fulfillment of a promise, the repayment of a debt, the showing of proper respect, and so on. It is evident that this definition is framed to apply to actions, and persons are thought to be just insofar as they have, as one of the permanent elements of their character, a steady and effective desire to act justly. Aristotle’s definition clearly presupposes, however, an account of what properly belongs to a person and what is due to him. Now such entitlements are, I believe, very often derived from social institutions and the legitimate expectations to which they give rise. There is no reason to think that Aristotle would disagree with this, and certainly he has a conception of social justice to account for these claims. The definition I adopt is designed to apply directly to the most important case, the justice of the basic structure. There is no conflict with the traditional notion. (TJ 1999, 9–10)

Certainly, there is no conflict between Rawls’s concept of justice and the traditional one. The two are consistent. Individuals can do what they owe to each other when the benefits and burdens of social cooperation 95

96

Peter de Marneffe

are properly distributed. But it does not follow from this that Rawls’s definition “tallies with tradition” if “tallies with” means “agrees with” or “corresponds to,” because it is in fact an entirely different concept. The paragraph just quoted seems to suggest that this is a quibble, but I think it has greater significance than Rawls explicitly recognizes. A century earlier, Mill had identified justice with perfect duty, writing, “justice implies something which it is not only right to do and wrong not to do, but which some individual person can claim from us as his moral right” (Mill 1979, 49). Mill recognizes other senses of justice as well, but his identification of justice with perfect duty captures the central concept of justice as traditionally understood – by Plato,1 Aristotle,2 Aquinas,3 1

2

3

The first definition of justice considered in Plato (1992) is “speaking the truth and paying whatever debts one has incurred,” 5 (331c). The second definition considered is “to give each what is owed him,” 6 (331e). Characteristically, Socrates is not satisfied with either definition and Plato goes on to characterize justice in several different ways, finally arriving at some characterizations that Socrates seems satisfied with in Book IV. However, the central notion of not wronging others remains central to the argument of Republic. Glaucon tells the story of the ring of Gyges to press the question of whether justice is non-instrumentally good for a person and in pressing this question with this story he presupposes that murder and theft are unjust and that they are the kind of unjust actions that people would commit if they thought they could get away with them, which he takes to indicate that people do not believe that justice is non-instrumentally good for them. Given the centrality of Glaucon’s challenge to motivating the argument of the book, we may infer that a person who is just in Plato’s sense of having a well-ordered soul will not commit unjust actions of this kind and that a society that is just in Plato’s sense of being well-ordered will be one in which the rulers do not commit unjust actions of this kind against each other or against the people. In other words, if one has justified true beliefs about what one ought to believe, do, feel, and want, and the emotional and desiring parts of one’s soul are properly regulated by these beliefs, one will not wrong others in these ways. One will not murder or steal or violate one’s agreements. One will not murder the king as Gyges did. One will not sentence Socrates to death as the Athenian assembly did. For passages supporting this interpretation (see Plato 1992, 118–121 [442a–44e]). In Aristotle (2019, 80 [1129a–29b]), Aristotle distinguishes two different notions of justice. One is “lawfulness,” which is “complete virtue in relation to another.” The other is fairness, understood as not taking more than what one is entitled to; see Irwin’s gloss on pleonexia, at Aristotle 2019, 388–89. Because taking more than one is entitled to does not violate a perfect duty unless one is taking it from someone else who is entitled to it, it is debatable whether fairness in Aristotle’s sense should be identified with perfect duty, but Rawls clearly interprets him this way in the passage quoted above in the text (TJ 1999, 9–10). In Aquinas (1947, II–II, 58, 11), Aquinas writes: “the proper act of justice is nothing else than rendering to each one his own” and makes clear that by “his own” he means “what is owed to him.” Following Aristotle, Aquinas also recognizes the concept of “distributive justice,” “which distributes common goods proportionately,” that is, in proportion to merit, (1947, II–II 61,1). This is a species of justice for Aquinas because each has “what is owed to him” (in some sense) only if common goods are distributed proportionately, and this notion of distributive justice does not fit the logic of perfect duty well because it is unclear who, if anyone, owes this to him. The requirements of perfect duty are nonetheless a central element of Aquinas’s understanding of justice as initially defined.

The Significance of Injustice

97

Hobbes,4 Locke,5 Adam Smith,6 Hume,7 and Kant,8 among others.9 Rawls, too, incorporates this traditional notion of justice into his theory, under the heading of “natural duties,” which apply to individuals, in contrast to “justice,” which, in his theory, applies to “social systems and institutions” (TJ 1999, 94).10 These natural duties, according to Rawls, include “the duty of helping another when he is in need or jeopardy, provided that one can do so without excessive risk or loss to oneself; the duty not to harm or injure another; and the duty not to cause unnecessary suffering” (TJ 1999, 98). Later, Rawls identifies a natural duty of mutual respect, as well, which requires us “to treat one another civilly and to be willing to explain the grounds of [our] actions, especially when the claims of others are overruled” (TJ 1999, 156). In light of recent work by Scanlon (1998), Darwall (2006), Wallace (2019) and others, we can characterize the significance of natural duties in the following way. When we fail to perform a natural duty, we fail to give some other specific person what we owe him and in doing so act in a way that grounds a grievance on his part. We flout a valid secondpersonal demand that he has the standing to make of us. We violate principles for the regulation of our conduct toward each other that no one can reasonably reject and so act in a way that we cannot justify to him. In this way, we fail to respect him as a person, or to value him

4 5

6 7

8

9

10

In Hobbes (1985, part I, chapter XV, 202), Hobbes writes: “the definition of injustice is no other than the not performance of covenant. And whatsoever is not unjust is just.” In Locke (1959, vol. II, book IV, chapter 3, section 18, 208), Locke writes: “`Where there is no property there is no injustice,’ is a proposition as certain as any demonstration in Euclid: for the idea of property being a right to anything, and the idea to which the name of ‘injustice’ is given being the invasion or violation of that right, it is evident that these ideas, being thus established, and these names annexed to them, I can as certainly know this proposition to be true, as that a triangle has two angles equal to two right ones.” In Locke (1983, 31), Locke characterizes “no violence nor injury” as “narrow measures of bare justice” in contrast to “charity, bounty and liberality.” In Smith (1869, part II, section II, chapter 1, 73), Smith writes, “the violation of justice is injury: it does real positive hurt to some particular persons.” In Hume (1978, book III, part II, section XI, 567), Hume identifies “the three fundamental laws of justice” as “the stability of possession, its transference by consent, and the performance of promises” and characterizes these as “duties of princes, as well as of subjects.” In Kant (1991, 65), Kant identifies “duties of right” with perfect duties. Consequently, if we equate “duties of right” with “duties of justice,” or translate “recht” as “justice,” as some translators do, e.g., Kant (1965), then Kant, too, identifies justice with perfect duty. See Kant (1965, XVII–XVIII), for John Ladd’s explanation of this translation choice. According to Ulpian (c. 170-223? 228?), “Justice is a steady and enduring will to render unto everyone his right …. The basic principles of right are: to live honorably, not to harm any other person, to render to each his own.” See Watson (1998, vol. I, 2.) Rawls also refers to the natural duties throughout as “principles of right” in contrast to “principles of justice,” as in the phrase “principles of right and justice” (TJ 1999, 167).

98

Peter de Marneffe

properly as a fellow rational being. This warrants resentment on his part, guilt on our part, and indignation on the part of others. Injustice in the traditional sense thus has a special kind of significance for our relations to each other, which, I suggest, the failure of our social institutions to conform to Rawls’s two principles of justice does not have. I should make clear that in questioning whether Rawls’s concept of justice “tallies with tradition,” I am not questioning whether anyone before Rawls had the concept of social justice understood as the proper distribution of the benefits and burdens of social cooperation. As Rawls observes in the paragraph just quoted, Aristotle had this concept of social justice, too, although his conception of social justice was of course very different. Sidgwick, too, recognizes this concept of justice in The Methods of Ethics, where he refers to “an ideally just distribution of rights and privileges, burdens and pains, among human beings as such” (Sidgwick 1981, book III, chapter 5, section 3, 274).11 My point is simply that this concept of social justice is very different from the notion of perfect duty that Rawls himself identifies as the traditional notion of justice in the paragraph above. To clarify further, I am not claiming that Rawls’s concept of distributive justice is different from the traditional concept of distributive justice, a claim that has been made by others. Fleischacker, for example, maintains that Rawls’s concept of distributive justice is fundamentally different from Aristotle’s. This is because he identifies what he calls “the modern concept of distributive justice” with a perfect duty on the part of “the state” to guarantee everyone a sufficient minimum of material goods (Fleischacker 2004, 4, 6, 76, 79, 123, 160–61), a duty that was not recognized by political philosophers prior to the late eighteenth century (Fleischacker 2004, 75–79), and because he attributes this modern concept to Rawls (Fleischacker 2004, 114). But what Fleischacker calls the “modern concept of distributive justice” is not in fact Rawls’s concept. Rawls’s concept of distributive justice is “the proper distribution of the benefits and burdens of social cooperation” and this concept is found in Aristotle. Fleischacker is quite right that Aristotle does not recognize a perfect duty on the part of the state to guarantee a sufficient minimum, but this concept of distributive justice is very different from Rawls’s concept. For one thing, the fact that the basic structure does not ensure a proper distribution does not entail that it fails to guarantee a sufficient minimum. Attributing the modern concept of distributive justice to 11

On the previous page, Sidgwick writes: “we seem to recognize an ideal system of rules of distribution which ought to exist, but perhaps have never yet existed, and we consider laws to be just in proportion as they conform to this ideal.”

The Significance of Injustice

99

Rawls, Fleischacker writes: “Rawls assumes that duties of distributive justice are duties that a state owes to its citizens” (Fleischacker 2004, 119), but he gives no page reference and I do not believe that Rawls ever claimed this. What Rawls claimed is that the basic structure is just if and only if it satisfies his two principles (TJ 1999, 14–15, 75, 78–79, 343). Where Rawls first introduces his concept of justice, he dubs it “social justice” (TJ 1999, 4) and by “Rawls’s concept of distributive justice” I mean his concept of social justice.12 But Rawls does not explicitly define “social justice” where he first mentions it. Instead, he writes that “the principles of social justice … provide a way of assigning rights and duties in the basic institutions of society and … define the appropriate distribution of the benefits and burdens of social cooperation” (TJ 1999, 4). Then on the next page he explains that “the concept of justice” is specified by this role, in contrast to “conceptions of justice” that are given by the different sets of principles that might play this role (TJ 1999, 5). This tells us what role principles of social justice are to play, but it does not actually tell us what social justice is. Several key passages, however, in his first two sections show that by social justice Rawls means the proper distribution of the benefits and burdens of social cooperation. Section 1 opens with the ringing sentence: “Justice is the first virtue of social institutions, as truth is of systems of thought” (TJ 1999, 3), which makes sense only if justice consists in the proper distribution of the benefits and burdens of social cooperation. In section 2 Rawls writes: “The justice of a social scheme depends essentially on how fundamental rights and duties are assigned and on the economic opportunities and social conditions in the various sectors of society” (TJ 1999, 7), which makes sense only if justice is the correct assignment of rights and duties and the correct distribution of opportunities. Then at the end of this same section, right before the paragraph quoted above, Rawls identifies “the concept of justice” with “the proper balance between competing claims” (TJ 1999, 9). These passages show that by social justice (the concept) Rawls means the correct assignment of rights and duties and the proper distribution of the benefits and burdens of social cooperation. Different conceptions of justice, then, constitute different views about what assignments of rights and duties are correct and what distributions are proper. This is what Rawls means when he says that Aristotle “has a conception of social justice” (TJ 1999, 10): he has a conception of the proper distribution of the benefits and burdens of social cooperation.

12

Fleischacker (2004, 1), too, uses the terms “distributive justice” and “social justice” interchangeably.

100

Peter de Marneffe

My main observation, then, is that this concept of social justice is very different from the perfect duty concept of justice, which is the concept that has been central to our philosophical tradition, and consequently that injustice in Rawls’s sense has a different significance for our relations to each other from injustice in the traditional perfect duty sense. This difference is revealed when we see that a society that is not just in Rawls’s sense might be one in which there is no injustice in the traditional sense. Rawls states his two principles of justice as follows: A. The Two Principles of Justice (in serial order): 1. The principle of greatest equal liberty 2. (a) The principle of (fair) equality of opportunity (b) The difference principle (TJ 1999, 107). He then identifies several “mixed conceptions” of justice by reference to this statement of his two principles: B. Mixed Conceptions. Substitute for A2 above 1. The principle of average utility; or 2. The principle of average utility, subject to a constraint, either: (a) That a certain social minimum be maintained, or (b) That the overall distribution not be too wide; or 3. The principle of average utility subject to either constraint in B2 plus that of equality of fair opportunity (TJ 1999, 107). These mixed conceptions are less egalitarian than Rawls’s conception because they do not include the difference principle, which prohibits social and economic inequalities that do not function to benefit the least advantaged. As a result, these mixed conceptions are less egalitarian, too, in doing less to “mitigate the influence of social contingencies and natural fortune on distributive shares” (TJ 1999, 63). But although they are less egalitarian, some of these mixed conceptions are nonetheless liberal conceptions. A liberal conception of justice, according to Rawls, has the following features: it specifies certain basic rights and liberties, assigns special priority to their recognition and protection, and ensures that all citizens have adequate all-purpose means to make effective use of these rights and liberties (PL, 223). The mixed conceptions satisfy the first two criteria by including A1 and in giving it priority over the B principles. Mixed conceptions B2 and B3 satisfy the third condition as well via 2(a) or 2(b). And even B1 might satisfy this third condition because, given the diminishing marginal utility of money, policies that maximize average utility might effectively ensure for everyone adequate all-purpose means. But although some, and perhaps all, of these mixed conceptions are liberal conceptions, a society that is effectively regulated

The Significance of Injustice

101

by one of them would not be just in Rawls’s sense. A society is just in Rawls’s sense only if its basic social institutions are effectively regulated by a conception of justice that identifies the proper distribution of the benefits and burdens of social cooperation and, according to Rawls, the proper distribution is the one identified by his two principles. Consequently, if the basic structure of society is effectively regulated by one of these mixed conceptions, it is not just in Rawls’s sense. What I now observe is that it does not follow from this that there is any injustice between individuals in this society. In a regime effectively regulated by a liberal mixed conception, private citizens might perform all their natural duties and, consequently, not flout any valid second personal demand, or act in a way that they could not justify to each other, or disrespect each other as persons, or act in a way that warrants resentment, guilt, and indignation. It is quite true that in a regime of this kind, even if I wrong no one, I might enjoy advantages as the result of luck that I cannot, in some sense, justify to others. But it is important to distinguish here two different senses of justification to others. In one sense, I can justify my good fortune to others by reference to the fact that I do not violate any natural duty to them in having it. In another sense, I can justify my good fortune to others only by reference to the fact that it is consistent with the proper distribution of the benefits and burdens of social cooperation. Assuming for the sake of discussion that Rawls’s two principles define the proper distribution, the advantages I have under an alternative liberal conception of justice might not be justifiable to others in this second sense. But it does not follow from this that they cannot be justified to others in the first sense. They might nonetheless be justified to others in the first sense because in having these advantages I violate no natural duty. Some might think that the justifiability of our basic social institutions is itself something we owe to each other.13 But the justifiability of our social institutions is not under anyone’s control. So the sense in which we could owe it to each other is obscure. Assuming that we have a natural duty to others only to do what it is possible for us to do, and assuming that it is not possible for any one of us to ensure that the basic structure is fully justified, the justifiability of our social institutions does not seem to be the kind of thing that any one of us could have a natural duty to provide any other. One of our natural duties, according to Rawls, is the natural duty of justice, which “requires us to support and comply with just institutions

13

Some remarks in Nagel (2005) might be taken this way. See 129–30.

102

Peter de Marneffe

that exist and apply to us” (TJ 1999, 99). This does not entail that someone wrongs someone else if just institutions do not exist, but Rawls goes on to write that this natural duty “also constrains us to further just arrangements not yet established, at least when this can be done without too much cost to ourselves” (TJ 1999, 99), and one might think that this natural duty requires citizens of a liberal society that is not yet just to vote, directly or indirectly, in favor of policies warranted by Rawls’s two principles, which I will call “justice as fairness policies.” But is it true that we have a natural duty to vote for such policies? Suppose you accept an alternative liberal conception of justice. Suppose you have considered Rawls’s argument from the original position and have concluded, upon due reflection, that parties in the original position would in fact choose one of the mixed conceptions that Rawls identifies instead of his two principles. Or suppose you accept a liberal conception of justice in which the original position plays no fundamental justifying role at all. Do you still have a natural duty to vote for justice as fairness policies? Several considerations, taken together, may lead one to doubt this. First, the conception you hold includes the principle of equal liberty or something like it. So in voting according to your conception of justice you are not voting in favor of slavery or religious persecution or any other grave violation of individual rights. Second, the conception of justice you hold commits you to the goal of ensuring that everyone has the means necessary to make effective use of their basic liberties, which entails a decent minimum. Third, your vote makes virtually no difference to the outcome. So you are not actually harming anyone by voting the way you do. Finally, because the vote is valuable to you partly as a means of expressing your sincere political convictions, and others have their own vote with which they can express their sincere political convictions, a demand from others that you vote according to their sincere political convictions seems out of place, even, perhaps, a form of pleonexia. Legislators in a liberal society that fails to ensure the proper distribution of the benefits and burdens of social cooperation might still have a natural duty to vote for justice as fairness policies, even if private citizens don’t, but some argument is needed here because the four points just made about private citizens apply to legislators too. And suppose that only a small minority of citizens endorses Rawls’s two principles and that legislators are elected through fair procedures and are open to defeat in the next election. Do legislators then have a natural duty to vote for justice as fairness policies even if those who elected them do not endorse them? Rawls himself does not claim that legislators in a non-well-ordered liberal society have a natural duty to vote according to his conception of

The Significance of Injustice

103

justice, and his discussion of the duty of civility in Political Liberalism (PL, 217) suggests that he does not in fact believe this. One might now suggest that if the basic structure is not effectively regulated by Rawls’s two principles of justice, then it is the basic structure itself that wrongs us. But the basic structure is not a person.14 Consequently, injustice of the basic structure cannot have the moral significance for us that being wronged by another person has. Some theorists seem to treat the state as an agent, which makes sense if the actions of state officials are determined by the directives of a single person or deliberative body.15 But our state – a federal system with multiple branches of government at both the national and state levels, and with semi-autonomous agencies such as the Federal Reserve Bank – is nothing like this.16 Moreover, for state actions to have the significance for us that the actions of individuals have, the state must be not only an agent but a person, too, and to be a person, an agent must be able to engage in practical reasoning, to consider pros and cons, and to make normative judgments that reliably guide its actions.17 State officials are persons in this sense and they can wrong us. Deliberative bodies might count as a quasi-persons, too, if certain conditions are met. But the modern state as such is not a deliberating agent and so cannot wrong us in this way. According to the Kantian interpretation that Rawls sketches in section 40, by living together in the well-ordered society of justice as fairness and by faithfully performing our natural duties and obligations, we express “our nature as free and equal rational persons” (TJ 1999, 226).18 This identifies something good about living in this kind of society, which might offer a way of understanding the personal significance of injustice in 14 15 16

17 18

Or, as Estlund (2020, 43) has recently put a closely related point: “An individual is not a basic structure, nor can any person be a fair system of social cooperation.” On the nature of collective agency, see Scanlon (2008, 162); and List and Pettit (2013). It makes even less sense to regard the basic structure as an agent. To count as an agent, an entity must have a coherent set of beliefs and desires that determines or explains its behavior. This is not true of the basic structure, which includes competitive markets, the educational system, and the institution of the family, because the behavior of persons acting within these institutions is not determined and cannot be explained by the coherent set of beliefs and desires of any single individual or group agent. For related discussion, see Scanlon (2008, 165); and List and Pettit (2013, 154–59). Rawls also writes that we express our nature as free and equal rational beings when we “act on” or “act from” his two principles of justice (TJ 1999, 222). This is slightly awkward because, as he makes clear elsewhere, the principles of justice apply to institutions and not to individual agents. But as he makes clear later in the book, we act “on” or “from” the principles of justice in the well-ordered society of justice as fairness by performing our natural duties and our obligations within this society. Thus, as he puts it toward the end of the book (TJ 1999, 495), “it is the principles of individual duty and obligation that define the claim of this ideal upon persons.”

104

Peter de Marneffe

Rawls’s sense that does not depend on a link to natural duty. Perhaps we fail to express our nature as free and equal rational persons by living in a different kind of society. But is this true? Suppose the basic structure is effectively regulated by a liberal conception of justice other than justice as fairness. Suppose you perform all your natural duties and obligations and vote for justice as fairness policies. Isn’t this enough to express your nature as a free and equal rational person? How would you express this nature any better in the well-ordered society of justice as fairness? In a less egalitarian liberal society, “the accidents of natural endowment and the contingencies of social circumstance … that seem arbitrary from a moral point of view” (TJ 1999, 14) will play a larger role in determining the distribution of goods, but how does this affect what you express by your own actions? The observation I am making here about Rawls’s concept of social justice can be expressed as a question: If you perform all your natural duties, how does living in an unjust society impair your moral relations with others? One way to understand the recent debate between relational and distributive egalitarians is to think that relational egalitarians are more troubled by this question than distributive egalitarians are. Anderson, for example, bases her critique of luck egalitarianism squarely on the traditional notion of justice, which she endorses as the correct one, and forcefully characterizes as follows: First, there can be no injustice without an injury to someone’s interests. Second, there can be no injustice without an agent who is (or was) substantively responsible for it – someone obligated to avoid, correct, or bear the cost of the injustice or of its correction or amelioration. Third, there can be no injustice without an agent who is (or was) entitled to complain to the responsible agent, to hold that agent to account and exact compliance with the demand. Fourth, there can be no injustice where all agents continuously and successfully comply with all the demands that everyone can reasonably make of them. (Anderson 2010, 5)

She goes on to conclude: It follows that justice as an evaluation applied to states of affairs is entirely derivative of justice as an appraisal of the conduct of agents. Where all agents conduct themselves justly – where they successfully comply with all reasonable demands – the state of affairs resulting from their conduct is just. (Anderson 2010, 5, italics in original)19

The challenge, then, for a relational egalitarian like Anderson is to explain the injustice of inequality solely by reference to the complaints that individuals can make on their own behalf to the conduct of others. Anderson herself suggests in a footnote that a complaint of this kind 19

This is similar to Nozick (1974)’s understanding of justice.

The Significance of Injustice

105

might be made against “the state” or “society at large” (Anderson 2010, 5), but for the reasons just given, neither “the state” nor “society at large” is a person and so is not the kind of entity to which second-personal demands can be addressed or with which we can have second-personal moral relations.20 To what extent, then, can egalitarian commitments be grounded solely on what we owe to each other as individuals? Is it possible, for example, to ground Rawls’s fair equality of opportunity principle – that “those of similar abilities and skills … should have the same prospects of success … irrespective of the income class into which they were born” (73) – solely on the objections that individuals can make on their own behalf to the conduct of other individuals? If not, the gap between Rawls’s concept of justice and the traditional one remains unbridged. This is not an objection to Rawls’s theory or to the concept of social justice as he understands it.21 It is simply an observation about the significance of injustice on this understanding and more specifically about the significance of injustice for our relations with each other. Because injustice in Rawls’s sense does not entail any form of interpersonal mistreatment or disrespect, injustice in his sense does not have the same significance for our relations with each other that injustice in the traditional sense does. One might go further and say that whereas the traditional notion of justice is identical to the morality of right and wrong that Scanlon, Darwall, and Wallace have written about, justice in Rawls’s sense is fundamentally discontinuous with it. A Theory of Justice does not present a distinctive theory of right and wrong in this sense. Rather it presents a vision of the ideal democratic society, something worthy of aspiration but not necessary for mutual respect between persons.

20

21

For Darwall, a presupposition of second-personal address is that those we address can make similar demands on us on their own behalf, which requires that they regard themselves as having the standing to make such demands. See Darwall (2006, 12–14, 28–29, 60). If so, then another reason why neither the state nor society at large is a person is that neither has the mental capacity to regard itself in this way. Because neither has self-consciousness, there is no mental state that could count as regarding itself as having such standing. For theoretical difficulties that might face this concept of social justice, see Estlund (unpublished).

Part II

Developments between A Theory of Justice and Political Liberalism

6

On Being a “Self-Originating Source of Valid Claims” Stephen Darwall

6.1

The Place of the Dewey Lectures in Rawls’s Corpus

Rawls first introduces the idea that persons are self-originating sources of valid claims in his 1980 Dewey Lectures as a central element of what he there calls “Kantian Constructivism.” The Dewey Lectures occupy a distinctive position in Rawls’s philosophical trajectory. Delivered nine years after A Theory of Justice, they constitute Rawls’s last publication before his “political turn,” first in “Justice as Fairness: Political, Not Metaphysical” (in 1985), and then in Political Liberalism (in 1993). Quite differently from political liberalism, Rawls’s project in the Dewey Lectures would appear to be metaethical. Kantian constructivism provides, he says, a theory of “moral objectivity,” holding that this can be “understood in terms of a suitably constructed social point of view that all can accept” (CP, 307). Rawls contrasts constructivism with “rational intuitionism,” the metaethical view that there is a “self-evident” “moral order that is prior to and independent of our conception of the person” or any “suitably constructed social point of view” that employs it (CP, 343). The Dewey Lectures aim to show how the argument from the original position, understood in terms of a “model conception” of the person as a selforiginating source of claims, can be seen to provide a metaethical framework for the normative theory of justice as fairness (and, implicitly, for a more general moral theory Rawls calls “rightness as fairness” (TJ 1971, 111) set out in A Theory of Justice. Constructivism plays an important role in Political Liberalism also, but there it is what Rawls calls “political constructivism,” which he explicitly distinguishes from Kantian “moral constructivism.” In Political Liberalism’s terms, Kantian or moral constructivism is a “comprehensive view” (PL, 89–130), which is to be bracketed in political theorizing about justice. By contrast, we might almost title the Dewey Lectures: “Kantian Constructivism: Metaphysical, Not Political.” In what follows, I want to consider how Kantian Constructivism and its claim that persons are self-originating sources of valid claims contrasts 109

110

Stephen Darwall

with the “Kantian Interpretation” in A Theory of Justice and to consider the philosophical implications of this development in Rawls’s view. Both the Kantian Interpretation and Kantian Constructivism share the Kantian view that, as Rawls put it in A Theory of Justice, “moral legislation is to be agreed to under conditions that characterize men as free and equal rational beings” (TJ 1971, 252). It is common to both works that the parties to the original position are conceived to be “representatives of free and equal moral persons” (CP, 322) and, therefore, that principles of justice (or morality more generally) must be choiceworthy from that point of view.

6.2

Two Aspects of Freedom

But the two works differ in what they take freedom in “free and equal” persons to consist in. In A Theory of Justice, Rawls understands freedom as the Kantian idea of “independence from the contingencies of nature and society,” a phrase that recalls Rawls’s thought that social shares should be independent of “social fortune” or “luck in the natural lottery” (TJ 1971, 256, 75). Moral agents are not simply subject to forms of motivation that are given by nature and society; they can step back from these, critically reflect on them, and choose their own ends. “Free persons conceive of themselves as beings who can revise and alter their final ends” (TJ 1999 131).1 This gets carried forward in the Dewey Lectures’ claim that persons have a “moral power” consisting in “the capacity to form, to revise, and rationally to pursue a conception of the good” (CP, 312). To make it more evident that parties in the original position represent free moral persons in this sense, the Dewey Lectures give the parties a “highestorder interest” to realize and exercise this moral power (CP, 312). (This highest-order interest is also included in the revised edition of [TJ, 1999]). In the Dewey Lectures, however, this moral power, which Rawls calls “independence,” is only one “aspect” of what he there calls freedom (CP, 330). To independence, Rawls adds a second aspect of freedom, namely, that “persons are self-originating sources of valid claims.” Rawls cites slavery as an illustration that this is indeed an aspect of freedom: “slaves are human beings who are not counted as self-originating sources of claims at all” (CP, 331). This “makes clear,” he says, “why counting moral personality itself as a source of claims is an aspect of freedom” 1

Note that this passage is from the revised edition (TJ 1999).

On Being a “Self-Originating Source of Valid Claims”

111

(CP, 331). In the Dewey Lectures, then, Rawls brings in the idea that persons have a fundamentally equal authority, being a self-originating source of valid claims, as a further essential feature of Kantian Constructivism. This played no role in A Theory of Justice (even, we might note, in the 1999 revised edition).2 But what difference does this addition really make? It is revealing here to consider points Rawls makes in the section on the Kantian Interpretation in A Theory of Justice in responding to a criticism he attributes to Sidgwick. Rawls develops there the idea that principles of justice that are chosen in the original position can be linked to the Kantian idea of autonomy. “Kant held,” Rawls writes, “that a person is acting autonomously when the principles of his action are chosen by him as the most adequate possible expression of his nature as a free and equal rational being” (TJ 1971, 252). Rawls then cites Sidgwick as remarking “that nothing in Kant’s ethics is more striking than the idea that a man realizes his true self when he acts from the moral law, whereas if he permits his actions to be determined by sensuous desires or contingent aims, he becomes subject to the law of nature” (TJ 1971, 254). However, Rawls continues, Sidgwick objects that “on Kant’s view the lives of the saint and the scoundrel are equally the outcome of a free choice (on the part of the noumenal self ) and equally the subject of causal laws (as a phenomenal self )” (TJ 1971, 254). Rawls comments that “Sidgwick’s objection is decisive” so long as one takes a noumenal choice of principles to be unconstrained by anything but consistency. A scoundrel could guide themselves unfailingly on consistent self-given principles and, indeed, take such principles to override strong motivations to the contrary, but, Rawls says, that would still fall short of what Kant’s idea involves.3 “The missing part of the argument,” Rawls adds, “concerns the concept of expression” (TJ 1971, 255). We can realize our true (rational) self only by “expressing it in our actions.” Kant himself “did not show that acting from the moral law expresses our nature in identifiable ways that acting from contrary principles does not.” However, the “conception of the original position” makes good this 2

3

Evidently, this thought was firmly anchored in Rawls’s thinking as early as the 1950s and partly derived from his serious engagement with contemporary Protestant theology. See Bok (2017)for a fascinating account of the roots and role of the idea of reciprocal recognition in Rawls’s early theorizing about morality. Here are some examples: “We do not choose our morality, we find that we have it, that it springs from our recognition of others” (Bok 2017: 167). “Morality, and in particular justice, is imbedded in the act of recognizing persons as persons: justice is the reciprocal recognition of persons as persons” (Bok 2017: 178–79). Cf. Gibbard on the possibility of a “thoroughly coherent Caligula” in criticizing Korsgaard in a similar context (Gibbard 1999).

112

Stephen Darwall

“defect.” If we think of “the original position as the point of view from which noumenal selves see the world” and think of the parties as having “a desire to express their nature as free and equal members of the intelligible realm” (roughly, I take it, the highest-order interest in exercising their power of “independence”), then, Rawls thinks, the defect to which Sidgwick objected can be eliminated. By acting on principles that they would choose from the original position (“the point of view from which noumenal selves see the world”), rational persons “express their nature as free and equal members of the intelligible realm.”

6.3

A Gap in the Kantian Interpretation: Independence an Insufficient Ground for the Principles of Right

Now suppose that Rawls is right about that. Suppose, that is, that acting on principles that would be chosen in the original position best expresses our nature as free and equal rational persons in the sense of the Dewey Lectures’ first “moral power” of independence: our “capacity to form, to revise, and rationally to pursue a conception of the good” (CP, 525). Even if this were true, that could not adequately ground or otherwise vindicate the chosen principles as principles of right or justice. Requirements of morality and justice have a distinctive normative profile, one that is connected conceptually to the idea of legitimate or justified claims in a way that being most rationally choiceworthy or the fullest expression of our independent and rational nature is not. The problem is akin to the one Prichard famously raised in “Does Moral Philosophy Rest on a Mistake” (Prichard 2002). Showing that acting morally or justly best expresses our rational independence, or is in our interest, cannot justify requirements of morality or justice in their own terms. To see the problem, consider what Rawls goes on to say regarding the claim that acting on principles we can affirm in the original position “expresses” our nature as free and rational: It is for this reason, I believe, that Kant speaks of the failure to act on the moral law as giving rise to shame and not to feelings of guilt. And this is appropriate, since for him acting unjustly is acting in a manner that fails to express our nature as a free and rational beings. (TJ 1971, 256)

In other words, in failing to “live up” to our free and rational nature, we act beneath ourselves and naturally feel shame (the feeling that one is fittingly looked down upon as contemptible or unpresentable in some way) (Darwall 2018). It is significant that Rawls links independence in the Kantian Interpretation to shame rather than guilt. Shame lacks a conceptual

On Being a “Self-Originating Source of Valid Claims”

113

connection to the right that guilt has. To feel shame is to feel as if one is not as good, as estimable or presentable, as one wishes to present oneself as being. Shame responds, broadly speaking to the good, conceived to include the estimable, and not (or not necessarily anyway) to the right, that is, to our having acted wrongly or violated someone’s just claims. Of course, wrongdoing can be an occasion for shame, as in the passage above. But shame does not respond to wrongdoing as such. Rawls’s Kant sees it as a failure to live up to our better or higher free and rational nature. Rawls himself makes this very point later in A Theory of Justice. He asks us to imagine someone who has cheated or given in to cowardice and who “feels both guilty and ashamed” (TJ 1971, 445). The person “feels guilty,” Rawls says, “because he has violated his sense of right and justice.” The person’s shame, on the other hand, responds to having failed “to achieve the good of self-command” being thereby “unworthy of” esteem, both that of “associates” and their own (TJ 1971, 445). Guilt is connected conceptually to the right and thereby to accountability and the authority to make claims and demands. To feel guilt is to feel as if one has culpably done wrong, that one is a fitting object of blame. Guilt is self-blame, the feeling through which we hold ourselves accountable for something for which we feel aptly held accountable. It is the connection to accountability that links guilt to the juridical or deontic concepts of obligation, right, rights, and justice. It is a conceptual truth, I claim, that actions are obligatory or wrong not to do if, and only if, omitting them would be blameworthy lacking excuse, where blame is a Strawsonian reactive attitude through we hold people accountable, others and ourselves, and bid for their recognition of our authority to do so (Strawson 1962; Darwall 2006, 2013a, 2013b). 6.4

Filling the Gap with the Second Aspect of Freedom: Persons as Self-Originating Sources of Valid Claims

This brings us to the Dewey Lectures’ second aspect of the freedom of persons, being self-originating sources of valid claims. That persons have a fundamental (second-personal) authority as sources of valid claims necessarily makes violating these claims grounds for guilt (rather than shame). Failing to respond adequately to others’ legitimate claims, and thereby to their second-personal authority to make them, is to fail to respect them as moral equals as morality and justice require. The deontic concepts of right, wrong, permissibility, rights, and justice all have a conceptual connection to accountability and the distinctive attitudes through which we hold ourselves and others accountable that

114

Stephen Darwall

no notion of the good has – whether of the desirable, the estimable, or the choiceworthy (being a good thing to do).4 Consider the idea of choiceworthiness, even of moral choiceworthiness. Being morally obligatory, or wrong not to do, is simply a different idea, having a different normative profile, than that of being what would be morally good or even morally best to do (Darwall 2016, 2019). To see this, consider two people who agree that moral reasons most highly favor a given action, so make it morally best to do, but disagree about whether that action is morally obligatory or wrong to omit. Imagine, for example, a disagreement between an act consequentialist and someone like Scheffler in The Rejection of Consequentialism (Scheffler 1982). Both reject agent-relative deontological constraints, and both could agree that morality always most recommends the optimific act. But they disagree about whether especially burdensome costs to the agent can render an action that would otherwise be morally obligatory morally permissible. The act consequentialist thinks that non-optimific acts are always morally wrong, whereas the Schefflerian holds that nonoptimific acts are sometimes not wrong when they are covered by an agent-relative prerogative or permission that defeats the charge of moral wrong when the personal costs are sufficiently high in relation to the impartial benefits. For such a disagreement to be so much as possible, the two must distinguish the deontic concept of moral obligation or right from the idea of moral choiceworthiness, from what it would be morally good, or even best, to choose. What the disagreement between the Schefflerian and the act-consequentialist is about, as I analyze it, is whether nonoptimific actions are always culpable when done without excuse. Considerations of the right occupy a different conceptual space than that occupied by any species of the good. This means that the way the “noumenal” conception of the person as independent of natural and social contingency operates within Rawls’s Kantian Interpretation in A Theory of Justice makes it impotent to vindicate principles chosen in the original position as principles of right rather than good. The idea that persons are self-originating sources of valid claims corrects this defect. To be a self-originating source of valid claims is, as I interpret it, to have an equal basic second-personal authority (Darwall 2006). Once we have this additional aspect of the freedom of persons in place, we can then treat the original position as a justification 4

I take good as well-being to be connected to the desirable through the idea of care or benevolent concern. What is for someone’s well-being is what is desirable for them for their sake, that is, out of care for them (Darwall 2002b).

On Being a “Self-Originating Source of Valid Claims”

115

for principles of justice and right in a way we could not with the materials of the Kantian Interpretation alone. Beginning with the idea that all persons have a shared basic authority to make claims and demands of one another, we then regard the parties to the original position as representatives of persons so conceived in order to work out more specifically what demands of justice and right persons actually have the authority to make of one another and of any political order that can justifiably govern in their name. And this grounds the chosen principles as principles of right and justice.5 Now it might be objected that Rawls was already assuming this additional idea from the beginning of A Theory of Justice when he said that the very idea of a public conception of justice involves a “common point of view from which [persons’] claims” on one another “to the advantages of social life” can be adjudicated (TJ 1971, 5).6 There is no doubt that Rawls was clear from the outset that the function of a public conception of justice is to adjudicate claims that have their “source” in persons in the thin sense that they are claims that people actually make. But so far this says nothing about anyone’s (de jure) authority to make these claims and any legitimacy their claims might inherit from that. To this, the Dewey Lectures adds the crucial thought that persons all share a default (de jure) authority to make claims and demands of one another and of any political order that can justifiably govern in their name. In saying that slavery denied enslaved persons this aspect of freedom, Rawls was clearly not saying that slavery made it literally impossible for the enslaved actually to make claims against their owners or planter society. Obviously, many did. Rather, he was saying that planter society recognized a de jure claiming authority of whites that it failed to recognize for Blacks.7 Of course, not even slavery could nullify Blacks’ equal de jure authority to make claims and demands. That is an essentially normative standing that nothing de facto can alter. Again, once we have the idea that persons are self-originating sources of valid claims, we can then treat the parties to the original position as representatives of persons so conceived. Doing so allows us to view principles that would be chosen from that perspective as principles that specify more determinately what people are entitled to expect from one another (principles of right) and from their political order (principles of justice). Without this aspect of the freedom of persons, however, we lack 5 6 7

For a more extended treatment, see Darwall (2006, 349–79). I am indebted to Paul Weithman for this objection. Even free Blacks were denied civil and political rights as a “badge of slavery” (Mason 2012).

116

Stephen Darwall

anything having the requisite conceptual shape to support fundamental principles of right.

6.5

The Priority of the Right to the Good

Rawls famously remarks in A Theory of Justice that “in justice as fairness, the concept of the right is prior to that of the good” (TJ 1999, 28). This is related to Rawls’s distinction between teleological and deontological theories of right and to his insistence that justice as fairness is a deontological rather than a teleological theory (TJ 1999, 22).8 The distinction between deontological and teleological theories is, it is important to see, within theories of right. The hallmark of teleological theories is that they begin with a theory of the good that is entirely independent of considerations of right and then attempt to ground a theory of right somehow in these. For any such theory, the good is prior to the right. In claiming that the right is prior to the good, Rawls is denying that the right can be grounded in an independently specifiable good. The point is not that the right is completely independent of the good. The argument from the original position must presuppose some account of the good to inform the parties’ choices. This is where Rawls’s “thin theory of the good” and his important theoretical idea of primary goods enter (TJ 1971, 395–99). Also, a notion of good, again Rawls’s primary goods, is necessarily involved in specifying the content of candidate principles of justice, for example, in defining benefits or advantages in the difference principle’s requirement that unequal advantages may attach to social positions only if this is to the greatest advantage of the worst-off group. The crucial point is that the relevant notion and measures of good are not given independently of, but are themselves shaped by, considerations of right. Considerations of right constrain those of the good, as Rawls understands it: “interests requiring the violation of justice have no value” (TJ 1999, 28). Indeed, the very idea of primary goods is one we need from the perspective of the theory of right. It is hard to see the point of the concept in theorizing about the good independently. Consider the theory of primary goods and the “highest-order interests” attributed to the parties to the original position in Kantian Constructivism. If we view the parties as representing persons as sources of valid claims, then we can see primary goods as what is good for persons so conceived, that is, as interests persons have as beings who lead their lives on terms of reciprocal recognition for their shared authority to make 8

Rawls takes this distinction from Frankena (1963).

On Being a “Self-Originating Source of Valid Claims”

117

claims of one another and hold themselves accountable to each other. Kantian Constructivism, as a contractualist theory of right and justice, conceives of persons as having two fundamental moral powers – the capacity for a sense of justice and the capacity “to form, to revise, and rationally to pursue a conception of the good” (CP, 312). These two moral powers work together within a Kantian Constructivist theory of right (justice) to reflect the two aspects of freedom that Rawls distinguishes in the Dewey Lectures – being a self-originating source of valid claims and independence, respectively. It is important that although Rawls also gives the parties a third, “higher-order” interest “in protecting and advancing their conception of the good as best they can, whatever it may be,” he is clear that this interest is “subordinate to the highest-order interests” (CP, 313). The reason is the priority of right. Independence differs crucially from rational pursuit of actual interests that are given naturally or socially, independently of and unconstrained by considerations of right. The function of independence as a highest-order interest having priority over the parties’ interest in pursuing their actual ends and conception of the good is to ensure the priority of the right over the good. “This priority of right of over good is characteristic of Kantian Constructivism” (CP, 319). Rawls puts the same point by speaking of the “strict priority of the Reasonable with respect to the Rational” (CP, 319). Again, the concepts of right and, we can now add, the Reasonable, carry conceptual implications of legitimate claims and demands, the authority to make them, and mutual accountability, that no concept of the good has, whether it be the desirable, welfare or well-being, the estimable, the choiceworthy, or even the morally choiceworthy.9 What is wrong is, as a conceptual matter, what it would be blameworthy to do, absent excuse, where blame is a Strawsonian reactive attitude through which we hold someone accountable, whether ourselves or others (Strawson 1962; Darwall 2006). Similarly, justice is not simply the “first virtue” of social institutions, in the sense of a foremost ideal. Rather, it concerns what we can justifiably claim or demand of institutions that govern in our name. It is only with the second aspect of the freedom of persons, that persons are self-originating sources of valid claims, that Rawls provides Kantian Constructivism with what is needed to ground the parties’ choices as principles of right having these conceptual implications. Persons’ shared equal basic claim-making authority thus fills a genuine gap in Rawls’s Kantian Interpretation of justice as fairness. Taken on 9

I argue that Rawls’s concept of the Reasonable should be understood in these secondpersonal terms in Darwall (2006, 23–24, 300–20). On the idea of the morally choiceworthy and moral reasons, see Darwall (2019).

118

Stephen Darwall

their own, the materials of the Kantian Interpretation in A Theory of Justice are impotent to support anything in the distinctive conceptual territory of the right, much less the priority of right. Rawls’s observation about Kant’s claim in the Kantian Interpretation that our failure to act on principles that most fully express our nature as free and rational is an occasion for shame rather than guilt can only ground a defect of good (the estimable), not a violation of right. 6.6

Does Kant Hold to the Priority of the Right?

In A Theory of Justice, Rawls says that “the priority of right is a central feature of Kant’s ethics” (TJ 1971, 31). In support, Rawls cites a section of The Critique of Practical Reason in which Kant says: The concept of good and evil must not be determined before the moral law (for which, as it would seem, this concept would have to be made the basis) but only … after it and by means of it (Kant 1996b, 5:63)10

Kant distinguishes the concepts of good (das Gute) and evil (das Böse) that figure in this passage from the “empirical” concepts of well-being (das Wohl) and ill-being (das Übel), respectively. The Critique of Practical Reason’s signature thesis is that “the objects of a practical reason,” including practical laws, cannot be derived from these latter empirical concepts (Kant 1996b, 5:59–60). A genuinely practical reason can only be determined “through freedom” and “formal laws of the will,” which are grounded in the “Fundamental Law of Pure Practical Reason”: “So act that the maxim of your will could always hold at the same time as a principle in a giving of universal law” (Kant 1996d, 57, 22, 30). If we take the priority of right to be that the right is prior to good as well-being, then this commits Kant to the priority of right. But when Kant says in the quoted passage that concepts of good and evil “must not be determined before the moral law,” he may seem committed also to the priority of right to das Gute. However, appearances here are misleading. The Fundamental Law of Pure Practical Reason is identical to what Kant calls the “moral law,” at least in its “universal law” formulation (FUL) (Kant 1991, 4:421).11 But it is important to be clear that Kant reserves “moral law” for the practical law as it is given to imperfect rational

10

11

Page references to Kant’s works are to page numbers of the Preussische Akademie edition. Rawls refers to 5:62–65. (See LHMP 230–32). In what follows, I draw from Darwall (forthcoming). “Act only in accordance with that maxim through which you can at the same time will that it become a universal law.”

On Being a “Self-Originating Source of Valid Claims”

119

agents in the form of imperatival commands. As he says, in the “corollary” to the Fundamental Law, “Pure reason is practical of itself alone and gives (to the human being) a universal law which we call the moral law” (Kant 1996b, 5:31). When Kant says in the passage above that “the concept of good and evil must not be determined before the moral law,” he means before the formal principle whose content is common to the moral law and the Fundamental Law of Pure Practical Reason. He does not mean that the concept of good (das Gute) must be determined subsequently to the moral law conceived in the distinctively obligating (imperatival) terms it is given to imperfect rational beings. Kant says, indeed, that “the only objects” of pure practical reason are “those of the good and the evil” (Kant 1996b, 5:58).12 “Imperatives,” he says, “say that to do or to omit something would be good, but they say it to a will that does not always do something just because it is represented to it that it would be good to do that thing” (Kant 1996b, 4:413). “Good or evil always signifies a reference to the will insofar as it is determined by the law of reason” (Kant 1996b, 60). The good that is the object of practical reason (das Gute) is what it would be good or best to will or do, that is, the choiceworthy. And obligating oughts are simply the imperatival terms in which the choiceworthy appears to less than fully rational agents. An imperative, that is, a rule indicated by an “ought,” expresses objective necessitation to the action and signifies that if reason completely determined the will the action would take place in accordance with this rule (Kant 1996d, 5:20).13

Whatever we would do, if we were fully rational, what it would be best to do, is what, as we actually are, we are obligated to do. Ironically, perhaps, it turns out that Kant probably does not hold that the right is prior to the good as he, Kant, understands the good (das Gute) in the passage that Rawls takes as evidence that he does. The question of

12

13

Cf. “Thus a perfectly good will would just as much stand under objective laws (of the good), but it could not be represented as thereby necessitated to actions that conform with laws, because it can of itself, according to its subjective constitution, be determined only be the representation of the good” (Kant 1996, 4:414). Note also: “They [imperatives] say that to do or to omit something would be good, but they say it to a will that does not always do something just because it is represented to it that it would be good to do that thing” (Kant 1996, 4:413). For Kant, obligation does not provide any object for the will, much less, one that is independent of the good. Rather, rational will invariably aims at what it is good or best to do, with obligation being the (imperatival) form thoughts about what it would be good to do take for an imperfect rational agent.

120

Stephen Darwall

pure practical reason is what is good or best to do in the sense of what there is most reason to do. That is independent of moral duty or obligation. The latter are simply the form that reason’s deliverances of what it is good or best to do take from the perspective of a will who can have motives to act otherwise. 6.7

Conclusion

Whatever Kant’s final verdict on the priority of right, there is no doubt about its fundamental role for Rawls. I have argued that it is only with the Dewey Lectures’ addition of the notion that persons are self-originating sources of valid claims that Rawls’s Kantian Contractualism is able to supplement the Kantian Interpretation so that it can adequately ground the parties’ choices in the original position as principles having the distinctive normative profile of the concept of right. If what I have just noted about Kant is along the right lines, it may follow that that this aspect of Kantian Contractualism goes beyond the practical philosophy of Kant himself. Rawls’s idea that persons are self-originating sources of valid claims thus does real work in his theory. Without that idea, we cannot say that persons can been treated unjustly and wronged. And with it, we can say that the dignity of persons includes a shared authority that makes us accountable to one another. In my view, this is the aspect of our moral freedom that most deeply underlies the idea of justice and, indeed, deontic morality more generally.

7

Moral Independence Revisited A Note on the Development of Rawls’s Thought from 1977–1980 and Beyond

Samuel Scheffler I will begin by addressing a question about the influence a short article of mine may have had on the development of John Rawls’s thought. The question arises because Rawls said explicitly that my article played a role in his thinking, yet readers have found it difficult to understand how or why it did so. The topic is of interest only insofar as it may help shed light on the evolution of Rawls’s thinking during the period between the publication of A Theory of Justice and the publication of Political Liberalism. That evolution, or rather the earliest stage of it, is the larger topic of this chapter. In the Introduction to Political Liberalism, Rawls includes the following acknowledgment: To Samuel Scheffler, who in the fall of 1977 sent me a short paper, “Moral Independence and the Original Position,” in which he argued that there was a serious conflict between the third part of my paper “The Independence of Moral Theory” (1975), which took up the relation between personal identity and moral theory, and my argument against utilitarianism in Theory. I remember this as the moment (I was on leave that year) when I started thinking about whether and how far the view of Theory needed to be recast. The decision to pursue this problem and not another topic led eventually to the Columbia lectures of 1980 and to the later essays elaborating the idea of political liberalism. (PL, XXXII–XXXIII)

Rawls says nothing further to explain how or why my essay influenced his thinking, and some readers have found its influence difficult to understand. For example, Paul Weithman writes: Scheffler’s is a very short piece to have motivated such significant changes in Rawls’s view. The reasons Rawls found the piece so provocative have remained somewhat obscure. The line of thought that Rawls followed from Scheffler’s article to PL is difficult to trace. (Weithman 2010, 36)

Weithman’s puzzlement is easy to understand. For one thing, the version of my essay that was eventually published is only seven pages long (Scheffler 1979), and although the pre-publication version I sent Rawls was a bit longer, it was still quite short. As Weithman suggests, it’s 121

122

Samuel Scheffler

difficult to understand how such a modest effort could have had a significant impact on Rawls’s work. But it’s not just the brevity of my essay that has created puzzlement. It’s also the fact that there is no evident connection between the content of my essay and the factors that Rawls identifies elsewhere in his Introduction as being responsible for the major differences between Theory and Political Liberalism. He says that to understand the nature and extent of those differences, “one must see them as arising from trying to resolve a serious problem internal to justice as fairness, namely from the fact that the account of stability in part III of Theory is not consistent with the view as a whole.” He adds, “I believe all differences are consequences of removing that inconsistency” (PL, XV–XVI). This makes it even more puzzling that Rawls should have attached significance to my essay, for stability is not so much as mentioned in that essay. If Political Liberalism was motivated by the attempt to remedy the deficiencies of the account of stability in A Theory of Justice, how could my seemingly unrelated essay have been what provoked Rawls’s second thoughts about Theory? Some commentators have answered this question by simply asserting that my essay was about stability. For example, Rex Martin says that “Scheffler’s paper argued that the account of stability developed in A Theory of Justice was in conflict with Rawls’s ‘The Independence of Moral Theory,’ section III” (Martin 2014, 281). But my paper argued no such thing. Other commentators have suggested, in effect, not that my paper was about stability but that Rawls thought it was. Thus, Paul Weithman, after describing a kind of case that, in his view, may have led Rawls to become dissatisfied with the account of congruence in Theory of Justice, writes: “this may be the kind of case that convinced Rawls that the Kantian Congruence Argument failed because it is the kind of case that is suggested by the article of Samuel Scheffler’s that Rawls credits with leading him to rethink his early treatment of stability” (Weithman 2010, 259). But Rawls does not credit my article with leading him to rethink his early treatment of stability. He credits it only with leading him to think – or, more accurately, with providing the occasion for his starting to think – about whether and how far the view in A Theory of Justice needed to be recast. He says nothing about stability. But if my article is not about stability and Rawls did not think it was, and if Political Liberalism was motivated entirely by the attempt to correct the account of stability in A Theory of Justice, then we are left with the question of why my article was of any interest to him at all. Why did it lead him to rethink anything? On one level, this is less mysterious than it may seem. I sent Rawls my essay in 1977. Political Liberalism was

Moral Independence Revisited

123

published sixteen years later, in 1993. Over the course of those sixteen years, he developed his view gradually, trying out many different ideas in many different lectures and essays, each of which went through numerous versions and drafts. Some of those ideas eventually found their way into Political Liberalism, often in modified form, while others were abandoned altogether. The new concern with stability emerged fully only in 1987, with the publication of “The Idea of an Overlapping Consensus.”1 And by Rawls’s own account in the Introduction to Political Liberalism, he took himself to have reached a clear understanding of political liberalism only in the last few years prior to the book’s publication. So he traveled a long road between 1977 and 1993. Any influence my own essay may have had was presumably an influence on some of his first steps along that road, and there is no reason to expect it to be directly legible in the book he produced sixteen years later. After all, a conversation with someone may have helped persuade you that you should leave Maine without doing anything to explain why you ended up, sixteen years later, in Los Angeles rather than in Nashville or Seattle. Yet we may still wonder what effect the essay had on his thinking at the time he read it. And determining this may help us to understand the early stages of the odyssey that eventually led him to Political Liberalism. I don’t propose to give an account of that odyssey as a whole. But what I can do is to provide some context for his remarks about my essay and, in the process, to remind people of what was on his mind in the late 1970s. This may help combat a certain tendency to anachronism: a tendency to read the doctrines of Political Liberalism back into the writings from that earlier period. My essay was written in response to Rawls’s essay “The Independence of Moral Theory,” which was published in 1975 and had been his Presidential Address to the Eastern Division of the APA in December, 1974 (“IMT” 1975).2 In that essay, Rawls draws a distinction between moral philosophy and moral theory. Moral theory is the comparative study of substantive moral conceptions and is one of the main parts of moral philosophy, which is a more general branch of inquiry that also includes what we think of as “metaethics.” Rawls’s thesis is that moral theory is substantially independent of other parts of philosophy, such as epistemology, the theory of meaning, and the philosophy of mind.

1

2

Although it was foreshadowed by some remarks at the very end of “Justice as Fairness: Political not Metaphysical,” published in 1985 (CP, 388–414). See “The Idea of an Overlapping Consensus” (1987). Editor’s note: page references to “IMT” will be to Rawls’s Collected Papers, where that essay is reprinted.

124

Samuel Scheffler

Indeed, the metaethical problems that are most closely connected to those areas of philosophy often depend on moral theory, on a proper understanding of the structure and content of substantive moral conceptions. Moral theory has its own problems and subject matter and, to a great extent, progress in other parts of moral philosophy depends on progress in moral theory, rather than the reverse. For the time being, then, moral theory should be pursued independently of questions in other areas of philosophy. Moral theorists should devote themselves to achieving a proper understanding of substantive views like utilitarianism, perfectionism, and Kantian ethics, without awaiting further progress in epistemology, the theory of meaning or the philosophy of mind. Rawls’s essay is divided into three main parts: the first arguing for the independence of moral theory from epistemology, the second for its independence from the theory of meaning, and the third for its independence from the philosophy of mind – especially the theory of personal identity. My essay was a response to that third part, which was itself a response to the arguments of Derek Parfit in “Later Selves and Moral Principles,” published in 1973. As Rawls interprets him, Parfit had argued that a conception of personal identity as consisting in certain empirical continuities and connections favors a utilitarian over a Kantian moral conception. In reply, Rawls argues that the “broadly empirical” (CP, 300) theses about personal identity established by the philosophy of mind underdetermine the choice among substantive moral conceptions. Although those theses represent constraints that any moral conception must satisfy, all of the leading conceptions do satisfy them. To be sure, these conceptions differ from one another in the complete criteria of personal identity they employ, but these differences derive from their differing conceptions and ideals of the person and not from disagreements about issues in the philosophy of mind. The feasibility of the different criteria depends on whether they can be realized in practice, and that is a matter of psychological and social theory, which is not addressed by the philosophy of mind. Admittedly, he says, “the utilitarian conception has less need for a criterion of identity than a Kantian view; or perhaps better, it can get by with a weaker criterion of identity” (CP, 299). By contrast, “a Kantian view is more dependent on personal identities; it relies, so to speak, on a stronger criterion” (CP, 299). That is largely because it “must conceive of identities as stretching over much longer intervals” (CP, 299). Nevertheless, he maintains, if a “Kantian well-ordered society is possible and workable,” then its members will satisfy the Kantian criterion of personal identity; they will “lead lives such that the necessary identifications can be made” (CP, 300). For “there is no degree of connectedness

Moral Independence Revisited

125

that is natural or fixed; the actual continuities and sense of purpose in people’s lives is always relative to the socially achieved moral conception” (CP, 301). This means the essential question is whether “the wellordered society corresponding to a moral conception generates in its members the necessary continuities and sense of purpose to maintain itself” (CP, 301). Rawls’s position is that not only the Kantian view but all of the leading moral conceptions, with their different criteria of identity, are probably feasible by this standard, so the theory of personal identity provides no grounds for choosing among them. In my essay, I maintained that this line of argument conflicts with Rawls’s arguments against utilitarianism in A Theory of Justice. Toward the beginning of Theory, Rawls famously says that “utilitarianism does not take seriously the distinction between persons” (TJ 1971, 27). In consequence, he argues, if we assume 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 1971, 29). That is because “the correct regulative principle for anything depends on the nature of that thing” (TJ 1971, 29), and utilitarianism does not respect the nature of persons as distinct beings with separate systems of ends. This informal, preliminary argument against utilitarianism is borne out by his subsequent, more elaborate argument that the parties in the original position would reject a utilitarian conception. For that argument turns on the point that the parties know that each of them has a rational plan of life (a “separate system of ends”) and a long-term interest in seeking to carry out that plan. They reject utilitarianism largely because they are convinced that it does not adequately protect those interests. But if what Rawls says in “The Independence of Moral Theory” is correct, there is a problem with this argument. The parties in the original position know the laws of human psychology, so they are aware that no degree of intrapersonal connectedness is natural or fixed and that the degree of connectedness that obtains is always relative to the socially achieved moral conception. Yet they don’t know what the socially achieved moral conception in their society is. In particular, they don’t know whether they live in a “Kantian society” or a “utilitarian society” (CP, 300): a society that has or has not generated the continuities and connections that a Kantian view requires. In consequence, they don’t know whether they have the longterm interests in carrying out a rational plan of life on which the argument against utilitarianism depends. More generally, if “the correct regulative principle for anything depends on the nature of that thing,” and if human nature is compatible with a person’s being either a “Kantian person” or a “utilitarian person,” and if, furthermore, which of these one is depends on

126

Samuel Scheffler

the socially realized moral conception in one’s society, then it is no argument against choosing utilitarian principles to regulate the institutions of one’s society that those principles don’t protect the interests people would have in a Kantian society. My essay concludes with a brief sketch of a couple of ways Rawls might try to resolve the conflict to which I had called attention. The first option, I suggested, would be to try to show that, contrary to what Parfit had argued, there are theory-independent facts about personal identity and the empirical connections within lives that support Rawls’s principles over utilitarianism. This would be to abandon the claim that moral theory is independent of the philosophy of mind. The second option would be to reaffirm that claim; to concede, accordingly, that his arguments against utilitarianism in A Theory of Justice cannot appeal to a theory-independent conception of the person; but to argue instead that the original position models an ideal of the person which “has great intuitive appeal and coheres with many of our deepest convictions” (Scheffler 1979, 402), and that what the arguments in A Theory of Justice do is to demonstrate why a commitment to that ideal is incompatible with utilitarianism. When I sent Rawls my essay, he replied in a typewritten letter of December 9, 1977, sent from The Institute for Advanced Study, where he was spending the term on leave. His letter reads in part: I want to thank you for sending me the copy of your paper. There is no doubt that you have pointed out a real difficulty. The arguments in my book against utilitarianism and the reply to Parfit in “Independence” are inconsistent, or appear so. I am particularly grateful to you for pointing this out, for as strange as it may appear to you, I was not aware of the inconsistency. Apparently I was so anxious to find some way in which moral theory is independent from the philosophy of mind that I overlooked its implications altogether. That paper of mine I never liked at all. Now that you’ve shown me the inconsistency, it at least has the merit of forcing some sort of reconciliation. Something has to go; I’m not sure now which …. Many, many thanks. I’ll have to think about it.”

The letter also contained a handwritten post-script, evidently written the next morning, which reads: I’ve thought about this some overnight and several possibilities along the lines you suggest seem to be worth pursuing. But my thoughts are too vague at the moment. I’ve come to realize recently the importance of the conception of the person for how the OP is described and made some changes in the German edition to this effect. Your paper greatly shows the necessity for working all this out, if possible.

Apart from Rawls’s intellectual generosity, two things about this letter stand out, because they provide an indication of what was on his mind in late 1977. The first is his emphasis on how eager he was to establish the

Moral Independence Revisited

127

independence of moral theory from the philosophy of mind. The second is his reference to his recent realization of the importance of the conception of the person. Let me comment on each of these in turn. First, it is clear that Rawls was alarmed by the prospect, which presented itself vividly to him as a result of Parfit’s striking and influential essay, of becoming mired in debates about personal identity. He was convinced that his project did not depend on claims about personal identity, and he was looking for principled ways to explain and defend that independence. More generally, he was trying to make methodological room for a concern with substantive conceptions of justice that was not unduly hostage to developments in other areas of philosophy. The distinction between moral philosophy and moral theory, which was crucial to his attempt in “IMT” to make such room, eventually disappeared from his work, as I will indicate below. But the underlying concern remained a central feature of his thought, which manifested itself in different ways at other times. In “Justice as Fairness: Political not Metaphysical” (1985) and “The Idea of an Overlapping Consensus” (1987), it manifested itself in his advocacy of “the method of avoidance,” understood as the attempt “to leave aside philosophical controversies whenever possible, and look for ways to avoid philosophy’s longstanding problems.”3 And in Political Liberalism, of course, it is reflected in the crucial claim that a political conception of justice does not depend on any single comprehensive moral or philosophical doctrine but may instead become the focus of an overlapping consensus among proponents of the leading comprehensive doctrines likely to be represented in a modern democratic society. As for personal identity, it disappears from his writing almost entirely after “IMT.” The clearest exception I am aware of is in a footnote in PL, where, after having distinguished between the “public or institutional” (PL, 31) identity of citizens and their “nonpublic or moral” identity (PL, 32n), he distinguishes both of these notions of identity from the idea of “personal identity” as traditionally understood by philosophers.4,5 The philosophical problem of personal identity, he says, “raises profound questions on which past and current philosophical views widely differ and surely will continue to differ. For this reason it is important to try to develop a political conception of justice that avoids this problem as far as possible” (PL, 32n.)

3 4 5

“Justice as Fairness: Political not Metaphysical,” 230–31. This footnote in PL is an expanded version of footnote 24 in “Justice as Fairness: Political not Metaphysical,” 242. But see also “Kantian Constructivism” (1980, 571).

128

Samuel Scheffler

This brings me to Rawls’s comments about the conception of the person. His reference to his having recently come to realize “the importance of the conception of the person for how the OP is described” is significant for a couple of reasons. In saying that my essay “showed the necessity for working all this out,” Rawls makes clear that he read my essay through the lens of a preexisting concern about the role played by ideas of the person in his theory. The essay made an impression on him, to whatever extent it did, because it reinforced a conviction he had already reached that he needed to think further about the role of the conception of the person and its relation to the original position. He does not explain in his letter why exactly he had reached this conclusion. It is worth noting, however, that “The Independence of Moral Theory” was just one of three essays that Rawls published in 1975. The other two were “A Kantian Conception of Equality” (CP, 254–66), which discussed the sense in which his theory is Kantian, and “Fairness to Goodness” (CP, 267–85), which addressed one of the most prominent criticisms of A Theory of Justice in the years immediately following its publication, namely, that its reliance on primary goods introduces a bias in favor of individualistic conceptions of the good.6 Ideas of the person play a role not only in “Independence of Moral Theory” but in the other two essays as well. So it is clear that Rawls had been thinking seriously about the conception of the person and its role in his theory during the years immediately prior to 1977. Two further points deserve mention. First, the two things that Rawls emphasizes in his letter, namely, his desire to secure the independence of moral theory from the philosophy of mind and his growing sense of the importance of the conception of the person in his theory, are in some tension with one another. For the more one emphasizes the role of a particular conception of the person in his theory, the more it may seem that the theory must depend on wider philosophical considerations about the nature of persons and personhood. One way of reading the argument in the third part of “Independence of Moral Theory” is as an attempt to dissolve the tension. If, as I argued and he evidently agreed, that attempt was unsuccessful, then he needed to find some other way of dissolving it. Second, the fact that Rawls had concluded by 1977 that he needed to rethink the role of the conception of the person provides support for his insistence in later years that, contrary to what many people believed then and what some believe even now, the changes in his view that are reflected in his later essays and in Political Liberalism itself, especially

6

See Teitelman (1972); Nagel (1973); and Schwartz (1973).

Moral Independence Revisited

129

with regard to the conception of the person, were not made in response to communitarian criticism of A Theory of Justice. One common assumption is that his eventual interpretation of the conception of the person as a “political conception” was a change made in response to Michael Sandel’s criticism in Liberalism and the Limits of Justice (Sandel 1982). Sandel had argued that the original position modeled a metaphysical conception of the self as “unencumbered” and shorn of any defining commitments and ties. In the Introduction to Political Liberalism, however, Rawls writes, “the changes in the later essays are sometimes said to be replies to criticisms raised by communitarians and others. I don’t believe there is a basis for saying this” (PL, XVII). Although he certainly wished to emphasize, in response to what he regarded as Sandel’s misreading of his view,7 the contrast between his conception of the person and a metaphysical conception like the one Sandel attributed to him, his 1977 letter makes clear that his desire to rethink the role of the conception of the person in his theory significantly predated the publication of Sandel’s book in 1982 and had nothing to do with the issues Sandel raised.8 If there was anyone whose views worried him back in 1977, it was Parfit and not Sandel. Indeed, I don’t believe he ever felt worried by Sandel’s metaphysical interpretation of the original position. He simply thought it was mistaken and that the mistake needed to be corrected. The idea of a political conception of the person, meanwhile, was his eventual solution to a concern dating back to the 1970s about how best to characterize the role of the conception of the person in his theory generally and in relation to the original position in particular. If we take Rawls at his word and assume that it was in late 1977 that he first began thinking about whether and how far the view defended in A Theory of Justice needed to be recast, then we may look at the work he did in the immediately succeeding years for insight into his earliest thoughts about how to recast it. During that period, Rawls gave a series of prominent lectures at various universities, including the Kant Lectures at Stanford in the spring of 1978, the Howison Lecture at Berkeley in the spring of 1979, the Dewey Lectures at Columbia in the spring of 1980, and the Tanner Lectures at Michigan in the spring of 1981.9 Both the 7 8

9

See “Political not Metaphysical” (239n) and Political Liberalism (27n). This is not to deny that Sandel’s book made the possibility of a metaphysical interpretation of the original position more salient to Rawls and so explained his eagerness to repudiate it, as reflected in the very title “Justice as Fairness: Political not Metaphysical.” I am grateful to Joshua Cohen for discussion of this point. Full disclosure: I attended the Howison Lecture, the Tanner Lectures, and some of the Kant Lectures; I was a commentator (along with Brian Barry and Anthony Kronman) for the Tanner Lectures; I discussed drafts of all four sets of lectures with Rawls on various

130

Samuel Scheffler

Dewey Lectures and the Tanner Lectures were published, in revised (and, in the case of the Tanner Lectures, expanded) form, not long after their delivery.10 The Kant and Howison Lectures were never published as such, although, as I will explain below, a good deal of material from those lectures was incorporated in one way or another into subsequent publications. For some years, Rawls evidently envisioned a book-length manuscript comprising revised versions of some or all of these lectures. By the summer of 1979, for example, he was circulating a manuscript comprising five “lectures” (hereafter referred to as “the 1979 manuscript”) of which the first was described as a revised version of the Howison Lecture and the second, third, and fourth were described as later versions of the Kant Lectures.11 His original Michigan Tanner Lectures on “The Basic Liberties and Their Priority” were described as Lecture VI in a draft that he circulated around the time of its delivery. When he eventually published Political Liberalism (in the Dewey Lectures book series of the Columbia University Press), he preserved the division of the book into lectures rather than chapters, but the content was quite different. He says in the Introduction that the first three lectures were heavily revised versions of the Dewey Lectures as they had appeared in the Journal of Philosophy and that he had originally planned to publish them with three additional lectures: “The Basic Structure as Subject” (first published in 1977, revised version published in 1978), “The Basic Liberties and their Priority,” and “Social Unity and Primary Goods” (published in 1982). However, he adds, he eventually decided that the resulting manuscript was insufficiently unified, and he subsequently wrote and incorporated into Political Liberalism four additional essays (or lectures), written between 1985 and 1989, although the previously published versions of “The Basic Structure as Subject” and “The Basic Liberties and their Priority” were also included as Lectures VII and VIII.12

10

11

12

occasions; and some of those discussions are acknowledged at a couple of points in the unpublished versions of the Howison Lecture and the published versions of the Dewey and Tanner Lectures. The Dewey Lectures were published as “KCMT” (1980). The published version of the Tanner Lectures originally appeared in The Tanner Lectures on Human Values, Volume III, ed. Sterling McMurrin (Salt Lake City: University of Utah Press, 1982), 3–87. They were subsequently included as Lecture VIII in Political Liberalism (PL, 289–371.) Rawls had previously given a Tanner Lecture, apparently with the same title, at Brasenose College, Oxford, in 1978. A copy of this manuscript, numbered as indicated in the text above, is apparently on file in a folder titled “Lectures by Rawls, 1979” in the Rawls Archive at Harvard University (Box 26, folder 1). I also have a copy on file. In the case of “The Basic Subject as Structure,” the published version that was included in Political Liberalism was the 1978 version, which originally appeared in Goldman and Kim (1978, 47–71).

Moral Independence Revisited

131

The upshot is that, although the idea of a book comprising a reworked series of lectures is finally realized in Political Liberalism, none of the lectures that are included there, except “The Basic Structure as Subject,” had even been written at the time he seems first to have conceived of a book with that structure (no later than 1979). The Kant and Howison Lectures, which had featured prominently in the 1979 manuscript, are not included as self-standing lectures, although Rawls had earlier stated that material from the Howison Lecture had been incorporated into Lectures I and III of the Dewey Lectures (see KCMT, 515), which were then in turn revised extensively before inclusion in Political Liberalism (see PL, XIII) Similarly, material from the portions of the 1979 manuscript that descend from the Kant lectures can be found, often much revised, throughout Political Liberalism, as well as in other publications, such as “Social Unity and Primary Goods,” which is the successor to Lecture IV of the 1979 manuscript (there entitled “Responsibility for Ends”).13 In addition, Lecture III of the 1979 manuscript, which descends from the second Kant lecture, itself incorporates material from “The Basic Structure as Subject.” Because the Kant Lectures, the Howison Lecture, and the 1979 manuscript were never published but were instead succeeded or superseded by other published writings, they may afford some insight that those other writings do not into the earliest phase of Rawls’s thinking about how to recast A Theory of Justice. I will focus on the original Howison Lecture, which was delivered in Berkeley on April 30, 1979, and was entitled “Constructivist Moral Conceptions.”14 It is one of the first places where Rawls argues explicitly and at length that his theory should be thought of as a form of constructivism. He frames the discussion in the lecture with reference to a modified version of the distinction between moral philosophy and moral theory that had loomed large in “IMT.” In “CMC,” he says again that moral theory is the comparative study of substantive moral conceptions, such as utilitarianism, perfectionism, and intuitionism, but he adds that a moral conception has three parts: a scheme of principles and precepts that defines its content, an interpretation of the social role of morality, and – of crucial importance, he says – a conception of the person. Moral philosophy, meanwhile, has “its focus on the problem of justification” (“CMC,” 2). Although it “comprehends” moral theory, it “has as its leading question the following: which moral conception, if any, is true, or else the most reasonable; if indeed either truth or reasonableness 13 14

“Social Unity and Primary Goods” was published in Sen and Williams (1982, 159–85). Copy on file with the author. Abbreviated hereafter as “CMC,” with page references given in the text.

132

Samuel Scheffler

applies to these conceptions, which is itself a question.” This, he contends, makes moral theory a preliminary but necessary investigation for moral philosophy. Against this background, Rawls describes his own moral conception, justice as fairness, as a constructivist conception, and he says that what is crucial for and distinctive about such a view is the way its conception of the person relates to the other two elements of the view: its content and its interpretation of the social role of morality. A constructivist conception regards persons as constructing moral principles and, in this process of construction, their conception of themselves plays an important and consequential role. Similarly, the conception of the person shapes the constructivist interpretation of morality’s social role. The rest of the lecture is devoted to fleshing out the characterization of justice as fairness as a constructivist view. There is a great deal that is of interest in this lecture. Looking backward, we may note some continuities with and departures from “IMT.” Looking forward, we may note the introduction of new ideas that anticipate future work. There are also some false starts: ideas that feature here and in the writings of the next few years but are retracted or renounced by the time of Political Liberalism, if not sooner. Consider first the backward-looking view. Although Rawls continues to say, in this lecture, that moral theory involves the study of substantive moral conceptions, as he had in “IMT,” there is no longer an attempt to establish that it is independent of other areas of philosophy. And having set aside the argument for independence, Rawls makes no mention at all of personal identity (he evidently felt no temptation to argue that facts about identity actually support justice as fairness, as suggested by my “first option”). On the other hand, the conception of the person, which played an important role in “IMT,” receives even greater prominence here, because of the central role it plays in a constructivist view. And Rawls mentions without quite endorsing the idea that his own conception of the person may have some special status in comparison with other such conceptions (something similar was proposed as my “second option,” but Rawls characterizes the relevant form of “special status” in a distinctive way). He elaborates on this point in the following passage: In justice as fairness … we assume that we live in a democratic society under modern conditions, and that we want to find a conception of right and justice for the more fundamental questions that arise in such a society. The intuitive idea is to draw upon the conception of the person that seems to be held by citizens in a democratic regime or the conception it seems that they would hold on reflection, at least so far as this conception is thought to be relevant in matters of political and social justice. The hope is that although the history of democratic thought

Moral Independence Revisited

133

shows long-standing disagreement over which principles of liberty and equality are correct, or most reasonable, there may prove to be no truly basic dispute regarding the appropriate conception of the person. Perhaps there are only a few viable notions of the person sufficiently general to be part of the formulation of a moral doctrine, and all but one may be excluded by other conditions essential to a democratic view. (“CMC”, 22)

This is a remarkable passage, for reasons I will discuss shortly. But let me first survey some other points of comparison and contrast with his earlier and later writings. Staying with the backward-looking view, one element of “IMT” that carries over to the Howison Lecture, as I have noted, is the distinction between moral theory and moral philosophy. However, the characterization of the substantive moral conceptions studied by moral theory as having three main elements, and of moral philosophy as having the question of justification as its focus, are new features that represent departures from the earlier essay. Moreover, the introduction of constructivism makes trouble for the moral theory/moral philosophy distinction, because constructivist moral conceptions, although substantive, are themselves said to include a distinctive view about justification. In particular, they hold that “the problem of justification is in the first instance a practical social matter, not an epistemological problem” (“CMC,” 3). This means that constructivism is a position that appears to belong both to moral theory and to moral philosophy. His recognition of this difficulty led Rawls to rework his account of the moral theory/moral philosophy distinction yet again, this time in the revised version of the Howison Lecture dated July 1979 that was included as Lecture I in the 1979 manuscript.15 But he abandoned the new version of the distinction by the time of the Dewey Lectures, and the distinction itself played a 15

In the revised version of the lecture, Rawls says that the moral conceptions studied by moral theory have four parts rather than three: in addition to its content, interpretation of the social role of morality, and conception of the person, a moral conception also has an account of justification. Moral philosophy, meanwhile, is said to be concerned with the question of justification in a different way. In a letter of July 19, 1979, he explained the new version of the distinction as follows: I finally decided to say that each moral conception normally has some account about how it is to be justified, at least in its own terms. Moral theory now includes, as it were in a descriptive manner, what these various accounts of justification are. Moral philosophy begins, when the problem of justification is systematically, & I think most fruitfully pursued, from moral theory in this more extended sense. Moral philosophy surveys the narrow views of justification in order to find which one is superior, all things considered. It is still a broader and more inclusive inquiry, since it connects with other parts of philosophy & much else, e.g. social theory is often relevant. And even when it uses the results of moral theory, it has a different focus – a different question to answer, & that systematically.

134

Samuel Scheffler

relatively minor role in those lectures. As far as I know, it largely disappeared after that. My conjecture is that, once it was no longer being used as the basis for a claim about the independence of moral theory from other parts of philosophy, it was a dispensable feature of his view, and one that was difficult to preserve once constructivism, which seemed to straddle the distinction, entered the scene. The ideas of independence that survived or resurfaced at other periods were, as I have already indicated and as I will reiterate below, formulated in other terms. Consider now the forward-looking perspective. The characterization of his view as constructivist and his emphasis on the role of the conception of the person remained important features of Rawls’s position during the entire period up to and including Political Liberalism. Yet his understanding of these ideas changed significantly over time. The constructivism Rawls articulates in this lecture developed into the “Kantian constructivism” of the Dewey Lectures, and only later was it was it transformed into a more limited “political constructivism,” which was contrasted with Kant’s more general moral constructivism. Similarly, the conception of the person articulated here is a feature of a moral conception generally. It is not yet the more narrowly circumscribed “political conception of the person” of Political Liberalism. Much the same is true of another element he introduces in the Howison Lecture, namely, the contrast between constructivism and intuitionism. As I have noted, Rawls says that constructivism interprets the problem of justification as “a practical social matter and not as a problem of knowledge” (“CMC,” 21). On a constructivist view, the “question of justification becomes: how might the members of a society reach an understanding that among the familiar moral conceptions, or natural variations thereof, there is one that is most reasonable for them in virtue of how they conceive of their persons. The search for the most reasonable conception for us as persons of a certain kind replaces the search for moral truth” (“CMC,” 3). The “primacy of the practical” (“CMC,” 3) that is thus a feature of constructivism is contrasted at some length in the lecture with epistemological intuitionism. This contrast is an important theme of Rawls’s later work, beginning with the Dewey Lectures. But it too is substantially reinterpreted by the time of PL, where the role of “the practical” is largely superseded by the emphasis on “the political,” and where the focal point of the contrast between constructivism and intuitionism is not the notion of justification but rather the notion of objectivity. But precisely because these novel elements of the Howison Lecture underwent such extensive revision before attaining final form in Political Liberalism, the passage I quoted four paragraphs ago, which says that

Moral Independence Revisited

135

justice as fairness is aimed at finding principles of justice for a modern, democratic regime, and that its conception of the person is one that is or would be affirmed by citizens in such a regime, is all the more noteworthy, since it strikingly anticipates the political turn that his work was to take in later years. So too Rawls’s observation in that passage that “the history of democratic thought shows long-standing disagreement over which principles of liberty and equality are correct,” and his statement that the aim of justice as fairness is to find some basis for agreement in the face of this long-standing disagreement. All these comments, which were toned down quite a bit in the corresponding passage in the Dewey Lectures,16 would have been completely at home in Political Liberalism. This means that, in the very same lecture where Rawls begins to articulate the philosophically robust moral constructivism from which he later sought to distance himself, he also provides clear anticipations of the political liberalism with which he was eventually to replace it. There is, however, one final passage of the Howison Lecture that seems to point in the opposite direction. In the course of explaining how a constructivist conception interprets the social role of morality, Rawls discusses the publicity condition (one of the “formal conditions” imposed on conceptions of justice in A Theory of Justice) and distinguishes several different “levels” of publicity. What is characteristic of a constructivist conception is that it insists on “the complete publicity condition” (“CMC,” 12), which means that: the full justification of a moral conception, as it would be presented in its own terms, must also be public. One way to apply this complete condition is to work out the consequences of a moral conception on the supposition that it is the public moral constitution of a society in which its full philosophical justification is public knowledge and accepted by everyone. Such a society I called the wellordered society paired with that conception; the parallel in Kant is the Realm of Ends. (“CMC,” 12)

This passage takes on special significance in light of Rawls’s subsequent work. When, in the Introduction to Political Liberalism, Rawls says that Theory’s account of stability was flawed, he adds that the ultimate source of problem lay in Theory’s “unrealistic” account of the wellordered society paired with a conception of justice. In such a society, of course, all citizens are assumed to accept the conception’s principles of 16

The corresponding passage appears in “Kantian Constructivism in Moral Theory: The Dewey Lectures 1980,” 569–70. All but one of "CMC"’s references to a democratic society or democratic thought have been eliminated from this passage, although such references are scattered throughout the Dewey Lectures, where they stand in some tension with the “Kantian constructivism” that is being defended.

136

Samuel Scheffler

justice, and furthermore, Rawls says in his Introduction, they do so on the basis of what in Political Liberalism he calls a “comprehensive philosophical doctrine.” This is the assumption that he now describes as unrealistic. Summarizing the problematic A Theory of Justice account, he writes: An essential feature of a well-ordered society associated with justice as fairness is that all its citizens endorse this conception on the basis of what I now call a comprehensive philosophical doctrine. They accept, as rooted in this doctrine, its two principles of justice. Similarly, in the well-ordered society associated with utilitarianism citizens generally endorse that view as a comprehensive philosophical doctrine and they accept the principle of utility on that basis. Although the distinction between a political conception of justice and a comprehensive philosophical doctrine is not discussed in Theory, once the question is raised, it is clear, I think, that the text regards justice as fairness and utilitarianism as comprehensive, or partially comprehensive, doctrines. (PL, XVI)

Despite what Rawls says here, there may be room to wonder whether it really is clear in A Theory of Justice that citizens’ acceptance of the two principles of justice is based on their shared affirmation of a comprehensive philosophical doctrine. By contrast, the passage from “CMC” leaves little room for doubt on that score. It asserts that, in a well-ordered society, “the full philosophical justification” of the “moral conception” (that is, of the principles of justice) must be “public knowledge and accepted by everyone.” If, as seems plausible, Rawls is thinking of the “full philosophical justification” of a conception of justice as depending on (what he later called) a comprehensive philosophical doctrine, then this is an instance in which his first steps toward revising the view in A Theory of Justice actually lead him further away from his eventual destination. In his characterization of the complete publicity condition, he emphasizes the very feature that he later identified as the ultimate source of the flaw that led him eventually to the views expressed in Political Liberalism. Stepping back from these specific points of comparison and contrast, let me make a more general observation about the relation between “CMC,” the work that preceded it, and the work that came later. Earlier I noted the existence of a tension, dating back to the 1970s, between Rawls’s desire to secure the independence of moral theory from wider philosophical issues and controversies and his strong sense of the importance of the conception of the person in his theory. “The Independence of Moral Theory” did not succeed in resolving that tension, but the constructivism introduced in “CMC” and developed more fully in the Dewey Lectures may appear to do better. On the constructivist view, moral principles are constructed on the basis of a conception

Moral Independence Revisited

137

of the person, for specific practical purposes, and are not construed as representing or providing knowledge of an independent realm of facts. This may seem to insulate them from wider philosophical debates, such as debates about truth, objectivity, knowledge, and justification. The difficulty, however, is that the robust moral constructivism that Rawls developed during this phase of his thought is itself a philosophical position that takes a stand on issues of moral truth, objectivity, knowledge, and justification. So the tension remains. It is only with the political turn in Rawls’s thought, as most fully developed in Political Liberalism, that the tension is finally resolved – at least to his own satisfaction. The views expressed in that book – especially the distinction between the political and the philosophical and the replacement of Kantian moral constructivism with a narrower political constructivism – at last give Rawls what he wants. His conception of the person is presented now as a “political conception,” which specifies how persons are to be conceptualized for legal, political, and institutional purposes, and is said to be implicit in the public political culture of a democratic society. So understood, it retains a clear and central role in his theory and in the construction of the principles that define his substantive conception of justice. Yet that substantive conception, recast as a political conception designed for a modern democratic society, is insulated from a wide range of traditional philosophical debates and controversies, including debates about personal identity and the metaphysics of the person. And this despite its reliance on the political conception of the person. That is because, whereas the political conceptions of justice and the person have a carefully circumscribed social role, and appeal to ideas implicit in the public political culture of a democratic society, the philosophical debates and controversies belong to the various “comprehensive philosophical doctrines” that are bound to flourish in a free society. Their resolution, accordingly, is something that a liberal theory of justice cannot be expected – and should not attempt – to provide. Moreover, he insists, the distinction between a political conception of justice, on the one hand, and the various comprehensive philosophical doctrines, on the other, is not itself a philosophical distinction. As he puts it in the Introduction to Political Liberalism: The dualism in political liberalism between the point of view of the political conception and the many points of view of comprehensive doctrines is not a dualism originating in philosophy. Rather, it originates in the special nature of democratic political culture as marked by reasonable pluralism. (PL, xxi)

Thus was “the independence of moral theory” – or something like it – eventually achieved. And it was achieved while at the same time insisting

138

Samuel Scheffler

on the indispensable role of the (political) conception of the person, thus dissolving the tension dating back to the 1970s between two of Rawls’s important aspirations: between his desire to represent his substantive theory as independent of wider philosophical controversies and his conviction of the importance to his theory of the conception of the person. I have discussed only one short period in the evolution of Rawls’s thought between 1977 and 1993. Moreover, Rawls was a tireless worker and an active philosophical correspondent, who worked out his ideas by producing multiple drafts of lectures and essays and discussing those drafts in person or by mail with a wide variety of students, colleagues, and friends. He normally tried out many versions of his ideas on many different philosophical interlocutors before arriving at formulations he regarded as satisfactory (at least until he stopped regarding them as satisfactory). So one shouldn’t place too much weight on any particular draft or letter or conversation. Still, the texts I have been discussing provide a snapshot of Rawls’s thinking during a period when, by his own account, he was just beginning to consider how the position he had defended in A Theory of Justice might need to be recast. They help to identify some of the problems that were troubling him during that period and some of the aspirations that informed his thinking about how to address those problems. And they reveal some of the diverse strands of thought that coexisted, sometimes uneasily, in the work he produced during this stage. Some of the early solutions he tried out foreshadowed the position he eventually arrived in in Political Liberalism, while others initially led him further away from that position and, indeed, some even intensified his commitment to the very features of his A Theory of Justice view that he eventually identified as the ultimate source of the problems. Of course, there is a great deal more than this in Political Liberalism. As the views expressed in that book developed over time, new problems attracted Rawls’s attention and the full elaboration of his “political turn” required him to introduce an array of new concepts and to address many issues that had not been considered in A Theory of Justice. Many of these new problems, concepts, and issues were scarcely on his radar in the late 1970s. The intellectual route he traveled between 1977 and 1993 was a long and sometimes winding one, and Paul Weithman is surely right to say that it is not always easy to trace. Still, it is a route that is well worth trying to understand. Coda In December of 1990, I received a letter from Rawls, most of which is not relevant to this discussion, but which contained the following postscript:

Moral Independence Revisited

139

“Do you by any chance still have a copy of the letter you wrote me in October or November of 1977 when I was at the Institute at Princeton. I can’t find it among your letters to me. I recall you said my address at the APA in 1974 implied a change in the conception of the person from TJ + that I was changing my view. This greatly started me thinking. I would like to have a copy, if possible.” The context in which Rawls raised this question is not clear. He may have been starting to write (or to think about writing) his Introduction to Political Liberalism. What is evident, though, is that by December, 1990, Rawls no longer had a clear memory of the essay I had sent him thirteen years earlier. He had only a vague recollection that I had said something that had caught his attention and started him thinking and that it had something to do with the role of the conception of the person. In replying to his letter, I suppose I must have reminded him about my essay and our subsequent exchanges. In any case, he had refreshed his memory by the time he wrote the final version of the Introduction, with its generous acknowledgment of that essay. But his question underlines the reasons for Paul Weithman’s puzzlement, and it provides amusing confirmation that any influence my essay may have had on Rawls was exclusively an influence on the early stages of his development of his ideas between 1977 and the publication of Political Liberalism in 1993.17

17

Thanks to Samuel Freeman and Paul Weithman for comments on earlier drafts, and to audiences at Berkeley, Braga, and Notre Dame for discussion.

8

The Method of Insulation On the Development of Rawls’s Thought after A Theory of Justice

Rainer Forst 8.1

A Theory in Motion

We tend to read Rawls’s work with reference to two fixed points, his great monographs A Theory of Justice (1971) and Political Liberalism (1993). A large number of interpretations have been devoted to analyzing the “political turn” in Rawls’s thought since around 1985,1 and increasingly the development from his earlier views to A Theory of Justice has been an object of scrutiny (and of attempts to critically historicize Rawls; cf. Forrester 2019; Nelson 2019; Smith et al. 2021). What tends to be overlooked by focusing on these fixed points is that Rawls’s theory was constantly developing, as attested by numerous writings including his comprehensive rewriting of his theory in 1975 already for the German translation. In the following, I want to highlight a certain direction of that development and a significant methodological characteristic I see at work there, what I call the “method of insulation.” Briefly put, it is the attempt to strengthen the theory by seemingly weakening it, that is, by reducing its core commitments to what is absolutely necessary for the theoretical enterprise to be successful, shedding additional baggage along the way. The metaphor of “insulation” expresses Rawls’s attempt, as it were, to preserve the energy of the enterprise by shielding it from the cold, to block the holes that might arise as a result of wide-ranging, foundational philosophical theorems and theses that were not essential for its core aims. The root of the word is the Latin insula, meaning island. So we could also call that method “islandization,” in the sense of creating an autonomous theoretical island. In philosophy, such a method comes in many forms – think of Ockham’s razor, the so-called lex parsimoniae, with its motto: numquam

1

See in particular the careful analyses by Weithman (2011) and in Freeman (2007) as well as my own in Forst (2002, chapter IV.2; 2011, chaper 4).

140

The Method of Insulation

141

ponenda est pluralitas sine necessitate2 – yet Rawls lent it a special twist. He developed the ideal of an autonomous moral theory (until the original Dewey Lectures) or of a “free-standing” political conception of justice within political philosophy (PL, 98). And he increasingly thought that this kind of “doctrinal autonomy,” as he called it, required the independence of the theory from other parts of philosophy. In what follows, I will briefly reconstruct the main “declarations of philosophical independence,” as I call them, that we find in his theory after A Theory of Justice until Political Liberalism, and I will highlight certain problems, or even contradictions, to which this development led. For there came a point at which the method of insulation endangered rather than strengthened the autonomous standing of Rawls’s theory. I will show this by pointing to a problem within his conception of public reason.

8.2

Declarations of Independence

Before examining some of Rawls’s texts from the 1970s, it should be noted that the beginning of the path of development can be traced back to Theory itself. That work is ambivalent with respect to the “foundational” philosophical claims it makes. On the one hand, in the paragraph that introduces the method of reflective equilibrium, Rawls emphasizes that this method does not claim to rely on or to produce moral truths: “I do not claim for the principles of justice proposed that they are necessary truths or derivable from such truths” (TJ 1999, 19). In §9, he suggests a modest form of moral theory “as the attempt to describe our moral capacity” (TJ 1999, 41). He adds that this is “not meant simply [as] a list of the judgments … that we are prepared to render” (TJ 1999, 41) but rather as a careful and “conscientious” reflection on such judgments that must lead to a set of principles fitting into a systematic unity and that might “involve principles and theoretical constructions which go much beyond the norms and standards cited in everyday life; it may eventually require fairly sophisticated mathematics as well” (TJ 1999, 42). On the other hand, we find a number of quite strong theoretical statements that go beyond the modest formulations connected to the method of reflective equilibrium. For example, in §5, the critique of utilitarianism, leading to the famous sentence that “utilitarianism does not take seriously the distinction between persons” (TJ 1999, 24), and further remarks on deontological theory express foundational and comprehensive moral commitments. Thus one is not surprised to find, in the 2

“Plurality should not be posited without necessity” (Ockham 1495, I, 27, 2, K).

142

Rainer Forst

context of the discussion of the priority of right over the good, that the theory will contain principles that entail guidelines about “what forms of character are morally worthy, and so upon what kinds of persons men should be” (TJ 1999, 28). These are phrases that the later Rawls would no longer use and would criticize as implying at least a partial comprehensive doctrine, but they abound in Theory, including the argument in §86 about our “nature” as free and equal rational beings, which is expressed by acting according to the principles of justice (TJ 1999, 501 et passim). This is why Rawls states in §40 that the principles of justice are “analogous to categorical imperatives” (TJ 1999, 222). In a number of essays in the 1970s, Rawls elaborates on this philosophical outlook. A major step in that development is his 1974 address as President of the Eastern Division of the APA on “The Independence of Moral Theory.” He defines moral theory as “the study of substantive moral conceptions” (CP, 286), claiming that “much of moral theory is independent from the other parts of philosophy” (CP, 286), such as “the theory of meaning and epistemology, metaphysics and the philosophy of mind,” which “contribute very little” to the task of moral theory (CP, 286f.). To be sure, Rawls does not deny that “moral philosophy” (CP, 287) in a comprehensive sense of the term is related to these fields of philosophical inquiry – for example, when considering “the existence of objective moral truths, and the nature of persons and personal identity” (CP, 287) – yet moral theory, as the essential part of moral philosophy, ought to proceed independently from these considerations and takes a certain priority, since progress in moral philosophy depends on this exercise. Reflective equilibrium (“wide and not just narrow” [CP, 289]) aims to “find a scheme of principles that match people’s considered judgments and general convictions” (CP, 288) and thus remains agnostic with regard to the “problematical” notion of moral truth (CP, 288). Rawls even goes so far to call his method “a kind of psychology,” which “does not presuppose the existence of objective moral truths” (CP, 290). This is the move toward philosophical agnosticism that insulates the theory against debates in other areas of philosophy, which Rawls then spells out with respect to discussions of meaning, metaphysics, and the philosophy of mind. As I indicated, in the context of the discussion of personal identity, one can see that Rawls’s strategy of being agnostic with regard to certain philosophical matters and metaphysical questions (one could almost call the approach “moral and not metaphysical”) is not just one of modesty; rather, he asserts the priority of moral theory with its own method that can then be used to move into these larger areas. Thus, for example, he asserts “the dependence of moral epistemology on moral theory” (CP,

The Method of Insulation

143

301). In particular, Rawls argues that a Kantian conception or “ideal” (CP, 299) of the person is more demanding with respect to criteria of personal identity; but that is no reason to argue against it, for one need not assume that a well-ordered society could not generate in its members such a sense of being moral: “What is decisive is the content of the moral view and its roots in human sensibility” (CP, 301). This shows how far Rawls’s claim that moral theory is a kind of empirical-conceptual study of human sensibility is more demanding than it sounds, for it does not just take stock of what people actually think; it also inquires into the possibilities of what humans could be, given their best moral judgments, leading to a conception of justice and of persons with a sense of justice that can be “brought about” (CP, 301) by a just society.3 In this context, the method of insulation shields itself against the charge of constructing an unrealistic and unfeasible, more than “minimalist,” notion of the just society by starting from considered judgments and creating a systematic whole that enlarges the realm of the socially possible. One might almost call it a “realistic utopia,” to use a phrase Rawls (LP, 4–7) later coined in a different context. Essentially, Rawls advances an argument that we can call social constructivism, since he states that what “sorts of persons we are is shaped by how we think of ourselves and this in turn is influenced by the social forms we live under” (CP, 300). In another important text from this period, “A Kantian Conception of Equality,” Rawls affirms his view that the conception of the person with the two moral powers and higher-order interests to develop them can be “reformulated within the scope of an empirical theory” (CP, 264). He was convinced that no other view besides justice as fairness, at least not the rivaling ones that contest it, could provide a better account of what justice demands in a modern society. Moreover, he argued that this could be shown by principled reflection, which is to say that the method of insulation does not tone down the conceptual and substantive normative claims the theory rests on. It recalibrates the foundations on which it builds a demanding theory, but it does not aim at something like the lowest common denominator of what people actually believe. Rather, it holds up a mirror to persons, challenging them to reflect on what they ought to think if they properly considered what justice in social life means and if they were not held captive by the existing social forms in doing so.

3

On this point, compare Scanlon’s “deliberative interpretation” of the method of reflective equilibrium (Scanlon 2002, 142).

144

Rainer Forst

8.3

“The Most Reasonable Doctrine for Us”

The 1980 Dewey Lectures elaborate on this and suggest a new methodological route, namely that of Kantian constructivism. They seek to preserve the Kantian core of the approach without a commitment to Kantian metaphysics or strong foundational claims. And even though Rawls later remarked that the title of those lectures was “misleading” (CP, 389) – it should have read “Kantian constructivism in political philosophy” rather than “Kantian constructivism in moral theory,” since it lays out a “political conception” (CP, 389) – it is important to read the lectures as a transitional text toward the “political” view Rawls later developed; it prepares the way for a further step of insulation but does not as yet fully take it. What is most important about the Dewey Lectures with regard to the method of insulation is an ambivalence that should arise from this point on. For, on the one hand, Rawls further reduces the philosophical “baggage” that supports his neo-Kantian view by stressing that the “aim of political philosophy … is to articulate and to make explicit those shared notions and principles thought to be already latent in common sense” (CP, 306) and thus to “discover and formulate the deeper bases of agreement which one hopes are embedded in common sense” (CP, 306). In this vein, he stresses that the conception of justice as fairness does not claim to be true according to some metaphysical or intuitionist standard of truth but is reasonable “for us” (CP, 307), given the public traditions in democratic societies. On the other hand, as the difference between the formulations “thought to be latent” or “implicitly affirmed” (CP, 306) and the formulation “one hopes are embedded” indicate, Rawls’s approach in moral theory does not simply take stock of what people actually think; rather, assuming that “our society is not well-ordered” (CP, 355), it reconstructs “considered convictions,” such as a moral conception of the person (most important in these lectures), which “stand up to critical reflection” (CP, 306). One sees a twofold move here – a move toward the public culture of a democratic society and its deep-seated convictions and a move away from using an elaborate apparatus of critical reflection. Given this twofold move, the notion of the “reasonable” that takes center stage in these lectures has at least three meanings, with Rawls continually going back and forth between them. The first is an appeal to the considered judgments of citizens of a democratic society, crafting a conception of justice as reasonable “for us” (CP, 307); this could be called a common sense conception. The second is a philosophical notion of “reasonable” as opposed to metaphysical truths of substantive principles; let us call this the constructivist conception. The third highlights the

The Method of Insulation

145

Kantian point of the conditions of right imposed on the parties of the original position, leading them to think of themselves as aiming at “full autonomy” (CP, 308), that is, as not being motivated by “lower-order impulses” (CP, 315) but by the highest-order interest to express their moral personality. In this sense, the reasonable, framing the rational, expresses, in Kant’s terms, “pure practical reason” (CP, 319). This is the pure conception of the reasonable, a thoroughly moral one: “The unity of practical reason is expressed by defining the Reasonable to frame the Rational and to subordinate it absolutely” (CP, 319). Rawls thought he could unite these three notions of the reasonable by the method of reflective equilibrium (CP, 321), appealing to a reflexive sense of the notion of moral personality and a public conception of justice in a well-ordered society. But the method of insulation gives rise to a potential dilemma. In its declarations of independence from metaphysical notions of moral truth, including Kantian metaphysics, the theory shields itself from debates on that level; but at the same time it seems to accept a certain dependence on the shared convictions of democratic citizens in contemporary societies: “We look to ourselves and to our future, and reflect upon our disputes since, let’s say, the Declaration of Independence” (CP, 306). But such dependence would go too far if the conception of justice that resulted were heteronomously to follow “low impulses” such as “certain particular affections for this or that group of persons, association, or community” (CP, 315), as Rawls stresses. Thus the potential dilemma implies that the autonomy from disputed philosophical doctrines could imply a heteronomy based on dominant social convictions not in line with a Kantian approach. What looks “most reasonable” for many people in democratic contemporary societies could look quite unreasonable from the critical standpoint of Kantian autonomy that the original Dewey Lectures sought to redefine. In short, the method of insulation reaches an impasse; it could turn out to be self-destructive. Why self-destructive? Because any normative theory of justice, such as that of Rawls, needs to provide a proper moral reason for why one ought to consider oneself and others in the morally right way – and the normativity of that kind of reason cannot be sufficiently grounded in what most people think or in what is implicit in a society’s general understandings. These are not moral reasons at all, properly speaking, especially from a Rawlsian, neo-Kantian perspective. 8.4

Political Not Metaphysical

The next declaration of independence to be considered is the text from 1985 about “Justice as Fairness: Political not Metaphysical.” Right at the

146

Rainer Forst

beginning, Rawls rejects the assumption that his conception of justice “depends on philosophical claims I should like to avoid, for example, claims to universal truth, or claims about the essential nature and identity of persons” (CP, 388). He spells out what he calls the “method of avoidance” (CP, 395) and suggests that we should “apply the principle of toleration to philosophy itself” (CP, 395), emphasizing how a “political” conception of justice “should be, so far as possible, independent of controversial philosophical and religious doctrines” (CP, 388). In this text, Rawls goes very far in his attempt to strengthen the theory by weakening it: In affirming that justice as fairness is a political and not a “general moral conception” (CP, 389f.), he distances himself from his earlier approach of founding the conception of justice within moral theory and stresses that it starts from “basic intuitive ideas that are embedded in the political institutions of a constitutional democratic regime and the public traditions of their interpretation” (CP, 390). In this way, the conception of justice can be supported by a plurality of “comprehensive philosophical, religious, and moral doctrines” (CP, 411) within an “overlapping consensus” (CP, 410), so that justice as fairness becomes an integral part of these various doctrines from within their view; hence it is morally affirmed by them but not on the basis of the same reasons. It borrows its normative force, so to speak, from them. In this context, Rawls mentions as a mere possibility what the theory used to claim generally, namely that “some may even affirm justice as fairness as a natural moral conception that can stand on its own feet” (CP, 411); but the standard way in which citizens will understand the normative foundations of that account of justice is to regard it as an immanent part of their comprehensive doctrine, thus precisely not as standing on its own feet. At the same time, Rawls does not renounce the deontological character of the conception, which leads him to assert that within his view the “concept of justice is independent from and prior to the concept of goodness in the sense that its principles limit the conceptions of the good which are permissible” (CP, 413). What that account cannot explain, however, is how that kind of independence and priority can go along with the moral dependence of justice as fairness on comprehensive doctrines. The deontological independence stressed here is undermined by the reliance on ethical comprehensive doctrines, or so it seems. In Political Liberalism, Rawls would appeal to the notion of the reasonable in an attempt to solve that puzzle, especially with regard to the idea of “reasonable” comprehensive doctrines. But what notion of the reasonable is being used there – the common sense one or the pure one? Or yet another, perhaps the notion of “public reason”?

The Method of Insulation

8.5

147

Doctrinal Autonomy

In the revised and expanded Dewey Lectures published in Political Liberalism, Rawls tries to lay out the foundations and contours of the “political conception of justice as a free-standing view” (PL, 10) which “offers no specific metaphysical or epistemological doctrine beyond what is implied by the political conception itself” (PL, 10). It is based on “political values,” that is, values that only refer to the basic structure of society and that can be combined with various reasonable comprehensive doctrines. Obviously, the notion of the reasonable carries a lot of weight here. A free-standing view has to stand on something; otherwise it is freefalling. As I explained elsewhere (Forst 2013, chs. 9 and 10; 2017), Political Liberalism answers the classic question of toleration in a way very much in line with the tradition of thinkers such as Castellio or Bayle: a justifiable view of toleration requires a basis from which the question of the grounds and the limits of toleration can be determined without prejudging this issue in a partisan way. Hence it needs to rely on a notion of moral and epistemological reasonableness that can be upheld against all particular comprehensive doctrines as essential to a just and reasonable way of living together despite deep disagreements about ethical values. That notion will form the core of public reason as the medium of democratic justification among citizens who accept the priority of justice; what is more, Rawls says that the conception of justice as fairness “expresses their shared and public political reason” (PL, 9). For thinkers of toleration in the relevant tradition, reason alone held the promise of defining the common ground that none of the competing doctrines could justifiably reject. The argument of Political Liberalism unfolds along a thin line. On the one hand, Rawls is at pains to stress that the conception of justice is not “political in the wrong way” (PL, 40) in the sense of a compromise or mere overlap of existing comprehensive doctrines, and he also stresses the deep divisions within contemporary democratic societies that make abstract theorizing and critical reflection necessary (PL, 44f.). That excludes forms of conventionalism. On the other hand, the political conception, which is “of course, a moral conception” (PL, 11) based on moral “ideals, principles and standards” (PL, 11, n. 11), must not be of a comprehensive nature itself. This rules out appeals to pure practical reason understood along Kantian lines. Still, if one compares the original and the revised Dewey Lectures, one finds a great deal of continuity, apart from the main difference of translating the argument from one within “moral theory” to one within

148

Rainer Forst

“political philosophy.” Hence the “Reasonable” (PL, 48ff.) still frames the “Rational” and contains the two essential aspects of the “willingness to propose fair terms of cooperation and to abide by them provided others do” (PL, 54) and of the “burdens of judgment,” which explain to what extent reasonable disagreement about ethical matters between comprehensive doctrines is not a fault but is in line with the faculty of proper reasoning, since many matters of life can be reasonably answered by values and beliefs that reason can neither prove nor refute. These are the two aspects of an advanced account of toleration. In connection with the idea of full public justification of the conception of justice as generally shared, these two aspects constitute what Rawls still calls “full autonomy” (PL, 77), that is, the autonomy realized by citizens acting according to principles of justice “they would give to themselves when fairly represented as free and equal persons” (PL, 77). This is the core of the “reasonable psychology” explaining the motivation of citizens aiming to realize an ideal conception of themselves as autonomous in that way. In the chapter on constructivism, Rawls stresses that the procedure of construction “embodies all the relevant requirements of practical reason and shows how the principles of justice follow from the principles of practical reason in union with conceptions of society and person, themselves ideas of practical reason” (PL, 90). In acting on the basis of such principles, citizens “show themselves autonomous, politically speaking” (PL, 90). Rawls goes on to argue that this way of conceiving of the justification of the principles of justice “enables us to state the meaning of an autonomous political doctrine as one that represents, or displays, the political principles of justice … as reached by using the principles of practical reason in union with the appropriate conceptions of persons as free and equal and of society as a fair system of cooperation over time” (PL, 98). Hence the “full autonomy” of citizens practicing public reason is the core idea of the claim to “doctrinal autonomy” (PL, 90) of the freestanding conception of justice. This implies that a conception of justice based on other grounds than those is “doctrinally heteronomous” (PL, 99). Doctrinal autonomy, to be sure, no longer rests on the claim that the constructivist procedure “constitutes” the order of moral and political values; but it does imply with Kant that reason is “self-originating and self-authenticating” (PL, 100) and that the conception of justice is fully justified by reference to practical reason alone. At this point, the ambivalence of the method of insulation I already pointed out in connection with other texts comes to the fore. On the one hand, the idea of a free-standing, doctrinally autonomous conception of justice

The Method of Insulation

149

expressing an ideal of political autonomy only works if it is based on noncomprehensive but sufficiently substantive grounds of theoretical and practical reason. These must be strong enough to trump ethical considerations of comprehensive doctrines that prove to be incompatible with the political conception; otherwise, the priority of right or of the reasonable would disappear. On the other hand, the method of reducing the comprehensive moral and metaphysical claims of the theory leads Rawls to say that the substantive moral grounds on which the conception rests in the eyes of citizens are those that derive from the respective comprehensive doctrines, such that “those who affirm the political conception start from within their own comprehensive view and draw on the religious, philosophical, and moral grounds it provides” (PL, 147). Eventually, then, this leads to a contradiction: the free-standing doctrine must be autonomous and can only be so if morally grounded on practical reason alone, while the same, non-comprehensive doctrine must borrow its normative force from the comprehensive doctrines and thus does not have an autonomous status with respect to them. In the end, the journey toward independence from comprehensive doctrines leads to dependence on them. The method of insulation turns against itself. 8.6

Public Reason

This aporia, which kept Rawls busy in subsequent texts, such as the new introduction to the paperback edition of Political Liberalism or the exchange with Habermas, plays itself out in various contexts, one of them being the discussion of public reason in Political Liberalism and in a later text, “The Idea of Public Reason Revisited.” As I pointed out, the idea of public reason is related to full autonomy, such that it exhibits an ideal of democratic citizenship and leads to citizens speaking a shared language of justice and not one based on a comprehensive doctrine when they discuss essential matters of justice. In political life, public reason expresses autonomy to the extent that citizens regard each other as free and equal agents of justice and conduct their arguments in such a way as to give priority to political values and not to their comprehensive views if the latter suggest something different from the political conception. From the way Rawls tries to spell out this language of justice, it is clear that the political conception provides its grammar, and the pluralistic idea that religious believers might argue for that conception as based on the word of God – while Millians, for example, might support it on the basis of utilitarian considerations – suddenly pales and disappears when citizens consider their duty “to conduct their fundamental discussions within the framework of what each regards as a political conception of

150

Rainer Forst

justice based on values that the others can reasonably be expected to endorse” (PL, 226), which is why truly public principles and guidelines of inquiry are required. And this is why (with certain provisos of translation) “reasons given explicitly in terms of comprehensive doctrines are never to be introduced into public reason” (PL, 247). At this point, I do not want to go into the complex debates that arose with respect to this issue and the way Rawls tried to respond (cf. Forst 2002, ch. III.1; 2011, chs. 4 and 7); rather, what I want to highlight is that this again is a context in which the ambivalence and the tensions, if not the contradictions, brought about by the method of insulation become apparent. Once engaged in reducing the philosophical claims of the conception of justice, a journey that began long before the “political” turn, Rawls continually struggled to find the right way to explain the independent, autonomous, non-metaphysical and yet moral grounding of that conception. In the end, however, he could not find the proper balance between the claim of doctrinal autonomy and that of a noncomprehensive political conception as a “module” of comprehensive views, to highlight the two essential, opposing aspects of the method of insulation. Dialectically speaking, the struggle for ever more independence led to the danger of philosophical and normative dependence, and the island that Rawls tried to create is in danger of being washed away.

9

The Stability or Fragility of Justice Japa Pallikkathayil

Rawls describes the development of his view in Political Liberalism as motivated by a problem with A Theory of Justice’s argument for the inherent stability of justice as fairness. In what follows, I begin by considering how and why stability is treated as relevant to the assessment of conceptions of justice in A Theory of Justice. After noting the problem Rawls identifies with his treatment of stability in A Theory of Justice and outlining the ways in which Political Liberalism is meant to address this problem, I step back from this discussion to consider its key presupposition, namely, that stability is an important feature of a conception of justice. I do this by considering a series of objections G. A. Cohen presses against this position. I argue that these objections largely do not have traction against Rawls’s view insofar as we grant Rawls the claim that the task of principles of justice is to provide fair terms of cooperation. But I then develop an objection to this claim by drawing on Cohen’s critique of Rawls’s treatment of the circumstances of justice. I argue that these circumstances are more capacious than Rawls allows. Nonetheless, contra Cohen, I do not think that recognizing the force of this objection requires us to give up one of the key insights of Rawls’s project, which is that justice is fundamentally about realizing a certain kind of relationship rather than realizing a particular distribution of goods. I close with a brief reflection on the role stability plays in an alternative conception of the relevant relationship. 9.1

Rawls on Stability

In A Theory of Justice, Rawls claims: It is evident that stability is a desirable feature of moral conceptions. Other things equal, the persons in the original position will adopt the more stable scheme of principles. However attractive a conception of justice might be on other grounds, it is seriously defective if the principles of moral psychology are such that it fails to engender in human beings the requisite desire to act upon it. (TJ 1999, 298) 151

152

Japa Pallikkathayil

Notice that Rawls moves in this passage from describing stability as a desideratum of a conception of justice to describing a conception of justice as seriously defective if it does not engender in human beings the requisite desire to act upon it. I will be largely focused on this second thought. I begin by unpacking a bit further what kind of stability is of interest to Rawls. Second, I examine why that kind of stability occupies his attention. Finally, I briefly describe how Rawls argues for the stability of justice as fairness. Let us begin, then, by considering the kind of stability that is of interest to Rawls. To assess whether conceptions of justice can engender in human beings the requisite desire to act upon them, we consider whether people living in a well-ordered society structured by that conception of justice would develop a sense of justice and come to regard having an effective sense of justice as part of their good. A well-ordered society is one in which “(1) everyone accepts and knows that the others accept the same principles of justice, and (2) the basic social institutions generally satisfy and are known to satisfy these principles” (TJ 1999, 4). Would people in such a society develop a sense of justice and regard it as part of their good in a way that would lead them to reject freeriding? Paul Weithman helpfully distinguishes this kind of inherent stability from what he calls imposed stability (Weithman 2010, 45). The Hobbesian state provides a helpful example of imposed stability. Here obedience to the law is generated by the threat of punishment. Rawls, by contrast, is interested in the stability that comes from people’s allegiance to the principles of justice. Although this kind of stability is obviously attractive, why is a conception of justice that does not produce it “seriously defective”? I suggest that the answer to this question lies in the role the idea of cooperation plays in Rawls’s view. Consider three closely related appearances of the idea of cooperation. First, Rawls takes the circumstances of justice to be “the normal conditions under which human cooperation is both possible and necessary” (TJ 1999, 109). Rawls’s characterization of society as “as a cooperative venture for mutual advantage” (TJ 1999, 4) pairs naturally with this conception of the circumstances of justice. And the same is true of Rawls’s characterization of the subject of justice as “the basic structure of society, or more exactly, the way in which the major social institutions distribute fundamental rights and duties and determine the division of advantages from social cooperation” (TJ 1999, 6). If principles of justice are to provide fair terms of cooperation, they must be compatible with willing compliance with the institutions governed by them. If compliance is brought about simply through the

The Stability or Fragility of Justice

153

threat of sanctions, we would not have genuine cooperation. Our actions would be coordinated and perhaps even in ways that were reliably replicable over time. But we would not be acting together on the basis of any shared principles or toward any shared ends. I take this to be an important reason why Part III of A Theory of Justice is directed at showing how people in a society structured by justice as fairness could come to willingly comply with its institutions and laws and hence how such a society could be stable in the relevant sense. There are two steps in this project. First, Rawls aims to show that the principles of moral psychology are such that people living in a society structured by justice as fairness would come to develop a sense of justice. Second, he aims to show that such people could view having an effective sense of justice as part of their good. If they could not, freeriding rather than compliance would be the rational response to the compliance of one’s fellow citizens and hence cooperation would not have been achieved. In Political Liberalism, Rawls identifies a problem with A Theory of Justice’s argument for stability. Although Rawls is less than fully explicit about what the problem is, it stems from the fact that under the kind of free institutions required by justice as fairness, Rawls believes people will reasonably come to hold different comprehensive doctrines. This reasonable pluralism unsettles A Theory of Justice’s argument that having an effective sense of justice is a part of one’s good insofar as that argument relies on claims about which reasonable citizens may disagree. For example, as Samuel Freeman emphasizes, some reasonable citizens may deny that the kind of Kantian autonomy Rawls argues is realized by having an effective sense of justice is in fact a part of one’s good (Freeman 2003, 29). Identifying this problem in A Theory of Justice’s argument for stability also seems to prompt a shift in the threat to stability that occupies Rawls’s attention. Whereas A Theory of Justice’s account of stability is focused on rejecting the rationality of freeriding, Political Liberalism’s account of stability is focused on securing what Rawls calls “stability for the right reasons” as contrasted with a mere modus vivendi. A modus vivendi is characterized by strategic rather than genuine commitment to liberal institutions. Participants in a modus vivendi accept that liberal institutions are the best that they can do for their own comprehensive doctrines given the current distribution of views in society. But if they found themselves in the majority, they would not regard themselves as having a moral reason not to impose their comprehensive doctrine on those with whom they disagree. For this reason, a modus vivendi is in tension with the idea of society as a fair system of cooperation, albeit in a rather different way than freeriding. While the free rider declines to do her part in the

154

Japa Pallikkathayil

cooperative scheme, the participant in a modus vivendi at least coordinates her actions with those of fellow citizens. But she does not do this from shared principles or for the sake of shared ends. In this way, she is also not cooperating with her fellow citizens. Rawls’s argument that liberal society can be more than a mere modus vivendi requires the introduction of a host of new concepts. He argues that stability for the right reasons may be secured through an overlapping consensus on a family of reasonable political conceptions of justice. Political conceptions of justice are freestanding, that is, expressed in terms implicit in the public political culture of a liberal democracy rather than in terms tied to a particular comprehensive doctrine. In virtue of being freestanding, political conceptions of justice may be genuinely and not merely strategically endorsed by adherents of different comprehensive doctrines. Rawls gives an account of how such an overlapping consensus might develop. And if an overlapping consensus obtained, that would enable adherents of different comprehensive doctrines to cooperate with one another on the basis of shared principles or values. Notice that in Political Liberalism’s account of stability the task of showing how having an effective sense of justice may be a part of one’s good is moved off stage. This work is internal to comprehensive doctrines and so no general account can be given. As Rawls puts it, “it is left up to each citizen, individually or in association with others, to say how the claims of political justice are to be ordered, or weighed, against nonpolitical values” (PL, 386). This means that there is no general response to the threat to stability posed by the possibility of freeriding. But perhaps any comprehensive doctrine that allows the kind of genuine commitment to a political conception of justice needed for an overlapping consensus would have to have a way of blocking this possibility. In any case, what I hope this discussion has brought out is that across both A Theory of Justice and Political Liberalism the kind of stability that is of interest to Rawls is the stability that comes from allegiance to principles of justice. Moreover, his interest in this kind of stability may be seen as flowing from his commitment to associating justice with the realization of fair cooperation. In Sections 9.2 and 9.3, I consider G. A. Cohen’s objections to these closely related commitments. 9.2

Cohen on Stability

In this section, I consider three objections to Rawls’s view pressed by Cohen. These objections are focused on the role stability plays in Rawls’s argument for justice as fairness. While I take each of these objections to fail, they point us toward the connection Rawls draws between justice

The Stability or Fragility of Justice

155

and fair cooperation, which I have argued underlies the significance of stability in his view. In Section 9.3, I draw on Cohen’s discussion of Rawls’s treatment of the circumstances of justice to call into question this connection. Cohen’s first objection focuses on a distinction he draws between principles of justice and rules of regulation. The parties in Rawls’s original position are attempting to identify rules to govern their shared social life. Cohen contends that the choice of such rules reflects more than considerations of justice: if rules of regulation “are soundly based, they will reflect both values other than justice and practical constraints that restrict the extent to which justice can be applied. That being so, justice, itself, could not be what is specified by such rules” (Cohen 2008, 3). Given the aim of the parties, it is entirely understandable that they are concerned to secure stability. But Cohen contends that justice is one thing and stability is another. For this reason, the parties’ concern for the stability of the principles they choose impugns the status of what they choose as principles of justice: “We of course don’t want justice to be permanently at risk. But the very coherence of that fear shows that stability is not a feature of justice itself” (Cohen 2008, 328). Cohen’s objection here reflects a deep disagreement between Rawls and Cohen about the relationship between justice and other values. As Rawls famously puts it at the beginning of A Theory of Justice: “Justice is the first virtue of social institutions, as truth is of systems of thought. A theory however elegant and economical must be rejected or revised if it is untrue; likewise laws and institutions must be reformed or abolished if they are unjust” (TJ 1999, 3). Rawls is clear that he does not treat justice as simply one among many considerations that bear on the rules that govern our social life. I am inclined to think that Rawls has the more compelling view of this matter. That is because treating justice as merely one among many considerations that feature in the rules that govern our common social life makes it difficult to treat acting justly as always laudatory in the way that even Cohen himself seems inclined to treat it. Consider an example that Cohen intends to demonstrate that even though unequalizing Pareto improvements on equality might represent sound policy, they do not promote justice: Imagine a peaceful anarchy … in which manna falls from heaven and gets shared equally because the sharers think that’s the right way to deal with manna from heaven. Now suppose that an extra piece of irremovable but destructible manna falls on Jane’s plot. Jane says: “I don’t want this extra manna, I’m going to make a big bonfire with it to which you’re all invited, because it’s not fair for me to have more than you guys do, for no good reason.” If you think Jane is being merely

156

Japa Pallikkathayil

foolish, then you can reject the claim that justice favors equality in this elementary case. But I for one would not think that Jane is being foolish. I would think that she is simply a remarkably just person, and I think we should commend her for being one …. (Cohen 2008, 317–18)

If allowing an unequalizing Pareto improvement were all things considered what one ought to do, it is mysterious why we would commend Jane for being especially just in this circumstance. Should she not instead be condemned for being unjustifiably wasteful and for fetishizing justice when considerations of justice have been outweighed by other considerations? It is difficult to maintain both that justice is merely one among many potentially relevant considerations and that being just is always or even generally commendable. In this way, it is difficult for Cohen’s view to explain the significance we normally attribute to considerations of justice. Of course, as I said, this disagreement about the nature of justice is a deep one and I do not take the brief foregoing discussion to settle the matter. But turning our attention to the second objection I want to consider will allow us to capture something of the spirit of the first objection without requiring us to settle whether justice should be identified with the rules that govern society. Here let us focus on Cohen’s suggestion that it is a coherent conceptual possibility that justice will not be stable. He claims that “to judge that principles of justice qualify as principles of justice only if, once instituted, their rule has a propensity to last, is absurd. It would mean that one could not say such entirely intelligible things as ‘This society is at the moment just, but it is likely to lose that feature very soon: justice is a fragile achievement’ …” (Cohen 2008, 327–28). In Section 9.5, I will gesture toward a Kantian view of justice that illuminates why, contra Cohen, there may be a tension between holding that justice obtains but that it might soon not. Here though I want to draw out two ways in which Rawls’s view is not actually committed to denying the possibility that just institutions may not last. First, notice that Rawls does not claim that just social arrangements must be entirely unshakeable. As he describes it “an equilibrium is stable whenever departures from it, caused say by external disturbances, call into play forces within the system that tend to bring it back to this equilibrium state, unless of course the outside shocks are too great.” (TJ 1999, 400, emphasis added) There are, then, circumstances in which we might not expect a just society to endure even though it could be described as stable in Rawls’s sense. Second, it is important to reflect on how Rawls assesses the stability of conceptions of justice. He does so by considering how they would be received in a well-ordered society structured by them. And one aspect of

The Stability or Fragility of Justice

157

a well-ordered society is that everyone accepts and knows that others accept the same principles of justice. But it is entirely possible that the basic structure of a society might satisfy a conception of justice without full acceptance of that conception or without its being common knowledge that full acceptance obtains. This might, for example, be the case in newly founded liberal democracies. If full acceptance and common knowledge do not obtain, it might well be that just institutions are fragile even though the conception of justice with which they accord is stable in the relevant sense. Thus, there seems to be room in Rawls’s view for holding that just institutions could, under certain conditions, be fragile. This is because he is not concerned with whether just institutions have a propensity to last come what may but with whether they would engender their own support in ideal conditions. This reply to Cohen’s objection highlights again the distinctive nature of the kind of stability that is of interest to Rawls. He is not focused on the mere propensity for institutions to persist whatever the reason and whatever the circumstances. Rather, he is focused on the very specific question of whether a conception of justice can be the subject of willing compliance when others are likewise disposed. This focus is the subject of Cohen’s third objection. Cohen presses why we should take a deficit in the relevant motivations to reflect poorly on a conception of justice rather than on the people who lack those motivations: “Is it an axiom that human beings are capable of justice? Is ‘original sin’ a contradiction in terms?” (Cohen 2008, 330) To put this objection in more Rawlsian terms, why think that identifying fair terms of cooperation requires that there be any circumstances in which people are actually disposed to cooperate fairly? Why should utilitarianism’s difficulty securing willing compliance tell against it as a conception of justice rather than tell against the insufficient altruism of human beings? I suspect that Rawls’s work suggests more than one answer to these questions, though perhaps those answers are compatible and even mutually illuminating. Perhaps a certain kind of constructivism about moral facts requires that practical reason not be at odds with itself and hence that the verdicts of collective and individual rationality cohere. Be that as it may, I want to focus on a reason for using the possibility of willing compliance in a well-ordered society to assess a conception of justice that stems from the nature of justice as Rawls conceives of it. How could I be offering you fair terms of cooperation if it would be irrational for you to do your part even if you know I am doing mine? Although I think raising this question provides a relatively straightforward reply to Cohen’s objection insofar as we accept Rawls’s identification of principles of justice with fair terms of cooperation, I want to

158

Japa Pallikkathayil

acknowledge that it sits ill at ease with another aspect of Rawls’s view. To introduce the problem, let us first consider how Rawls describes the aim of his argument for the congruence between the right and the good. This argument, which he takes to be crucial in establishing the stability of justice as fairness, is supposed to proceed as follows: Consider any given person in a well-ordered society. He knows, I assume, that institutions are just and that others have (and will continue to have) a sense of justice similar to his, and therefore that they comply (and will continue to comply) with these arrangements. We want to show that on these suppositions it is rational for someone, as defined by the thin theory, to affirm his sense of justice. The plan of life which does this is his best reply to the similar plans of his associates; and being rational for anyone, it is rational for all. (TJ 1999, 497)

So far, Rawls’s aims cohere well with the reply to Cohen I have been developing on his behalf. But then a few pages later, Rawls seems to limit the ambition of the congruence argument. He allows for the possibility that there may be some people in the well-ordered society of justice as fairness “for whom the affirmation of their sense of justice is not a good. Given their aims and wants and the peculiarities of their nature, the thin account of the good does not define reasons sufficient for them to maintain this regulative sentiment” (TJ 1999, 503). The possibility of such people raises a pressing question about whether “those who do affirm their sense of justice are treating these persons unjustly in requiring them to comply with just institutions” (TJ 1999, 504). Rawls largely sets aside this question since he has not fully developed a theory of punishment. But he does not think that the parties in the original position would insist that “a person can be required to do only what is to his advantage as defined by the thin theory” (TJ 1999, 504). Moreover, he claims that such persons would still have to acknowledge the collective rationality of the conception of justice chosen in the original position. Of these people, he says, “It is, of course, true that in their case just arrangements do not fully answer to their nature, and therefore, other things being equal, they will be less happy than they would be if they could affirm their sense of justice. But here one can only say: their nature is their misfortune” (TJ 1999, 504). This strand of thought does not support the claim that fair terms of cooperation must be the possible object of willing compliance. But for that very reason, Cohen’s objection has force against it. If the inability of some to rationally abide by the terms given by a conception of justice in the absence of sanctions can tell against their nature rather than against the conception of justice, why think that inability ever tells against the conception of justice? For this reason, I suggest that Rawls’s commitment to treating stability as a crucial aspect of a conception of justice

The Stability or Fragility of Justice

159

requires maintaining the tight connection between fair terms of cooperation and the possibility of willing participation that I suggested above. In this section, I have argued that Rawls has the resources for plausible replies to Cohen’s three objections to treating stability as an essential feature of a conception of justice. These replies, however, rely on Rawls’s association of principles of justice with fair terms of cooperation. In Section 9.3, I develop an objection to this association. 9.3

Cohen on the Circumstances of Justice

Recall that Rawls takes the circumstances of justice to be the normal conditions under which cooperation is both possible and necessary. Although Rawls lists a number of such circumstances, both he and Cohen focus on two circumstances in particular. As Rawls puts it, “the circumstances of justice obtain whenever persons put forward conflicting claims to the division of social advantages under conditions of moderate scarcity” (TJ 1999, 110). Cohen describes Rawls’s circumstances of justice as consisting roughly in “limited altruism and limited scarcity” (Cohen 2008, 331) and he aims to show that even when these circumstances do not obtain, the concept of justice still has application. As we will see, conflicting interests and limited altruism are not quite the same. In part for this reason, the first half of Cohen’s objections largely miss the mark. But I will suggest that Cohen’s discussion of moderate scarcity can be developed to effectively undermine Rawls’s association of principles of justice with fair terms of mutually advantageous cooperation. Let us begin by considering a circumstance of limited altruism. Rawls invites construing limited altruism as among the circumstances of justice by describing himself as adding nothing essential to Hume’s account of those circumstances (TJ 1999, 110). And Hume straightforwardly treats limited altruism as among the circumstances of justice. Cohen claims that Rawls should have held “that if altruism were unlimited, then justice would indeed be unnecessary (though not impossible), and that if selfishness were unlimited, then it would be necessary if it were possible but that it would almost certainly be impossible” (Cohen 2008, 334). First, consider the circumstance of unlimited altruism. In a different argumentative context, Rawls considers the possibility of unlimited altruism: A perfect altruist can fulfill his desire only if someone else has independent, or first-order, desires. To illustrate this fact, suppose that in deciding what to do all vote to do what everyone else wants to do. Obviously nothing gets settled; in fact, there is nothing to decide. For a problem of justice to arise at least two persons must want to do something other than whatever everyone else wants to do. (TJ 1999, 165)

160

Japa Pallikkathayil

Although it is tempting to think, with Cohen, that a condition of unlimited altruism is a condition in which justice is unnecessary but still possible, Rawls offers a strikingly different assessment. A circumstance of unlimited altruism is one in which mutually advantageous cooperation is impossible. In such a condition, no one has any desires that could be advanced by cooperation because no one has any substantive desires at all.1 Hence, for Rawls, considerations of justice are inapplicable in such a circumstance. Since Cohen’s own view is that justice is a property of distributions rather than of the principles governing institutions, he might still insist that justice is possible in this circumstance. But I doubt this claim has much intuitive pull for anyone not antecedently committed to Cohen’s construal of justice. A society of perfect altruists would be so dysfunctional that no one would have any use for the resources available to them. In such a circumstance, it is difficult to see why the distribution of resources would matter. Although I side with Rawls in thinking that justice has no place in a society of perfect altruists, as I mentioned at the outset, I do not think that limited altruism is quite the same as the conflict of interests Rawls emphasizes when he presents the circumstances of justice. He elaborates the relevant kind of conflict as follows: While the parties have roughly similar needs and interests, or needs and interests in various ways complementary, so that mutually advantageous cooperation among them is possible, they nevertheless have their own plans of life. These plans, or conceptions of the good, lead them to have different ends, and purposes, and to make conflicting claims on the natural and social resources available. (TJ 1999, 110)

The kind of conflict of interests Rawls describes seems more aptly contrasted with a complete harmony of interests and a complete opposition of interests rather than with unlimited altruism and unlimited selfishness. Notice that a complete harmony of interests does not immediately suggest the same kind of dysfunction that beset the society of perfect altruists. There is no reason to think that mutually advantageous cooperation would be impossible among people with completely harmonious interests. Indeed, if anything, it seems like mutually advantageous cooperation would be commonplace in such a circumstance. Since there would be no conflict about the distribution of the benefits and burdens of social cooperation, Rawls seems to doubt that principles of justice would 1

Of course, one might conceive of unlimited altruism differently, perhaps as directed at fulfilling others’ interests rather than their desires. But in such a case, it seems likely there would be the kind of conflict of aims that Rawls takes to be necessary for questions of justice to arise. He would thus agree that justice has application in such a circumstance.

The Stability or Fragility of Justice

161

be applicable. But it is unclear why, by his own lights, this should matter. Rather, it seems as though this is a circumstance in which justice – something like “social institutions should be arranged to everyone’s advantage” – is easily achieved. Notice, though, that this objection leaves the connection between justice and cooperation intact. The objection merely identifies a possibility of mutually advantageous cooperation that Rawls overlooked. So, although Rawls might have been wrong to exclude a complete harmony of interests from the circumstances of justice, acknowledging this would not undermine his larger project. Having considered the circumstances of unlimited altruism and completely harmonious interests, we may now turn our attention to the circumstances of unlimited selfishness and complete conflict of interests. My discussion of these circumstances will be comparatively brief. Whether mutually advantageous cooperation is possible among the completely selfish depends on how complementary their interests are. For that reason and since Rawls’s own emphasis is on conflicting interests, it will be more productive to set aside unlimited selfishness and focus on completely conflicting interests. A circumstance of extreme scarcity provides an excellent example of a circumstance in which people’s interests are completely opposed. I will therefore turn my attention to Rawls’s treatment of extreme scarcity. I am going to argue that, contra Rawls, justice has application in such a circumstance. This will simultaneously demonstrate the applicability of justice to a circumstance in which people’s interests are completely opposed. Rawls describes the kind of moderate scarcity that he takes to be among the circumstances of justice as follows: “Natural and other resources are not so abundant that schemes of cooperation become superfluous, nor are conditions so harsh that fruitful ventures must inevitably break down. While mutually advantageous arrangements are feasible, the benefits they yield fall short of the demands men put forward” (TJ 1999, 110). Although Rawls seeks to exclude even circumstances in which “fruitful ventures must inevitably break down,” it will be dialectically more useful to consider an even more extreme condition in which mutually advantageous cooperation is altogether impossible. Cohen gives an example of extreme scarcity in which this is the case. Suppose there is only one life preserver and many people who will die if they do not get it. Mutual advantage cannot be realized in this circumstance as there is only one benefit to be distributed and there are no future gains for those who do not receive the benefit, as they will be dead. Nonetheless, Cohen suggests that the egalitarianism that informs Rawls’s view supports rolling dice to determine who gets the life preserver. And if this is the just procedure, Cohen claims that there is no reason to think

162

Japa Pallikkathayil

that it is impossible for it to be followed. He invites us to consider “that rare and powerful someone who could have seized the only life preserver but who instead rolled the dice to determine who would get it” (Cohen 2008, 333). Cohen suggests that such a person would not merely be generous but especially just. One might worry that this example does not make sufficient contact with Rawls’s project since he is considering principles of justice for the regulation of the basic structure of society rather than principles governing individual conduct. But we can reformulate the example so that it is focused on how political, economic, and social institutions distribute scare resources. It might well be that such institutions could not exist in conditions of extreme scarcity without the significant use of coercion. But I see no reason to assume that the use of coercion to maintain a public lottery in conditions of extreme scarcity would be unfair. Of course, to fully evaluate this institutional scheme, we would need to compare it to others. My point here is that the kinds of questions about the fair division of rights and responsibilities and benefits and burdens that are familiar from Rawls do not seem out of place even when cooperation for mutual advantage is off the table. One might object that social institutions cannot be maintained by coercion alone. Some people must be genuinely committed to those institutions for them to function. Hobbes brings out this concern beautifully: “For if men know not their duty, what is there that can force them to obey the Laws? An Army, you’l say. But what shall force the Army?” (Hobbes 1990, 59).2 This is an important observation and one to which I will return in Section 9.4. For now, notice that although coercion might be needed to assure individuals that others will comply with the terms of a public lottery, we need not assume that those committed to principles of justice would themselves need to be moved by coercion. Rawls recognizes that commitment to principles of justice may come at great personal cost: “we may in the end suffer a very great loss or even be ruined by it. As we have seen, a just person is not prepared to do certain things, and so in the face of evil circumstances he may decide to chance death rather than act unjustly.” (TJ 1999, 502) Of course, Rawls has in mind here commitment to justice as fairness, which is formulated for the circumstances of justice as he conceives of them. But since Rawls takes it to be possible to maintain a commitment to justice even at the cost of one’s life, I see no reason to think that it would be impossible to maintain such a commitment in conditions of extreme scarcity. We thus arrive at

2

I am indebted to Sharon Llyod for this reference.

The Stability or Fragility of Justice

163

the following conclusion: even when cooperation for mutual advantage is impossible, questions about the fair distribution of rights and responsibilities and benefits and burdens remain. I take this to suggest a more capacious understanding of the circumstances of justice. These are simply the circumstances of human interaction. Examining the circumstance of extreme abundance provides further support for this conclusion. Rawls excludes the condition of extreme abundance from the circumstances of justice because cooperation would be “superfluous.” Cohen claims that even in a circumstance of extreme abundance, there would still be an egalitarian rule that could be followed, namely: “take whatever you like. People’s opportunities would then be radically identical, and no inequality that would disturb an egalitarian could ensue” (Cohen 2008, 333). Cohen takes this to show that the concept of justice still has application even in extreme abundance. That is, we can still ask what the fair distribution of rights and responsibilities and benefits and burdens is in a circumstance of extreme abundance even if the answer to that question is immediately obvious. This objection is fine as far as it goes. But a deeper objection may be identified by considering what both Rawls and Cohen seem to overlook, namely, that even in circumstances of extreme material abundance we might still get in each other’s way. Your barbecue may interfere with my enjoyment of fresh air. My music may interfere with your peace and quiet. Perhaps extreme material abundance means either of us could easily move. But this fact does not settle which, if either of us, must. Notice that these conflicts need not be the product of spiteful motivations. Our plans may simply be at odds even in circumstances of extreme material abundance. The examples above concern the ways in which our actions may affect those around us. But the point runs deeper. Just as you may be an impediment to my plans, you may also have a role to play. If I am besotted with you, it may not matter to me that there are countless other people quite like you whom I could pursue instead. And extreme material abundance may not help you avoid me if I am always at liberty to follow you. Our plans may be inexorably at odds precisely because they involve each other. Surely justice is not silent about the issues regarding privacy and nonconsensual contact raised by the foregoing example. I take this to show that extreme material abundance is irrelevant to the applicability of the concept of justice because there are benefits and burdens made possible uniquely through human interaction and that are not easily characterized in terms of abundance or scarcity. Perhaps mutually advantageous cooperation is possible with respect to these benefits and

164

Japa Pallikkathayil

burdens. But just as in the case of extreme scarcity, we can ask about the fair distribution of these benefits and burdens regardless of whether that is so. Before closing this section, I want to highlight a circumstance of justice that Rawls mentions but does not emphasize. He describes the individuals for whom cooperation is both possible and necessary as “roughly similar in physical and mental powers; or at any rate, their capacities are comparable in that no one among them can dominate the rest. They are vulnerable to attack, and all are subject to having their plans blocked by the united force of others” (TJ 1999, 109–10). This curiously Hobbesian moment in Rawls’s discussion should give us pause. Someone who is substantially stronger or smarter than the rest of us might not need to cooperate with us. But surely institutions that allowed such a person to dominate the rest of us would be unjust. I suggest that Rawls’s focus on cooperation leads him here to focus on the wrong kind of equality for the purposes of justice. What matters for justice is, to borrow a phrase from Rawls’s later work, that we are “self-authenticating sources of valid claims” (JF, 23). That is, the fact that we are individuals living out our own lives and carrying out our own plans ought to constrain how others interact with us.3 And this is so regardless of whether mutually advantageous cooperation with us is possible or necessary. 9.4

An Alternative

One might worry that if we reject Rawls’s description of the circumstances of justice in favor of a more capacious focus on human interaction as such, we might lose the ability to see justice as fundamentally about realizing a certain kind of relationship rather than realizing a particular distribution of goods. But I do not think this is the case. The relationship of cooperation is not the only relationship with which we might associate justice. To take a concrete example, consider Kant’s claim that the task of political institutions is to enable us to realize mutual independence.4 Given the Kantian pedigree of Rawls’s view, this relational ideal has unsurprising resonance with some important elements of his view, for example, that social institutions must respect the separateness of persons. 3

4

I take it that this is why the society of perfect altruists is beyond the scope of justice. Those in such a society are incapable of carrying out any plans. This is much as it would be if one were to aim to do what an oracle tells one to do, but the oracle never speaks. The problem with the society of perfect altruists is not specifically their altruism but that altruism in their circumstances renders them incapable of action. Kant develops this view in Part I of (1991).

The Stability or Fragility of Justice

165

But mutual independence does not straightforwardly require cooperation and for that reason I think it is instructive to consider how and why the issue of stability arises within it. Among the barriers to mutual independence Kant takes the state to need to resolve is the assurance problem. If I do not have assurance that you will not violate my rights, I am not fully independent from you.5 For this reason, the problem of freeriding and compliance with the law more generally is one that the Kantian view must take seriously. But since the Kantian view is not aiming at cooperation, relying on coercive sanctions to resolve the assurance problem is a potentially viable strategy. In other words, in this respect imposed rather than inherent stability is adequate for the Kantian’s purposes. But the assurance problem posed by a modus vivendi is not so easily dispensed with. If we suppose that legitimate political institutions must have a democratic character, then the security of our rights will depend importantly on whether our fellow citizens are genuinely rather than strategically committed to those institutions. This is why, I take it, there might be some tension between thinking that justice has been achieved and yet that just institutions are fragile. If our rights might be done away with tomorrow by legislators who are inadequately committed to them, then we lack the kind of assurance of our rights needed for mutual independence. Institutional checks and balances can help in some respects with this issue, but since all branches of government are run by fellow citizens, the basic problem remains. In the spirit of the passage from Hobbes above, one might ask: If legislators do not respect our rights, who can force them to? The judiciary, you will say. But who can force the judiciary? This suggests that inherent rather than imposed stability might be important even if one does not accept Rawls’s association of justice with fair cooperation. Even without that association we have an important reason for offering public justification in order to facilitate genuine commitment to our laws and institutions by people who have different comprehensive doctrines. But I suspect that this reason may not go all the way to supporting Rawls’s ideal of public reason, at least as he presents it. To see why, I will briefly sketch Rawls’s description of this ideal and then note how focusing on mutual independence rather than cooperation might reshape that ideal. Rawls describes the ideal of public reason as realized when government officials and candidates for public office “act from and follow the 5

For discussion of this aspect of Kant’s view, see Ripstein (2009, 159–168) and Pallikkathayil (2017, 40–44).

166

Japa Pallikkathayil

idea of public reason in terms of the political conception of justice they regard as the most reasonable. In this way, they fulfill what I shall call their duty of civility to one another and to other citizens” (CP, 576). Likewise, citizens who are not in government realize the ideal of public reason when “they think of themselves as if they were legislators” and hold governmental officials and candidates for public office to the ideal of public reason (CP, 577). By requiring us to justify laws and policies through political conceptions of justice rather than through our comprehensive doctrines, the duty of civility facilitates moving beyond a mere modus vivendi and to an overlapping consensus. The duty of civility serves the aim of realizing fair cooperation by directing us to rely on principles and values that may be shared with those with differing comprehensive doctrines. But if we aim at mutual independence rather than cooperation, different strategies for sustaining genuine and not merely strategic commitment to our institutions may be appropriate. Rather than offering justification via political conceptions of justice, we might offer justification framed in terms of others’ comprehensive doctrines. This is something like the form of discourse Rawls calls “conjecture” (CP, 594). But he describes conjecture as directed at showing people how their comprehensive doctrines might allow them to endorse a political conception of justice, whereas I am suggesting finding support within others’ comprehensive doctrines directly for the relevant political institutions. Although this strategy does not provide a shared set of values from which to deliberate, it may nonetheless enable citizens with diverse comprehensive doctrines to be genuinely committed to those institutions. Thus, although inherent stability may remain an important desideratum even without Rawls’s focus on securing the conditions of cooperation, the mechanisms used to support inherent stability might end up looking somewhat different than those that Rawls emphasizes.6

6

I develop an account of different strategies for engagement with citizens who endorse different comprehensive doctrines in Pallikkathayil (2019); see also Eberle (2002).

Part III

Rawls, Ideal Theory, and the Persistence of Injustice

10

The Circumstances of Justice Erin I. Kelly

Political philosophy is shaped by an understanding of the circumstances of justice, which describe the need for justice. John Rawls draws the outline of his understanding of these circumstances from David Hume, who claimed that the instability of material possessions, along with their scarcity, is the chief impediment to social peace and harmony and that the security of possessions is the chief advantage of society. “Such possessions as we have acquir’d by our industry and good fortune,” writes Hume, “are both expos’d to the violence of others, and may be trasferr’d without suffering any loss or alteration; while at the same time there is not a sufficient quantity of them to supply everyone’s desires and necessities” (Hume 1978, 487–88). According to Hume, the virtue of justice is the remedy it provides to the problems of material scarcity and insecurity. The Humean circumstances of justice rest on a claim about the origins of political society in relation to human nature. Rawls’s description of the circumstances of justice, as we will see, are broader than Hume’s and emphasize the insecurity of important liberties and opportunities, as well as material possessions. Yet in his classic work, A Theory of Justice, Rawls shares Hume’s aspiration to generalize across time and place in philosophizing about justice through an appeal to the abstraction of a prepolitical state of nature. I will argue for a more historically sensitive understanding of the circumstances of justice than the one Rawls presents. I propose that the circumstances of justice should be understood to include contingent facts about the history of the society to be governed, ideally, by principles of justice. I argue that the parties to Rawls’s original position thought experiment should consider, in their deliberations, their society’s history of war and other forms of politically organized violence and domination, at least when that history has resulted in enduring forms of social inequality that call for rectification, on grounds of justice. I will take it for granted that social inequalities that are premised on norms of group-based inferiority call for rectification because they are fundamentally at odds with any reasonable theory of democratic forms of social organization. 169

170

Erin I. Kelly

Regrettably, history is replete with examples of social orders premised on the inferiority of racial, ethnic, religious, and sexual minorities, as well as women, the mentally ill, the poor, and the disabled. I am especially interested in forms of collectively organized violence and domination that are perpetuated through social and political institutions, or what Rawls refers to as “the basic structure” of a society. I will refer to institutionalized forms of violence and domination, and the social inequality they have produced, as structures of domination. My focus will be the racial caste system in the United States because it has been central to the economic and social order of the United States since the birth of the nation. My argument is that a theory of justice should be designed to address the significance and persistence of structures of wrongful subordination in a society’s history. Reckoning with a history of collectively organized structures of domination and subordination is typically delegated, by Rawlsian and nonRawlsian political philosophers alike, to the realm of “nonideal” or applied theory. The idea is that we formulate ideals of justice in theory, apart from our knowledge of a particular society’s history of individual and collective wrongdoing, and the distortions those structures may inflict on people’s motivations and concern for one another, and then use ideal principles of justice as a guide to real-world social change (TJ 1999, 215). I resist this approach and, instead, locate structures of domination within the circumstances of justice that motivate the theoretical task itself. I advocate this theoretical revision to emphasize that the social significance of collective wrongdoing and its legacy shape the problems of justice that the resources of political philosophy, including the rich resources of Rawls’s theory, can and should help us to think about.1 In short, I am proposing to reorganize the question to which Rawls provides insightful answers. When we revise the description of the circumstances of justice, we revise our understanding of the importance of the chosen principles. This is because our understanding of the circumstances of justice shapes our ideas about why justice matters. I will not take issue with Rawls’s two principles of justice. Instead, I emphasize that exploring the application of the Rawlsian “original position” thought experiment to the problem of remedying historically rooted wrongs enables us to add something to what Rawls has to say about the good of justice. One aspect of the good of distributive justice is 1

On the relevance of the circumstances of justice to the framing of problems of justice, see Murphy (2017b, 84). I have benefited from Murphy’s discussion of the circumstances of justice in transitional societies.

The Circumstances of Justice

171

that it provides a basis for moral repair in the wake of collective wrongdoing. *** I arrive at a more expansive understanding of the circumstances of justice by engaging with Rawls’s thoughts about the historical origins of political liberalism. According to Rawls, liberal conceptions of justice have three features: first, they are committed to certain basic rights, liberties, and opportunities, namely, he says, those familiar from constitutional democracies; second, they accord a special priority to these basic rights, liberties, and opportunities; and third, they are committed to providing all citizens with the resources people need to make effective use of their basic rights, liberties, and opportunities (PL, 6). Rawls locates the origins of these features of liberal thought in the history of war and violence, specifically, in early modern Europe. The American context differs in important respects, but common threads emerge, upon reflection. This is because group-based violence presents difficult challenges to the pursuit of justice as well as increasing its urgency. Though Rawls would not disagree, he does not accord structures of domination much attention in the theoretical representation of the problem of justice. Nevertheless, I believe his two principles of justice can address them. The generality we give up in framing the deliberative problem to incorporate concern with historical wrongdoing is recovered in convergence, across different social settings, on the selection of liberal principles of justice. In his later writings, Rawls emphasized that liberal conceptions of justice should be political in that they should not assume or require moral consensus among people of differing religious and secular values. Justice does not require moral consensus among people who disagree about questions of the good. Though members of a pluralistic democracy will disagree about matters of ethics and religion, they might be brought to share political values – in particular, those connected with the three defining features of liberal conceptions of democratic justice. Rawls proposes that a conception of democratic justice representing these values could be the object of an overlapping consensus. An overlapping consensus is a point of convergence among people’s diverse life plans – a module of shared political values that fits into a variety of reasonable life plans. Rawls’s interesting proposition might seem, however, to push the critical question back. How is an overlapping consensus possible? That is, how might those who endorse different and even conflicting moral and

172

Erin I. Kelly

religious views come to identify and commit to a common set of political values? Where could the will to identify this formula for peace and justice come from? Rawls’s answer to this question draws on historical experience: an overlapping consensus might emerge from the exhaustion of war, as it did in Europe after centuries of religious fighting. People who are tired of fighting have reason to identify common values. Out of the ashes of war’s destruction, a modus vivendi or peace of exhaustion might emerge. A modus vivendi might evolve into a collective commitment to a liberal core of political values. This argument locates the possibility of political liberalism in what is called transitional justice. Transitional justice describes a political setting in which members of a society lack a commitment to mutual rights and obligations. There is no social contract. People are not mutually respecting members of a cooperative scheme. They are survivors. They are recent enemies, injured and traumatized. They approach one another in grief and with mistrust and resentment. They lack a common sense of justice, but they have reason to find common ground. The ending of their war presents an opportunity, indeed, their circumstances make it necessary for them to forge a common future out of a destructive past. It is necessary for them to identify common values if they are to live together. The alternative is political splintering and the persistence of conflict and injustice or disintegration and abandonment through political secession of the possibility of common society. The necessity of reaching agreement does not, however, show how it is possible to achieve it. How might wounded peoples find their way forward together? Let’s return to Rawls’s thought. He writes, “the historical origin of political liberalism (and of liberalism more generally) is the Reformation and its aftermath, with the long controversies over religious toleration in the sixteenth and seventeenth centuries. Something like the modern understanding of liberty of conscience and freedom of thought began then. As Hegel saw, pluralism made religious liberty possible, certainly not Luther’s and Calvin’s intention” (PL, XXVI). In considering Rawls’s reference to “the long controversies over religious toleration in the sixteenth and seventeenth centuries,” we might think of Locke’s letter on toleration and other important intellectual contributions to liberal political thought. But we might also concentrate on the circumstances that gave birth to the liberal values that Locke and others debated. Here, I submit, we need a wider understanding of the circumstances of justice than Rawls presents. From the vantage point of history, we can see that the circumstances of justice include organized forms of domination

The Circumstances of Justice

173

and oppression. Rawls did not articulate this fully enough. His description is more individualistic. Following Hume, he describes the “objective” circumstances of justice to consist in resource scarcity, a rough equality of people’s physical and mental powers, their mutual vulnerability to attack, and the need to share geographical territory. These circumstances describe the plight of individuals in a state of nature. Rawls describes them as, “the normal conditions under which human cooperation is both possible and necessary” (TJ 1999, 109). He remarks that because individuals are similar in physical and mental powers, “no one among them can dominate the rest.” While this might be true, strictly speaking, it is also true that groups of people are capable and have organized together, with great commitment, to dominate other groups. This is the case even though the underlying instability of oppressive social structures leads over time to eruptions of violence and war. We should conclude from the long history of war and violence that the objective circumstances of justice include the familiar dynamics of group domination and subordination. In the United States, group-based domination includes centuries of slavery followed by civil war, Jim Crow segregation, and mass incarceration. These structures of domination have inflicted group-based harms on Black Americans, including the social stigma imposed on them as members of a formerly enslaved caste. The stigma of group-based harms damages what Rawls refers to as the social bases of self-respect. As Rawls recognizes, the social bases of selfrespect are important components of democratic equality. This implies that democratic inclusion demands a remedy to the harms of groupbased stigma. We might come to see group-based stigma as not merely an unfortunate consequence of slavery but a persisting social injustice. Democratic inclusion is a form of belonging. At several junctures in his theory, Rawls emphasizes the importance of the search for belonging. This search may, of course, reach beyond democratic membership. A religious quest for belonging connects human existence to a transcendent and eternal reality. By contrast, nontranscendent versions of the possibilities for belonging track the accomplishments of human society and, indeed, the value of humanity itself. Rawls describes the outer reaches of a secular search for human belonging in this way. To take up the conditions that are “reasonable in our conduct with regard to one another” would be, he says, “to view the human situation not only from all social but also from all temporal points of view,” which is, he says, “not a perspective from a certain place beyond the world, nor the point of view of a transcendent being; rather it is a certain form of thought and feeling that rational persons can adopt within the world.” He continues. Such persons would, “whatever their generation, bring together into one

174

Erin I. Kelly

scheme all individual perspectives and arrive together at regulative principles that can be affirmed by everyone as he lives by them, each from his own standpoint” (TJ 1999, 587). This mode of normative thinking, as Rawls so beautifully describes it, is a form of impartiality that displays sensitivity to place and time. It is responsive to the variations in human experience across geography and history. I submit that the passage from A Theory of Justice that I have just quoted can be seen as a highly abstract way of describing a search for reconciliation between people who stand in unjustifiably unequal social positions. Furthermore, one way to describe Rawls’s insight, though he does not put it this way, is to say that justice should be understood to include a remedy for group-based harms that will facilitate moral repair between people whose relative social standing has been marred by a history of collective wrongdoing. I turn now to what Rawls refers to as the “subjective” circumstances of justice. As Rawls describes them, the subjective circumstances of justice comprise conflicts of interest that arise from people’s limited knowledge, biased reasoning, selfish tendencies, and doctrinal differences, which span a diversity of philosophical, religious, social, and political views. I agree that these conflicts of interest are real and important, but they under-describe the historically familiar subjective circumstances that demand a just remedy. I have in mind not simply ordinary biased judgment, selfish tendencies, and philosophical disagreements but extraordinary efforts groups of people have made to erect economies and social structures aimed at the exploitation of other human beings. I would also include the deliberate efforts that members of dominated and exploited groups have undertaken to resist their oppression. These efforts appeal to the conscience of all persons who are willing, in good faith, to reckon with the past and its legacy. So, to sum up so far: we should recognize that the circumstances of justice include not only people’s present needs, limited resources, selfconcern, and conflicting beliefs but also their shared history. The circumstances of justice should be understood to call for a collective search for reconciliation in the aftermath of collective wrongdoing, including war and other forms of organized violence. Those who have not recently been at war do not share that experience of hardship to motivate their efforts. Seriously disadvantaged people, of course, feel the urgency of the need for change, but privileged people do not typically feel a sense of urgency, because they enjoy substantial benefits from social inequality. Still, persons of conscience among them often search at the very least for some rationale for what cannot be justified. This may lead to some confusion and miscommunication, of the sort lamented by Martin Luther King Jr., who remarked:

The Circumstances of Justice

175

Negroes have proceeded from a premise that equality means what it says, and they have taken white Americans at their word when they talked of it as an objective. But most whites in America in 1967, including many persons of goodwill, proceed from a premise that equality is a loose expression for improvement. White America is not even psychologically organized to close the gap – essentially it seeks only to make it less painful and less obvious but in most respects to retain it. Most of the abrasions between Negroes and white liberals arise from this fact. (King 1967, 8)

What King describes as a lack of “psychological organization” by whites to rectify unjustifiable inequalities should be counted among the circumstances of justice. There are other familiar obstacles to repair and reconciliation between persons of goodwill, beyond obvious differences in their socioeconomic interests. One obstacle is the sense of social dishonor and alienation experienced by persons who have been oppressed. King described this as the “bleakness of corroding despair” (King 1963, 1964, 92).2 The internalization of oppression is an obstacle to the just demand for redress. King found that this obstacle must be overcome and that it could be overcome through collective action. The strategy he advocated was the practice of “nonviolent direct action.” It surpassed his expectations. King was astounded by the power of the Montgomery bus boycott to express its participants’ “new found faith in themselves” (King 1961). The realization of a sense of self-worth through collective forms of political action unsettled the corrosion of despair with a demand for democratic equality. Collective political action is one strategy among possible others for countering the psychological harms of oppression. A second obstacle is found in the sense of loyalty that beneficiaries of historically rooted inequality often feel to one another and to their ancestors, namely, to the socio-ancestral group that has perpetuated forms of domination from which its beneficiaries have benefited. This may lead to a distorted sense of history. For example, a Confederate monument in front of the Lafayette County Courthouse in Oxford, Mississippi, reads, “In memory of the patriotism of the Confederate soldiers of Lafayette County, Mississippi. They gave their lives in a just and holy cause.” The back panel reads, “the sons of veterans unite in this justification of their fathers’ faith.” While this commemoration reeks of bad faith, there are other examples that may be harder to dismiss.3

2 3

King later revised this statement from “Letter from a Birmingham Jail,” to refer to an “abyss of despair” (King 1964, 92). Some have criticized the World Trade Center memorial and museum for remembering 9/ 11 with no attention to the history of US imperialism in the Middle East.

176

Erin I. Kelly

Consider these words, penned by Thomas Jefferson and inscribed on the walls of the Jefferson Memorial in Washington, DC: “We hold these truths to be self-evident, that all men are created equal, that they are endowed by their Creator with certain inalienable rights, that among these are life, liberty, and the pursuit of happiness. That to secure these rights governments, are instituted among men.” Over the course of his life, Jefferson enslaved more than 600 people on his estate. He fathered six children with Sally Hemmings, who was one of his slaves. The Jefferson-Hemmings children were enslaved by Jefferson. Memorialization produces a dilemma for persons of conscience and, indeed, for the conscience of a nation. It is a dilemma that involves choosing between justice and loyalty. The difficulty, in choosing justice, lies in how to recalibrate loyalties while maintaining a sense of authenticity. How can one acknowledge the flaws and injustices of one’s forebearers – as one’s forebearers – while maintaining a sense of their connection to oneself? How can one memorialize them without embracing their injustice? It is tempting to abstract from moral flaws in the thought and behavior of one’s forebearers and to focus instead on the better angels of their nature – to appeal, for example, to abstract principles of justice, equality, and freedom, which might serve to link the past and present of a society’s public political culture with the possibility of a better world. But this exercise in selection threatens to constitute an evasion of the historical specificity of how those principles were ignored, violated, misappropriated, and twisted into a weapons of degradation and war. When Thomas Jefferson wrote about the equality of all men, his philosophical position could not be squared with his political commitments. It simply could not be done. The universal rights and equality of “all men” were not compatible with the American institution of slavery. Democracy has always been understood, by its practitioners, as compatible in fact with the freedom of some to dominate others. This is the central argument in sociologist Orlando Patterson’s celebrated book on the history of the concept of freedom. The “freedom to dominate,” Patterson says, is one dimension of a tripartite, Western concept of freedom that emerged under conditions of slavery. Philosophically celebrated notions of “freedom from interference” and “the freedom of collective self-government” have always been historically conjoined with a third aspect of freedom – the freedom to dominate. Historical reckoning opens the possibility of an authentic conversion from loyalty to justice. Reckoning with historical injustice includes an examination of democracy’s contradictions and the various rationalizations, including ideas of racial hierarchy and other lies about inequality,

The Circumstances of Justice

177

that have been used to obscure them. Here I think Rawls can help us. The need to reckon with historical injustice presents an opening for us to learn from what Rawls taught us about how to think about justice. Rawls proposed the original position as a hypothetical choice situation that is constrained in ways that make the choice of principles under those constraints fair. The constraints on the original position, represented by the veil of ignorance, force each party to consider the interests and perspectives of all others as they consider possible solutions to the problems evident in the circumstances of justice. As Rawls presents those circumstances, the relevant problem is a problem about how fairly to distribute the benefits of social cooperation between free and equal persons. Since all cooperating members are understood to contribute to the collective scheme, the parties conclude that the gains should be enjoyed reciprocally. If we broaden the original position thought experiment to include information about group-based oppression in a society’s history and the harms that oppression has caused, this shifts the nature of the problem the parties seek to remedy with principles of justice. The parties would consider the impact of past wrongs on the disadvantaged over time. They would consider how the beneficiaries of those wrongs have evaded responsibility. They would take up the viewpoint of people who resisted wrongs committed against them and concede that they did so understandably and justly. They would understand that democratic justice not only produces a fair system of social cooperation in which people mutually benefit from the production and distribution of goods, but also that distributive justice between them is needed to achieve moral repair in the aftermath of historical wrongdoing. They would seek justice to underwrite the possibility of reconciliation between groups whose relationship has not been a relationship of reciprocity and mutual trust. They would understand that acknowledgment of past wrongs is a way to remediate the social stigma associated with groups who have been wronged. These ethical thoughts would be stimulated by requiring the rational and self-interested parties in the original position to consider their society’s history of group-based forms of oppression from behind a veil of ignorance that obscures their current social position. Entering the original position thought experiment would encourage a privileged person to find common ground with those who struggled against the oppressive forces that have benefited a select, undemocratic few, without requiring the privileged person to agree that he would have, had he been in his ancestors’ place, relinquished his unjustly achieved privilege. Reckoning with the past by considering the perspective of people who were wronged – how they felt and what they did in response to their

178

Erin I. Kelly

unjust treatment – permits people to stand together without denying differences of social position. It enables all to see clearly something that we know: that the universal truth of certain ethical principles – of justice and equality – is compatible with the abuse of those principles. We also see that the history of resistance to collective wrongdoing is a resource for all members of society. I have been describing, through the Rawlsian thought experiment, the possibility of conversion from a familiar defense of some version of the status-quo by the privileged to an imagined identification with people who have resisted oppression. This involves, I am suggesting, a projection of oneself into a (possibly ongoing) historical conflict in an effort to view the conflict from the perspective of persons who we know have been wronged. This imaginative exercise opens the possibility of granting their claim to reparative justice. Rawls argued that the original position thought experiment brings us to the two principles of justice. The two principles include, importantly, the requirement of fair equality of opportunity. Fair equality of opportunity is the demand that those who are similarly motivated and talented should have equal chances of success in attaining their desired social positions (JF, 44). As Tommie Shelby has persuasively argued, fair equality of opportunity must be understood as sensitive to a variety of obstacles, including group-based favoritism and status inequalities, that interfere with open positions (Shelby 2004, 1697–1714). An unfortunate litany of historical wrongs shows us that fair equality of opportunity can be threatened in multiple ways. It is threatened not only by how the better off may leverage their wealth for further advantages, as Rawls discusses, but also by how they maintain their social standing by withholding important forms of recognition from other social groups. In other words, they may behave in ways that undermine equality in the social bases of self-respect. Interference with the social bases of selfrespect unsettles fair equality of opportunity, not only by directly undermining the openness of positions but also by compromising the social support people normally rely on to develop their talents and motivation. A historically sensitive description of the circumstances of justice brings these dynamics into view. My revision of the Rawlsian circumstances of justice overlaps with a line of thought suggested by Charles Mills, who has been highly critical of the project of Rawlsian ideal theory. The idealizing abstraction of liberal theory is, as Mills sees it, an exercise in avoidance. Despite 350 years of slavery and 100 years of Jim Crow racial segregation in the United States, liberal political philosophies contain almost no discussion of racial injustice. Furthermore, liberal political philosophers have failed to interrogate

The Circumstances of Justice

179

the fact that the heroes of liberal political thought – Locke, who invested in African slavery, and Kant, the father of scientific racialism – articulated ideals of justice that were, to their minds, compatible with white supremacy. Mills thinks liberal theory’s failure to reckon with this apparent contradiction is the result of its misdescription of social reality. Despite the allegedly firm distinction between ideal and nonideal theory, Mills argues that liberal political philosophers elide the distinction between the ideal and the actual by conceptualizing political society as, fundamentally, a cooperative system. This erases the social position of members of society who are excluded from the benefits of social cooperation and exploited for white benefit (Mills 2017b, 29–39). Still, Mills does not abandon the framework of liberal political philosophy altogether. Instead, he proposes a version of the Rawlsian thought experiment described by a thinner veil of ignorance. The thinner veil – allowing the parties to the original position to know that they will emerge into a society “whose basic structure has historically been shaped by white supremacy” – would permit the original position to serve as a device of representation for arriving at principles of corrective justice (Mills 2017b, 213; see also Mills and Flickschuh 2018, 85). I follow Mills in favoring a thinner veil, but I differ with him on the need for additional principles of corrective justice. I am urging that we view Rawls’s two principles themselves as principles of corrective justice, properly interpreted as sensitive to the harm structures of domination have inflicted on the social bases of self-respect, as well as the material interests, of oppressed people. The thinner veil, which permits an expanded understanding of the circumstances of justice, brings into view the urgency of substantiating fair equality of opportunity by closing the wealth gap and repairing the social bases of self-respect.4 A historically sensitive original position helps us to understand that moral accountability is needed to address persisting inequalities that are the result of historical wrongdoing. It is especially important when the social bases of trust and mutual respect have been damaged. Accountability is achieved through public efforts by the state to address past wrongs and to deliver the promise of democratic equality. Rawls saw philosophy itself as a form of reconciliation and here is an example of its possible resources. Though this is not what Rawls meant by reconciliation, I submit that philosophical reflection, with hindsight, might lead us to the view that those who have suffered wrongs have a claim to reparative justice. Closing the wealth gap and ensuring the rights 4

In 2016, the median family wealth in the United States for whites was $171,000 and $17,600 for Blacks. See, e.g., Lee (2019).

180

Erin I. Kelly

and opportunities to which all members of society are entitled achieves reparative justice when and because it represents an effort to correct wrongful forms of domination that have shaped the social distribution of wealth and the dynamics of social status. My point is that public reasoning about justice matters. In particular, public efforts to acknowledge past wrongs are critical to achieving a just social order because they are needed in order to recalibrate loyalties among the privileged, to correct the harms of stigma wrongfully imposed on historically marginalized groups, and to repair the social bases of self-respect. My extension of the Rawlsian thought experiment permits us to view Rawls’s principles of justice in a new way: as principles of acknowledgment and redress. In turn, this understanding of fair equality of opportunity sheds light on the underlying good of justice. An important aspect of the value of distributive justice is that it offers the possibility of recognition, reparation, and reconciliation in the aftermath of historical injustice. I conclude with a thought about the Rawlsian notion of a “people.” The idea of a people threatens to stand for a troublesome romantic notion that is highly subject to ideological distortion. But it is less troublesome when it is rooted in history, which incorporates multiple perspectives. The notion of a people that is rooted in history is not a romantic ideal. Rather, it is a historical fact that must be confronted as a condition of hope for the future.5

5

For comments on an earlier draft, I am grateful to Vincent Chiao, Chad Flanders, Joshua Kleinfeld, Christopher Lewis, Cynthia Stark, Paul Weithman, and participants at a conference at the University of Notre Dame, “John Rawls’s A Theory of Justice at Fifty: An Anniversary Conference,” September 24–25, 2021.

11

Why Rawls’s Ideal Theory Leaves the Well-Ordered Society Vulnerable to Structural Oppression Henry S. Richardson

Inspired in large part by the work of the late Charles Mills, this chapter brings some of its lessons to bear on John Rawls’s ideal theory of justice.1 Mills called for abandoning ideal theory and instead developing nonideal theories of corrective and reparative justice (Mills, 1997). He also claimed that Rawls’s ideal theory simply did not apply to racist societies such as the United States. While I fully accept the need for theories of nonideal justice, I am convinced that Rawls’s view applied to the United States: it assesses the United States and condemns it as unjust.2 Here, however, I set aside these debates and instead argue that Rawls’s ideal theory is defective. The characterizations of social structure and of power in Mills’s work and in the Black Radical tradition on which he drew help reveal a fundamental flaw therein: Rawls’s ideal, well-ordered society lacks adequate bulwarks against structural oppression and structural domination arising within it.3 In my view, it is important to remedy this flaw, as ideal theory and nonideal theory work best in tandem.4 1

2

3

4

I am grateful to the students in my Spring 2021 seminar on Ideal and Nonideal Justice; to the participants in Sabine Döring’s Spring 2021 Oberseminar on practical philosophy at the Eberhard Karls University of Tübingen, in the panel, Where Are We at with Republicanism at the MANCEPT Workshops, September 9, 2021; and in the conference on John Rawls’s A Theory of Justice at Fifty, September 24, 2021, for very useful comments on earlier drafts. I am particularly indebted to conversations and correspondence with Paul Weithman, David Estlund, Rainer Forst, Erin Kelly, Andrew Koppelman, and Leif Wenar. For evidence that Rawls viewed US racial injustice as a serious problem, see Terry (2021). Terry also finds that Rawls was too unwilling to rock the boat in his negotiations with the Harvard faculty about these issues. Here avoiding psychological speculation, I argue (with hindsight) that Rawls’s ideal theory was insufficiently radical. I am differentiating oppression from domination roughly along the lines set out by Young (2011a, 37): oppression arises from social configurations that systematically and significantly limit certain groups’ capacities to live well; domination arises from social power imbalances that systematically and significantly limit some people’s ability to have a voice or a hand in determining the social conditions within which they act. I will here focus on dominant power as leading to oppression. In work in progress – the erstwhile second half of the present essay – I offer constructive suggestions about how Rawls’s ideal theory could be modified to remedy this defect.

181

182

Henry S. Richardson

Even if Mills would have been flattered by my taking inspiration from his writings, he surely would have found my setting about to improve ideal theory perverse. I am devastated that he is no longer with us either to rebuff or to improve these efforts – or to ignore them and carry on his own pathbreaking work.5 The long-lasting, pervasive, and multitentacled structural racism afflicting the United States can readily be recognized as unjust without appeal to any ideal theory. As Mills and others have argued, discerning its full horrors does require an awareness of history. Further, understanding its persistence may require considerable psychological and sociological sophistication.6 Plumbing the depths of its injustice may require innovations in nonideal normative theorizing.7 Nonetheless, as Tommie Shelby suggests, when structural injustice is grossly apparent, as I take the racial injustice in the United States to be, there is no need to invoke an ideal theory to call it out as unjust.8 Whether an ideal theory of justice is useful in determining what should be done about structural injustice in a particular place and time, such as the United States at this moment, is a more controverted question. Mills and Elizabeth Anderson argue that it is not (Anderson 2010, chapter 1) Shelby, to my mind, demonstrates that it is (Shelby 2016). Here, I assume that good ideal theory is potentially useful to nonideal theorizing, even if only in the somewhat indirect ways illustrated in Shelby’s work. In this chapter, I aim to pick up the relationship between ideal theory and structural injustice from the other end. I do so partly to honor John Rawls’s pathbreaking theory of justice and the fiftieth anniversary of his masterpiece, A Theory of Justice. The ideal theory set out and defended in that book insisted that its principles were fully universal in their application and must be fully general in their formulation (TJ 1999, 113–15). As a consequence, Rawls’s theory, “justice as fairness,” was to apply to all societies characterized by the circumstances of justice: moderate scarcity and a pluralism of conceptions of the good (TJ 1999, section 22).9 As our 5

6 7 8

9

Professor Mills was due to have participated with Tommie Shelby and me in the fiftieth anniversary conference’s September 24, 2021, session on Rawls, Race, and Non-ideal Theory. He passed away, still at the peak of his powers, on September 20, 2021. See, e.g., , Clark (1989); Tilly (1999); and Du Bois (2009 [1903]). See, e.g., Boxill (1992); and McGary (1999). I owe the distinction, in this context, between “gross injustices” and “less manifest injustices” to Shelby (2016, 13). As Shelby there indicates, what theoretical understanding is needed in order to settle what should be done to remove and correct a gross injustice is another matter. In Mills (2020), Mills relates as an “epiphany” his coming to understand that Rawls’s characterization of society as a “cooperative venture for mutual advantage,” taken

Rawls’s Ideal Theory and the Well-Ordered Society

183

own society exemplifies, societies in these circumstances can contain structural racial injustices that constitute serious oppression. Rawls’s ideal theory would thus apply to them.10 The question that I want to raise about Rawlsian ideal theory is whether satisfying its requirements would provide an adequate bulwark against structural oppression arising where it does not already exist. I will be assuming for the sake of argument that the Rawlsian well-ordered society – that is, the society well-ordered by justice as fairness – contains at the outset no significant structural oppression, whether on lines of race, ethnicity, class, or religion. This is an importantly substantive assumption, one not guaranteed by the features of a society well-ordered by justice as fairness.11 I will show that, even on this assumption, a Rawlsian ideal society is unduly vulnerable to new forms of structural oppression arising. I will show how and why serious structural injustice could well arise in a society well-ordered by justice as fairness. This concern can be seen as broadening Rawls’s pioneering focus on stability. In addition to enjoying “stability for the right reasons,” in which individuals’ conceptions of the good help buffer the well-ordered society from perturbations, the well-ordered society should also be well-buffered against incipient oppression. A proper ideal theory of social justice should include prophylactic principles that would inhibit its arising. Here, my thesis is a negative one: Rawls’s ideal theory is defective because it lacks sufficient bulwarks against new forms of bigoted oppression arising.12 The range of possible types of oppression is wide and, I fear, extends beyond those forms experienced in human history. Groups that have

10

11

12

descriptively, characterizes no nation yet seen. In his earlier work (2018), however, Mills allowed – I believe correctly – that this phrase might be intended as a normative description. Mills (2020, 21) notes Rawls’s statement that “from the start, then, we view a democratic society as a political society that excludes … a racist [state]” (JF, 21). Mills there takes this statement to imply that Rawls’s theory “does not apply to the United States.” To the contrary, as the preceding lines indicate, Rawls believes that a racist state is excluded on normative, not conceptual or stipulative grounds: such a state cannot support the “amity and concord” at which a democratic society constitutively aims. Assuming (plausibly) that the United States has what Mills calls a “racist basic structure,” Rawls’s ideal theory applies to it and finds it unjust. This assumption is an expository heuristic that my argument will end up calling into question. For arguments that a Rawlsian well-ordered society may have an unjust past, see Shelby (2004, 1711) and Kelly (2017, 77). Olúfemi O. Táíwò (2019) powerfully questions the empirical viability of demarcating _ domestic “basic structures” from transnational structures. I here abstract from this question and from discussion of Rawls’s much-disputed simplifying assumption that there is only one society. Whatever basic structure(s) exist need to include bulwarks against oppression arising to count as just.

184

Henry S. Richardson

been and are being subjected to structural oppression include the working class, women, LGBTQ+ people, and racial, religious, and ethnic groups. To simplify my exposition and to contribute to rectifying the long silence of Rawlsians – including me – about structural racial injustice, I will refer to the horrific structural racial injustice in the United States in illustrating how a Rawlsian ideal society lacks sufficient guards against structural oppression arising. We have some reason to hope that oppression as total, widespread, and long-lasting as the enslavement of African people and the ensuing oppression of their descendants in the United States and other countries will never recur. Today’s world would ill tolerate the unbridled imperialist exploitation or the extreme racist ideology it required to get going. I do not argue that anything like this could arise from a Rawlsian well-ordered society that starts out free of any oppression. Rather, my claim is that bigoted oppression could arise. We have seen uncountably many and varied such forms of such oppression in human history. To mention just two additional cases of structurally perpetuated oppression occurring today in democratic countries, consider the marginalization faced by the Roma in Europe and of Muslims in India. I recognize that any society well-ordered by the principles of justice as fairness would contain many provisions that would help prevent oppression from arising. The assurance of equal basic liberties to all will reduce the likelihood that anyone is dominated. The guarantee of the fair value of the equal political liberties will mitigate the danger of the political apparatus being hijacked by factions bent on oppression. The principle of fair equality of opportunity and the difference principle, working in tandem, will do much to keep wealth, and the power that comes with it, from being overly concentrated in the hands of a few over the course of generations.13 For the sake of argument, I will grant that all these effects hold. What, then, could go wrong? I will argue that the efforts by critics of oppression to grasp and to expose the nature of the structural injustices we are now facing also point us to two serious defects in Rawls’s ideal theory of justice. The first is his overly juridical conception of the basic structure of society. The second – a corollary of the first – is his overly moralized conception of power. The question about the basic structure is whether, as Rawls conceives it, it can do enough. Although I celebrate Rawls’s institutional turn as a pioneering and progressive move in 13

As Rawls pictures things, individual wealth will either be widely dispersed, as in property-owning democracy, or individual wealth will be minimized, as in liberal socialism (JF, Part IV).

Rawls’s Ideal Theory and the Well-Ordered Society

185

normative political theory, I will argue that recent discussions of nonideal theory help reveal that his conception of the basic structure of society and the corollary understanding of power are far too narrow. This negative point will be no news to anyone steeped in the Black Radical tradition carried forward by Mills and Shelby. Yet each of them has put this complaint to use largely in their development of nonideal theory. I seek to add a new twist by taking this negative point back into the heart of Rawlsian ideal theory. In what follows, I will argue that these two defects – the narrowness of Rawls’s conception of the basic structure of society and of his conception of social power – leave a society well-ordered by the principles of justice as fairness significantly vulnerable to new outbreaks of oppressive structural injustice. 11.1

The Narrowness of Rawls’s Conception of the Basic Structure of Society

Rawls defines the “‘basic structure’ of society” as comprising “a society’s main political, social, and economic institutions, and how they fit together into one unified system of social cooperation from one generation to the next” (PL, 11).14 The basic structure of society sets an important background for social cooperation; yet Rawls’s conception of the basic structure unduly narrows the scope of his inquiry. As he says in introducing the idea, “the scope of our inquiry is limited … I am concerned with a special case of the problem of justice. I shall not consider the justice of institutions and social practices generally” (TJ 1999, 7). G. A. Cohen, in complaining of Rawls’s selecting the basic structure as the first subject of social justice, described two consequential features of this narrowness. First, as is well known, in criticizing Rawls’s incentives-based argument for the difference principle, Cohen correctly emphasized that in defining the basic structure as institutional, Rawls excluded from “the primary subject of social justice” the individual economic choices made within the rules of these institutions and any ethos that may guide those choices.15 In addition, Cohen 14

15

Familiar though this idea has become, it is not transparent. Rawls’s distinction between “local justice (principles applying directly to institutions and associations) [and] domestic justice (principles applying to the basic structure of society)” (JF, 11) is helpful. This distinction between considering institutions and associations severally (as local justice does) and considering them jointly (as domestic justice does) clarifies that how they fit together into one unified system is the essential point. I am grateful to Sabrina Leeds for her pointed questions about this distinction. Or, more precisely, any ethos not arising from the basic structure. G. A. Cohen (2008) argued at length that Rawls’s conception of the basic structure of society offered too narrow a focus for a theory of social justice because it leaves out the ethos guiding

186

Henry S. Richardson

helpfully noted that the basic structure of society leaves out norms that are non-institutionalized. As he mentioned, these informal practices – “conventions and usages” – can sustain sexist gender oppression as well as racial oppression (Cohen 2008, 125).16 At the core of any society’s basic structure, as Rawls uses the term, is a society’s legal order. Rawls did not assert that the legal order exhausts the basic structure.17 Nonetheless, as he put it, “the law defines the basic structure within which the pursuit of all other activities take place” (TJ 1999, 207). Cohen expressed some doubts as to whether or not it is essentially a coercive legal order (Cohen 2008, 133, note 40). That doubt is well-taken but not for the reasons Cohen gave. Rawls saw the law as going beyond being merely coercive because it can also empower, permit, and grant immunities. This range of functions is reflected in Rawls’s definition of “an institution” as “a public system of rules which defines offices and positions with their rights and duties, powers and immunities, and the like” (TJ 1999, 47). Although the basic structure includes legal rules that are coercively backed, this is not an essential feature of all institutions belonging to a society’s basic structure. Rather, two other features are more important. The first is the publicness of the rules, a feature that implies that “a person taking part in an institution knows what the rules demand of him and the others. He also knows that the others know this and that they know that he knows this, and so on.” For a society-wide institution, this publicity must apply on a society-wide basis (TJ 1999, 48). The second feature is that Rawls defines institutions, and hence the basic structure of society, in terms drawn from the Hohfeldian tradition, which analyzes a legal system’s norms in terms of rights correlative to duties and legal powers correlatively held over those lacking specific immunities.18 These Hohfeldian relational concepts, which helpfully elaborate what it means to “follow the rules,” also shape Rawls’s narrow conception of power, to which I will turn shortly.

16

17

18

individual economic choices. His proposed remedy was not to broaden the idea of the basic structure but to reject Rawls’s suggestion that the first subject of a theory of justice should be the basic structure (Cohen 2008, 150). See also Kelly (2017, 90), emphasizing that discrimination can live on in “diffuse attitudes, habits, and practices.” Here, I argue that, by the same token, discrimination can arise in a Rawlsian well-ordered society via diffuse attitudes, habits, and practices. G. A. Cohen noted many passages in which Rawls described the basic structure without emphasizing or implying the coerciveness of the legal order (Cohen 2008, 133, note 40). In one of these passages (PL, 301), Rawls seems to treat the law as one of several component institutions of the basic structure. Rawls does not mention Hohfeld either in A Theory of Justice or, apparently, in his Collected Papers; but Kenzie Bok tells me that Rawls had learned about Hohfeld’s discussion of first-order rights and duties and second-order powers and immunities from H. L. A. Hart (private communication). See Hohfeld (1978, [1919]).

Rawls’s Ideal Theory and the Well-Ordered Society

187

One might think that while legal institutions well fit his definitions, other sorts of institutions, such as the family and private associations, might also enter into the basic structure of society as he conceives it. Yet this is not exactly what we see. To explain why not, I will first focus on Rawls’s understanding of “the regulative primacy” of principles of justice before turning to the cases of the family and private associations. Highlighting this idea will help explicate how, in Rawls’s view, all elements of any society’s basic structure are legally articulated. At the outset of his later lecture on “The Basic Structure as Subject,” Rawls explains that he takes the basic structure as “the first subject of justice” because of its “regulative primacy”: it governs how “the major social institutions … assign the fundamental rights and duties and shape the division of advantages that arise through social cooperation” (PL, 257–8, emphasis added). This conception of the basic structure enables Rawls to define the valuable notion of “background justice,” which seems useful to an account of structural justice or injustice and explains why it would not be adequate simply to set out a set of rules or principles applicable to individuals (PL, 267). The regulative primacy of the principles of justice applicable to the basic structure is well modeled by Rawls’s account of the “four-stage sequence” (TJ 1999, section 31). The first step of this hypothetical process is the original position, which selects principles of justice. The subsequent stages proceed, under a progressively thinning veil of ignorance, to hypothetical delegates to a constitutional convention, hypothetical legislators, and finally hypothetical judges who rationally and reasonably implement the principles. Implementing the idea of “regulative primacy” is the stipulation that each step in this process “inherits the constraints adopted at the preceding stage”: the principles of justice constrain the constitution, the constitution constrains the legislators, and legislation constrains the judges and administrators (TJ 1999, 176). As this sequence makes plain, Rawls’s principles are primarily intended to regulate a society’s laws and their interpretation. The basic institutions that they constrain and shape are legally defined and legally oriented ones. Against this suggestion, one may object that Rawls includes religious associations and the family within the basic structure of society (PL: 440–90; see also PL, 258 on the family). By including them within the basic structure, it might be thought, Rawls broadens it beyond legal institutions, modifying his view so as to address gender-based oppression by including informal “conventions and usages” as part of the basic structure. In response, while there would have been ways of including such informal norms and practices within the basic structure, somehow

188

Henry S. Richardson

understood, Rawls’s way of addressing the family and religious associations does not have this upshot. He insists that “the principles of political justice [do] not apply directly to the internal life of the many associations within it, the family among them” (PL, 468). Rather, he argues, families and other associations are subject to a number of external legal constraints. Churches may not prevent their congregants from leaving or punish them as apostates (PL, 468). Families may not violate the basic rights and liberties that are due to all citizens (PL, 469). These constraints derive from the principles of justice and bear on a family “as an institution” in ways incorporated into “family law” (PL, 469, 470). In other words, just as no official is above the law, neither is any religious association or family. It remains the law that embodies justice. I conclude, then, that Rawls’s pioneering conception of the basic structure, though useful in characterizing some forms of background injustice, is quite narrowly defined as including only legally constituted institutions and the legally constrained aspects of any other “major social institutions [that] distribute fundamental rights and duties and determine the advantages from social cooperation” (PL, 257–58). Of course, any such claim of narrowness is implicitly comparative. An excellent example of a broader or more inclusive conception of the basic structure is Mills’s, which he called “the real ‘basic structure’” (Mills 2007, 117). This conception adds to the economic and juridico-political elements on which Rawls focuses four further dimensions: the somatic, epistemic, cultural, and ontological (or personhood-defining).19 I will shortly turn to elaborating why the comparative narrowness of Rawls’s conception of the basic structure seriously limits its ability to address structural injustice. Before doing so, however, I will set out the corollary narrowness of the conception of power central to Rawls’s theory of justice. The dominant conception of sociopolitical power in Rawls’s A Theory of Justice is the Hohfeldian one, on which a power is an ability, deriving from a system of (moral, social, legal, political …) norms or rules, to alter a person’s rights, duties, liberties (i.e., lacks of duty), and no-rights.20 I base this interpretive claim on having reviewed all of the 19

20

This is the list given in Mills (2018, 85). See also Mills (2003). Mills’s cultural dimension could partially subsume G. A. Cohen’s concern with the ethos on which people act, namely insofar as it is culturally generated, as in his example of the “powerful democratic ethos that was formed in the experience,” in Britain, of World War II (Cohen 2008, 378). Powers may also operate to alter the higher order normative “incidents” (as they’re called): a person’s powers, liabilities (to being affected by such powers), immunities (against being affected by such powers), and disabilities (or lacks of power). For instance, a legislature may be empowered to establish immigration courts to adjudicate the claims of asylum seekers.

Rawls’s Ideal Theory and the Well-Ordered Society

189

places in A Theory of Justice where Rawls uses the word “power” and as many of the places in his other published works that I have been able to find. Since my focus is on group oppression and how to block it, I set aside Rawls’s frequent mention of individuals’ psychological powers as rational and reasonable beings. His most frequent mentions of sociopolitical power refer to the Hohfeldian legal powers of legislators, majorities, or holders of political office to effect legal changes. (TJ 1999, 172, 186, 190, 195, 197, 200, 211, 230, 261, 264, 296, 321, 336, 340f., 430f.) Tellingly, in the revisions for the revised edition, in four such instances he dropped the term “power” for either “authority” or “rights,” thereby more plainly indicating that he was talking about Hohfeldian normative power, not ordinary causal power (TJ 1971/1999, 64/56, 70/61, 93/80, 97/83). In the United States, legal power was exercised to authorize enslaving people, to authorize the recapture of escaped enslaved persons, to repossess the property of newly freed Black people, and effectively to nullify the political power of Black people at the ballot box.21 Much more recently, US statutes discriminatorily imposed severe penalties for crack cocaine possession and oppressive mandatory sentencing guidelines. Rawls’s theory is well positioned to decry these sorts of injustice or at least to support the efforts of those who aim to decry them. The fourstage sequence, properly carried out in its idealized way, would clamp down on such laws. Yet Rawls’s conception of sociopolitical power, insofar as it is developed in constructively laying out his own theory, is unduly narrow in a way that parallels his legalistic conception of the basic structure of society. It misses at least two other aspects of sociopolitical power. First, by focusing on the sort of power that depends on legitimate authority, it misses the simple sort of de facto power described by Thomas Hobbes, which flows from having “riches, reputation, [and] friends” – particularly, we might add, well-placed friends, such as friends in high office (Hobbes 1985, I.10.2). Rawls does occasionally mention economic power in connection with the fair value of the political liberties and in one place speaks more broadly about the need “to forestall accumulations of power and property” (TJ 1999, 246; see also 245 for the same point).22 Tantalizingly, he once refers to a feeling of powerlessness as the 21

22

David Walker gave a vivid description of the ruthless legal maneuvers used in the early nineteenth century to separate freed Black people from their property; see Walker (2015 [1830], 9–10). For twentieth-century examples, see Ture and Hamilton (1992, chapter 5). Other passages refer more broadly to the power flowing from investment (TJ 1999, 263), the importance of discrepancies of bargaining power (TJ 1999, 251), and the fact that markets help to disperse economic power (TJ 1999, 241).

190

Henry S. Richardson

main root of envy (TJ 1971/1999, 535/469) but he does not develop the sort of broad account of power that might help explain this sense of powerlessness as something other than an illusion. Second, since Rawls concentrates on public, legal norms, he underrates the importance of the normative powers that are supported, not by the society-wide and institutionalized public norms, such as those of the law, but by the norms – often informal ones – of dominant in-groups.23 Third, Rawls ignores the constitutive power of social discourse and its shifting stock of concepts and presuppositions.24 The narrowness of the account of power that Rawls puts to work in justice as fairness exacerbates the narrowness already noted in his conception of the basic structure, which leaves aside important dimensions on which dominant social power can be accumulated and then exercised with oppressive effect. His conception of power additionally prevents the theory from being sensitive to these concentrations of power as such. Let me further develop each of these mutually reinforcing facets of Rawls’s theory’s vulnerability to new forms of oppression arising, starting with the narrowness of his conception of the basic structure. As G. A. Cohen incisively remarked: “One reason why the rules of the basic structure … do not by themselves determine the justice of the distributive upshot is that, by virtue of circumstances that are relatively independent of [legal] rules, some people have much more power than others to determine what happens within those rules” (Cohen 2008, 138).25 How, specifically, can such disparities of power arise? Rawls was concerned that a society well-ordered in important respects might devolve into a society with sharp class divisions.26 Yet he argued that the difference principle, causally working in tandem with the principle of fair equality of opportunity, suffices to put serious limits on the accumulation of wealth. We may suppose that the difference principle functions to 23

24

25 26

A wide range of other theorists have developed conceptions of social structure and social power that extend it well beyond legal structure and legal power. Consider, for example, Foucault (1983), MacKinnon (1989); and Young (2011b), in addition to Charles Mills. On such constitutive power, see, in addition to the sources cited in the previous note, Haslanger (2012). It is somewhat ironic that Rawls neglects this, since he had put forward an account of constitutive power in his 1955 article, “Two Concepts of Rules” (CP, 20–46). I conjecture that he had not forgotten the point but later viewed its relevance in a positive light, as helping solidify the educative role of his principles (see, e.g., JF, 56) and did not sufficiently attend to how competing cultural forces could pull the other way. Based on arguments given above, I omit Cohen’s mention of the law’s being coercively enforced. Rawls considered such a possibility in discussing the meritocracy objection to the principle of fair equality of opportunity in A Theory of Justice (TJ 1999, 91–92). I discuss Rawls’s response to the meritocracy objection in Richardson (2021, 157–58).

Rawls’s Ideal Theory and the Well-Ordered Society

191

reinforce the principle of fair equality of opportunity in this way. Yet in a society well-ordered by justice as fairness, the serial priority of the principles of equal basic liberties and of fair equality of opportunity implies that the difference principle addresses only disparities in “the rights and prerogatives of authority and income and wealth” (TJ 1999, 80).27 Given that his conception of power centers on Hohfeldian legal powers, I think it’s fair to assume that the authority to which he is referring is legal authority, which is capable of reshaping public, legal institutions. These are important resources that support social power. The difference principle’s remit, however, leaves out additional, crucially important resources that have their home outside the legal system. The narrowness of Rawls’s conception of power shows, for instance, in his neglect of influential forms of social and cultural capital.28 There are many reasons that those who attend elite colleges generally have higher income expectations than those who do not; but one of these arises from the social capital embodied in their network of alumni contacts, often actively fostered by the colleges in question. Fairly admitted, they may have been; but these networks accelerate their advantages beyond the level of mere admission and give their members a lifetime leg up. Insofar as these informal networks are used to gain a position in graduate school or a job, they will violate the principle of fair equality of opportunity. Yet not all positions can come under the control of this principle, for the priority of liberty will rule that out. For instance, Rawls’s first principle of equal basic liberties will protect the Catholic Church’s limitation of priest’s positions to males and could also protect certain private associations’ rights to limit their memberships in various ways, so long as their grounds are not unreasonable in the Rawlsian sense. For now, the broader point is that there are important sources of power that Rawls’s principles will not regulate. Cultural capital is another one of these. Anderson characterizes one element of cultural capital as “consist[ing] of cultural habits acquired by adaptation to the social environment” (Anderson 2010, 35). She observes that “facility with elite leisure activities such as golf, familiarity 27

28

The principle of fair equality of opportunity does allow opportunities for positions to vary with people’s talent. To simplify the problem of indexing advantage in interpreting the difference principle, Rawls assumes that people’s access to positions of “authority and responsibility” is strongly correlated with their income and wealth (TJ 1999, 83). In Justice as Fairness (JF, 59) – perhaps relying again on this correlation – Rawls seems to narrow the remit of the difference principle to addressing inequalities of income and wealth. I am grateful to Megan Ritz and Paul Weithman for having called these complexities to my attention. One could also view the narrowness of the difference principle as a narrowness about the nature of advantage.

192

Henry S. Richardson

with cultural reference points that are the basis for small talk with coworkers, mastery of elite norms for circulating among people at a company party, and similar cultural skills have no inherent economic value [but] grease the wheels of commerce merely in being shared among those who make business decisions” (Anderson 2010, 37). Some exposure to these elite cultural norms can be gained by attending an elite college. Now, Anderson is of course discussing our far-from-ideal society. Yet just as new forms of bigotry might arise in an ideal society, so, too, might new forms of elitist snobbism. Obviously enough, access to both social capital and cultural capital enhance one’s power in Hobbes’s sense: one’s ability to get what one wants. Slightly less obviously, Hobbesian power of these types typically becomes bound up with Hohfeldian powers of an extra-legal kind. This is easiest to see in the case of cultural capital, which is central to the sort of oppressive bigotry on which I have chosen to focus. Cultural power can breed elite snobbism; and with elite snobbism can come Hohfeldian powers grounded in subgroup norms, including the powers to snub, to disinvite, to blackball, or otherwise to exclude from privilege. When people adapt to a social environment, such powers and the subgroup norms to which they give rise, embodied in social clubs and groupings, both formal and informal, are an important part of that to which the newcomers must adapt. Conceptually adjacent to these purely informal social interactions are interactions with private businesses such as banks. Banks are legally regulated, but they also have their own internal, extralegal norms. These latter may allow problematic levels of discretion. Well-known vectors of racism in the United States have been the bank officers who, not liking the look of someone who has come asking for a loan, have been empowered by their employers simply to say – overtly as an assertion, but really as a veiled, lying refusal – “I’m sorry, but I can’t do anything for you.” Note that the bank officer may have considerable discretionary power of this sort even if the bank is subject to fairness-in-lending statutes. We know this from the history of litigation against banks for racial discrimination in lending.29 While it is possible for lawsuits complaining about statistically significant patterns of racial bias to prevail, such verdicts generally do not sanction any particular bank officer. As Erin Kelly has

29

This kind of banker’s discretion showed itself in the discriminatory way that, in the first decade of this century, Wells Fargo employees intentionally and covertly targeted Black and Latinx families in Baltimore for subprime (i.e., needlessly expensive) mortgages. See Broadwalter (2012). It was the Bank that had to compensate for these wrongings, not the individual bankers.

Rawls’s Ideal Theory and the Well-Ordered Society

193

pointed out, because structural racism can live on in “diffuse attitudes, habits, and practices,” merely assuring that the basic structure of a society is just in Rawlsian terms will not suffice to root out such oppressive racism (Kelly 2017, 90). By the same token, I am arguing, ideal theory as Rawls sets it out, with its legalistically defined basic structure, will fail adequately to prevent new forms of racism and bigotry – and indeed, as I shall shortly argue, oppression – from arising. It may be objected that I am neglecting the radicalism of Rawls’s principle of fair equality of opportunity, which requires that those with equal natural talents have equal chances at success in attaining positions in colleges and university, jobs, and political positions. Here, the objector may reproduce Shelby’s 2004 line of reasoning: If this is the required outcome, then “it seems clear that [the principle of fair equality of opportunity] would require, at a minimum, considerable redistribution of wealth, the expansion of educational and employment opportunities, and aggressive measures to address discrimination in employment, housing, and lending” (Shelby 2004, 1711). This is a glass-half-full view. My glass-half-empty response is that since the principles of justice as fairness apply only to the basic structure, and since Rawls defines the basic structure legalistically, the reach of the principle of fair equality of opportunity cannot extend beyond the legalistic operation of the four-stage sequence. That being so, its effectiveness remains vulnerable, as Kelly argues, to being defeated by existing diffuse racist attitudes, habits, and practices that have a social basis independent of the law. Moreover, new forms of bigotry that might well arise in a Rawlsian well-ordered society could have the same result. This effect could be exacerbated even by legally regulated functionaries such as bank officers. The largely unregulated interactions of people in the social and cultural networks of everyday life pose even greater dangers. Elements of Rawls’s own discussion of fair equality of opportunity equivocally lend support both to a fuller and to a more restricted way of reading this principle. The section of A Theory of Justice that introduces the idea of pure procedural justice (section 14) begins with a ringing statement that if positions – and, we might add, bank loans – “were not open on a basis fair to all, those kept out would be right in feeling unjustly treated” (TJ 1999, 73). That seems correct; but does Rawls’s principle of fair equality of opportunity carry this thought through to the retail or micro level? No: he immediately goes on to explain that “the role of the principle of fair opportunity is to ensure that the system of cooperation is one of pure procedural justice” (TJ 1999, 76). This implies that this principle will operate only at a structural level, working only in the background so as to settle legitimate expectations rather than intervening

194

Henry S. Richardson

with micro transactions.30 Attractive as a principle of equal opportunity applying directly to individual outcomes might be, seeing Rawls’s principle that way conflicts with his central claims that the basic structure is the subject of his theory and that his principles of justice operate in the background. Rawls’s juridical understanding of the basic structure of society and his corresponding conception of power restrict the sensitivity of his liberalism to new dangers of oppression and bigotry. We now must examine how this insensitivity undercuts his principles’ ability to keep these dangers at bay.

11.2

How a Rawlsian Well-Ordered Society Is Vulnerable to Fresh Oppression Arising

There have been oppressed groups for all of human history. Oppression perpetuates itself, morphs, and generates oppressive counter-reactions. I continue to assume that a Rawlsian well-ordered society contains no oppression to start with. How could it open the door to new oppression, while still technically remaining well-ordered? To sketch how structural oppression could newly arise within a Rawlsian well-ordered society, I will focus on the implications of Rawls’s narrowly legalistic conception of power. As we’ve seen, this conception of power pays little attention to resources other than money and political office and ignores the kinds of power that informal norms can establish. Let me spin out a hypothetical, in three stages, illustrating one way that oppression might easily arise in a Rawlsian well-ordered society. My aim, here, is not to set out a hypothetical trajectory leading all the way from a Rawlsian well-ordered society to slavery or apartheid. I am concerned, here, with the need for an ideal theory adequately to put in place means to prevent any serious slide in that direction. At a first stage, although there is no oppression, one might find that many of the highly educated and successful network very well among themselves, while simultaneously – exercising their basic liberties – 30

Whereas Rawls’s difference principle applies to the relative expectations of relevant social groups, which – as Rawls briefly indicates – might include groups disadvantaged by their race or sex (TJ 1999, 84–5) – the principle of fair equality of opportunity applies directly to individuals. Of course, if it factored through, beyond what is settled by the basic structure, to individual outcomes, the individualized version of the principle would, if fully satisfied, also imply satisfaction by any group-based version of the principle; but if I am right that the principle applies only to what the basic structure can affect, this implication cannot be presumed to hold. The basic structure constrains outcomes but does not fully determine them.

Rawls’s Ideal Theory and the Well-Ordered Society

195

avoiding networking significantly with others, thereby hoarding some cultural capital and potentially magnifying the effect of their educational attainments. The difference principle will limit the growth of their economic wealth intergenerationally; but Rawls’s principles have no principle limiting their accumulation of cultural capital. This cultural capital could easily renew their economic wealth with each successive generation. In this way, they are very likely to become able to dominate other groups that cannot match their power; but at this stage, they are not yet using that power to oppress anyone. It would not be hard to imagine some elitist snobbery creeping into this doubly successful group. If it happened, that would take us to the second stage. Borrowing categories from Sally Haslanger’s critical theory, such attributes might show up in their unconscious habits, selfunderstandings, and their identification with the group’s norms (Haslanger 2012, 8).31 Perhaps if the moral psychology of their wellordered society worked as A Theory of Justice imagined it would work, their public acceptance of justice as fairness and its justification would undercut any deep snobbism; but Rawls was, I believe, correct in concluding that the moral psychology assumed in that book needed to be weakened in order to accommodate the fact of reasonable pluralism. Since the members of our imagined successful group thus need not all be “little Rawlsians,” and instead will (in the presumed well-ordered society) support justice as fairness on the basis of disparate comprehensive doctrines, some cultural variation is allowed. Cultural snobbism that directly contradicted viewing citizens as free and equal would be ruled unreasonable; but my hypothesized divergence might begin less committally, with self-centered cultural norms that demarcate the in-group’s distinctive characteristics and celebrate its supposed superiority. This second stage may sound rather innocuous so far; but history shows that an in-group with superior power and a sense of its own superiority is a potent and dangerous social force. As Max Weber trenchantly remarked, when people “differ in their … health or wealth or social status or what not[,] simple observation shows that … he who is most favored feels the never ceasing need to look upon his position as in some way ‘legitimate,’ upon his advantage as ‘deserved’, and the other’s disadvantage as being … the latter’s ‘fault’” (Weber 1978, 953). Because the differential power and status in our hypothetical has arisen non31

Haslanger’s expounds her well-known self-reinforcing loop set out there as beginning with roles and proceeding to norms, then symbolism, and then identities. That fits her concern with existing unjust role divisions. Since my aim is to describe how unjust role divisions might newly arise, I enter the loop at a different point.

196

Henry S. Richardson

justly, the public conception of justice (justice as fairness) can neither give the in-group the validation it wants nor reconcile the others to the ingroup’s increasing dominance. Hence, as time passes, it is highly likely that the in-group will come to harbor animus either against those outgroups that fail to recognize in word or deed that the in-group is superior or that the in-group’s advantages are deserved. Which of these routes to nastier interaction is taken will depend on accidents of history. Economic downturns might make the less affluent more strident, demanding, and resentful, leading the powerful in-group to adopt defensive discriminatory attitudes. Alternatively, the well-ordered society’s assurance of the fair-value of the political liberties might come to frustrate our snobby ingroup, spurring them falsely to attribute unfair maneuverings to others. In today’s political climate in the United States, such a breakdown of mutual trust and respect is not hard to imagine. As Rawls notes, “every society contains numerous unreasonable doctrines” (PL, 441n.; cf. 488). Hence, it would not immediately knock the society out of its wellordered status for our successful group to develop bigoted doctrines, so long as they can still sign on to the society’s liberal basic structure. Although this is worrisome, this second step does not yet constitute oppression. Oppression, on my story, would arise only in a third stage, in which the now more openly prejudiced in-group flexes its muscles. Still compatibly with the assumption that they have been within a society otherwise wellordered by justice as fairness – given the above-noted narrownesses of its defining terms – we can imagine the group leveraging the combined success of its members, achieved within the bounds allowed by the difference principle, so as to reinforce its accumulated forms of economic and cultural power. Recent theorists of injustice have elaborated a plethora of pathways whereby this group may take advantage of its dominant position. An initial way that it might do this, even without explicitly trying to do so, would be to use its economic power to throw around its cultural influence in ways that wreak epistemic injustice (Fricker 2007). This could start by taking the form of their instinctively belittling the credibility of anyone not part of the in-group, thereby starting to shunt the outsiders into subordinate roles. In the minds of those within the in-group, this would start to build stereotypes about those outside of it. In the minds of those less-advantaged on the outside, it might begin to lead to a further level, that of hermeneutic injustice.32 This might arise in part because the outsiders come to identify, under social and cultural

32

See Fricker (2007, chapter 7) and the literature to which it has given rise.

Rawls’s Ideal Theory and the Well-Ordered Society

197

pressure, with the subordinated roles to which the powerful in-group has largely relegated them. As the dominant group moves toward hegemonic cultural status, those in other groups may come under its influence to the extent that their own self-interpretive resources begin to erode. That, thirdly, would pave the way for the most insidious form of nonmonetary power, which is variously described as power over discourse (Foucault 1978), concepts (Haslanger 2012), or self-understandings (MacKinnon 1989; Mills 1997). Since these shifts are taking place in informal norms and understandings not part of the basic structure of society, as Rawls has defined it, his principles will not block them. To be sure, at some point in this third stage, the society will cease to be well-ordered by justice as fairness; but that result underlines my point. The narrowness of Rawls’s conceptions of power and of the basic structure of society do not merely leave his well-ordered society vulnerable to such oppression arising: in addition, that oppression can be sufficiently serious as to drag down the features that had made it well-ordered in Rawls’s terms. Once this three-step process has run its course, it seems likely to yield group oppression, perhaps of the sort that Young calls “cultural imperialism.” As she explains, “to experience cultural imperialism means to experience how the dominant meanings of a society render the particular perspective of one’s own group invisible at the same time as they stereotype one’s group and mark it out as the Other” (Young 2011a, 58–59). By such a progression, a society that began as well-ordered by justice as fairness will have degraded into one marked by religious, ethnic, classist, or racial oppression. As my schematic hypothetical illustrates, Rawls’s conceptions of the basic structure of society and of power reveal not merely theoretical defects but also practically worrisome vulnerabilities. 11.3

Conclusion

In this chapter, I have not claimed that a Rawlsian well-ordered society could give birth to a form of oppression as bad as the racial oppression now existing in the United States, let alone that of the centuries of mass enslavement that preceded it; rather, my point is that an ideally just society needs to be better resistant to oppression newly arising than this. My hypothetical scenario – representing just one pathway whereby new oppression might arise within a Rawlsian well-ordered society – suffices to indicate that even such an ideal society is vulnerable to new forms of oppressive bigotry arising. Rawls’s ideal theory is vulnerable in this way because of his having ignored types of capital beyond the political and economic and his having failed to address the important power lurking in

198

Henry S. Richardson

systems of informal norms and constitutive conceptual understandings. As I argued at the outset, it is a serious defect in a society well-ordered by a given ideal theory to be significantly prone to degrading into a society blighted by oppression. It is not easy for even an ideal liberal society to be more resistant to new forms of oppression arising; but my arguments suggest that any credible attempt to address this issue would need to start with a less legalistic conception of the basic structure than Rawls’s and a broader conception of social power. The upshot, in fact, is worse than this. It is time to revisit my working concession that a Rawlsian well-ordered society starts out without containing any oppression. We are now in a position to see that this assumption is wholly unrealistic. Just about every existing society contains some significant level of oppression – class-based, gender-based, sexual preference–based, racially based, or ethnically based, if not some of each. Since, as I have argued, Rawls’s principles are not adequately sensitive to oppression, there is no reason to think that now rendering a society wellordered by justice as fairness would erase all of the oppression inherited from its past. Thus, we cannot expect any Rawlsian ideal society to be free of oppression. Ideal theory, I am convinced, is crucially helpful to orienting attempts to promote social justice – if only in the indirect and contextually mediated ways explored by Shelby. Yet we should want an ideal theory whose principles effectively exclude oppression.33 Rawls’s theory of justice made great progress, but it did not get that far.

33

Seeking such an ideal theory is by no means incompatible with pursuing theories of reparative justice and other kinds of nonideal theories.

12

Race, Reparations, and Justice as Fairness Tommie Shelby

Although John Rawls offered an elaborate and wide-ranging account of justice in A Theory of Justice, he did not directly address well-known problems of racial injustice. His efforts were largely devoted to explaining and justifying the principles that would regulate a fully just society, what he called a “well-ordered society.”1 Rawls was convinced that ideal theory was the most fundamental part of a theory of justice. He thought we needed first to establish what justice requires before we could gain a systematic grasp of the principles that should guide our responses to injustice (TJ 1999, 216). The parts of nonideal theory he did discuss were limited to “nearly just societies” and focused on the political morality of resistance, international relations, and just war (TJ 1999, 319–43; LP, 89–120). Rawls’s ideal theory would be seriously defective if its principles did not condemn and rule out racial injustice. But they do. The veil of ignorance prevents the parties in the original position from selecting principles that would favor one racial group over another. Rawls explicitly states that all racial groups possess the moral powers necessary to be full and equal participants in a just society (TJ 1999, 443). Justice as fairness, the name Rawls gives to his theory, is obviously incompatible with slavery, genocide, colonial subjugation, land expropriation, political disfranchisement, lynching, segregation, and racial discrimination in education, employment, housing, law-enforcement, and lending. Elsewhere, I have argued at length that justice as fairness is also incompatible with institutional racism, ghetto poverty, and mass incarceration (Shelby 2016).

1

“Now let us say that a society is well-ordered when it is not only designed to advance the good of its members but when it is also effectively regulated by a public conception of justice. That is, it is a society in which (1) everyone accepts and knows that the others accept the same principles of justice, and (2) the basic social institutions generally satisfy and are generally known to satisfy these principles” (TJ 1999, 4).

199

200

Tommie Shelby

Yet it is less clear what justice as fairness has to say about how we ought to rectify past racial injustice. Rawls provided little explicit guidance about how we should think about reparative justice claims for racial oppression. Since he did not address the question directly, some might conclude that the issue of reparations cannot be taken up productively within Rawls’s framework or that Rawls thought the question unimportant. Charles Mills (2017b, 2018), for example, has famously argued that the pressing need to rectify racial injustice cannot be adequately understood or defended within the framework of Rawls’s theory. Mill’s arguments are provocative and challenging but, I believe, can be met. In this chapter, I argue that reparations for racial injustice can be defended within the framework of justice as fairness. In previous work, and specifically in response to Mills, I suggested that reparative justice is compatible with Rawls’s theory but did not attempt to justify this claim (Shelby 2013). I have also speculated that if African Americans had the full benefits of a just society, in particular the advantages that come with equal basic liberties and fair equality of opportunity, few would demand reparations (Shelby 2004). Yet I have said little about whether such a demand would nonetheless be justified. So here I offer a more extended discussion of the political morality of reparations and its relation to racial justice.

12.1

Reparations and Rawls’s Framework

I will situate the familiar concerns about race and reparations within a broadly Rawlsian framework. By a “broadly Rawlsian framework,” I mean, principally, Rawls’s methods of justification (agreement in original position, reflective equilibrium, and public reason).2 I mean his two principles of justice with their priority rules. I have in mind the centrality he gives to the basic structure of society. The account of reparative justice must be compatible with his deontological (or non-teleological) approach to political morality. And it must properly observe Rawls’s distinction between ideal and nonideal theory. I think the idea of reparations for racial injustice coheres with all these ideas. This doesn’t necessarily mean that, say, land restitution to Native Americans for settler colonialism or reparations to African Americans for slavery can, in the present moment, be justified on Rawlsian grounds. Showing that reparative justice for past racial injustice can be justified, in principle, within the framework of justice as fairness is not the same as showing that any specific reparations claim can be so 2

On Rawls’s methods of justification, see Scanlon (2003).

Race, Reparations, and Justice as Fairness

201

justified. Reparations for racial injustices of the distant past – say, more than three generations ago – pose well-known special problems of justification within any theory of justice, not only justice as fairness. Rawls insists that his two principles are not merely standards for evaluating institutions and social systems but represent goals of reform or revolution that individuals, social groups, and state agents should try to realize (TJ 1999, 215). This might involve writing, rewriting, or amending a constitution and instituting various laws and policies that bring the basic structure more in line with the principles of justice. The point is to create a just basic structure that is stable from one generation to the next and that can garner the warranted allegiance of those who are born and live their lives within the social order. The four-stage sequence of application could be a practical guide (TJ 1999, 171–76). But Rawls’s theory seems most useful for identifying abstract goals of institutional or legal reform and specifying the general duties of individual citizens and obligations of government officials. On its face, it does not give much guidance about how to respond to injustices of the past. In A Theory of Justice, Rawls twice mentions that he considers “compensatory justice” to be part of nonideal theory (TJ 1999, 8, 309). Yet he doesn’t explain what he had in mind by that phrase. He does explicitly invoke the language of “reparation” when discussing the moral psychology of citizens within a just social order (TJ 1999, 411–12). And these brief remarks are a helpful clue to how reparative justice fits within justice as fairness. Rawls tells us that in a just society members develop robust bonds of civic friendship and mutual trust, bonds that tie them securely to their system of cooperation. When these bonds are established, the members of the society feel a sense of guilt when they fail to comply with public rules of justice and when they fail to do their fair share in maintaining the social system of cooperation. Such feelings of civic guilt, as we might call them, exhibit themselves as a familiar set of moral dispositions. First, those who wrong others seek to repair the harm they have caused. It is this repair of harmful wrongdoing that Rawls calls “reparation.” There is also the disposition to acknowledge that one has treated others wrongly. Persons who wrong others feel compelled to apologize for having done wrong. They agree, says Rawls, that censure and perhaps punishment would be appropriate responses to this wrongdoing. They also find it more difficult to be indignant when others fail to act justly (presumably because they would then feel like hypocrites). When these moral dispositions are absent, Rawls claims, this suggests that bonds of civic friendship and mutual trust are also absent or severely damaged. This brief discussion is suggestive. Yet it takes place within the development of ideal theory and Rawls’s early defense of the stability of a well-

202

Tommie Shelby

ordered society. It is not readily apparent what implications, if any, these remarks have for the cases that mainly concern us. For example, what is their relevance for thinking about wrongdoing that occurs in a highly unjust society, a society that is not “nearly just” by Rawls’s standards? What implications, if any, do these ideas about interpersonal wrongs have for wrongs that occur because the society is itself unjust? What do they mean for the attitudes citizens should have toward past injustice during a period of transition to a more just social system? And what is their significance for wrongs committed by one nation or people against the members of another society? However, Rawls could and probably should say, at a minimum, that these dispositions – toward reparation, acknowledgment, apology, and self-reproach – would be present, at least in embryonic form, among citizens sincerely committed to creating and maintaining a just society. Citizens so committed would seek to make amends for the harm caused by the unjust social structure. They would publicly acknowledge the injustices of the past. They would recognize that an apology to the victims of these injustices is fitting. And they would accept that some penalty or sacrifice (particularly for those directly involved in the injustice) is an appropriate response to these past failings. Rawls introduces several principles besides his two main principles – for example, the principle of redress, the principle of the common interest, the principle of fairness, the principle of paternalism, the principle of equal political participation, the principle of political settlement, and the principle of self-determination. This gives us another reason to think that there might be room for a principle of reparation within justice as fairness. However, such a principle would presumably be part of nonideal theory. Thus, the case for a principle of this sort must be like the ones he made for civil disobedience and just war, the most developed parts of his nonideal theory. Rawls can easily accommodate a reparation principle for interpersonal wrongs or for wrongs committed by a private organization against an individual or group, provided these wrongs occur within a basically just society. The familiar principles of criminal law and tort law would apply in a just constitutional democracy. A just state would be bound to enforce such rightful claims. The question is whether Rawls can also accommodate a reparations principle that applies to an unjust action when, at the time of the injustice, the action was not legally prohibited or even widely regarded as wrong. During the era of US slavery, for example, if a free person were unlawfully enslaved, they could claim reparations for the harm done (McDaniel 2019), as could a slaveholder who had a slave stolen or wrongly harmed by another. Yet slavery itself

Race, Reparations, and Justice as Fairness

203

was a legally permitted practice. To claim reparations for a legally permitted practice, one would have to appeal to natural law or perhaps to what Rawls sometimes calls “natural duties.” It is not difficult to understand why reparations might be owed when state officials violate existing laws or regulations. Something like this happens when, for instance, a government pays compensation to a victim of police brutality. So maybe Rawls would regard the state just like a liable private actor when it comes to reparations for publicly recognized wrongs. Things are more challenging when the injustice is a consequence of an unjust basic structure (according to the two principles) but is not out of step with the prevailing public conception of justice. Consider the evils of the Jim Crow regime, where many wrongs against Black people were perpetrated by state officials or were legally permitted. Can the public incur a debt for a wrong that is compatible with its recognized conception of justice? Again, there would need to be a valid natural duty to repair harms due to wrongdoing. I will return to this.

12.2

Rectification of Injustice

When injustices occur, they should be corrected, rectified, or redressed in some way, taking the most serious injustices first where feasible. When it comes to rectification (or what is sometimes called “corrective justice”), I believe there are important moral and practical differences between working to establish a just basic structure and protecting the legitimate interests of the unjustly disadvantaged as we carry out this necessary work. The first kind of action concerns institutional reform and ultimately building a well-ordered society. The second entails considering what changes are feasible now and then doing our best to defend the justice-based interests of all concerned, as we work toward full social justice. This second aim might involve some temporary compromises with injustice or difficult trade-offs. If we were to achieve a well-ordered society (which might be a longterm project), these legitimate interests would then also be adequately secured. However, the most disadvantaged should not have to wait (and often cannot wait) for a fully just or even nearly just society to have their basic interests protected and their life prospects enhanced. As we make our way toward, for example, fair equality of opportunity, it may be necessary to institute temporary remedial policies aimed at evening out undeserved advantages and disadvantages. These policies would be aimed at bringing about (or approximating) conditions where those with similar talents have a roughly equal chance to attain valuable social

204

Tommie Shelby

positions regardless of their race, gender, or class background. Some affirmative action policies might fall into this category. Reparations are different, as they have a different goal. Reparative justice is not a matter of bringing the basic structure – the system of public rules – in line with what justice requires. Nor is it a matter of temporary ad hoc policies meant to improve the life prospects of the unjustly disadvantaged. Thus, when it comes to corrective justice, we need a distinction between rectification aimed at (a) fully implementing the principles of justice system-wide, (b) preventing or attenuating systemic injustice on the path to realizing a well-ordered society, and (c) repairing harms due to injustices that have already occurred. Injustice, in its many forms, causes harm to individuals and social groups. This harm can be caused by an existing unjust social structure. For example, people are harmed when they are denied basic liberties or their liberties are wrongly curtailed. They are harmed when denied opportunities or social benefits they should have or when unjust obstacles prevent them from taking full advantage of the opportunities they do have. They are also harmed when they are denied or deprived of economic resources to which they are entitled. Injustices in the distant past can cause harm in the present. For example, people may have inherited disadvantages (e.g., with respect to wealth or property) because their parents or grandparents were treated unjustly (e.g., suffered employment discrimination, had their property stolen, or were prevented from acquiring wealth). They may grow up in families that are deeply disadvantaged because their parents or grandparents suffered educational, physical, or psychological harms and thus were unable to provide a healthy home environment. They may suffer from a stigma associated with their group’s previous inferior civic status, including degrading stereotypes and unconscious biases. The same individuals or groups can be harmed by both ongoing and past injustices. And some ongoing injustices are long-standing, reaching back generations, what Jeff Spinner-Halev (2007) has termed “enduring injustice.” When structural injustice is ongoing, the correct response is naturally to end the unjust practice and to establish practices or institute policies that prevent the injustice from recurring. If the problem is employment discrimination, then anti-discrimination measures should be instituted. Yet is that all? What about the harm already done because of the unjust practice? Certainly victims should be free from further unjust treatment. But it wouldn’t be reasonable to expect victims to be satisfied that the unjust practice has ended if that practice has left them greatly injured or significantly set back their legitimate interests. Some effort to repair the harm is called for, assuming it is reparable.

Race, Reparations, and Justice as Fairness

205

What about when the unjust practice ended some time ago (for example, US slavery) but people are still suffering because of its aftereffects? The appropriate form of rectification will depend on why these harms continue. For instance, it could be, appearances notwithstanding, that the unjust practice was never fully dismantled. Debilitating remnants remain, though perhaps are not generally recognized as such. These too would need to be remedied as part of an effort to completely eradicate the injustice. Or it could be that other unjust practices, old or new, prevented victims from fully recovering from their injuries. Perhaps those previously oppressed by the racial regime were inhibited from escaping poverty and political marginalization by injustices related to class, gender and the family, or law enforcement. This should be a reminder that true rectification cannot be limited to a single domain like racial justice. Yet some would insist that current disadvantages linger because reparations were not made for past injustices. Such compensation might have enabled victims to recover fully from their injuries. Or if reparations were offered, those compensatory measures were far too meager to truly repair the harm done. Much of the contemporary US debate over reparations turns on whether reparative justice is the way to respond to this problem of continuing racial disadvantage.

12.3

Reparative Justice

What are reparations? Not all compensation is reparation (Boxill 1972, 2003). A person may be owed compensation for damages even if the harm is not due to injustice. If I rent a car and then accidentally crash it into a tree, I may owe the owner compensation for the damage even if the accident was not due to any wrongful act on my part. (Maybe I swerved to avoid hitting a crossing deer.) However, a valid reparations claim is based on having been harmed through another’s unjust actions. Reparation, in the strict sense, is compensation for harm due to injustice. Let’s distinguish four ways of thinking about the compensation that reparation involves. All concern repairing harm due to injustice. They differ in who should repair the harm and why. On the first conception, compensation may be paid by persons who were not themselves at fault for the unjust harm. Such reparations might be paid on behalf of those who did owe them, that is, the perpetrators of the injustice (who may now be deceased or otherwise incapable of paying). But those who pay them cannot then be said to owe them, to have a “debt” that they must pay to avoid wrongdoing. It would thus be a stretch to say that those who receive such compensation had a claim-right against those who paid,

206

Tommie Shelby

though there might be good consequentialist reasons for such compensation to be offered nonetheless.3 On another conception, those who pay reparations may owe them but not because they perpetrated the injustice that caused the harm. They owe them because they received (perhaps unwittingly) ill-gotten gains or undeserved advantages, which they would not have received had it not been for unjust harms to others.4 They possess something they have no rightful claim to and thus should forgo these gains or advantages to compensate those who suffered the wrongs.5 It follows from this conception that reparations are not owed if no beneficiaries of the injustice exist, even if victims were harmed in ways that could be repaired. On a third conception of reparative justice – and the one I will primarily rely on – those who have reparations claims have them only against those who committed the injustice that harmed them. The debt isn’t really paid unless the unjust actor pays the costs. Here the point is not simply to repair the harm (which an innocent third party might be able to do) but to set things right between the wrongdoer and the wronged, to make amends. Only then is the grievance fully resolved and the case closed. It is also worth noting that on this conception, a moral debt can be created even if the perpetrator acquired no tangible benefits from their wrongful action. The harmful wrongdoing alone creates the duty of repair, not the benefits (if any) that accrued to the wrongdoer. There is a fourth way of thinking about reparative justice that merits brief mention. Some are just as concerned (or perhaps more concerned) that victimizers pay for what they have done as they are that victims recover from their injuries. They don’t want those who perpetrated the wrong to “get away with it” – that is, to have done the wrong without it costing them in some way. They want those who wronged them to suffer some loss or to be deprived of something valuable. On this view, those who do wrong should pay a penalty even if so doing doesn’t repair the harms or prevent future such wrongs. This makes reparative justice 3

4

5

Martin Carcieri (2010) offers a Rawlsian defense of reparations for African Americans relying on a conception of reparations along these lines. Invoking the idea of collective responsibility for corrective justice, he does not believe it is crucial to assign fault or liability for the harms in need of repair. Nor does he make a sharp distinction between reparative justice and distributive justice. For defenses of reparations for African Americans relying on this conception, see Hugo Adam Bedau (1972), Howard McGary (1999), and William A. Darity and A. Kirsten Mullen (2020). If ill-gotten gains are sufficient to owe reparations, then many African Americans owe reparations to Native Americans for benefits acquired by living on stolen land. Such a debt might be owed, in some form. But the point is rarely noted, nor its ramifications explored.

Race, Reparations, and Justice as Fairness

207

too much like retribution. I don’t believe this conception of reparative justice is compatible with justice as fairness, as it suggests that those who wrong others deserve to suffer. We should distinguish material reparations (e.g., money, property, or land) from reparations that do not involve the transfer of economic resources. Sometimes the claim of reparation is not for money or property that was stolen, destroyed, or damaged but for lost opportunities to acquire money or property. In this case, the victims were unjustly prevented from taking advantage of opportunities (educational, workrelated, or financial) that should have been theirs. We can think here of slaves who were prevented from earning money, owning land, or starting a business. But we can also think of gross violations of fair equality of opportunity (e.g., the denial of access to schools or discrimination in employment and lending). Sometimes the past injustice is cultural imperialism or otherwise involves a loss of a cherished culture or collective identity. African Americans who are descendants of slaves have been largely cut off from their native cultural heritage and their connection to a specific ancestral place in Africa. The harm done is incalculable, made worse by the fact that many whites refuse to treat them as equals in the developing American cultural mix. This kind of harm is not best rectified by money or property, if it can be rectified at all. It is like what Native Americans have lost due to war, conquest, and colonialism. It is not unusual for the descendants of victims of injustice to make a claim of reparations, particularly when they regard themselves as having been negatively affected by the wrongful harm to their ancestors. There will also be many persons who are not descendants of the victims of past injustice but who nonetheless have been negatively affected by the injustice. I believe there is a moral difference between being treated unjustly and being negatively affected by the fact that others have been treated unjustly. Reparations is most obviously due if a person or group has had their basic rights violated and were thereby harmed. It is less clear that reparations are due to a person or group who is harmed because some other person or group had their rights violated. When someone is unjustly convicted of a crime or injured through police brutality, they are treated unjustly, and their family, friends, and neighbors may also be negatively affected by the wrong. I suspect that only the one who has been treated unjustly has a right to reparations, but I am not confident about this. Much will turn on how tightly the causal connection between an injustice and its harms must be. Must the injustice be the direct and proximate cause of the harms, or are all harms traceable to an injustice – including indirect harms and harms years, perhaps decades, after the

208

Tommie Shelby

injustice occurred – a suitable basis for reparations claims? I am not sure and will take no firm position on the matter.

12.4

Closure, Reconciliation, and Restorative Justice

There is a more expansive conception of reparative justice that is not principally about compensation. Rather than seeking merely to repair harms (psychological, physical, cultural, or material), the idea is to repair moral relationships between people that have been broken or damaged due to injustice. Margaret Urban Walker calls this aim “restorative justice” to distinguish it from conceptions of reparative justice centered narrowly on compensation.6 There may have been an existing relationship that was damaged or made worse by an injustice. At other times, the perpetrator and victim had no real relationship prior to the injustice. They might even have been strangers. In that case, we can speak of the bad relationship that was created by the wrong. Previously unacquainted, now they are at odds, perhaps enemies. In both cases, there is often resentment, mistrust, and sometimes hostility and enmity that stand between the parties. These negative sentiments can fester and harden in the absence of concerted efforts to repair the damaged or broken relationship. Through what legitimate and effective means can these negative sentiments be dispelled and the relationships repaired? Compensation alone is generally insufficient. The perpetrator must also acknowledge the wrong and accept responsibility for the harm done and maybe even apologize. Some might wonder that if full compensation is made, what further goal is achieved by acknowledging and accepting responsibility for the wrong. In other words, what do such attempts at moral repair add to the compensation? There is reason to believe that it could bring closure to a tragic episode and at least lessen resentment. To refuse to acknowledge the wrong is an insult. It suggests that the victim didn’t merit better treatment or that the perpetrator might again engage in similar wrongful acts if they believed they could get away with it. Moreover, a payoff to silence complaints or quell anger is not the same as justice and generally won’t be received as such. There are injustices to which compensation would not be an appropriate response or whereby the harm done is irreparable. Though material reparations would have no place, restorative justice could still have a role to play. Public acknowledgment, acceptance of responsibility, and formal 6

See, for example, Margaret Urban Walker (2006).

Race, Reparations, and Justice as Fairness

209

apology can mend a broken relationship.7 However, these acts of moral communication generally need to be backed up with concrete actions that demonstrate sincere regret and repudiation of the wrong.8 Otherwise, they won’t be believable. This show of good faith is especially important when those involved have a relationship that requires trust, cooperation, and solidarity, such as the relationship among fellow citizens. The actions that function to demonstrate sincere regret and commitment to refrain from similar injustice can resemble compensation, which is why some call them “symbolic reparations” (Meyer 2006). But the aim is not to return the victim to the well-being they would have enjoyed if the wrong had not occurred. Thus, we can think of restorative justice as having two distinct goals, one more demanding than the other. Sometimes the objective is simple closure. The aim is to end resentment and conflict so that all sides can move forward with these past wrongs, not necessarily forgotten but behind them – no longer serving as a source of strife. It is a matter of closing the books on a grievance. There need be no expectation of ongoing cooperation or even future interaction. At other times, the objective is, or should be, reconciliation.9 The aim is a renewed relationship on terms of equal respect and mutual trust. Depending on the nature of the wrong and how long it goes unaddressed, closure may be as much as can reasonably be expected, but reconciliation is often worth striving for. I should note that, on some conceptions of restorative justice, moral relationships between groups are not fully repaired unless a collective narrative of the past is forged and affirmed by those in need of closure or reconciliation (Kutz 2004; Spinner-Halev 2007; Espindola and Vaca 2014). Here, not only is the moral relationship broken but collective memory is damaged, if not lost. The conception of restorative justice that I would endorse, and that I think is compatible with justice as fairness, includes public acknowledgment of past injustices. Certain critical and well-established historical facts should be widely known and rarely if ever denied. For example, the facts that constitute the wrong and (when possible) the identity of those who bear primary responsibility 7

8 9

It might seem that an apology is appropriately given only by those at fault. But perhaps it can be offered by those in the right relationship to those at fault (e.g., those who share a social identity with those responsible or those who have benefitted from the wrongdoing of others). For discussion of the value of such symbolic gestures, see Waldron (1992) and Walker (2006). For a discussion of why reparative justice does not require reconciliation and need not aim at it, see McGary (2010).

210

Tommie Shelby

for that wrong should be public knowledge. We also need mutual recognition of sufficient historical detail to grasp what made the wrong a wrong. However, we must allow that reasonable people, acting in good faith and with a healthy sense of justice, can disagree about the precise details of history – about exactly who did what, to whom, when, and where – and about the broader moral significance of any given historical fact. And sometimes it’s not possible to establish what really happened, at least not using commonly accepted standards of historical inquiry. The requirements of public reason may place some practical constraints on how robust such a collective narrative can reasonably be expected to be. 12.5

Justifying Reparations

I propose that we view reparative justice as a component of a complete conception of right, which Rawls defines as “a set of principles, general in form and universal in application, that is to be publicly recognized as a final court of appeal for ordering the conflicting claims of moral persons” (TJ 1999, 117). I also propose that we regard reparative justice as a natural duty – a moral requirement that holds regardless of whether one has voluntarily accepted the benefits of a just social arrangement or institution (TJ 1999, 303). Like the duty of justice and the duty of mutual respect, a natural duty of reparative justice would hold in a variety of circumstances. It would seem to apply at least to three distinct subjects of justice. It applies to injustices between individuals within a society and is thus a matter of private law or criminal law. It applies to the injustices perpetrated by the society itself (say, through its laws or public officials’ actions) against individuals or groups within the society. And because justice includes international relations, the duty of reparative justice would apply to injustices perpetrated by one state against the people of another state or against a stateless people (e.g., when a state engages in war for glory, power, revenge, material gain, or territory). The society of peoples would share such a principle, just as it would have principles of just war and rules governing the punishment of war crimes. Reflecting on the natural duties of justice and mutual respect, one can already see why Rawls would want to make a place for reparative justice. The duty of justice enjoins us not only to support and comply with institutions that are already just but also to rectify injustice by helping to bring about just social arrangements where they fail to exist (TJ 1999, 99, 293). The duty of mutual respect requires us to show everyone the respect due them as a person with a sense of justice and a conception of the good (TJ

Race, Reparations, and Justice as Fairness

211

1999, 297). Everyone benefits from the public acknowledgment of these duties. The same is true, I believe, of the duty of reparative justice. To see this, let’s turn to how reparative justice could be justified within a Rawlsian framework. The original position is the natural place to turn. The principles of reparative justice are to be chosen only after the fundamental principles of justice have been selected, and these ideal principles place constraints on and shape the content of other principles of justice. For the methodological purposes of ideal theory, the parties in the original position assume strict compliance – that is, all are presumed to act justly and to do their part in upholding just institutions – when choosing the principles of justice that apply to the basic structure taken as a whole (TJ 1999, 8, 216). Yet the parties will have to allow that serious injustices might have occurred in the past and could occur in the future, given what we know human beings to be like. What is more, though they don’t know their particular place in history or their specific circumstances, the parties will have to allow that injustices may be presently ongoing, and thus they will need to agree on required and permissible ways of responding to past and existing injustice. The need for such principles should be clear from the case of civil disobedience, which Rawls argues can be justified in a nearly just society when serious violations of equal liberty or fair equality of opportunity are persistent (TJ 1999, 326). But, again, first the parties need a public conception of justice to settle what is to count as “injustice” and to give reform efforts a clear objective. They can then choose principles of nonideal theory – principles of rectification – that are consistent with their shared view of justice. Why would the parties in the original position acknowledge a natural duty of reparation? Not even a well-ordered society could ensure that injustices never occur. Since anyone could be the victim of injustice and harmed as a result, it would be rational for the parties to agree to such a principle to protect their legitimate interests. They wouldn’t want to be forced to absorb the costs of repairing harms wrongfully imposed on them by others. The parties would agree that those who wrongfully harm others should pay the costs of repair. As each is presumed to have an effective sense of justice, each can avoid the burdens of repair by refraining from violating the rights of others. One often cannot avoid being the victim of another’s unjust actions. But as free and rational agents with a sense of justice, we can avoid acting unjustly. So principles of reparative justice adequately protect everyone’s interests. Recall also that the parties have reason to create and maintain just institutions that are stable. If individuals or groups were harmed due to

212

Tommie Shelby

injustice but had no recourse for getting their injuries repaired, this would likely weaken their resolve to support the social arrangement. For the sake of stability, then, the parties will want to avoid this situation if they can. Agreeing to a duty of reparative justice is a rational way to do so. And as with criminal sanctions, the threat of having to compensate others can serve as an incentive to take due care to respect others’ rights (Wenar 2006). A just social arrangement will have greater stability with a mutual acknowledgment that each should repair harms caused by their own unjust actions. These considerations can, I think, be extended to collective agents, such as peoples and governments. Such collectivities, too, would naturally seek to avoid involuntarily internalizing the costs of repair for wrongs that others have committed, and they have the capacity, qua collective agent, to refrain from injustice. When political conditions are favorable (for example, when governance is responsive to public demands and democratic processes), the public is responsible for the consequences of government actions and policies, even when these actions and policies run afoul of what justice requires.10 There is a collective responsibility, shared among citizens, to create and maintain a just social system. The collective liability for reparation should be shared accordingly. But it is less clear that current citizens should be held responsible for an unjust basic structure in the distant past, as they couldn’t have done anything to change that structure.

12.6

Rectification and Distributive Justice

Can Rawls’s focus on the basic structure of society teach us anything about reparative justice? Does the theoretical primacy of the basic structure entail a practical primacy, in terms of what is most urgent to rectify? Does the primacy of the basic structure place limits on what form reparations can take? To explore these questions, I will concentrate on the case of injustices against African Americans in the United States. 10

We must distinguish harm caused by government action permitted by the basic structure (the system of public rules, including the constitution) and harm caused by government action contrary to the public conception of justice. The latter type of harm is sometimes (partly) the responsibility of the specific government officials rather than (solely) the responsibility of the public. So, for example, there is legally permitted slavery and Jim Crow segregation and the harms they caused. Then there is police-assisted lynching and other wrongful acts by officials that are contrary to law. Still, some unlawful acts by government officials could and should have been anticipated and efforts made to prevent them. Here the public may share responsibility for the harmful wrongdoing despite having prohibited the wrongs.

Race, Reparations, and Justice as Fairness

213

When individuals or groups within the same society wrong one another under otherwise just social conditions, it is easy to see why material reparations may be due and how they could be implemented. Reparative justice questions are more challenging when the wrongful harm took place under seriously unjust social conditions. Returning a victim back to their pre-harm state of wellbeing (should this even be possible) would still leave them living within an unjust basic structure. Often the violated property claims were part of a system of legal rights that was itself unjustified. It is not that no one can have legitimate expectations about property and its transfer under unjust circumstances. It’s just that when such violations are compensated, ill-gotten gains and arbitrary advantages inevitably remain. If the harm done through past injustice is economic, there is also a question about whether material reparations would be called for once distributive justice for the social structure was secured.11 For instance, there are reparative justice claims that seem to presuppose that some form of welfare-state capitalism is just. But Rawls maintains that such a society is incompatible with his principles of justice. Justice as fairness in a property-owning democracy requires (1) wealth to be widely dispersed (to maintain political equality), (2) highly limited intergenerational wealth transfers within families (to maintain fair equality of opportunity and to ensure the fair value of political liberties), and (3) limited economic inequality (to satisfy the difference principle). Liberal socialism requires means of production and finance capital to be held as public assets and eschews the very idea of a right to inherit productive assets. Both forms of political economy would secure an adequate minimum income so that all can participate in political life on fair terms. So when some African Americans protest “racial disparities” due to past racial injustice, the complaint is often best interpreted as a charge that there is ongoing injustice that should be ended, that there is an existing failure of justice quite apart from any past racial injustices that impact the present. The racial disparities that call for remedy are forms of group-based oppression or political subordination, examples of one group using its collective advantage to dominate or exploit another. This is one reason that special voting protections for Black US citizens are needed to ensure fair political participation.

11

For a helpful discussion of how some reparations claims are best interpreted as distributive justice claims or as a strategy for achieving distribute justice, see Wenar (2006, 401–2). For a good discussion of why distributive justice often takes priority over reparative justice, see Kutz (2004, 291–96).

214

Tommie Shelby

Sometimes the focus on racial disparities centers on differences in wealth between groups. African Americans are, on average, greatly disadvantaged with respect to wealth when compared to their white counterparts. From the standpoint of justice as fairness, though, there are many affluent African Americans who have unjust advantages over other Black Americans and, indeed, over some white Americans. These are forms of socioeconomic inequality that cannot be said to improve the prospects of the most disadvantaged in society. When people speak of the need to rectify disparities between racial groups, they sometimes ignore these other socioeconomic disparities between individuals and households, which also call for redress. We must avoid entrenching economic injustice as we redress injustices of the past. That would mean, for example, scrutinizing the right of inheritance and seeking to establish egalitarian economic arrangements that benefit everyone. One trouble with material reparations claims is that they often presuppose claims of inheritance that may themselves be unjust or otherwise indefensible. If individuals have limited inheritance rights (not according to existing law but according to what justice requires), then members of historically oppressed groups will also have limited claims to material reparations insofar as these claims are based on the idea that their ancestors were owed reparations that were never paid.12 The thought here is that descendants of those unjustly treated inherit their ancestors’ claim to reparations. Yet we would first need to know how much each can legitimately inherit before we can say how much in reparations should be paid. These problems don’t seem to arise within a historical entitlement theory of justice of the sort Robert Nozick proposed (1974, 150–53). If economic justice is fully defined by natural rights to acquisition, transfer, and rectification, then inheritance rights are effectively unlimited. There is no concern about how material reparations might upset an egalitarian social structure or inhibit our ability to establish such a structure. Here private ownership of the means of production is treated as a matter of basic justice, and individuals are permitted to transfer all their assets to their descendants (or to whomever they please). Things are otherwise with justice as fairness. If a strong egalitarian principle holds for social and material advantages (e.g., fair equality of opportunity and the difference principle), then wealth transfers as reparations will be severely limited. These transfers would have to be compatible with creating and maintaining a just social 12

For a Rawlsian defense of a limited right of inheritance and an explanation of its bearing on reparative justice, see Thompson (2001).

Race, Reparations, and Justice as Fairness

215

structure. To ensure that class background does not hurt anyone’s chances to acquire valued social positions, there cannot be great economic inequality. Otherwise, those within wealthy families would have an unfair advantage in competing for these positions. If socioeconomic inequalities are justified only if they improve the life prospects of the least advantaged, we must be mindful of how material reparations would impact the broader social structure. Things are different when those who should pay reparations and those who have a reparations claim are not, or do not expect to be, fellow citizens within a single polity. Sometimes the injustice has been perpetrated by a foreign power against a culturally cohesive people with a right of self-determination.13 The victims reasonably want their injuries repaired – to be returned to their previous level of well-being –and to live a life apart from those who wronged them. Creating a just basic structure inclusive of victims and perpetrators is not an aim. So victims can be somewhat indifferent to how the transfer of wealth would affect economic relations among those who wronged them. Some “peoples” themselves have been made possible by injustice. When a social group faces oppression over many generations, they may develop bonds and systems of organization as strong as any people. Perhaps this is (or once was) true of African Americans. Maybe they have (or at one time did have) something akin to a right of selfdetermination, as many black nationalists – Martin Delany, Marcus Garvey, and Malcolm X – have claimed. A demand for material reparations makes perfect sense when the plan is emigration or secession. Things are also different when there is little hope that a (nearly) just social structure is on the horizon. When achieving a just society doesn’t appear to be feasible anytime soon, those who have been harmed by past injustice will naturally have a difficult time committing to the creation of just social arrangements if that means suffering unfair hardships now and for the foreseeable future. They might demand their injuries be repaired while we work for structural transformation. Indeed, they may refuse to work toward creating a just social structure until they receive adequate reparations. In their most pessimistic moments, they demand reparations now and adjust their expectations to living with broader systemic

13

For an argument that African Americans are not the right kind of “cultural entity” to make a collective claim to reparations on this basis, see Hill (2002). Hill does believe, however, that African Americans can claim reparations for slavery as a right of inheritance, because their ancestors were owed a debt for unpaid wages. Here she relies on a helpful distinction between harm to the group qua collective and harm to the group’s individual members (perhaps because they are a part of the group).

216

Tommie Shelby

injustice. There is a question of whether all these attitudes are reasonable even if they are understandable.14 It might be that what attitude one should have toward reparations for racial injustice depends on how unjust the existing social structure is. A nearly just society (as Rawls calls it) could encourage those with valid reparation claims to moderate their demand for material compensation. If they don’t face poverty and have something approaching fair equality of opportunity, insisting on greater material resources as reparations may even be unreasonable. When African Americans press a claim for material reparations, they are largely responding to their current economic disadvantage in US society. It is difficult to flourish without some measure of wealth, and many African Americans lack sufficient economic resources to function as equal citizens in America. They often attribute this economic disadvantage to past injustice, sometimes tracing this back to slavery or Jim Crow segregation. But insofar as the concern is with improving their present economic position in society, we don’t need to settle whether this disadvantage is traceable to some past injustice or is attributable to ongoing injustice.15 Black people certainly resent the suggestion that their current disadvantages are their own fault or that these disadvantages exist because they are failing to take advantage of adequate educational and economic opportunity. To avoid this implication, a public acknowledgment of past injustice and the harm it has caused is critical. But so is a public recognition of serious existing structural injustice. To summarize, for the domestic case under unjust conditions, which is the situation of contemporary African Americans in the United States, I believe that justice as fairness calls for the following approach to rectification (corrective justice): (1) Strive, individually and collectively, to create and maintain a just basic structure, prioritizing this objective over reparative justice. (2) Institute temporary remedial measures to attenuate existing injustices and to approximate just conditions, thereby enlarging the

14

15

Refusing to help create a just social structure is not only incompatible with our duty of justice, but it could also potentially rob future generations of a society that would enable their flourishing. The oppressed have children and grandchildren, too. What are we to say to these descendants about our refusal to make a better world for them? Can we say we were standing on principle, defending our self-respect? I don’t know how compelling that answer would be. For a discussion of some of the perils and obstacles that defenders of reparations face when sorting out the different causes of existing racial disparities, see Darby (2010).

Race, Reparations, and Justice as Fairness

217

freedom and enhancing the material life prospects of the unjustly disadvantaged. (3) Practice restorative justice (including symbolic reparations where appropriate) to bring closure to past racial injustice, thus laying the groundwork for ultimate interracial reconciliation. (4) Pay reparations to victims of recent, unlawful wrongdoing (whether perpetrated by private actors or public officials), provided such payments are compatible with creating or maintaining a just basic structure. The view defended here shares much in common with the one David Lyons’s advanced in Confronting Injustice (2013). Lyons’s rectification project for African Americans cannot be reduced to a general call for reparations. It is, however, meant to be a response to the underlying concerns of those who advocate for reparations – namely, the deep and enduring disadvantages that many blacks continue to face because of the legacy of slavery and Jim Crow. I agree with much of Lyons’s argument. Yet, I’m not so sure that the National Rectification Project he proposes responds to all the legitimate concerns of those who make reparations claims. At times, Lyons seems to elide the difference between correcting an ongoing injustice and rectifying a past injustice. Reparation is a moral project – it is about paying moral debts, making amends, and acknowledging past wrongs. It is not fundamentally about making people materially equal or reforming an unjust practice. In some ways, then, Lyons has changed the subject. We have reasons of basic justice to ensure equality of opportunity and to combat racism quite apart from whether the federal government has played a role in past racial subjugation. Insofar as Lyons is calling for fair equality of opportunity regardless of the source of existing inequalities, the backwards-looking dimension of reparations appears to be lost. Lyons could respond that the fact that the government contributed to and sustained a system of racial hierarchy provides a further reason to hold it accountable for these inequities. But this wouldn’t give Blacks a greater claim for redress than others (including many whites) who also do not have fair equality of opportunity. Lyons does say, “an essential element of the required rectification is an informed acknowledgement of the moral as well as material aspects of the wrongs of slavery and Jim Crow” (2013, 111). This points us in the right direction. But this public acknowledgment should be taken two steps further – that is, if genuine closure and interracial reconciliation are to occur. Yes, there should be a general acknowledgment of these wrongs and their consequences, as restorative justice requires. But this

218

Tommie Shelby

acknowledgment should include further symbolic acts to demonstrate sincerity. For instance, following a suggestion by Jeremy Waldron (1992), we might publicly affirm those groups who have suffered injustices – and, as a result, whose identities have become stigmatized – and publicly recognize their sacrifices in making America wealthy and more just. This could be done through commemorations, national holidays, museums, and monuments. Indeed, symbolic reparations could be paid to unjustly disadvantaged groups, not in a futile attempt to make them “whole” but to demonstrate that this public acknowledgment is offered in good faith. This would be analogous to offering a gift with an apology for some wrongful act. Public actions of this sort would include the backwards-looking dimension of rectification and thus could have a salutary effect on damaged interracial relationships. A second necessary aspect of this public acknowledgment is that Blacks should not be blamed for their disadvantages and whites should not resent efforts to rectify these disadvantages even if they must pay some of the costs and won’t materially benefit from these temporary corrective justice programs. African Americans might reasonably demand, not only public recognition of the history and legacy of white supremacy but also some public acknowledgment that their existing disadvantages have been caused by past and ongoing practices of racial subordination. Too often, people insist that Black disadvantages are self-imposed due to the moral failings and blameworthy improvidence of Blacks themselves. Blacks generally interpret this as an affirmation of racist ideology and a refusal to take responsibility for rectifying racial injustice. In light of this, the trouble with “universal” or “race-neutral” policies aimed at creating conditions for fair competition is that they implicitly suggest that Blacks are not actually due equal concern and respect. For they are allowed to benefit from social policies only insofar as whites benefit as well. Anything more targeted at improving Black life chances is generally met with hostility, resentment, resistance, and backlash. 12.7

Conclusion

The view of reparations and racial justice I have sketched and defended, though perhaps controversial, is not new. For the most part, the ideas I have presented already exist in the literature on reparative justice and historical injustice. What I have tried to show is how these ideas might fit together within the broader political morality of justice as fairness. One of the great virtues of Rawls’s conception of justice is that, though incomplete, it is highly systematic and wide-ranging, a true attempt to offer a philosophical theory with practical implications.

Race, Reparations, and Justice as Fairness

219

Like Mills, I wish Rawls had said more about how questions of racial injustice should be handled within his theory. Also like Mills, I would urge more philosophers to attend to white supremacy and its enduring legacy. But I remain convinced that justice as fairness cannot only account for but also illuminate racial injustice and its remedies, including historical injustice and the question of reparations.16

16

For helpful feedback, I would like to thank the participants at the “John Rawls’s A Theory of Justice at Fifty” Conference at Notre Dame. I’m also grateful for comments and questions from audiences at UC Berkeley and the University of Kansas. Special thanks to Joshua Cohen for particularly perceptive comments and to Paul Weithman for organizing this whole effort and editing the volume.

13

On the Role of the Original Position in Rawls’s Theory Reassessing the “Idealization” and “Fact-sensitivity” Critiques

Laura Valentini 13.1

Introduction

Over the past fifty years, no other work in analytic political philosophy has been as influential as John Rawls’s 1971 opus: A Theory of Justice. Rawls can be credited with reviving normative theorizing about politics, making the question of social justice central to contemporary political philosophy. Many influential books published after 1971 on the topic of justice are reactions, responses, or competitors to Rawls’s own work.1 The mark left by A Theory of Justice on the terrain of political philosophy is not limited to substance but extends to method too. It is in fact no coincidence that the method of “reflective equilibrium” – central to A Theory of Justice – has arguably become the dominant one in analytic political philosophy. The influence of A Theory of Justice is also reflected in the number of criticisms it has received. Dozens of articles have been written scrutinizing and challenging this or that aspect of Rawls’s work, ranging from his method, to the plausibility of the substantive principles he proposes, from the restricted scope of his theory (which excludes, e.g., the family and the global arena), to its seeming acontextual character.2 One aspect of A Theory of Justice that has been subjected to particularly intense critical scrutiny is the thought experiment Rawls constructs in order to justify his preferred principles: the original position (OP for short). To briefly recall, the OP denotes a hypothetical choice situation in which (representatives of ) citizens, stripped of specific knowledge about their place in society, are asked to select the principles that will govern

1 2

For instance, Nozick (1974); Walzer (1983); and Dworkin (2000) . Even an illustrative list would likely take up too much space here, which is why I am not offering specific references.

220

On the Role of the Original Position in Rawls’s Theory

221

their political community. The principles selected in this hypothetical choice situation, Rawls argues, are the correct principles of justice. Critics have found many alleged faults with the OP. Some have cast doubt on the validity of a hypothetical contract in general, others have taken issue with the specific characterization of Rawls’s choice situation, others still have expressed doubts about whether, even granted that characterization, the parties would indeed choose Rawls’s preferred principles.3 It would of course be impossible – and probably rather boring – to review and discuss all of these criticisms in the space of a single chapter. My aim here is therefore more modest. I wish to focus on two broad families of criticisms of Rawls’s thought experiment, pulling in exact opposite directions. These might be labeled the idealization critique and the fact-sensitivity critique. In a nutshell, the former holds that Rawls’s OP is too hypothetical, too detached from the real world to deliver the correct principles for us. The latter holds that the OP is too rich in realworld facts; it concedes too much to reality to tell us what real justice requires. I wish to reassess these critiques by bringing into sharper focus the precise role the OP plays in Rawls’s work. To do so, I shall rely on basic concepts familiar from the philosophy of science and social science and apply them to the normative domain. This will allow me clearly to distinguish between (i) Rawls’s theory, (ii) the arguments offered in support of it, and (iii) the desiderata applying to that theory. I will show that once these distinctions are kept in view, and the OP is put in its proper place, the aforementioned criticisms lose much of their apparent force. I proceed as follows. In Section 13.2, I lay out the main claims of Rawls’s A Theory of Justice and offer a more precise description of the OP thought experiment. In Section 13.3, relying on joint work with Christian List, I offer a broad characterization of normative theories and distinguish such theories from the arguments offered in their support and the desiderata they should satisfy (List and Valentini 2016). In Section 13.4, I present the idealization and fact-sensitivity critiques. In Section 13.5, I reconsider those critiques suggesting that much of their persuasiveness stems from misunderstandings about the role of the OP in Rawls’s broader theory or about the aims of – hence the desiderata applying to – that theory. Section 13.6 concludes. Before I start, let me make one clarificatory remark. My discussion will exclusively focus on A Theory of Justice (and, specifically, on the 1999 3

For collections of essays commenting on these and other aspects of the OP, see Daniels (1975), especially Part I and Hinton (2015).

222

Laura Valentini

edition). The OP of course features centrally in other important works by Rawls, such as Political Liberalism, The Law of Peoples, and Justice as Fairness: A Restatement. Examining its role in those works is beyond the scope of this chapter.

13.2

A Theory of Justice and the Original Position

Rawls’s topic in A Theory of Justice is the design and defense of principles of social justice, namely principles “for choosing among the various social arrangements which determine the division of advantages [of social cooperation] and for underwriting an agreement on the proper distributive shares” (TJ 1999, 4). These principles have the very practical task of establishing how society – and specifically its “basic structure” – should be arranged, so that its distributive outcomes are fair or just. Since the role of justice is eminently practical, Rawls is explicit that “we cannot, in general, assess a conception of justice by its distributive role alone” (TJ 1999, 6). All other things equal, a conception of justice that does better in terms of efficiency, coordination, and stability is, in Rawls’s view, to be preferred. But how are we to go about selecting the relevant principles? Rawls’s answer consists in a reinterpretation of the social contract tradition. Like social contract theorists, Rawls believes that the correct principles of justice are “the principles that free and rational persons concerned to further their own interests would accept in an initial position of equality” (TJ 1999, 10). However, instead of construing that position in terms of a pre-political state of nature, he opts for an alternative thought experiment, the original position. This is a hypothetical deliberative forum in which citizens (more precisely, their representatives)4 are meant to unanimously select those principles. What is distinctive about this position is that those in it are placed under a “veil of ignorance,” which “excludes the knowledge of those contingencies which sets men at odds and allows them to be guided by their prejudices” (TJ 1999, 17). In so doing, the OP embodies a fair choice situation, such that the resulting principles will also be fair: hence Rawls’s decision to name his theory “justice as fairness.” But how, exactly, are the parties characterized in this OP? And what do they know and not know? In Rawls’s own words, behind the veil of ignorance:

4

Rawls originally suggested “heads of families” (TJ 1999, 111 and 126).

On the Role of the Original Position in Rawls’s Theory

223

no one knows his place in society, his class position or social status; nor does he know his fortune in the distribution of natural assets and abilities, his intelligence and strength, and the like. Nor, again, does anyone know his conception of the good, the particulars of his rational plan of life, or even the special features of his psychology. (TJ 1999, 118)

That said, the veil does not remove all knowledge. In particular, the parties: know the general facts about human society. They understand political affairs and the principles of economic theory; they know the basis of social organization and the laws of human psychology. (TJ 1999, 119)

Such knowledge is available to them because “conceptions of justice must be adjusted to the characteristics of the systems of social cooperation which they are to regulate” (TJ 1999, 119). The practical task of conceptions of justice necessitates making such information available to the parties. More generally, the parties themselves are described as “mutually disinterested,” namely “as not taking an interest in one another’s interests,” and as “rational” in the standard economic sense of “taking the most effective means to given ends” (TJ 1999, 12 and 127). Finally, the parties are to make a decision assuming that everyone will comply with the ensuing principles (the “strict compliance” assumption) and that society is characterized by moderate, rather than absolute, resource scarcity (the “favorable conditions” assumption) (TJ 1999, 8). Rawls is convinced that the principles that would be chosen in this counterfactual deliberative forum are the correct principles of justice for us. While directly defending the principles would be both difficult and highly controversial, showing that such principles would be chosen under conditions that many would regard as fair counts strongly in their favor. The principles Rawls claims would be chosen in the OP are, famously, the following: (1) First Principle: Each person is to have an equal right to the most extensive scheme of equal basic liberties, compatible with a similar scheme of liberties for others (equal liberty) (TJ 1999, 53). (2) Second Principle: Social and economic inequalities are to be arranged so that they are both: (a) to the greatest expected benefit of the least-advantaged (difference principle); (b) attached to offices and positions open to all under conditions of fair equality of opportunity (fair equality of opportunity) (TJ 1999, 72).

224

Laura Valentini

The principles are lexically ordered, with (1) taking priority over (2) and (2b) taking priority over (2a). While the OP is of course central to Rawls’s case for his principles, it should be stressed that the OP is itself embedded into a broader methodology, that of reflective equilibrium. According to reflective equilibrium, we seek to achieve coherence between the description of the initial choice situation, the principles that would be chosen in that situation, and our “considered judgments” about justice – namely our strongly held convictions (the injustice of slavery, of religious intolerance, of racial discrimination, etc.). In Rawls’s words: In searching for the most favored description of this situation we work from both ends. We begin by describing it so that it represents generally shared and preferably weak conditions. We then see if these conditions are strong enough to yield a significant set of principles. If not, we look for further premises equally reasonable. But if so, and these principles match our considered convictions of justice, then so far well and good. But presumably there will be discrepancies. In this case we have a choice. We can either modify the account of the initial situation or we can revise our existing judgments …. By going back and forth … I assume that eventually we shall find a description of the initial situation that both expresses reasonable conditions and yields principles which match our considered judgments, duly pruned and adjusted. (TJ 1999, 18)

To sum up, Rawls’s argument for his two principles of justice centers around the OP thought experiment, which, in turn, is embedded in the method of reflective equilibrium. As I have anticipated in Section 13.1, the OP has been the target of considerable criticism. My strategy to “rescue the OP” from such criticisms will involve better locating it within a broader account of what a normative theory is, as distinct from the arguments in its support and the desiderata the theory should satisfy given its function. To do this, I must therefore first sketch an account of what a normative theory is. 13.3

Normative Theories (in General)

In political philosophy, and in no small part due to Rawls’s influence, one often hears scholars speaking of theories of justice, legitimacy, the state, autonomy, toleration, equality, freedom, and much else. The notion of a theory is so often invoked that virtually any systematic reflection on an important moral-political matter is often “anointed” with the title of “a theory.” There is, of course, nothing wrong in this use of the term, so long as it is understood to be a loose, nontechnical one. But there is a more targeted and interesting use of the notion of a theory, one that is consistently found in the natural and social sciences.

On the Role of the Original Position in Rawls’s Theory

225

In what follows, drawing on joint work with Christian List, I will lay out an understanding of normative theories that is inspired by these other scientific disciplines. (List and Valentini 2016; McDermott 2008) 13.3.1 Theories: Positive and Normative On this understanding, a theory is a set of propositions or statements that plays a particular functional role in an agent’s conceptualization of reality. Depending on the function of a given theory, we may distinguish between positive theories and normative ones. The former seek to explain empirical reality and to allow us to predict future events. Familiar theories in the natural sciences (e.g., physical theories, biological theories) and in the social sciences (e.g., political theories, economic theories) do precisely that. They help us predict, for example, the trajectory of celestial objects or the growth of certain organisms, they can tell us how electoral systems affect party systems, or how demand and supply relate to each other. Normative theories, by contrast, seek to tell us how we ought to act. They do not predict but prescribe. For example, a normative theory of workplace relations tells us how individuals should behave in the workplace and how management structures should be organized. Alternatively, a normative theory of personal ethics tells us how we should behave in our personal lives: as friends, partners, parents, etc. A theory of justice, as Rawls understands it, tells us how we ought to structure our institutions so as to generate a fair division of the benefits and burdens of social cooperation. Theories so conceived – whether positive or normative – tend to be expressed in the form of axioms, laws, or principles. Think about the laws of physics or economics, or of Kant’s Categorical Imperative and Rawls’s aforementioned principles of justice. The corresponding theories consist of the body of implications of those principles as applied to specific circumstances. I have very briefly sketched an account of what a theory is. Theories in this sense should be distinguished from the evidence offered in their support and the desiderata applying to them. After all, theories can be well or poorly supported, and they can be good or bad at performing the function they are meant to perform. Calling something “a theory” by itself carries no evaluative implications. 13.3.2 Evidence and Desiderata In the positive domain, the evidence we typically marshal in support of this or that theory takes the form of reliable empirical observations. It is

226

Laura Valentini

the presence of such observations that allows us either to corroborate a theory – when the observations match the theory’s predictions –or to discredit it – when the observations are at odds with the theory’s predictions. Things are trickier when we switch to normative theories. There are in fact different views about what counts as “normative evidence” and what qualifies as “normative reality.” (Indeed, some even doubt the existence of such a reality.) Without getting bogged down in normative ontology and epistemology, for present purposes, I shall simply refer to what Rawls himself considers evidence in support of his theory. As it happens, Rawls’s views – especially about reflective equilibrium – are shared by many practitioners of analytic political philosophy. My focus on them should thus not excessively restrict the appeal of my discussion. As we have seen, at the broadest level, Rawls believes that (i) consistency with considered judgments and (ii) being the output of the initial choice situation count as evidence in support of his theory. Neither the judgments nor the initial choice situation is set in stone, however. In the method of reflective equilibrium we work, as he says, “back and forth,” and should be prepared – within limits – to revise our choice situation as well as to reconsider some of our judgments, in search for coherence. Once coherence is attained, what we have is at least a unified account of our political-moral sensitivities, independently of whether those can be said to track an independent, perhaps timeless, normative reality.5 Let me now consider a few more desiderata, beyond theories’ being well-supported by evidence. Often, whether in the positive or in the normative domain, we want our theories to be coherent, noncontradictory. In fact, a contradictory theory would be ipso facto false. We also typically wish our theories to be parsimonious, at least holding explanatory or justificatory power constant. That is, a theory that offers an elegant, noncomplex but compelling explanation of a phenomenon is typically to be preferred to a convoluted one. Relatedly, theories appear more illuminating when they are axiomatizable in the form of a limited set of principles, as opposed to consisting of long and cognitively demanding lists of propositions. Finally, we want theories to play their intended function well. For example, if the function of a theory is to predict weather events with relative accuracy but little available data, then an extremely

5

In A Theory of Justice Rawls has been interpreted as seeking to identify precisely such timeless principles, possibly due to his remarks about the OP allowing us to look at our place in society “sub specie aeternitatis” (TJ 1999, 514). The later Rawls of Political Liberalism is instead explicit about the context-specific nature of his theory, which is meant to articulate ideas already shared in contemporary liberal democracies.

On the Role of the Original Position in Rawls’s Theory

227

precise theory that requires copious amounts of data won’t do. Similarly, if the function of a theory of justice – as Rawls understands it – is to tell us how to distribute the benefits and burdens of cooperation within a stable social arrangement, a theory that could never hope to garner the support of citizens or that would be incompatible with basic psychological or economic laws would be unsatisfactory, since it would be unable to perform its intended function. In this section, I have sketched an account of what a normative theory is and distinguished theories from the evidence supporting them and the desiderata they should meet. I have said that this sketch will help me reassess some prominent critiques of Rawls’s OP, but, so far, I have said very little indeed about those critiques. I move on to this task in Section 13.4.

13.4

The Idealization and Fact-Sensitivity Critiques

As we have seen, the OP contains a mix of real-world facts and counterfactuals. The veil of ignorance removes some information from the parties while leaving some available to them. The critiques I am about to explore take issue with both. The first, the idealization critique, alleges that the parties in the OP are too removed from real-world human beings, and the circumstances for which they need to deliberate too removed from real-world societies, for Rawls’s thought experiment to deliver plausible principles. The second, the fact-sensitivity critique, insists that the parties, by taking into account certain limits of human nature and facts about human motivation and society, end up selecting suboptimal standards of justice: they compromise what real justice is for the sake of accommodating problematic empirical facts.

13.4.1 The Idealization Critique It is widely accepted that the general theoretical operation of abstraction – that is, of assuming away certain features of reality in our theorizing – is virtually unavoidable. As Onora O’Neill points out, if we want our theories to cover more than a single concrete case, abstraction is a “must” (O’Neill 1987, 55). There is, however, a related theoretical operation, which is allegedly neither equally unavoidable, nor equally innocuous: this is “idealization.” In O’Neill’s words: An assumption, and derivatively a theory, idealizes when it ascribes predicates – often seen as enhanced, “ideal” predicates – that are false of the case in hand, and so denies predicates that are true of that case. For example, if human beings are

228

Laura Valentini

assumed to have capacities and capabilities for rational choice or self-sufficiency or independence from others that are evidently not achieved by many or even by any actual human beings, the result is not mere abstraction; it is idealization. (O’Neill 1996, 41)

In short, abstraction relative to a subject matter X involves silence or neutrality with regard to X, idealization involves falsehoods about X.6 Manifestly, Rawls’s A Theory of Justice involves several idealizations so understood, and these are crystallized in the OP thought experiment. As we have seen, unlike real-world human beings, the parties are assumed to be mutually disinterested, independent, rational, and not wedded to a particular conception of the good. Equally, their clients are assumed to all comply with the ensuing principles, when we know that no real-world society benefits from universal compliance with its laws. Critics of Rawls have variously expressed dissatisfaction with the presence of such idealizations in the OP – though not all of them have used this precise language. This line of objection has taken different shapes over the years. It was first presented by communitarians, like Charles Taylor and Michael Sandel, who accused the liberal individual – as allegedly represented by the parties in the OP – to be excessively “atomistic” and detached from broader social processes (Sandel 1982; Taylor 1985). Sandel famously described Rawls’s parties as “unencumbered selves” detached from their ends and went on to suggest that it would have been virtually impossible for such empty “loci of volition” to choose any principles, let alone ones that real-world, context-embedded people should accept (Sandel 1984). More recently, the idealization critique has been put forth by skeptics about ideal theorizing generally. These skeptics oppose Rawls’s and others’ attempt to identify what a perfectly just society would look like, namely one where “everyone is presumed to act justly and to do his part in upholding just institutions” (TJ 1999, 8), In these critics’ views, principles designed to capture perfect justice – that is, principles designed in “ideal theory” – cannot plausibly guide action in the real world.7 The point is put quite well once again by Onora O’Neill (though here she isn’t explicitly referring to Rawls’s work): Practical reasoning that assumes that ‘ideal’ predicates are satisfied will not reach conclusions safely and soundly for cases where they are not satisfied. (O’Neill 1996, 41)

6 7

For critical discussion, see Ronzoni (2010). I have previously discussed this critique of ideal theorizing, though from a somewhat different angle, in Valentini (2009).

On the Role of the Original Position in Rawls’s Theory

229

Critics of Rawls have not only suggested that, due to the fullcompliance idealization, his theory may not be able to tell us anything useful about how to address serious injustices, such as racial and gender discrimination (Baier 1985, 225; McCarthy 2004). Some have even argued that ideal theorizing, à la Rawls, may be “ideological” in the pejorative sense of the term. By distorting individual capacities and setting aside oppression and injustice, ideal theory, says Charles Mills, is “really an ideology, a distortional complex of ideas, values, norms and beliefs that reflects the nonrepresentative interests and experiences of a small minority of the national population … who are hugely over-represented in the professional philosophical population” (Mills 2005, 172, italics original). The idealizations contained in ideal theory thus obscure existing injustice and thereby contribute to its continued existence. Having outlined the alleged flaws with the idealizations contained in the OP, let me now turn to the fact-sensitivity critique. 13.4.2 The Fact-Sensitivity Critique At the opposite end of the critical spectrum lies what I have called the “fact-sensitivity” critique. This takes issue not with the facts Rawls distorts or removes from the OP but with the facts he allows the parties to consider. As we saw earlier, these include the broader circumstances in which human societies find themselves, as well as social scientific and psychological knowledge about economics, politics, etc. The availability of these empirical facts in the OP, proponents of the fact-sensitivity critique argue, renders Rawls’s theory unable to really get at what justice (with a capital J) is. Whatever ends up being selected in the OP is thus a compromise between the demands of justice and other, nonjustice-based considerations, such as stability and efficiency. The most prominent proponent of this line of critique is G. A. Cohen.8 Cohen finds Rawls’s OP problematic insofar as it makes the selection of “fundamental” principles of justice sensitive to facts. This factssensitivity, Cohen alleges, leads Rawls’s principles to be neither fundamental nor genuine principles of justice (Cohen 2003, 2008). In support of the former claim, Cohen advances the view that considerations of empirical fact – for example, “Policy X makes people happy” – are made relevant to the selection of normative principles only in virtue of some further normative principles that do not themselves depend on those facts. So, for example, suppose we were to ask whether we should adopt 8

David Estlund has also taken issue with what he calls “concessive” theorizing, though he is generally more sympathetic to Rawls than Cohen is; see Estlund (2020).

230

Laura Valentini

policy X and answered that we should “because such policy makes people happy.” In doing so we’d invoke an empirical fact (i.e., the fact that the policy makes people happy), but the relevance of this fact to our decision is not freestanding. Presumably, the fact is relevant by virtue of a further consideration, this time of a normative nature, namely that “we should seek to make people happy.” Again, suppose we were to inquire about why we ought to make people happy. Some may think this is some sort of ultimate, self-evident “ought.” But others might be prompted to offer a ground or reason for it, such as “because making people happy also makes them more productive” (I am not suggesting this is a good answer but merely illustrating the logic of Cohen’s argument). Once again, this statement of fact can only affect our selection of the normative injunction to make people happy by appeal to a further principle, along the lines of “one ought to promote productivity.” Here again, we may raise the “why” question, and so on and so forth. Cohen claims that, at some point, any such concatenation of facts and normative principles will culminate in a fundamental, fact-free principle: a principle that does not include empirical facts among the grounds for endorsing (i.e., selecting) it. Normative principles, for Cohen, are fundamental when they’re factfree in this way. This, in Cohen’s view, shows that Rawls’s theory does not provide us with fundamental principles of justice. This is because such principles are selected by the parties in the OP in light of certain empirical facts, about society, economics, human psychology, etc. This, in turn, makes them ill-suited for being fundamental principles of justice. They are, to some extent, already “applied” principles, the validity of which is constrained by a set of empirical assumptions. To that extent, Rawls’s theory hasn’t really told us what, at bottom, justice is. In Cohen’s view, the inability of the OP to get us to fundamental principles of justice is related to its alleged inability to yield principles of justice, proper. This is because the parties in the OP are asked to choose principles for regulating their social arrangements, such that the distributive consequences of social cooperation are fair. As we have seen, selecting these principles requires some appeal to facts, since whether a principle is good or bad at such “regulation” will partly depend on the consequences of its adoption (Cohen 2003, 241). So, suppose that a principle were proposed that, based on our knowledge of basic psychology and social science, would be too complex for individuals and institutions to follow. That principle would be ruled out in Rawls’s OP. Or suppose that a principle were proposed that would appear too demanding for individuals to be reliably expected to adhere to. That principle would also be ruled out by the OP. Rawls’s discussion of the socalled strains of commitment explicitly confirms this.

On the Role of the Original Position in Rawls’s Theory

231

The parties must weigh with care whether they will be able to stick by their commitment [to the chosen principles] in all circumstances. Of course, in answering this question they only have a general knowledge of human psychology to go on. But this information is enough to tell which conception of justice involves the greater stress. (TJ 1999, 152)

But all of these considerations, about individuals’ ability or likelihood or willingness to follow certain rules, seem – Cohen suggests – extraneous to the question of what justice really is. At most, they are relevant to “social regulation.” To sum up, as David Estlund puts it (commenting on, but not endorsing, Cohen’s view): “the original position’s regulatory approach lets the choice of principles be determined by nonjustice considerations, and so does not reliably identify justice” (Estlund 2015, 149).

13.5

Reconsidering the Critiques

The idealization and fact-sensitivity critiques target Rawls’s OP. They are, in turn, taken by their proponents to deal a significant blow to Rawls’s project – namely his theory of justice – as a whole. But this move from criticizing the OP to criticizing the ensuing theory is too quick. This is because, while the OP is part of Rawls’s A Theory of Justice, the OP is not part of Rawls’s theory in the more precise sense proposed earlier in the essay. The two principles, with their body of implications, constitute Rawls’s theory. The fact that the principles would be chosen in the OP is just an argument, an important consideration, in support of those principles. It counts as “evidence” (Sangiovanni 2016, 22; Valentini 2016, 28). Whether the features of the OP the critics take issue with make the theory itself problematic is therefore an open question. In what follows, I will suggest that, for the most part, they do not.

13.5.1 Reconsidering the Idealization Critique The idealization critique, recall, holds that, since they involve falsehoods, idealized normative theories are either useless or counterproductive. If the idealizations are found at the level of the theories themselves, this claim is correct but uninformative. For to say that a theory – strictu sensu – is idealized is to say that its propositions are false. An idealized theory about how we should organize society is, by definition, a theory containing falsehoods about how society should be organized: it gets the relevant facts – in this case, normative facts – wrong (List and Valentini 2016, 546). But the idealization critique, as we have seen, targets the OP

232

Laura Valentini

thought experiment. And the presence of idealizations there need not problematically affect the principles (i.e., the theory) eventually selected by the parties. To determine whether they do, we need to better understand the role played by these idealizations in the OP. As Onora O’Neill remarks, if theorists of justice “do not offer reasons for starting from these idealizing assumptions … their theories will, strictly speaking, be inapplicable to the human case” (O’Neill 1996, 41). At least when it comes to Rawls’s OP, reasons are indeed being offered. In particular, the aim of such idealizations is twofold. First, they are meant to allow us to describe an initial choice situation that is fair, one where factors we would naturally consider unjust – such as bias or self-interest – are assumed away. In other words, the OP models a somewhat impartial point of view, and in order for it to do so, it must make counterfactual assumptions. After all, if we were to ask ourselves what principles would be chosen by real-world agents, including rich and poor, powerful and powerless, I doubt we’d converge on the conclusion that the outcome of that agreement reflects the demands of justice. The idealizations in the OP, therefore, are not meant to offer an improved version of the circumstances to which Rawls’s theory is meant to apply. Indeed, those broad circumstances are explicitly not idealized in the OP.9 The parties decide in light of those circumstances: including moderate scarcity, limited altruism, natural and social scientific laws, etc. The principles selected are assumed to apply to human beings as they are, and the parties know this. The description of the parties in conjunction with the veil of ignorance, in turn, is designed to capture our sense of justice: what would be a fair decision situation, not what flesh and blood human beings, to whom the principles should apply, are like.10 Second, the OP, as a justificatory device, is meant to embody premises that are less controversial and more widely shared than the conclusion they lead to, namely the two principles of justice. Suppose Rawls had presented those principles straightaway, putting them in front of us. We may well have responded: “Why these principles, and not some of their competitors (e.g., Marxist, libertarian, utilitarian, etc.)?”. The OP is designed to offer a response to this kind of challenge. If we suppose, as Rawls did, that readers find the fairness of the OP more immediately plausible than that of the principles selected in it, showing that those 9 10

Barring society being a closed system in an increasingly globalized era, which admittedly comes with its own limitations. For discussion of how the OP embodies counterfactuals that may well be true of existing humans (qua counterfactuals), see Lawford-Smith (2010, 362–64).

On the Role of the Original Position in Rawls’s Theory

233

principles are the output of the OP lends credibility to them. In light of this, it is unclear why the presence of idealizations in the OP should make Rawls’s theory – that is, his two principles – inapplicable to its intended subject matter. Of course, one might take issue with the specific idealizations characterizing the OP. In other words, one might question Rawls’s claim that the OP should be the favored characterization of the initial choice situation and suggest that, instead of Rawls’s idealizations, others should have been selected. This more specific complaint, while still potentially damaging, is also consistent with Rawls’s broader method of reflective equilibrium, which, as we have seen, is by its very nature open-ended. The OP and its ensuing principles may not stand in reflective equilibrium with some people’s considered convictions, such that they would prefer a different description of the initial situation. But this is not to question the use of idealizations, or Rawls’s methodology more broadly, but to challenge the idea that the specific idealizations Rawls uses accurately reflect “our” sense of justice. As I mentioned earlier, one such specific idealization has been the object of particularly intense criticism, namely Rawls’s assumption that the parties in the OP select principles in light of full compliance. Discussion of partial compliance – that is, of cases involving violations of justice – is confined to a minor role in Rawls’s opus. The worry, here, is that focusing on full compliance is equal to “abstracting away from realities crucial to our comprehension of the actual workings of injustice in human interactions and social institutions, and thereby guaranteeing that [justice] will never be achieved” (Mills 2005, 170). This is why ideal theory may then become perniciously ideological: unwittingly supporting rather than challenging injustice. There is something to this critique, but this “something” – I think – does not invalidate Rawls’s enterprise. If the point of the enterprise is to identify what justice for creatures like us requires, then it seems necessary for the selectors in the OP to assume that their counterparts will indeed comply with the relevant principles. If we want to know what justice, with a capital J, is, we need to ask what everyone’s “fair share” would be, supposing everyone else will do their part (See Simmons 2010, 10). The question of what ought to be done when some violate the demands of justice, as these are understood under full compliance, is a separate matter.11 11

Rawls believes that knowing what justice under full compliance requires is a prerequisite for addressing questions of partial compliance. Whether this is so, however, is a matter of contention.

234

Laura Valentini

In short, if one wants to pursue the question about “full” or “perfect” justice, as opposed to the question of what remedial measures should be taken when full justice is not achieved, then the full-compliance assumption seems perfectly justified. A critique such as Mills’s is thus best interpreted as targeting the very choice of question Rawls, and others following him, have made. This is where the “ideology” worry comes in: depicting perfect justice, and setting aside inquiry into injustice, can only serve the interests of the powerful. Why else focus on perfect justice, when there is so much to say about injustice and oppression? This framing of the critique in terms of “ideology,” while intriguing, is perhaps too quick. After all, a certain choice of focus may be the fruit of intellectual curiosity (e.g., one may want to know what justice is), or of the sincere but likely mistaken belief that manifest injustices do not need much theorization, or of the controversial but not implausible view that an account of perfect justice is necessary in order to address questions of injustice. Even if the framing in terms of ideology can be questioned, I do think Mills has a point when he denounces the excessive focus on “full compliance/perfect justice” characterizing a certain strand of analytic political philosophy as a whole.12 And it may well be true that this focus has largely been caused by the influence of Rawls’s work on the field. But this, in itself, seems to be no critique of the particular way in which Rawls conducts his own inquiry into the nature of justice. The problem is not Rawls’s use of certain idealizations in the OP but his focus on justice rather than injustice. To sum up, the use of idealizations in the OP does not make Rawls’s theory itself idealized or inapplicable to real-world circumstances. Questions remain about Rawls’s specific choice of idealizations and particularly about the broader impact on the discipline of his use of the full-compliance assumption. But these questions, so far as I can see, do not pose a threat to the integrity of Rawls’s theory qua theory of justice.

13.5.2 Reconsidering the Fact-Sensitivity Critique As we have seen, some critics charge Rawls’s theory with being too idealized, too focused on perfect justice. The fact-sensitivity critique does exactly the opposite: it holds that the OP method, requiring the parties to select in light of various facts, renders Rawls’s theory not a theory of justice proper. Instead of getting us to fundamental justice, the OP leads us to 12

For a more recent critique of the “perfect justice” paradigm, see Sen (2009) .

On the Role of the Original Position in Rawls’s Theory

235

principles apt for regulating society’s institutions, in light of values that include but are not limited to justice. Efficiency, stability, etc. also matter. But these, so the worry goes, are all concessions to empirical facts. I fear that, once we keep in mind the particular function of a theory of justice as Rawls understands it, the fact-sensitivity critique misses its target. In Rawls’s view, the function of that theory is to provide principles lending a stable basis for a fair division of benefits and burdens of social cooperation. If that is the function of those principles, then the principles must be designed based on general social scientific and psychological facts about society and humans. If the principles were not able to engender their own support, or to govern a stable community, they would just be poor principles. For a simple but hopefully effective analogy, consider the following. Suppose I wanted to bake chocolate cookies, and suppose I like them chunky. I proceed to write down a recipe according to which chocolate chunks must be added at the very end of the process. Someone then comes along and insists that my baking efforts cannot possibly deliver chocolate cookies proper, because such cookies have to be smooth: one must melt chocolate already in the cookie mix instead of adding chunks at the end. I must confess that I would not take such a complaint very seriously. Whether the recipe is right or wrong depends on the particular outcome we want. If the recipe is to be a guide to baking chocolate cookies that are chunky, then I’ve done a good job. If the cookies are meant to be smooth, then I haven’t. Suppose my inflexible gourmet friend says: “But chocolate cookies are by definition smooth and not chunky!” Well, then I’d say we are just disagreeing about terminology. If he’d prefer that I put a different label on my cookie box, say “biscuits with chocolate chunks,” I’d be happy to do that. But there is, in fact, nothing wrong with my recipe, given what I wanted to produce. And how exactly we should name a baked good is mostly a matter of convention: there isn’t an immutable truth of the matter about chocolate cookies. I think the same goes for justice. Fact-sensitivity critics may believe that Rawls’s enterprise should have been labeled something other than “a theory of justice.” Everything he did was right, according to the aim he set himself, he just chose the wrong label for that aim. Instead of principles of justice, or social justice, perhaps he should have labeled his principles something like “fair rules of social regulation.” Admittedly, the labeling is not quite as appealing, but apart from linguistic appeal, I’m not sure anything would be lost via this label-switch. What is more – though this claim may be controversial – it isn’t obvious to me that there

236

Laura Valentini

is a truth of the matter about what justice is.13 There are truths about what we ought to do, but what we place under the rubric of justice vs. some other value is, to a large extent, a matter of bookkeeping. Some ways of organizing our conceptual space may be more or less illuminating than others, but so long as we are clear about what we mean by “justice,” different functional roles may well coexist for the same concept, just like the expression “chocolate cookies” can denote both chunky and smooth ones: what matters is being clear about which one we intend to bake. The fact-sensitivity critique too, therefore, doesn’t appear to deal such a lethal blow to the OP and Rawls’s theory, once we keep in mind the selfprofessed function of that theory. 13.6

Conclusion

I have reconsidered two prominent critiques of Rawls’s original position thought experiment, each pulling in a different direction. I have then argued that once we keep firmly in view the role that the OP plays in Rawls’s broader theory, and the function that theory sets itself, the force of those critiques can be significantly reduced. What my discussion has highlighted, however, is the fact that those critiques, while superficially directed at the OP, are perhaps best understood as targeting the aim of Rawls’s theory. Fact-sensitivity critiques ultimately boil down to the claim that a theory of justice should not be concerned with stable rules for social cooperation but with something else. As already mentioned, I am not sure that there is a truth of the matter about this. But suppose you disagree and believe that there is a truth about what should be called “justice.” If so, it still remains true that this is the terrain over which advocates of Rawls and champions of the fact-sensitivity critique should “battle it out.” The OP should be left out of the fight, so to say. Similarly, idealization critiques seemed to start from the assumption that the job of a theory of justice is, primarily, to denounce and help mitigate injustice in a somewhat direct way. If this is what we want, then admittedly Rawls’s theory doesn’t quite deliver. But this again doesn’t highlight a problem with the OP. The broader question is what the function of a theory of justice should be, or, perhaps more aptly put: what questions political philosophers who are broadly interested in “what we owe to each other” should address. But this question, however important, is not one the answer to which can be used to criticize the methodology Rawls employed to answer his own question. 13

For a defense of the claim that concepts – unlike propositions – cannot be true or false, see List and Valentini (2016, section 3).

Part IV

Pluralism, Democracy, and the Future of Justice as Fairness

14

Public Reason at Fifty Kevin Vallier

This volume commemorates the fiftieth anniversary of A Theory of Justice. Some philosophers worry that A Theory of Justice no longer merits the attention it receives. The profession has understood and digested Rawls’s work. Research on A Theory of Justice has become moribund and scholastic (Enoch 2015). We should move on to new projects. Some aspects of Theory have run their course. But one of Rawls’s projects continues to bear fruit: the project of public reason liberalism. The public reason project begins before A Theory of Justice and continues past Political Liberalism. Rawls raised some critical lines of inquiry and left them unexplored. Gerald Gaus’s approach to public reason refreshed the project.1 Public reason remains a fertile research program fifty years on. I show the fertility of the public reason project by reviewing seven models of public reason. I start with the model in Theory and explore the transition to Political Liberalism and the introduction to its paperback edition. I then explore Rawls’s unaddressed problem of justice pluralism. That is, Rawls did not contain reasonable disagreement about justice (Vallier 2019). Failing to stop it requires formulating a fourth model of public reason. Justice pluralism places the Rawlsian at a crossroads. If she denies it, she must rely on reasonableness to secure agreement on justice. Depending on the idea of the reasonable has well-known drawbacks. The chief worry is that relying on the reasonable threatens to settle too many questions by stipulation. That will make the public reason project moribund.2 Yet if Rawlsians accept justice pluralism, the public reason project must change. It must follow Gerald Gaus’s line of inquiry (Gaus 2011a). 1 2

Gaus (2011a, 2016, and 2020). But Gaus (1990, 1996) also engages the idea of public reason. Gaus (1996, 131–32) points out grave ambiguities in Rawls’s conception of the reasonable.

239

240

Kevin Vallier

Since we should acknowledge justice pluralism, I turn to review Gaus’s work. I start with the model of public justification in The Order of Public Reason, our fifth model of public reason. I then examine two more models. The sixth draws on The Tyranny of the Ideal, the seventh on The Open Society and Its Complexities.3 The seventh model has only begun to bear fruit. Indeed, it generates a research program, which Gaus called the New Diversity Theory.4 My aim here is not to address problems in Gaus’s work but to show how the public reason project has evolved as a fertile research program. I will not review the doctrine of public reason. Readers of this piece know the basics or will pick up the details as I review the models.5 Nonetheless, the overarching problem is this. Human beings are free and equal. Yet we must direct one another to engage in certain lines of conduct rather than others. We tell each other what to do. If we are free and equal, how can we have authority over one another? For Rawls and Gaus, we have authority over one another if we only accept shared rules of behavior.6 When we demand that others follow these rules, we only insist that they live up to their standards. Freedom, equality, and authority coexist. In short, moral and legal rules derive their authority from their mutual acceptability. Laws gain authority if and only if they are publicly justified for citizens. The character of justification varies with the model of public reason we use. Let’s turn to the first model now. 14.1

A Theory of Justice

Theory’s project extends beyond identifying two principles of justice. Rawls meant his theory of justice – justice as fairness – to play two different roles.7 First, justice as fairness should fit or “congrue” with a shared conception of the good life.8 Complying with just institutions should congrue with each person pursuing their good. Establishing this fit requires showing that citizens of a well-ordered society develop 3 4 5 6

7 8

I will refer to Gaus’s The Order of Public Reason, The Tyranny of the Ideal, and The Open Society and Its Complexities as OPR, TI, and OSC. Gaus (2018) contains some of these ideas in utero. Vallier (2018) reviews core concepts and conceptions of public justification. Yet we must not elide differences. Rawls publicly justifies constitutional essentials and matters of basic justice. Gaus, by contrast, focuses first on what he calls “social morality.” He then develops an approach to publicly justifying political institutions. Compare here Political Liberalism, xlvii and Gaus (2011a, 2–13). Or rather, many roles. See TJ 1999, 3–6. TJ 1999, section 86. Or, more accurately, Political Liberalism’s reading of what A Theory of Justice presupposed. See PL, XVI.

Public Reason at Fifty

241

morally. They will regard acting justly as good for its own sake and as part of their broader good (TJ 1999, section 86). Call this the rationality condition. Second, each person will see acting justly as good for her as long as others do. She can act justly with others by following the directives of joint institutions.9 Citizens cooperate from a sense of reciprocity and enjoy the good of justice. Suppose then that just institutions coordinate the pursuit of our respective interests. In that case, our society becomes inherently stable. Such stability comes from moral motives rather than state imposition. Call this the stability condition (TJ 1999, 436; PL, 140–44). Justice as fairness, then, provides us with two principles of justice. They meet the rationality and stability conditions. To stably realize the good of justice, a conception of justice must fit our visions of the good. We can thus construct A Theory of Justice’s model of public justification as follows. “J” refers to justice, and “CD” refers to a comprehensive doctrine. The arrow represents citizens endorsing justice as fairness as congruent with their good. We can say that J is publicly justified for people that adopt CD. In A Theory of Justice, the well-ordered society model includes all or most members of the public.

Figure 14.1 Model 1: A Theory of Justice

In Model 1 (Figure 14.1), a single comprehensive doctrine endorses a single conception of justice. I oversimplify a bit here. Rawls thinks that justice must be “the public basis of the terms of social cooperation” (TJ 1999, 122). But in Political Liberalism, he argues that A Theory of Justice “regards justice as fairness … as comprehensive” or partly so (PL, XVI). Justice as fairness includes goods that reasonable people might contest as goods. But we can still distinguish between justice as fairness and the goods it provides. Justice as fairness was partly comprehensive. But it 9

“The hazards of the generalized prisoner’s dilemma are removed by the match between the right and the good” (TJ 1999, 505).

242

Kevin Vallier

never answered most questions about the good life. Instead, Rawls thought justice as fairness fit with other shared intrinsic goods. 14.2

Political Liberalism

Rawls grew dissatisfied with the model of a well-ordered society in A Theory of Justice (PL, XVII–XXX). Its well-ordered society contains a destabilizing dynamic: the burdens of judgment (PL, 54–58). These factors lead to systematic disagreement about politics and morality. They lead people to develop distinct but reasonable comprehensive doctrines (PL, 58–67). Some of these doctrines might embrace new goods that override the good of justice as fairness. A paradigmatic challenger is Roman Catholicism. Catholics might regard eternal beatitude – union with God – as the greatest good (Catholic Church 2012, 426). Beatitude outranks worldly justice in cases of conflict. If so, Catholics may begin refusing to follow justice as fairness. Since cooperation is conditional, others may refuse to cooperate if others do not. Non-Catholics may refuse to obey just institutions when they see Catholics dissent. The regime will destabilize, or the state will impose stability by force. Force will create a modus vivendi regime, one that is stable but not for the right reasons. Both outcomes undermine moral stability, as Rawls understands it. To forestall this outcome, Rawls recast his theory. The recasting begins by grappling with the fact of reasonable pluralism. Reasonable people disagree about the good life due to the burdens of judgment. A conception of justice may not even fit all reasonable comprehensive doctrines. And it will not fit them all in the same manner. Justice as fairness should fit like a “module” into reasonable views. But diverse fits will abound (PL, 144–45). Another revision: justice as fairness becomes a political conception of justice (PL, 11–14). It governs political life and citizenship, not moral life as a whole. Justice as fairness must become the object of an overlapping consensus (PL, 144–50). Once reached, rational and reasonable people will follow just institutions (PL, 387–91). Almost. Even if we reach an overlapping consensus, we may miss stability for the right reasons. We now must assure those with different values that we accept the same conception of justice. Weithman argues that Rawls meant public reasons to serve as assurance signals.10 When we use public reasons, we assure others that we 10

Weithman (2015) provides an excellent summary of this role for public reason, first developed in Weithman (2010).

Public Reason at Fifty

243

Figure 14.2 Model 2: Political Liberalism

embrace justice as fairness. Society builds mutual assurance. Its members learn that they all adopt justice as fairness. The well-ordered society becomes stable for the right reasons. Note that in this model (Figure 14.2), conceptions of the good vary, but conceptions of justice vary not at all. Everyone shares the “very same” principles of justice. (PL, 35) We now see many comprehensive doctrines as reasonable yet incompatible with one another. Justice as fairness fits with each reasonable comprehensive doctrine but in distinct ways.

14.3

The Paperback Edition of Political Liberalism

Rawls became dissatisfied with Political Liberalism’s model of a wellordered society. The problem of justice pluralism loomed (PL, XXXVI; see also Gaus and Van Schoelandt 2017; Weithman 2017; and Vallier 2019). If reasonable people disagree about the good, why not justice too? Rawls acknowledged the challenge in the paperback edition. Justice as fairness becomes one of a family of reasonable political conceptions (PL, xlvi). We must now grapple with Rawls’s same concerns under conditions of justice pluralism. In Model 2, we agreed on justice, and our agreement had a single object. No longer. In response, Rawls develops a third model of public justification. This one allows for justice pluralism. In the second model, public reason contains several conceptions of the good. In the third model, public reason has several conceptions of justice too (Gaus 2011b). Rawls limited the second set to help societies coordinate on one conception. If members disagree modestly, a disagreement might not destabilize wellordered societies. Rawls places three limits on the set of reasonable conceptions of justice. All members must:

244

Kevin Vallier

(i) identify a wide range of liberties characteristic of a democratic regime, (ii) assign those liberties a special (though not lexical) priority over other considerations, (iii) guarantee all-purpose means that people need to enjoy their liberties’ worth. (PL, xlvi) These conceptions of justice vary. Some may omit the difference principle, for example (PL, XLVII). And reasonable people can adopt different rankings of conceptions of justice. Hence Model 3 (Figure 14.3) from the paperback edition of PL.

Figure 14.3 Model 3: The paperback edition

Each reasonable comprehensive doctrine endorses one reasonable political conception as best. But all reasonable comprehensive doctrines accept all members of the set as reasonable. Yet they rank conceptions of justice differently. And they share no principles that generate a social ranking. We have what Gaus calls an optimal eligible set of conceptions of justice (Gaus 2011a, 323). Each conception is reasonable, but we cannot derive a collective ranking. 14.4

Justice Pluralism

A new question arises. How big is the set? If the group of conceptions of justice is small, then Model 3 is close to Model 2. Rawls could rest easy. But if the collection is extensive, problems arise immediately. Consider an example. William Edmundson reads Rawls as a liberal socialist (Edmundson 2017). Rawls thought that liberal socialism was one way to realize justice as fairness (PL, 416). Let’s assume justice as fairness is in the optimal eligible set. By realizing justice as fairness, liberal socialism becomes legitimate for a well-ordered society.11 11

An economic regime is legitimate under two conditions: (1) morality permits imposing the regime on citizens and (2) citizens have no moral right to resist those impositions.

Public Reason at Fifty

245

Now consider John Tomasi’s free-market fairness (Tomasi 2012). Tomasi’s Rawlsian approach to justice meets Rawls’s three conditions. It contrasts with Rawls by treating more economic liberties as basic. For Rawls, economic rights include personal property and freedom of occupation. Tomasi includes rights to own productive property and freedom of contract (Tomasi 2012, 68–84). Since democracies protect such rights, Tomasi may treat them as fundamental liberties.12 But free-market fairness generates its own ideal regime types – capitalist regimes (Tomasi 2012, 203–14). If free-market fairness occupies the optimal eligible set, liberal capitalism is legitimate. If both conceptions are reasonable, socialism and capitalism become legitimate for citizens who adopt the opposing economic ideology. Legitimacy floats too freely from justice.13 Legitimacy helps explain why morality sometimes permits unjust state actions. States often pass unjust policies. But a state retains its right to rule even if it sometimes errs in particular, and it may still make laws and policies for all. The idea of legitimacy explains how the state retains its right to rule when it acts unjustly. The regime is legitimate, and so it has the authority to enforce. And so, legitimacy is looser than justice. But it cannot be too loose. Legitimacy gives the state leeway to err, but massively unjust regimes lack legitimacy. Any theory that legitimizes every regime between liberal capitalism and socialism is deficient – legitimacy is too broad. Rawls should have defended his sharp limits on reasonable conceptions of justice. But he did not. He instead offers a list of constraints, some of which seem arbitrary (PL, LVI–LVII).14 A political conception is unreasonable if it lacks public financing for campaigns.15 Any society without public funding, even egalitarian ones, is unreasonable. That seems questionable. So, the burdens of judgment stuff the eligible set of conceptions of justice. Model 3 must allow deep justice pluralism, transforming it into Model 4 (Figure 14.4). Model 4 legitimizes too many regimes by equating justice pluralism with “good” pluralism. Jonathan Quong defends narrow limits on liberal political conceptions (Quong 2011, 192–220). Justice pluralism has a “justificatory” rather 12

13 14

See Gaus (2011a, 511–21). That liberty prevails in democracies may not suffice to make it “characteristic.” But Rawls does not provide an account of which liberties are “characteristic.” I am grateful to Paul Weithman for discussion on this point. One may argue that free-market fairness is a reasonable liberal conception. As far as I know, no Rawlsian has said as much. 15 PL, LVIZ–LVII. PL, LVI.

246

Kevin Vallier

Figure 14.4 Model 4: Symmetry with wide legitimacy

than a “foundational” character. In brief, disagreements about justice aren’t as deep as disagreements about the good. I deny this (Vallier 2019). Too many doctrines interweave justice with the good. Consider Catholicism again. If we disagree profoundly about Catholicism, we can thereby disagree profoundly about justice. A reasonable Catholic perspective yields a disagreement about justice. Quong’s strategy cannot end deep justice pluralism. If we accept Model 4, we face another challenge. It is no longer clear why we cannot legislate a reasonable conception of the good. We all think one conception of the good is best. But most acknowledge one or more alternatives as reasonable. Catholics may find Eastern Orthodoxy reasonable, and Sunni Muslims might see a Shia regime as reasonable. In Model 4, we can impose liberal capitalism on socialists and vice versa and justify a perfectionist regime along the same lines if we accept broad legitimacy. Rawls’s work on public reason ended here. Neither he nor his students solved the problem of justice pluralism, so we must shift to another great theorist of public reason, Gerald Gaus.16

14.5

The Order of Public Reason

Gaus started working on public reason in the mid-1980s.17 He arrived at the topic through many figures, not only Rawls. Gaus defended public justification three years before Political Liberalism (Gaus 1990). Gaus used a public reason framework to defend democracy and liberal rights 16

17

Some may say that Gaus is not a true public reason theorist. Public reason addresses citizens as such. In this case, public reasons are shared or shareable reasons. But Gausian public justification parallels overlapping consensus justifications, which resembles what Rawls calls “full justification.” So, the objection to Gaus rules him out by stressing some parts of Rawls’s doctrine of public reason and not others. See Political Liberalism, 386. Gaus (1986) is his first published essay on the topic.

Public Reason at Fifty

247

three years later (Gaus 1996). By 1996, Gaus had written 800 pages on public justification in morality and politics. Gaus’s project, then, is not an extension of Rawls’s. Gaus’s case for a public justification principle differs from Rawls’s. Gaus does not restrict public justification to the domain of the political.18 He allows nonpublic reasons to figure into public justifications (Gaus 2011, 38–42). Gaus does not set aside controversial social scientific information.19 And he defends market order, though eschewing libertarianism (Gaus 2011a, 506–8). Fifteen years passed between Justificatory Liberalism and Gaus’s magnum opus, The Order of Public Reason. During that time, interest in public reason waxed and waned in certain quarters of the profession. But Quong’s Liberalism without Perfection and OPR revitalized discussion. We can read OPR as addressing difficulties in Model 4. Gaus accepts the symmetry of disagreements about goodness and justice. He places them on a par. Indeed, he doesn’t separate justice and the good as sources of justifying reasons. Free societies face pervasive “evaluative” pluralism (Gaus 2011a, 279–83). This pluralism applies to many moral and nonmoral disputes. Gausian models of public reason use more fine-grained concepts than Rawlsian models. For Rawls, public justification syncs conceptions of the good with conceptions of justice (PL, 385–86). Both Rawlsian concepts consist of many norms and values. Gaus preferred to use justificatory reasons and moral rules.20 Public justification proceeds in a fine-grained way. We justify each rule in accord with our justificatory reasons. Sometimes public justifications congeal into justifications for principles of justice. But they need not do so. We publicly justify constituent rules piece by piece. A public rationale for the whole might arise as spontaneous order.21 It will have no single system-level justification. But each piece will have a local justification. While Gaus’s model accepts symmetry, it rejects wide legitimacy. Legitimate states must coerce far less than the Rawlsian state (Gaus 2011a, 521–28). Gaus effectively treats justice like Rawls treats the good. No one ideal has moral authority over members of the public. Instead, members can converge on shared norms of justice but not whole conceptions. Rawlsians must follow Gaus here to respect justice pluralism 18 19 20 21

Gaus (2011a, 2–13) makes clear Gaus’s focus is on social morality broadly and not mere political morality. Gaus (2011a) appeals to social scientific findings throughout. Gaus (2011a) uses this language throughout. Gaus (2021, 128–38) explains how a publicly justified order is a kind of selforganized system.

248

Kevin Vallier

while avoiding broad legitimacy. Rawlsian states must only advance shared goods and shared norms of justice. Both Rawls and Gaus have their own set of terms of art in describing public justification. For ease of understanding, I will stick to Rawlsian jargon. For our purposes, we can represent Gausian ideas with Rawlsian labels. But recall that Gausian public justification allows for finer-grained public justifications. We can now depict Model 5, the first Gausian model (Figure 14.5).

Figure 14.5 Model 5: The Order of Public Reason

In OPR, individuals have their conceptions of goodness and justice, and both vary. Public justification justifies constitutional norms, laws, and policies to diverse perspectives. In Model 5, we represent everyone with varied conceptions of goodness and justice. They then pursue reconciliation between people with different views. Notice the new arrows between comprehensive doctrines. These represent a critical but often missed feature of Gaus’s work. Some have objected that Gaus’s view courts anarchy.22 Due to deep evaluative pluralism, we cannot show that any rules are publicly justified for a diverse public. In some cases, an empty eligible set is either no problem or a good thing. The optimal eligible set of racist laws is null – an excellent outcome. The more significant worry is that many important optimal eligible sets are empty. In that case, we cannot publicly justify institutions that seem pre-theoretically justified. The horizontal arrows answer the challenge. We have reasons to pursue our values and ideals. But we also have reasons to preserve moral relations with those who reject our worldview.23 Gaus’s doctrine of moral relations is complex. He often understates its centrality to his project. But he arguably defined moral relations as social relations bound by reactive 22 23

See Enoch (2013). For a compelling response, see Schultz-Bergin (2021). For a discussion of the centrality of moral relations to Gaus’s work, see Vallier (2022).

Public Reason at Fifty

249

attitudes. Reba and John have moral relations under two conditions: (1) when they may hold one another responsible for moral violations and (2) when they aptly experience resentment and indignation in response to violations. Moral relations are the thinnest nonstrategic relations between persons that recognize moral personality. Gaus grounds public justification in moral relations.24 Relations endure only if our practice of responsibility uses publicly justified rules. If rules lack public justification, we cannot coherently hold others responsible for violations (Gaus 2011a, 193–204). Others cannot see the rule as justified, even on reflection. We cannot aptly blame or punish others if they violate the rules. They cannot determine if the rationale binds them. Societies without publicly justified rules face two problems. First, large-scale social cooperation will break down. Large social orders cannot rest on instrumentalist motivations.25 Second, members struggle to form love, friendship, and trust relations.26 Yet large-scale social cooperation and closer moral ties benefit us. We have integration reasons for moral relations with diverse others.27 We value moral relations on both grounds. So, we have a solid reason to preserve them under diverse conditions. Gaus thus assigns each person integration reasons besides reasons to pursue their ideal. Each member has a two-variable function that evaluates a rule’s public justifiability. Owing to integration reasons, members will often accept morally suboptimal moral rules. They want to preserve moral relations, even if their ideal reasons rank other ethical rules as best. Noting integration reasons, members will populate the essential optimal eligible sets. And they will often populate it with more than one member. In the case of a non-singleton optimal eligible set, we have a choice problem. How do we converge on a member of the group? In some cases, we can use publicly justified decision procedures.28 But in other cases, we disagree about those too. Disagreements about decision procedures raise worries about a justificatory regress. Agents can now coordinate within the optimal eligible set to avoid the regress (Gaus 2011a, 395–97). They conform without a decision procedure. Gaus’s Kantian Coordination Game explains how convergence occurs. Agents may adopt rules and interact with others under those rules.

24 25 26 27 28

See Gaus (2011a), but the theme is pervasive in Gaus’s work. Again, see Vallier (2022). Gaus (2011a, chapter 2) provides a detailed argument for this claim. See Gaus (2011a), but Gaus draws on Gaus (1990, chapter 6). For this term and rich analysis of issues in Gaus (2011a), see Vincent (2021). Gaus (1996, chapter 13) explores this strategy.

250

Kevin Vallier

They will gradually converge on a rule because practiced rules have increasing returns (Gaus 2011a, 398–400). The more community members observe a rule, the more they have reason to adopt it. They can reconcile under the rule with an increasing number of others. And so, our integration reasons to abide by an extant norm grow as people adopt it. The model identifies a bandwagon effect. With time, the rule governs the whole community. Public justification is complete. We can represent the optimal eligible set graphically as shown in Figure 14.6. Condorcet Paradox preferences work here. The public will not agree on social choice rules to shrink the set further. Less controversial principles, like a Pareto principle, cannot shrink the set (Gaus 2011a, 321–32). Yet, if Gaus is correct, once social interaction begins, a bandwagon effect should form all the same. It can be represented as shown in Figure 14.7. Alf x

Betty y

Charlie z

y

z

x

z

x

y

Condorcet Paradox Preferences

Figure 14.6 Condorcet Paradox preferences

Here society has two rules in its optimal eligible set {x, y} and agents may choose to abide by either one. But, with time, more agents pick x, creating the increasing returns dynamic. 14.6

The Tyranny of the Ideal

Model 5 has a problem too. Our justice ideals label social worlds as just or unjust. They dictate how to organize society, in contrast to many (but not all) conceptions of the good. But Gaus treats ideals of justice like ideals of the good. Standards of justice become partially privatized. They play a different role than in Rawls’s theory. Justice ideals do not order society. But they help groups identify their best social world. (Gaus 2016, 42–50) They should also help those groups arrive at the ideal. But in a diverse society, people have different ideals, and communities fall far short of them. Gaus argued, on this basis, that we don’t know much about what our standards of justice recommend (Gaus 2016, 130–32). We also don’t know how they will work when we arrive. We

Public Reason at Fifty

251

All x

50/50

All y Time

Figure 14.7 Increasing return dynamics

do not know how to reach them, as the path may pose such great dangers that we should stick to what we know. Gaus altered his model of a publicly justified polity based on these considerations. The Tyranny of the Ideal model is now an “open society” with a few critical updates from Model 5 (Gaus 2016, 150–205). First, we assume members of the public do not know how to reach their ideals. And they do not understand how those ideals will take shape in the future. Second, any open society is a complex adaptive system (Gaus 2016, 118–27). The system has many parts. These parts have causal feedback relations with one another. Predicting the results of system-level changes is difficult (Gaus 2016, 138). Prediction is often impossible. Gaus builds epistemic humility into the model of a publicly justified order. We can now publicly justify moral rules if they ease social discovery. We must support institutions that form communities of inquiry into justice. Gaus calls these “republican communities,” following Fred D’Agostino (Gaus 2016, 97–8, drawing on D’Agostino 2010, 138–41). Gaus thought that these communities could learn from their compatriots. He also thought we learn from other communities. Though crosscommunity social learning is tricky (Gaus 2016, 114–20). For these reasons, Gaus supported markets, democracy, and freedom of association (Gaus 2016, 198–207). These institutions help people form republican communities and share information.

252

Kevin Vallier

Figure 14.8 Model 6: The Tyranny of the Ideal

In Gaus’s open society, diverse communities uncover critical insights through their interactions. Diversity becomes a resource for discovering and preserving moral relations. We reach TI’s model of public justification – Model 6 (Figure 14.8). Public reason now includes considerations about how to organize society for social discovery. TI offers an account of public justification. Yet Gaus discusses preserving relations of accountability, a term he uses elsewhere in association with public reason and maintaining moral relations (Gaus 2016, 218).29 TI implicitly and modestly modifies the optimal eligible set and Kantian Coordination Game. And so, it changes the Kantian Coordination Game. Gaus argues that moral norms in open societies tend to function as prohibitions (Gaus 2016, 187–97). They prohibit some lines of conduct but permit many ways of observing them. The rules may allow for norm innovation. The optimal eligible set might get new members. 14.7

The Open Society and Its Complexities

In OSC, Gaus explores social complexity.30 Open societies are complex adaptive systems. Their development is unpredictable. We struggle to determine which norms achieve public justification (Gaus 2021, 6–10). We must remain humble about our grasp on our ideals. Otherwise, we will squelch complexity rather than manage it. Gaus thinks open orders benefit us. Yet they exhibit autocatalytic diversity: disagreement selfamplifies (Gaus 2021, 109–17). Sometimes diversity and disagreement resolve into more straightforward points of agreement or dissensus. But the dangers of diversity remain. 29

As I argue in Vallier (2022).

30

Gaus (2021, 118–27) goes to the heart of the matter.

Public Reason at Fifty

253

Gaus also recognizes that the Kantian Coordination Game made an unrealistic homogenizing assumption (Gaus 2021, 147–58). It assumes that people assign equal weight to their ideal and integration reasons. All adopt the same trade-off rate between the two sets of reasons. Gaus argues that this assumption is unrealistic due to cognitive diversity. Some people prefer to conform, but others pursue moral change. And so Gaus modifies the Kantian Coordination Game. It now allows different trade-off rates between ideal reasons and integration reasons. Suppose agents vary the weight of their integration and ideal reasons differently. If so, OPR’s bandwagon effect may not occur, and it might not even appear. If some prefer their ideal, they may discount shared returns to moral requirements, and they place little value on coordinating with others vs. establishing their preferred rule. If moral reformers disagree, the Kantian Coordination Game may polarize the population. Members coordinate on different and conflicting eligible rules. We sometimes see this in polarized societies like the United States. To address the problem, Gaus built an agent-based Kantian Coordination Game (Gaus 2021, 147–58). In an agent-based model, the modeler programs individual agents. She then observes system-level patterns. Gaus’s model outlines when cognitively diverse agents will coordinate on a rule. Gaus thought societies with shared weights reconcile better than societies with diverse weights. His model showed otherwise. Communities with various rankings coordinated within the eligible set more often (Gaus 2021, 147–58). Diversity helps us reconcile.

Figure 14.9 Model 7: The Open Society and Its Complexities

Here I allow different weightings of ideal reasons and integration reasons, and I represent the different weightings through the thickness of the various arrows. The first member cares more about reaching his ideal than integration with others. The third member cares more about integration than reaching his ideal. Model 7 is the newest model of public justification in the public reason literature, the apex of a half-century of thought (Figure 14.9). Disagreement and diversity have forced modification after modification.

254

Kevin Vallier

14.8

The New Diversity Theory

Gaus’s final two books feature two discoveries. In TI, republican communities help open societies discover the truth about justice. OSC adds that conflicting weightings of ideals and integration further aid reconciliation. In both cases, diversity serves as a resource for social cooperation rather than a threat. Seeing this, Gaus asked whether the fundamental insight generalizes beyond his project. Around 2018, Gaus began developing a research program, The New Diversity Theory (NDT).31 The slogan: diversity is a resource for social cooperation, not only a challenge. But we need more than a slogan to move forward. Different theorists will develop Gaus’s program in different directions. In my view, NDT follows the close of social contract theory’s secondwave. Social contract theory’s first wave includes the great modern political philosophers. The key figures are Thomas Hobbes, John Locke, Jean-Jacques Rousseau, and Immanuel Kant.32 It dies in the early nineteenth century. The second wave begins in the 1950s. Seminal theorists include James Buchanan, John Harsanyi, Rawls, and David Gauthier.33 The tide rises with the development of the idea of public reason in the 1980s and 1990s. Chief figures include Jurgen Habermas, Rawls, Charles Larmore, D’Agostino, Gaus, and others.34 The second wave began to decline in the 2000s, but Gaus, Quong, and Paul Weithman gave it new life.35 Their books split the public reason project into three traditions. Weithman continues Rawls’s theory, though in a more sophisticated form. One might call this the original project. Quong’s project preserves the Rawlsian apparatus to make it internally coherent. Reaching coherence requires resolving some critical problems with the original project. Ryan Muldoon and I have called this the coherence project (Vallier and Muldoon 2021). Gaus focuses on allowing diversity to run its course. What we call the diversity project explores when diversity helps justify liberal order.36

31 32 33 34 35 36

Gaus (2018) uses the phrase for the first time. Gaus draws this theme out of these four thinkers in Turner and Gaus (2018). Harsanyi (1955); Buchanan and Tullock (1962); Gauthier (1986); Theory of Justice. Political Liberalism; Larmore (1987); D’Agostino (1996); Gaus (1996); Habermas (1998). Weithman (2010); Gaus (2011); Quong (2011). Ryan Muldoon and I describe and contrast these projects in Vallier and Muldoon (2021).

Public Reason at Fifty

255

Gaus played a prominent role in social contract theory’s second wave. But Models 6 and 7 have moved beyond it. Gaus abandons identifying principles of justice as an object of an agreement. He thinks of that project as planning social order (Gaus 2021, 139–46). Institutions receive public justification because they help us discover what justice requires (Gaus 2021, 241–46). The second-wave social contract theorists thought they could divine agreement points. Contracting parties must find agreement points within the choice process. In OSC, institutions receive public justification for several reasons. They harness cognitive diversity, aid social learning, and manage complexity. Parties no longer choose a shared social plan. They choose rules for a complex social system that they struggle to understand (Guas 2021, 147–59). Gaus’s project remains contractarian in a broad sense. Political philosophy models free and equal persons who are choosing norms and institutions. But parties no longer choose a shared plan, nor do they even select simultaneously. They choose rules to govern a complex and ever-changing social system, which they barely understand. Gaus thought that few saw the challenge that modern diversity poses for cooperation. But Gaus designed NDT to show that all is not lost. Take Gaus’s discoveries in TI and OSC to illustrate. The open society can improve itself and reach public justification by organizing social institutions for social discovery. It welcomes diverse cognitive types into moral disagreements. We achieve more reconciliation and insight with diversity than without it. As I see it, NDT is a research program with the following features: (1) Contractarian Problematic. Identify terms of social life that resolve our competing claims. These terms must preserve our freedom and equality. (2) Social Scientific. Identifying terms of social life requires that theorists use models and data from the social sciences. (3) Deontological Moral Psychologies. Humans act from moral considerations, such as fairness. Proper contractors are fair dealers. (4) Moral Systems. Humans create moral structures – systems of moral norms that govern interaction, and these systems figure into any plausible social agreement. Constitutional choice can create, reform, or abolish ethical systems. (5) Diversity. Moral agents have a wide range of conflicting values and beliefs. The rules of ethical systems vary between moral communities because diversity destroys the prospects for agreeing on moral principles.

256

Kevin Vallier

(6) Spontaneity. Diverse moral systems often arise spontaneously from interactions between diverse moral agents. (7) Complexity. Moral structures exhibit social complexity (many parts, with feedback relations). Agents choose rules to govern a complex moral system. (8) Complex Social Choice. Contractors choose rules through interaction, embedded in complex and partly spontaneous moral systems. (9) The New Diversity Claim. Diverse moral psychologies and moral systems improve our capacity for complex social choice. NDT proposes that diverse perspectives and moral systems help people cooperate. NDT tries to vindicate the New Diversity Claim. NDT overlaps with the aims of public reason liberalism, but they are not the same. Public reason liberals reject tenets of NDT. And New Diversity Theorists can reject public justification requirements. Nonetheless, NDT encourages public reason liberals to abandon the coherence project and follow the diversity project. The coherence project has value. But we must not waste too much effort on its more counterproductive aspirations. 14.9

Public Reason, Alive and Well

I have reviewed seven models of public reason and public justification. They stretch across half a century. We see how diversity and disagreement force each stage between Rawls’s first and Gaus’s last books. The continuity enabled Gaus to create a new research program. In sum, public reason is not a derelict line of research. Some of its branches may have reached a dead end. But not all. That is a testament to Rawls’s work on public reason – and Gaus’s too.

15

Reasonable Political Conceptions and the Well-Ordered Liberal Society Samuel Freeman

John Rawls sets forth two distinct accounts of the well-ordered society. First there are the familiar versions of a well-ordered society of justice as fairness Rawls describes in A Theory of Justice and the first edition of Political Liberalism. But in the 1996 introduction to the paperback edition of Political Liberalism, Rawls introduces the idea of a well-ordered liberal society,1 and a well-ordered constitutional democratic society in the 1997 “The Idea of Public Reason Revisited” (PL, 490). I focus on these less familiar ideas. A liberal or constitutional democratic society is wellordered for Rawls when governed by one or another reasonable liberal political conception realized in its laws and institutions and accepted and normally complied with by its citizens. Reasonable liberal political conceptions began to assume a prominent position in political liberalism after 1996. Indeed, in the unpublished 1998 revisions to Political Liberalism, Rawls says: “political liberalism is to be understood as a freestanding family of reasonable liberal political conceptions.”2 Here I discuss in Section 15.1 the well-ordered liberal society and then in Sections 15.2–15.3 the primary features of reasonable liberal conceptions and the centrality of the criterion of reciprocity within them. I contend in Section 15.4 that, while some reasonable conceptions may be nonegalitarian, liberal conceptions are not reasonable if they fail to impose any formal restrictions on permissible inequalities or fail to limit the effects of wealth on democratic politics. This includes not only libertarianism but also a wide range of positions I call “neo-liberal,” including the classical and libertarian-liberal positions of Hayek, 1

2

Political Liberalism, XXXVI. Except for the two introductions, the pagination of the 2005 edition is the same as the 1993 original and 1996 paperback editions. Unfortunately, the index was not changed in 2005 to reflect this difference in pagination. 1998 revisions to Political Liberalism, Lecture IV, §3, 2d par., 115, referred to here as “PL 1998 rev.” with lecture, section, and (where available) page numbers. Rawls proposed a revised edition in his 1998 letter to the editor, reproduced at PL, 438–39. These seven revised lectures remained unfinished and are in the Harvard Rawls Archives.

257

258

Samuel Freeman

Friedman, Buchanan, Epstein, Gaus, Tomasi, Schmidtz, Brennan, and others. None of these are reasonable liberal political conceptions according to Rawls’s criterion of reciprocity, and none satisfy his liberal principle of political legitimacy, since all permit and even encourage unrestricted inequalities and individuals’ unlimited accumulation of wealth and economic powers. 15.1

The Well-Ordered Liberal Society

A well-ordered society in Theory is a society whose basic institutions conform to the principles of justice and whose members accept and normally comply with requirements of justice. A well-ordered society aims to be a “perfectly just society,” wherein “everyone is presumed to act justly” (TJ 1999, 8) within “a perfectly just basic structure and the corresponding duties and obligations of persons” (TJ 1999, 216). These institutions, laws, and moral requirements are determined through the four-stage sequence and are perfectly just as a matter of pure procedural justice. Rawls in Theory occasionally alludes to a well-ordered society as being itself perfectly just (TJ 1999, 8, 216, 417–18). Here Rawls must be referring, not to any realistically attainable well-ordered society but to the hypothetical ideal of society whose basic structure is determined by the principles of justice through the four-stage sequence, which provides the criterion of moral correctness for our judgments of justice. For he says, “The best attainable scheme is one of imperfect procedural justice” (TJ 1999, 198). In Political Liberalism, he explicitly recognizes that in realistically ideal human conditions, there still will be mistakes of reasoning, factual beliefs, and other “burdens of judgment,” implying that any actualization of the ideal of a well-ordered society cannot itself be perfectly just in all respects. Some laws, conventions, and conduct will depart from justice as fairness, even among reasonable citizens all of whom endorse and seek to comply with justice as fairness. It is this realistic ideal of a well-ordered society that is the best attainable scheme – a “realistic utopia” – which is as close to we can get to a perfectly just society. In Political Liberalism, Rawls revised his account of the stability of a well-ordered society of justice as fairness. Though everyone accepts, as in TJ, the principles of justice and their requirements, stability and social unity are no longer grounded in citizens all endorsing the same Kantian comprehensive doctrine but in an overlapping consensus of different and often conflicting reasonable comprehensive religious, philosophical, and moral doctrines. This is familiar to readers of Political Liberalism. Less familiar are the implications of the 1996 introduction to Political

The Well-Ordered Liberal Society

259

Liberalism, where Rawls expresses doubts about whether an overlapping consensus on justice as fairness is realistic, given the burdens of judgment. Ideally the most reasonable [basis for social unity is when] society is effectively regulated by the most reasonable conception and citizens are in wide and general reflective equilibrium about this. Practically the most reasonable basis, one that might actually come about, is one in which all citizens agree that the regulative political conception is reasonable and some think it the most reasonable. This is sufficient for political society to be stable for the right reasons: the political conception can now be honored by all citizens as at least reasonable and for political purposes that usually is the most we can expect.3

Rawls appears to suggest here that a WOS where all citizens endorse justice as fairness as most reasonable is an ideal, but may not be practically realistic – not a realistic utopia. The concluding phrase – “that usually is the most we can expect” – may leave the door open to the real possibility of all reasonable citizens agreeing on the most reasonable conception.4 But even if a WOS where all accept Justice as Fairness as the most reasonable conception is not realistically possible, still a well-ordered liberal society whose institutions all conform to Justice as Fairness, and where Justice as Fairness is accepted by all citizens as one among several reasonable liberal conceptions, is realistically possible. This in effect is Rawls’s reply to the first “main aim of Political Liberalism [which] is to say, how a well-ordered society of justice as fairness … is to be understood once it is adjusted to the fact of reasonable pluralism” (PL, XXXV– XXXVI). Rawls’s implicit reply is that a WOS of Justice as Fairness is to be understood now as a well-ordered liberal society that conforms to justice as fairness, with which all citizens normally comply and recognize as reasonable, even if not the most reasonable conception – this at least is still a realistic utopia. Notable here too is Rawls’s letter to the editor of July 1998 referring to “a revised edition of Political Liberalism” that incorporates changes to political liberalism made in “Public Reason Revisited” (PL, 438). Rawls says, “political Liberalism is about a family of reasonable liberal ideas of political justice” and “justice as fairness itself now has a minor role as but 3 4

PL, XLVIII (emphases added); l, in 1996 paperback ed. That Rawls sought to keep the door open to the realistic possibility of a well-ordered society of justice as fairness is suggested in Lecture IV, 2nd par., 105 of the 1998 revised manuscript, which retains the italicized sentence from PL, 133 and adds the new material that follows: “Social unity is based on a consensus on a single reasonable political conception, or on a mix of such conceptions. [footnote] I speak here of a mix because one of a family of the reasonable conceptions might win on some political issues, while another at other times carries the day” (emphases added).

260

Samuel Freeman

one political conception among others” (PL, 439). Rawls was trying to present political liberalism, not as a justification of the most reasonable conception of justice for a democratic society (as in the first edition of Political Liberalism) but rather as an account of the “family of reasonable liberal political conceptions” that would satisfy the criterion of reciprocity and meet the conditions of liberal political legitimacy. Understood as an account of liberal political legitimacy and the public justification of the exercise of political authority through public reason, political liberalism need take no position on what is the most reasonable conception of justice or whether a well-ordered society of justice as fairness wherein all reasonable persons accept it as most reasonable is itself realistically possible. Given Rawls’s remarks that A Theory of Justice and Political Liberalism are “asymmetrical” (PL, 490), this appears to be his final understanding of the relationship between his two main works. Justice as fairness aside, I am concerned here with the second “main aim of PL,” which “is to say how a well-ordered liberal society containing a number of reasonable political conceptions is to be understood” (PL, XXXVI). What does Rawls mean by a “well-ordered liberal society” and a “well-ordered constitutional democratic society” (PL, 490)? Rawls says of a well-ordered liberal society (WOLS), that “in this case there is both the fact of reasonable pluralism and a family of reasonable though differing liberal political conceptions” (PL, XXXVI). In a well-ordered liberal society, there is no longer agreement among all reasonable and rational persons on the same political conception of justice. Instead, there is agreement on “the family of reasonable liberal political conceptions.” What do the members of this “family of reasonable political conceptions” share in common that entitles Rawls to say that a liberal society can be “well-ordered,” when there is no longer agreement on the most reasonable conception or even on the same reasonable conception? Sections 15.2 and 15.3 set forth what I take to be the main formal and then substantive features of members of “the family of reasonable liberal political conceptions” that have political authority and bestow political legitimacy on laws and institutions within a well-ordered liberal society. 15.2

Formal Features of the Well-Ordered Liberal Society

(1) Reasonable Persons and Reasonable Political Conceptions. Rawls says necessary and sufficient conditions of what is reasonable (or rational) cannot be specified. He utilizes the term “reasonable” in several different ways depending on its context. “The idea of the reasonable itself is given in part…for our purposes by the two aspects of persons being

The Well-Ordered Liberal Society

261

reasonable” (PL, 94). A reasonable liberal conception is then one that would be accepted by a reasonable person, defined by two basic elements as (1) a person who has an effective sense of justice and is prepared to propose fair terms of social cooperation that “they consider the most reasonable conception of political justice” and then act on them even if contrary to their own interests (PL, 446; see also, 48–49, 81, 94, 375) and (2) a reasonable person recognizes and accepts the consequences of the burdens of judgment “for the use of public reason in directing the legitimate exercise of political power” (PL, 54, 81, 94, 375; PL 1998 revised, 44) .5 Rawls also defines “correct judgments of justice” as reasonable (rather than true) when they are supported by a preponderance of reasons of the right kind, namely “the preponderance of reasons specified by the principles of right and justice issuing from an [objective] procedure” that itself incorporates all the relevant reasons and principles for justifying these principles (PL, 111).6 The objective procedure is the original position (OP). As discussed later, Rawls sees the OP within Political Liberalism as identifying a “family of reasonable liberal conceptions.” Even if conflicting judgments of justice that issue from these reasonable conceptions are all reasonable, they cannot all be most reasonable and hence “correct judgments.” The idea of “correct judgment” within political liberalism could then refer either to judgments that are shared by all reasonable conceptions or to those judgments that are implied by the preponderance of reasons within the most reasonable conception of justice. (2)The Criterion of Reciprocity. Reasonable persons in a WOLS accept the criterion of reciprocity, that when fair terms of cooperation are proposed that they believe are the most reasonable, they “must also think they are at least reasonable for others to accept, as free and equal citizens, and not as dominated or manipulated or under pressure of an inferior political or social position” (PL, 446, emphases added). This criterion assumes a fundamental role in Political Liberalism once Rawls recognizes a plurality of reasonable political conceptions. Of reasonable conceptions, he says, “reciprocity thus becomes a basic idea” (PL 1998 rev., I, 5

6

There are other characteristics of reasonable persons Rawls discusses in Political Liberalism: they take account of the consequences of their actions on others’ well-being (PL, 49n); they want to be fully cooperating members of society and recognized as such (PL, 81); and they have a reasonable moral psychology (PL, 82) – but the two in the main text are especially important in setting forth the features of reasonable liberal political conceptions. Rawls says in Political Liberalism, lecture III, “procedure that correctly formulates the principles of practical reason in union with appropriate conceptions of society and persons” (PL, 111). But he later says that the 1998 revisions omit any reference to “principles of practical reason” or “practical reason” (PL, 438).

262

Samuel Freeman

§6, 26, n. 40). This criterion grounds the liberal principle of legitimacy (PL, 137, 216), which is modified in “Public Reason Revisited” to reflect the criterion of reciprocity: Our exercise of political power is proper only when we sincerely believe that the reasons that we would offer for our political actions – were we to state them as government officials—are sufficient, and we also reasonably think that other citizens might reasonably accept those reasons. (PL, 446–47)

Here it is important that Rawls’s criterion for what others might reasonably accept is specified within political liberalism and not derived from some reasonable comprehensive philosophical, moral, or religious doctrine. To mark this difference, he refers to the “politically reasonable.”7 Reasonable persons – who have an effective sense of justice and seek to cooperate on fair terms even when contrary to their own interests – often do not always accept many philosophical and moral truths. Thus, we cannot reasonably insist that other citizens might reasonably accept as political reasons the truths of comprehensive doctrines. What we can reasonably expect them to accept as reasons in their capacity as free and equal citizens are the political values of justice and public reason and accept as principles those that satisfy the criterion of reciprocity. (3)The Ideal of Free and Equal Citizens. Any reasonable political conception and thus any well-ordered liberal society must have at its core an ideal of reasonable and rational free and equal citizens, and some idea of the fair terms of social cooperation that are most reasonable for them to accept as such citizens. Each of these liberalisms endorses the underlying ideas of citizens as free and equal persons and of society as a fair system of social cooperation over time. Yet since these ideas can be interpreted in various ways, we get different formulations of the principles of justice and different contents of public reason. (PL, 450–51)8

And in the 1998 revisions: “The family of reasonable liberal political conceptions of justice for society’s basic institutions articulate the conception of reasonable and rational citizens as free and equal” (PL 1998 rev., II, §7, 70). (4) Public Reason and the Political Values of Justice and Public Reason. Any reasonable political conception accepts the expression of the criterion of reciprocity through public reason and accepts the political values of justice and 7 8

See PL, 378, 441, 481, 488 on the politically reasonable. Rawls says earlier on PL, 450: “The content of public reason is given by a family of political conceptions of justice, and not by a single one …. The limiting feature of these forms is the criterion of reciprocity, viewed as applied between free and equal citizens, themselves seen as reasonable and rational.”

The Well-Ordered Liberal Society

263

public reason (PL, 447).9 Rawls says “public reason is a way of reasoning about political values shared by free and equal citizens” (PL, 490, emphasis added). The role of a reasonable political conception is to give “content” to public reason, and the “political values of [justice and] public reason.” Rawls says: “the political conception is a reasonable expression of the political values of public reason and justice between citizens seen as free and equal” (PL, 246, emphasis added). It is important that the political values of public reason are shared among reasonable free and equal citizens with different reasonable comprehensive views.10 That political values are shared is a consequence of the criterion of reciprocity. Reciprocity requires that the political values that govern a democratic society be generally accepted and shared among all its members. This differs significantly from Jerry Gaus’s and related accounts of public justification that reject shared public reasons and regard public reason as the “convergence” on laws of different and conflicting nonpublic reasons. By contrast, in a well-ordered liberal society, there is no disagreement about the political values of justice and public reason among reasonable citizens in a well-ordered liberal society. When they disagree, it is about how to interpret political values and the appropriate balance of public reasons, as determined by different reasonable political conceptions they affirm. What then are these shared political values of public reason that all free and equal RR citizens accept? They include the values of justice and equal political and civil liberty, fair equality of opportunity, economic reciprocity – the social bases of mutual and self-respect between citizens (PL, 139, 202, 203); social equality and values of the common good are also political values (PL, 224). In “Public Reason Revisited,” Rawls mentions the political values in the preamble to the Constitution, including: justice, a more perfect union, domestic tranquility, the common defense, the general welfare and the blessings of liberty for ourselves and our posterity. Under the values of justice he includes “equal basic liberties, equality of opportunity, also ideas concerning the distribution of income and taxation, and much else” (PL, 453). Other political values mentioned are political autonomy, the family and the “orderly reproduction of society over time,” arrangements for the education of children, and institutions of public health (PL, 456); also the freedom and equality of women, the equality of children as future citizens, the freedom of 9

10

Emphasizing the connection between reciprocity and the political values of public reason in the 1998 revised PL Rawls says “the criterion of reciprocity is an essential ingredient specifying public reason and its content” (PL 1998 rev., IV, §1, 111). “These values provide public reasons for all citizens” (PL, 474).

264

Samuel Freeman

religion, and again the value of the family in securing the orderly reproduction of society.11 In PL chapters 5–6, Rawls includes under the values of public reason shared guidelines for public inquiry and rules for assessing evidence, which keep inquiry free and public, informed and reasonable; also the virtues of reasonableness and fairmindedness (PL, 139). Finally, Rawls even mentions political values relating to “the claims of animals and the rest of nature” in order to “further the good of ourselves and future generations.” This would include conserving the environment and preventing destructive climate change, as well as other measures to preserve “the natural order and its life-sustaining properties (PL, 245). (5) The Fundamental Interests of Free and Equal Citizens. Reciprocity requires that the political values be interpreted in ways that citizens reasonably believe other citizens can reasonably accept in light of their fundamental interests as free and equal moral persons. A formal feature of any reasonable political conception then is that it has an account of the fundamental interests of free and equal persons in their capacity as citizens. These fundamental (or “higher-order political interests”) guide the political conception in its justification and application of its principles of justice and interpretation of the political values of public reason. Rawls says in the “Reply to Habermas” that “The essential idea is that deliberative democracy, and political liberalism also, limit relevant human interests to fundamental interests of certain kinds, or to primary goods, and require that reasons be consistent with citizens’ recognition as equals” (PL, 430, emphasis added). The fundamental or higher-order political interests Rawls says characterize reasonable political conceptions are the moral powers and other powers of reason that enable citizens to engage in practical reasoning, pursue their aims, and also take part in social cooperation. Reasonable political conceptions recognize that it is in virtue of the moral powers and other powers of reason that persons are free and equal.12 So the development and effective exercise of the moral powers should be primary among the fundamental interests for any reasonable political conception, as Rawls continues to maintain in the 1998 revisions. Rawls also suggests (in PL, ch. V) that the list of primary social goods he sets forth in Political Liberalism are also recognized by all reasonable political conceptions. There is not then any significant departure from justice as fairness in

11

12

PL, 474. The values of effectiveness and efficiency of the political institutions of the basic structure also appear to be liberal political values (PL, 454). Other places where political values are mentioned in addition to those listed here are PL, 30, 147, 171, 194f., 204f., 347, 400, 402. PL 19, PL 1998 rev., 13

The Well-Ordered Liberal Society

265

Rawls’s account of reasonable and rational citizens that make up a wellordered liberal society and their fundamental interests. (6) The Original Position. Rawls says in the 1998 revisions to Political Liberalism that the “family of reasonable liberal political conceptions” is selected in the original position by rational representatives of citizens.13 Agreement by rational representatives identifies the class of liberal political conceptions that are reasonable insofar as they satisfy the criterion of reciprocity and its extension in the liberal principle of political legitimacy. Rawls’s reasons for employing the original position here are open to conjecture. One reason might be to emphasize that the class of reasonable political conceptions satisfying the criterion of reciprocity all meet the strong impartiality and equality constraints of the original position. This distinguishes them from the class of mutual advantage and other views, which do not satisfy these constraints of the OP. Second, as argued below, Rawls in the end envisioned that reasonable political conceptions do not depart too far from the liberal egalitarian principles set forth in justice as fairness. By identifying this “family” by means of the OP, he perhaps intended to narrow the range and emphasize the unity of reasonable liberal political conceptions that satisfy the substantive constraints required by the criterion of reciprocity. Finally, Rawls’s use of the original position to select the family of reasonable political conceptions might seem to conflict with a passage in “Public Reason Revisited” – which Rawls even retains in chapter VII of the 1998 revisions of Political Liberalism – that some reasonable political conceptions might dispense with the original position entirely.14 The context, however, suggests that Rawls is addressing here a different question and use of the original position than its use in identifying the 13

14

“The thought is that once, if ever, reflective equilibrium is to be attained, the family of reasonable liberal political conception of justice may be seen as the outcome of the procedure modeled by the original position in which rational agents, as representatives of citizens and subject to reasonable conceptions of symmetry and the veil of ignorance, select the family of reasonable political conceptions of justice to regulate the basic structure of society” (PL 1998 rev., 75). The continuing centrality of the original position to identifying reasonable political conceptions is also evident in Rawls’s definition of correct judgments “as reasonable: that is as supported by the preponderance of reasons specified by the political principles of right and justice issuing from a procedure that correctly formulates a family of reasonable political conceptions” (PL 1998 rev., 93, emphases added). “A citizen engages in public reason, then, when he or she deliberates within a framework of what he or she sincerely regards as the most reasonable political conception of justice …. I have proposed that one way to identify those political principles and guidelines is to show that they would be agreed to in what in Political Liberalism is called the original position. Others will think that different ways to identify these principles are more reasonable. Thus, the content of public reason is given by a family of political conceptions of justice, and not by a single one” (PL, 450).

266

Samuel Freeman

family of reasonable political conceptions. He says, in effect (as is clear from the first sentence of the paragraph just referred to), that justice as fairness relies upon the original position to determine which reasonable conception is the most reasonable, unlike other reasonable conceptions that use other means to contend why they are the most reasonable conception. Reasonable persons will then disagree about whether the OP has a role in establishing the most reasonable conception, whether in public or nonpublic reason. But Rawls still claims they can agree that the family of reasonable conceptions satisfying the criterion of reciprocity and political legitimacy nonetheless can be identified within public reason by the original position.15 I have listed six formal features of any reasonable political conception that governs a well-ordered liberal society. There are others: reasonable political conceptions are freestanding from reasonable comprehensive doctrines; they address the institutions that constitute the basic structure of society; there is an overlapping consensus among reasonable comprehensive doctrines on political values and the family of reasonable political conceptions, etc. But these six features are especially important because of the way they combine to determine the substantive requirements of reasonable political conceptions.

15.3

Substantive Features of Reasonable Political Conceptions in a WOLS

Rawls says all reasonable conceptions must include three substantive principles requiring (a) equality of certain basic rights, liberties, and opportunities; (b) the priority of these rights, liberties, and opportunities, especially with respect to the general good and perfectionist values; and (c) a social minimum described as “adequate all-purpose means to make intelligent and effective use of their liberties and opportunities” (PL, xlvi; see also PL, 450). It is noteworthy that these three substantive features resemble and generalize the specific requirements of Rawls’s own principles of justice: the first principle of equal basic liberties, its priority over fair equality of opportunity, which in turn has priority over the difference principle. Rawls says that a conception that guarantees only formal basic liberties and opportunities and their priority does not satisfy the criterion of reciprocity. It is at most an “impoverished form of liberalism,” he says, 15

I discuss the role of the original position in the revised manuscript of PL in greater detail in a forthcoming essay, “Ideal Theory, Perfect Justice, and the Well-Ordered Society,” §5, to appear in The Journal of Social Philosophy.

The Well-Ordered Liberal Society

267

since it lacks a criterion of reciprocity and allows excessive social and economic inequalities as judged by that criterion. Positions that guarantee only formal equal liberties and equal opportunities are not in fact liberal at all, he says, but are forms of libertarianism (PL, lvi; see also PL 1998 rev. VII, §6, 240). Rawls clearly thinks “impoverished forms of liberalism” are unreasonable. One might understand this to rule out only orthodox libertarianism, such as Nozick’s view. But Nozick did not even guarantee formal equal opportunity, so impoverished forms of liberalism must include a broader class than orthodox libertarianism. Rawls’s gloss on the three substantive features of reasonable political conceptions – equal basic liberties and opportunities, their priority, and a social minimum – is revealing. Rawls says these features must be interpreted to “satisfy the criterion of reciprocity” and require a basic structure that “prevents social and economic inequalities from being excessive as specified by that criterion [of reciprocity]” (PL, LVII, emphasis added; also, PL 1998 rev. VII, §6, 1st par., 239). Certain institutions must be in place to prevent excessive inequalities: for without these arrangements Rawls says “reasonable political liberalisms hold that these excessive inequalities tend to develop” (PL, LVII, emphasis added; PL 1998 rev. VII, §6, 1st par., 239). What are these “essential prerequisites for a basic structure” that a reasonable liberal political conception must endorse to meet the criterion of reciprocity and hence be reasonable and stable for the right reasons? He lists five (PL, LVIII–LIX): (a) First, the political system must be “sufficiently independent of social and economic interests” and there must be “public financing of elections” and other measures to insure independence (PL, LVIII, emphases added). Property and other irrelevant qualifications to exercise the right to vote are prohibited since they do not satisfy the criterion of reciprocity. This assumes not simply a universal franchise and that the political system is democratic – which can be satisfied by a number of views – but more importantly “the political liberties [themselves] and fair access to the political process” also must be included among the equal basic liberties, hence on a par with the rest (PL, LVII, n. 36 emphasis added). (b) The second institution Rawls lists to satisfy reciprocity and prevent excessive inequalities are those guaranteeing “a certain fair equality of opportunity, especially in education and training” (PL, LVII, emphasis added). Formal equality of opportunity, like merely formal protection of equal basic liberties, is not sufficient. By a “certain fair equality of opportunity” Rawls relaxes the stringent demands of fair

268

Samuel Freeman

equality of opportunity required by “democratic equality,” in A Theory of Justice, §§ 12–13, which requires stricter substantive equality of educational and employment opportunities, and also fair access to “the benefits of culture.” Perhaps fair equal opportunity as depicted by the position Rawls calls “liberal equality” is sufficient, with educational and employment opportunities at reduced levels (TJ 1999, 63). (c) “A decent distribution of income and wealth,” that assures all citizens of “the all-purpose means necessary for them to take intelligent and effective advantage of their basic freedoms.”(PL, LVII, emphasis added) Rawls says this goes well beyond adequate public provisions for food, clothing, and housing and simply meeting basic needs. It must enable the informed and effective exercise of all “the basic liberties and opportunities,” including the political liberties and having fair access to the political process (PL, LVII, n. 36). (d) “Society as employer of last resort.” This enables citizens to have “a sense of long-term security and the opportunity for meaningful work and occupation,” which Rawls says is a condition of “citizens’ selfrespect.” This means that the social bases of self-respect are among the fundamental interests or primary social goods for citizens generally in all reasonable political conceptions. For without “a sense of longterm security and the opportunity for meaningful work and occupation,” citizens feel that they are not members of society but instead “are simply caught in it. This leads to bitterness, self-hatred, and resentment” (PL, LVII). (e) Finally, “basic health care assured [for] all citizens” is required of reasonable political conceptions. Like adequate income and wealth, it too is needed for the effective exercise of basic liberties and fair opportunities. (PL, LVII). What is revealing about these substantive features is that they indicate how Rawls understands the criterion of reciprocity as applied to free and equal moral persons with fundamental interests in exercising their moral powers and fair access to the primary social goods, including self-respect, all of which are emphasized in the 1998 revisions. We would need to take the substantive features of a WOLS one by one to get a clearer idea of why each is required by the criterion of reciprocity and other formal features. But roughly, for reasons similar to those Rawls gives in Political Liberalism, Lecture VIII, he regards equal political liberties with fair access to the political process that neutralizes the effects of wealth, along with the institutions of a deliberative democracy, as necessary for the “free, full and informed exercise” of the capacity for a sense of justice

The Well-Ordered Liberal Society

269

and also to maintain social equality and the sense of self-respect of citizens (PL, 342; see also PL, 333, 368, 417). Some degree of fair equality of opportunity, especially in adequate public funding of education, is among the conditions for enabling citizens to engage in meaningful work and also in understanding and exercising public reason and meaningful participation in democratic deliberation and voting (PL, LVII). Finally, a decent distribution of income and wealth enables all to take “intelligent and effective advantage” of basic freedoms as well as fair opportunities, and are also justified by the political values of economic reciprocity and self-respect, and so on. The general point is that the social institutions Rawls says are essential to any liberal political conception can be understood as grounded in the several formal features of any reasonable liberal political conception listed in section 15.2 above. 15.4

Impoverished Forms of Liberalism Are Not Reasonable Political Conceptions

Now to turn to the question: What does Rawls mean to rule out in saying that “impoverished forms of liberalism” and “forms of libertarianism” are not reasonable political conceptions? Given the stringent substantive constraints Rawls imposes on inequalities and his understanding of the criterion of reciprocity, I contend these should be broadly construed to include not just Nozick’s orthodox libertarianism but also classical and libertarian-liberal views. It is difficult to see how the social and economic views of such classical liberals as Hayek, Friedman, James Buchanan, David Gauthier, Richard Epstein, and more recent liberal-libertarians such as Jerry Gaus, David Schmidtz, John Tomasi, Jason Brennan, and similar theorists could meet these stringent conditions for reasonable political conceptions within a well-ordered liberal society. Even if such neo-liberal views (as I will call them) all provide some degree of fair access to opportunity by publicly funding education and a social “safety net” that meets basic needs, and even some degree of basic health care, still there is too much else that is incompatible with classical and libertarian liberalism. The Basic Liberties. First, Rawls clearly regards the basic rights and liberties that constitute any reasonable conception as fixed and not admitting of variation between conceptions. Basic rights and liberties … can be specified in but one way, modulo relatively small variations. Liberty of conscience and freedom of association, and the political rights of freedom of speech, voting and running for office are characterized in more or less the same manner in all free regimes. (PL, 228 and PL 1998 rev. VI, §4, p.194, emphases added)

270

Samuel Freeman

His claim that “the criterion of reciprocity is normally violated whenever basic liberties are denied” (PL, 447) also implies that reasonable political conceptions must agree on the same basic liberties essential to the full exercise of the moral powers and their priority. For disagreement on a basic liberty – such as the equal political liberties that underpin social equality and the sense of self-respect – would mean that those who reject it could not reasonably expect other reasonable persons to accept their denial of the fundamental status of that liberty. And indeed, with respect to political equality, most neo-liberal conceptions, even if they endorse democracy, do not regard equal political rights and liberties as basic liberties on a par with basic personal liberties such as freedom of conscience, expression, and association. Moreover, even those neo-liberals willing to formally incorporate political liberties as among the equal basic liberties still reject “fair access to the political process” with public financing of elections and the independence of the political process from the influence of wealth and social and economic interests. Moreover, Rawls says that the equal basic liberties, including the political liberties, fair equal opportunity, and an effective social minimum are necessary for the ideal of public reason and “public political deliberation,” which in turn is “essential for a reasonable constitutional regime.” The centrality of public political deliberation on shared political values of public reason within a constitutional democracy is not a governing idea in classical and liberal-libertarian thought. More often democratic deliberation on the common good is something to be avoided because of neo-liberals’ lack of confidence in democratic political processes. Neo-liberals’ rejection of equal basic political liberties is related to a second fundamental departure from the conditions for reasonable political conceptions. They regard laissez-faire capitalist economic liberties as equally important and on a par with the personal liberties of conscience, thought, association, and so forth. Full capitalist liberties are in effect if not by right among the basic liberties. This equivalence of capitalist with personal liberties and their priority, when combined with neo-liberals’ downgrading equal political liberties, implies that there can be little if any democratic regulation of economic powers and transactions nor restrictions on the accumulation of economic wealth, because of the priority of basic capitalist liberties over democracy. It is difficult to see how there can be either reasonable agreement on basic capitalist liberties among free and equal citizens who seek to cooperate on grounds of reciprocity and mutual respect in a well-ordered liberal society or rational agreement in the original position that such liberties should have a place in the family of reasonable political conceptions.

The Well-Ordered Liberal Society

271

Fair Equality of Opportunity. Contemporary classical liberals sometimes provide “a certain fair equality of opportunity, in education and training,” as represented by “liberal equality” in Theory. But this is not normally for the sake of the political values of social equality and economic reciprocity (PL, 224) or to enable citizens to “take part in the debates of public reason or contribute to social and economic policies.” More often, among neo-liberals, equal opportunity is weakly construed as “open positions” and “careers open to talents” (TJ 1999, 73) and, like political equality, is touted for instrumental reasons: Public funding of basic educational opportunities enables society’s members to achieve the degree of literacy and other basic competencies needed for productive work and subsequently to train them in more refined employment skills that are a prerequisite to achieving greater economic efficiency. Distributive Justice. No classical liberal seeks “a decent distribution of income and wealth” that meets the three conditions Rawls specifies: (1) to guarantee “all-purpose means” that go beyond meeting basic needs and that enable all citizens to take “intelligent and effective advantage of their basic freedoms”; (2) to prevent “those with wealth and income” from controlling the political process and dominating those with less income and wealth; and (3) “to prevent social and economic inequalities from being excessive.” The neo-liberal “safety net” is not designed to prevent social and economic inequalities from becoming excessive as specified by the criterion of reciprocity. Indeed, classical and other neo-liberals do not have a conception of excessive economic inequalities. Rather, their conception of economic justice is that any distribution is just so long as it is the outcome of legitimate market exchanges and other consensual transfers of strong private property rights. The classical liberal account of distributive justice is, we’ve seen, often grounded in the presumption that capitalist economic liberties with full ownership rights in means of production are on a par with basic personal liberties, which take priority over democratic liberties. The consequences of basic capitalist liberties, including unlimited rights to wealth accumulation, combined with the extraordinary social and economic powers that attend them, fail any reasonable test by the criterion of reciprocity and economic fairness among free and equal citizens. Reasonable citizens cannot reasonably expect other reasonable citizens to accept the consequences of an economic system in which enormous economic gains to the wealthiest need not improve the position of the least advantaged to any degree at all, and which may even make them worse off. To rely on the invisible hand and trickle down as the fundamental criterion of economic justice does not fulfill the political values of economic reciprocity nor citizens’ fundamental interests in maintaining their self-respect as social equals, hence does not meet the criterion of reciprocity.

272

Samuel Freeman

Full employment with Opportunities for Meaningful Work and Occupation. Neo-liberals also reject government as the employer of last resort that provides “opportunity for meaningful work and occupation” to maintain citizens’ sense of self-respect. It’s notable that this condition incorporates into reasonable liberal conceptions the same reasons and institutions Rawls relies upon in Justice as Fairness that distinguish property owning democracy from capitalism, including even welfare state capitalism (JF, 140). This is not to say that welfare state capitalism cannot satisfy the conditions of reasonable political conceptions. But it must mitigate excessive inequalities and the influence of wealth on democracy to protect the fair value of political liberties, and go well beyond merely meeting basic needs, and incorporate substantially more than the traditional social safety net if it is to enable all members of society to “make intelligent and effective use” of their freedoms and opportunities as Rawls understands them. Universal Health Care. Some but not all neo-liberals accept publicly funded basic health care as a public good in order to maintain public health or as public charity that is to be adjusted and balanced off with other competing goals. But none would regard basic health care as a moral/political right or requirement of justice that is needed for the full and informed exercise of citizens’ basic liberties or to enable them to take advantage of fair and diverse educational, employment, and cultural opportunities, as required by reasonable political conceptions. The formal and substantive conditions Rawls imposes on reasonable liberal political conceptions single out a rather restricted class of political conceptions, all of which endorse (1) Rawls’s first principle of justice and the same basic liberties essential to full exercise of the moral powers that Rawls specifies and their priority, together with conscientious efforts to restrict the influence of wealth on democratic politics; (2) some conception of fair and not merely formal equality of opportunity in education and employment; (3) a robust social minimum necessary for citizens to take intelligent and effective advantage of their basic liberties and opportunities;16 and (4) substantive limits on economic inequalities. It is difficult to see how classical or libertarian liberal property rights with laissez-faire capitalist liberties could be among the basic rights and liberties, even when more constrained non-basic capitalist liberties could be part of many reasonable political conceptions. Any reasonable liberal political conception also puts limits on the unchecked economic inequalities tolerated by capitalism because of their adverse effects on political equality, fair equal opportunity, and economic reciprocity.

16

PL, LVII and LVII, n. 36

The Well-Ordered Liberal Society

273

As a matter of conjecture, I think a main reason Rawls, in later editions of Political Liberalism, turns to the idea of a family of reasonable political conceptions as constituting a well-ordered liberal society is that he thinks that, because of the burdens of judgment, reasonable and rational free and equal democratic citizens will not, even under ideal social conditions, all agree and accept the difference principle as the most reasonable conception of economic justice; also they may differ on the substantive requirements of fair equality of opportunity and the measures required to guarantee the fair value of political liberties.17 Rawls’s appreciation for the growing literature on distributive justice that evolved in response to A Theory of Justice and the fact that many liberal egalitarian philosophers Rawls respected rejected the difference principle, perhaps played a role here. Rawls perhaps accepted after the first edition of Political Liberalism that reasonable persons, as he defines them, may reasonably endorse mixed conceptions that incorporate restricted utility instead of the difference principle, as well as priority and sufficiency views with substantial social minimums; others may reasonably accept socialist or luck egalitarian accounts that equalize inequalities grounded in brute luck, and reward people according to their efforts, or according to option luck; and there may be some desert-based views – such as “to each according to effort” – that reasonable citizens endorse that satisfy the reciprocity conditions and institutions of a well-ordered liberal society. Still, it’s notable that all reasonable political conceptions impose limits on economic inequalities needed to maintain political equality and the fair value of political liberties, fair equality of opportunity, social equality and economic reciprocity, and the self-respect of free and equal democratic citizens. Rawls explicitly mentions these political values as constraints on economic inequalities required by the criterion of reciprocity of any reasonable political conception (PL, LVI–LVII, 139, 224). So even if there are certain economically non-egalitarian liberal conceptions that qualify as reasonable according to Rawls’s criteria, classical and libertarian liberal conceptions are not among these. They reject the conditions of social and political equality and mutual and self-respect that are part of any reasonable liberal political conception.18 17

18

See PL, 228–30. Rawls says “Political discussion of the reasons for and against fair opportunity and the difference principle… fall under basic justice and are to be decided by the political values of public reason” (PL, 230, n.10 and PL 1998 rev., 194) In the 1998 revisions to Lecture VII of Political Liberalism on “The Basic Structure as Subject,” Rawls reemphasizes these points: §6 restates the five substantive conditions in the 1996 Introduction and §7 emphasizes that restrictions on economic inequalities are essential to the fair value of political liberties and fair equality of opportunity in any reasonable liberal political conception.

274

Samuel Freeman

15.5

Concluding Considerations

Here one might reply that what is unreasonable in a well-ordered liberal society – such as neo-liberal conceptions with no limits on inequalities – may be reasonable in nonideal circumstances and vice versa. Rawls thought preferential treatment to compensate for a history of racial injustice is reasonable, even though it may unreasonably violate fair equality of opportunity in a well-ordered society. Similarly, neo-liberals would argue that unrestricted economic inequalities in our nonideal capitalist society may be reasonable enough, assuming a social safety net, publicly funded schools, public highways, public health measures, and other essential public goods that neo-liberals recognize. Hayekians argue that restrictions on inequalities prevent economic efficiencies and maximizing overall wealth, and impedes economic growth, which inevitably disadvantages the less advantaged members of society. Vast inequalities allegedly are mutually advantageous since they ultimately benefit the less advantaged, as evidenced by recent capitalist growth in China and other parts of Asia. This is not an unreasonable argument to make within developing countries, and if so, then it might appear that neo-liberal conceptions that work from these premises should themselves be regarded as reasonable political conceptions. The problem is that what might seem reasonable measures to many of us to alleviate poverty under circumstances of vast inequalities does not render the political conception that justifies those inequalities as itself reasonable. Even if the vast inequalities neo-liberal views encourage eventually benefit the least advantaged in some circumstances, these mutual advantages are not the same as reciprocity, as Rawls emphasizes (PL, 17). The ultimate test for what satisfies this criterion and counts as a reasonable political conception is the family of liberal conceptions that can be agreed to behind the veil of ignorance by representatives of free and equal moral persons who aim to cooperate on grounds of reciprocity and mutual respect as citizens. Even if some classical and libertarian liberals accept some measures of the liberal welfare state, this does little to alter the unreasonable status of these views in endorsing everincreasing extreme economic disparities resulting in social and political inequalities. Moreover, the political values of economic reciprocity require not simply a robust social minimum that is hardly satisfied by the neo-liberal social safety net but also the fair distribution of an increasing social product. Wealth maximization taken alone, with the theoretical possibility of trickle down to the least advantaged but no guarantee of a fair share for each, is not a political value of public reason nor does it satisfy the

The Well-Ordered Liberal Society

275

criterion of reciprocity. A political conception is unreasonable if it permits unconstrained inequalities where the more advantaged increasingly accumulate wealth, while the least advantaged gain nothing additional and may even be made worse off – in spite of their ongoing efforts and increasing contributions to the social product. These have been the circumstances of capitalist accumulation and distribution over the past generation, as Thomas Piketty and others argue (see Piketty 2014), due largely to increasing tax cuts for the wealthiest and economic deregulation during the Reagan, Bush Jr., and Trump presidencies. Even had the basic needs of the homeless and other least advantaged citizens been adequately met during that period (they were not), this would not alter the unreasonableness of the classical liberal and libertarian economic policies of the past forty years. Constantly increasing economic inequality has also resulted in increasing social and political inequality and the denial of fair reciprocity at all levels of social and political life. A liberal political conception whose standard of distributive justice requires no more than a safety net that barely meets basic needs and that undermines fair access to opportunities and the political process in a wealthy capitalist society is an unreasonable liberal conception. Finally, it is objected that questions of the nature and extent of property rights and the fair distribution of income and wealth are just the sort of questions that reasonable people will always disagree about in a democratic society; so, it is self-defeating for egalitarian and high liberals to contend that classical and libertarian liberals have unreasonable views that are without liberal legitimacy. Rawls himself says: “we seek a sharable public basis of justification for all citizens in society” (PL 1998 rev., 213). So, what would be the point or benefit in denying political legitimacy to classical and libertarian liberal views, when reasonable people always disagree in these matters? The debate here is about what it means to say a person is “reasonable” within political liberalism. Rawls defines reasonable persons not simply as having a sense of justice and endorsing the political values of freedom and equality but also in terms of citizens accepting the principle of reciprocity and seeking to publicly justify their positions by political values of justice and public reason. I’ve argued that classical and libertarian liberals do not endorse an understanding of economic reciprocity and related political values that are reasonable in a sense acceptable to reasonable moral persons in Rawls’s sense. So, their views about economic justice are unreasonable as defined within political liberalism, even though they may be reasonable persons in other respects and as that term is normally understood. No doubt, neo-liberal views are reasonable and legitimate in terms of their own accounts of reasonableness. Some of

276

Samuel Freeman

them rely on neo-Hobbesian accounts of public justification and legitimacy. But these accounts openly reject Rawls’s account of reasonableness and the grounding of public justification in shared political values and public reasons.19 Neo-liberal views are not unreasonable to the degree they accept the basic personal liberties and their priority and some conception of fair equal opportunity. What makes them unreasonable is downgrading the equal political liberties, the priority given to basic capitalist rights and liberties, the unrestricted inequalities that inevitably result, with deleterious effects on fair equality of opportunities and social equality itself. Given these differences, a modus vivendi between high liberals and neo-liberals is the most that can be expected. Such a compromise, however mutually advantageous it may be under the circumstances, is not a reasonable agreement between liberal political conceptions that satisfies the criterion of reciprocity.20

19 20

See, for example, Jerry Gaus’s neo-Hobbesian account of public justification in Gaus (2011). I am grateful for their helpful comments to participants at the 2021 Notre Dame Conference in honor of John Rawls, to participants at a panel on Rawls at the 2021 American Political Science Association meetings, and to Joshua Cohen, T. M. Scanlon, Chris Melenovsky, and Paul Weithman.

16

Religious Pluralism and Social Unions* Paul Weithman

Public life in contemporary liberal democracies is marked by division and hostility. Some anti-liberal thinkers have claimed that the interminability of our public debates, and the rancor which characterize them, are natural manifestations of deeper facts about these societies and the civic relationships they allow – facts that these thinkers describe in stark and provocative terms. The modern liberal state is, they say, an arrangement that, like a public utility, is valuable only for the goods and services it delivers. Moreover, it is said, contemporary liberal democracies are too large and diverse for their political arrangements to be anything but instrumentally valuable and for relations among citizens, just as such, to be anything but transactional. The view that political life is only instrumentally valuable has deep roots in Western Christianity (Augustine 2004, XIX.17). It is endorsed, often tacitly and sometimes explicitly, by many religious citizens in our own time.1 But if the instrumentalist view is wrong, and members of large pluralistic democracies could find some noninstrumental value in relating to one another just as such, the rancor of our public life might be greatly ameliorated. It may seem quixotic to argue for that possibility because its realization seems so remote and because of the intellectual obstacles the argument would have to overcome. Yet a number of liberal theorists – T. M. Scanlon, Samuel Scheffler, Seana Shiffrin – have claimed that the relationship among fellow citizens can be inherently * Ancestors of this essay were delivered at an APA session on “Religious Philosophers on Neutralist Liberalism: 25 Years of Rawls’s Political Liberalism,” at a Harvard conference on “Inequality, Religion and Society: John Rawls and After,” at “Global Issues in Ethics III: Religion and Democracy” sponsored by the Australian Catholic University, and at the Nuffield Political Theory Workshop. I am grateful to session participants and audience members – especially Henrik Dahlquist, Cecile Laborde and Tony Taylor – for their helpful remarks. I am indebted to Ruth Abbey and Lori Watson for written comments on the antepenultimate version and to Peter deMarneffe, Jonathan Quong, and Leif Wenar for oral comments on the penultimate one. 1 Alasdair MacIntyre famously likened modern states to public utility companies in MacInytre (1994, 303).

277

278

Paul Weithman

valuable. They have shown the fruitfulness of that claim and have seen clearly the costs of denying it.2 In section 79 of A Theory of Justice, Rawls argues for a version of the claim that Scheffler, Shiffrin, and Scanlon later defended. I phrased that claim as one about a civic relationship. Rawls’s version concerns the activities in which that relationship consists. He argues that citizens of a well-ordered society who have a sense of justice would value for its own sake the collective activity of maintaining just institutions. As we would expect, he also draws out the implications of that claim for the quality of their relationship. For Rawls’s argument for that claim is part of his larger argument that a well-ordered society would be what he calls – in the title of section 79 – “a social union of social unions.” That it would be has implications for public life that Rawls describes in surprising terms: “the public realization of justice,” he says “is a value of community” (TJ 1999, 463). Despite the work that the idea of a social union of social unions had done for Rawls, that idea largely disappeared from his work after “The Basic Liberties and Their Priority.” And in a footnote to “Reply to Habermas” that has escaped attention, Rawls said “the conception of a social union of social unions … is no longer viable as a political ideal once we recognize the fact of reasonable pluralism” (PL, 388 note 21). Though Rawls came to think the idea of a social union of social unions was not viable, his treatment of it remains of interest. Some of its interest lies in a psychological assumption that is crucial to Rawls’s argument that a well-ordered society is a social union of social unions – an assumption I call “the security assumption.” Rawls never said why he came to think the idea of a social union of social unions was not viable in light of reasonable pluralism. I shall contend that reasonable religious pluralism casts doubt on the security assumption and on the argument that appeals to it, and that Rawls weakened the assumption as part of his turn to political liberalism. Seeing why the dubitability of the security assumption makes the idea of a social union of social unions nonviable, we can come to a better understanding of the development of Rawls’s thought. For as is well known, Rawls said that his deepened appreciation of reasonable pluralism led him to reject the stability arguments he had offered in A Theory of Justice and to recast justice as fairness as a political liberalism. One of 2

Samuel Scheffler (2018) argues that if membership in a political society is intrinsically valuable, then its citizens have an obligation to obey the law; see also Shiffrin (2021, 54 note 40 and accompanying text). Scanlon discusses the cost of denying the value of civic relations at Scanlon (2003b, 193–95).

Religious Pluralism and Social Unions

279

Theory’s stability arguments appealed to the idea of a social union of social unions. Seeing why Rawls thought that conception was not a viable ideal helps us understand Rawls’s political turn. Equally if not more important is the relevance of the security assumption for our own politics. That assumption, in both its original and its weakened forms, is a conditional. It identifies a condition that must be satisfied if members of a liberal democracy are to find their collective activity as citizens inherently valuable. It suggests why some members of liberal democracies as we know them deny the inherent value of relations with their fellow citizens. 16.1

Social Unions

Rawls appealed to the claim that a just society would be a social union of social unions in both A Theory of Justice and “Basic Liberties.” His treatments of a social union of social unions in the two works proceed in three stages:  First, Rawls lays out the idea of a social union and the goods that social unions make available, and he argues that a society well-ordered by his principles of justice would be a social union of social unions. One of the key claims at this stage is the one I highlighted in the introduction: the claim that citizens of the well-ordered society would value their collective activity for its own sake.  At the second stage, Rawls explains the key claim of the first stage: he argues that members of a well-ordered society would value their collective activity for its own sake because the principles that well-order their society are principles of reciprocity. The move from reciprocity to the claim about how citizens would value their activity depends upon the security assumption.  The treatments diverge at the third stage, where the results of the first and second stages are pressed into service to support different conclusions. In Theory, the results of the earlier stages feed into an argument for the stability of a well-ordered society. In “Basic Liberties,” they feed into a novel argument for Rawls’s principles of justice. Though the security assumption is appealed to at the second stage, we need to review the first stage to see what claims it is introduced to support. Because I am primarily concerned with the second stage, I shall say less about the third, though we need to review it to appreciate the changes that are required once the security assumption is questioned. The lines of thought that run through the first two stages in Theory of Justice and “Basic Liberties” are fundamentally the same, but that line is

280

Paul Weithman

sometimes clearer in one work and sometimes in the other. I shall trace that line as if it were straight, but readers who are familiar with both works will see that my exposition moves back and forth between the two. 16.1.1 The First Stage Rawls’s treatment of social unions begins with the observation that every human being is born with a variety of capacities, not all of which can be realized in any one life. For the realization of our talents takes time and training. Even the most gifted of us must choose which ones to develop. Moreover, though such choices may be ones each person makes for herself, the realization of talent is social or collective. We enter into activities with others, not only because it is enjoyable to do things in company or because we need others’ material support, but also because there are some ways of exercising our talents that depend upon others’ exercising theirs. Rawls illustrates this last point with a musical example. Someone with the talent to play a variety of instruments has to decide which one to play. However accomplished she becomes on her instrument of choice, there are some ways of exercising her talents on that instrument that depend upon others’ exercising their talents by playing other instruments to perform pieces that are scored for all. Of course, the musicians’ collective activity has to be well structured if it is to allow them to realize rather than frustrate their talents. When it is, the musicians realize very great goods. They enjoy the exercise of their own talents that the collective performance makes possible. They also enjoy others’ exercise of their talents in accordance with the companion effect of Rawls’s “the Aristotelian principle” (TJ 1999, 463). Moreover, they appreciate the performances of others as alternative developments of a capacity, which, if described in suitably abstract terms, can be seen as one they all share: musical talent. Finally, each values her own and others’ contributions to making the diverse exercises of musical talent possible. Realizing these goods depends upon participants entering into the performance in the right spirit and knowing that all the others do. Rawls says they must share a final end – in this case, the end of a good performance. But not just any final end will do. The activity in which the end is pursued has to be coordinated by rules or conventions that are common knowledge and that insure that “the excellences and enjoyments of each are complementary to the good of all” (TJ 1999, 461). To take part in the activity in the right spirit is to give those rules or conventions their proper role in practical reasoning. Moreover, the collective activity of pursuing the final end must be valued for its own sake.

Religious Pluralism and Social Unions

281

Groups that have shared final ends pursued through collective activities valued for their own sake are social unions. Many kinds of activities can be pursued in groups that have the defining features of a social union. In Theory, Rawls cites religion, the family, and games. Though I shall return to religion, games are important because they illustrate a feature of “entering into the performance with the right spirit” that is not explicit in the musical example. That is that the shared final end, and the rules or conventions for achieving it, must regulate participation in the collective activity (See TJ 1999, 461). The game of baseball, for example, can give rise to social unions in which all enjoy the diversity of talents displayed by the players. But it will do so only if those taking part as players or fans regulate their participation by a desire that the game be well played according to rules and conventions that they observe, and are known to observe, with a sense of fairness. In saying that the desire for a well-played game must “regulate” the activity of the participants, I mean that it governs their practical reasoning categorically, so that the desire outweighs desires to cheat or to treat governing conventions as mere opportunities for gamesmanship. What is a social union of social unions? Social unions exist within a society with a basic structure. That structure is ordered by principles of justice that govern society’s activities and groups. If members of that society had a shared final end, realized in activities valued for their own sake, governed by rules that all treat as regulative and that allow for the complementarity of the social unions, the whole society would satisfy the defining conditions of a social union. Members of that society would then realize the goods of a social union through their participation in it, as well as through their participation in the constituent social unions. And Rawls does indeed think that well-ordered society itself qualifies as a social union. For he argues that “both characteristic features are present: the successful carrying out of just institutions is the shared final end of all the members of society, and these institutional forms” – by which I take him to mean “the activities involved in sustaining these institutional forms” – “are prized as good in themselves” (TJ 1999, 462). Rawls’s arguments for the presence of the second feature are the ones that interest me, since they show how citizens value their collective activity. Rawls offers three arguments for the presence of that feature. I focus on the third, since it is the one that depends upon the security assumption. Recall that Rawls’s Aristotelian principle says that “other things equal, human beings enjoy the exercise of their realized capacities (their innate or trained abilities), and this enjoyment increases the more the capacity is realized, or the greater its complexity” (TJ 1999, 374). The argument

282

Paul Weithman

from section 79 to which I want to attend is based on the “companion effect” of that principle. According to the companion effect: As we witness the exercise of well-trained abilities by others, these displays are enjoyed by us and arouse a desire that we should be able to do the same things ourselves. We want to be like those persons who can exercise the abilities that we find latent in our nature. (TJ 1999, 375–76)

Assuming the companion effect, Rawls claims that the diversity of activities that is allowed when citizens all have a regulative sense of justice would be seen as the alternative exercises of our common moral powers. Rawls seems to think that this variety is a “comprehensive good.” And he says in “Basic Liberties” that “participation in this more comprehensive good can greatly enlarge and sustain each person´s determinate good” so that “individual and group accomplishments are no longer seen as so many separate personal or associational goods” (PL, 320). As we shall see in Section 16.1.2, Rawls says citizens participate in that comprehensive good by upholding just institutions. That they do so is why they find inherent value in that activity, and therefore why the well-ordered society has the second feature of a social union.

16.1.2 The Second Stage The argument with which I ended the previous section needs to be filled in, but I shall leave the details aside. What matters for present purposes is that Rawls’s claims about the availability of the goods of a social union of social unions are conditional upon the basic structure’s being wellordered by his two principles of justice. The question, addressed at the second stage of the social unions argument, is what it is about his principles that makes the goods available. The answer, as Rawls puts it in “Basic Liberties,” is that his principles “contain a notion of reciprocity appropriate to citizens as free and equal persons engaged in social cooperation over a complete life” (PL, 322). He thinks that when everyone has and is known to have a sense of justice informed by such principles, the psychological conditions of experiencing the goods of a social union of social unions are satisfied. I take Rawls to reason as follows: As we saw in the first stage, experiencing the goods of a social union of social unions depends upon the companion effect of the Aristotelian principle. The companion effect is implied by a psychological principle. But that principle is not true in all circumstances. Certain conditions must hold if we are to enjoy the exercise of others’ abilities and the resulting diversity of the public culture. What makes the companion effect operative in the well-ordered society of justice as fairness is that

Religious Pluralism and Social Unions

283

each citizen sees the public culture as hers in this sense.3 Each sees “the richness and diversity of society´s public culture as the result of everyone´ s cooperative efforts for mutual good [and each] appreciate[s] this culture as something to which [she] can contribute and in which [she] can participate.” (PL, 323) The reason each member of a Rawlsian well-ordered society sees the diversity of the public culture as the result of everyone’s efforts, including her own, is that in such a society, everyone has and knows that everyone else has a sense of justice. Everyone cooperates and is known to cooperate to secure the basic liberties and the resources needed to make effective use of them, and it is the exercise of the basic liberties that makes a diverse public culture possible. Moreover, the secure possession of basic liberties and resources is itself an essential condition of each person’s pursuit of her good. So when everyone has and is known to have an effective sense of justice informed by Rawls’s principles, each person knows that everyone else contributes to securing her good. That is why Rawls says that the reciprocity of his principles is what makes the goods of a social union of social unions available (PL, 322–23). Rawls’s argument as just stated depends upon a psychological assumption: that the security of the basic liberties in a social union of social unions makes the operation of the companion effect possible and so makes possible the enjoyment of a diverse public culture. This is the assumption I referred to earlier as the security assumption. Rawls states a version of that assumption early in section 79 of Theory: SA1: “When men are secure in the enjoyment of the exercise of their own powers, they are disposed to appreciate the perfections of others, especially when their several excellences have an agreed place in a form of life the aims of which all accept.” (TJ, 1999, 459)

The security assumption asserts the presence of a disposition. It raises the question of what triggers the disposition, so that we actually “appreciate the perfections of others.” Though Rawls does not say so, I assume that because of each person’s experiences and interests, some forms of diversity will be more salient to her than others and so more likely to elicit enjoyment. Thus some people may find a diversity of athletic pursuits salient, and will enjoy the excellences exercised in a variety of sports, while remaining unmoved by the diversity of musical styles of which they

3

The foregoing interprets “To appreciate something as ours, we must have a certain allegiance to it. What binds a society’s efforts into one social union is the mutual recognition and acceptance of the principles of justice; it is this general affirmation which extends the ties of identification over the whole community and permits the Aristotelian Principle to have its wider effect” (TJ 1999, 500).

284

Paul Weithman

become aware. For others, the diversity of the arts may be salient. For still others, the diversity of cuisine. And so I assume that one condition of the disposition’s being triggered by acquaintance with a perfection of others is that the perfection is realized in a form of activity that is salient. It is not immediately clear what the words “appreciate” and “perfections” in the security assumption mean. By “perfections,” I take Rawls to mean “well-developed abilities.” The question of what the word means arises because abilities can be appreciated under various descriptions. Some of those descriptions abstract away from the particular activities in which the abilities are displayed. A pianist, for example, may appreciate a horn player’s musicality, abstracting away from the fact that it is a perfection realized in horn-playing. But if the companion effect results in enjoyment of society’s “richness and diversity,” then perfections have to be appreciated under descriptions that refer to the distinctive and diverse activities in which they are realized. So the perfections to which the security assumption refers must be activity-specific. Appreciation can be a dispassionate but sophisticated understanding of its object, as the appreciation of artistic accomplishment is sometimes said to be. But if the security assumption refers to a disposition that makes the companion effect possible, and the companion effect results in pleasure, then “appreciate” must mean “enjoy.” We do sometimes enjoy things that we know to be bad for us. But since Rawls appeals to the companion effect to show how the goods of a social union of social unions are possible, I take him to think that when a member of a wellordered society appreciates the activity-specific perfections of others, she regards the perfections she enjoys as good for her. If I am right that the security assumption is about our disposition to appreciate salient diversity, and about what “appreciate” and “perfections” mean, then the assumption can be more clearly restated. It also needs to be restated so as that it explicitly reflects Rawls’s assumption that the relevant forms of diversity result from activities pursued in social unions. The resulting reformulation says: SA2: When someone is secure in her basic liberties, she is disposed to enjoy those activity-specific perfections of others that are realized in social unions whose activities are salient to her and so to regard those perfections as good for her.

I shall assume that the disposition to which the assumption refers is triggered by sufficient acquaintance with the activities in question, while also assuming that the threshold of sufficiency can be left vague, since it no doubt varies among persons and with the salience of the activity. The security assumption, when conjoined with the assumption of how the disposition to which it refers is triggered, implies an enjoyment of so

Religious Pluralism and Social Unions

285

many perfections that it seems implausibly strong. Surely, we may think, some of those for whom music is salient could fail to see some genres of music as good for them, whatever the level of musicality displayed by its performers. The problem this possibility poses for the security assumption cannot be evaded by redescribing the salient activity more finely, by taking it to be a musical genre rather than music itself, since it seems possible that those for whom the genre is salient will fail to find some subgenres to be good for them – and so on, for ever-finer redescriptions of the salient activity.4 Rawls does not say enough for us to know how he would address this objection. Perhaps he thinks that when we are secure in our liberties, we exercise a charitable selective attention, seeing as salient just those activities we are disposed to enjoy, while the others fade into the background leaving no disvalue behind. If so, then, though the security assumption may need some revision, I shall assume that Rawls relied on SA2 or some slight variant of it. What matters here is that it is a psychological principle implying that a generous response to pluralism is natural when one’s liberties are secure. But why does the security of the basic liberties dispose us to respond to pluralism generously, seeing others’ qualities as perfections? I believe Rawls’s answer would appeal to the fact that, as we saw, the security of the basic liberties results from everyone’s acting from her sense of justice and so acting to uphold the liberties. If everyone knows that everyone else is permanently committed to acting from her sense of justice, then each is “secure in her basic liberties” in one important sense of “secure”: each can be confident that she will be able to exercise her liberties into the indefinite future. But that commitment makes each person secure in another sense of the word as well. For everyone’s willingness to uphold the basic liberties amounts to a collective affirmation that everyone’s pursuit of her good is worthy of social support, so long as her good is consistent with the demands of justice. Since selfrespect includes the sense that one’s good is worth pursuing (TJ 1999, 386), this affirmation provides a social basis of self-respect. Those who respect themselves are confident of – hence secure in their sense of – their own worth. Confident of their own worth, they feel no need to elevate themselves in their own eyes by denigrating others. They are therefore disposed to be generous in their judgments because they can afford to be. That, perhaps, is why Rawls says “One who is confident in himself is not grudging in his appreciation of others” (TJ 1999, 387).

4

I am grateful to Henrik Dahlquist for pressing this point.

286

Paul Weithman

Because each knows that the conditions of her security are provided by others, the response to pluralism asserted by SA2 is a manifestation of what Rawls thinks is a deep human tendency to reciprocity (TJ 1999, 433). For in assessing others with generosity, citizens of the well-ordered society return good for good. Moreover, if SA2 is true, then reciprocity is doubly motivated in a well-ordered society. Not only does each know that everyone else is advancing her good by sustaining her liberties, but each also knows that those whose activities are salient to her are contributing to her good by realizing activity-specific perfections that she appreciates. In Section 16.2, I will raise some doubts about the security assumption. First, I want to look briefly at the third stage of Rawls’s treatments of social unions for, as I note above, that is the stage at which the treatments diverge. Seeing the arguments Rawls offers at the third stage shows how much would have to change once the security assumption is called into question. 16.1.3 The Third Stage At the end of Theory, section 79 and again in Theory, section 86, Rawls uses the conclusions reached at the first and second stages to argue that members of the well-ordered society would judge that their sense of justice is good for them. They would do so because having a sense of justice makes it possible for them to enjoy their society’s “richness and diversity” (TJ 1999, 500). That conclusion, in turn, functions as a premise in Rawls’s larger argument that a well-ordered society would be stably just. For the sense of justice is an effective desire to honor principles of right. When members of the well-ordered society judge that that desire is good for them, they find a “match” rather than a conflict “between the right and [their] good” (TJ 1999, 505). Finding the match, citizens affirm the regulative role of their sense of justice in their plans of life. The upshot, Rawls says, is that “a well-ordered society is as stable as one can hope for” (TJ 1999, 350). In “Basic Liberties,” Rawls puts the conclusions reached at the first and second stages to a quite different use. Instead of employing them as premises in further arguments about the value of a sense of justice and the stability of the well-ordered society, he uses them as premises in an argument for his two principles. He says, in effect, that parties in the original position would know the conclusions reached at the first and second stages. That is, they would know what goods can be had through a social union of social unions and they would know that if those they represent had a sense of justice informed by Rawls’s two principles, they

Religious Pluralism and Social Unions

287

would enjoy those goods. They would also know that none of the other principles on offer in the original position would make those goods available because none of the others expresses the right conception of reciprocity. Since it would be rational for parties to want those they represent to enjoy those goods when the veil is lifted, the Rawls of “Basic Liberties” concludes that parties in the original position would choose his two principles. This argument is not found in Theory. There, Rawls famously defended his principles by arguing that they would be preferred to average utilitarianism by parties who used the maximin rule for choice under uncertainty. The argument of “Basic Liberties” does, however, seem to anticipate an argument for the principles Rawls offered in Justice as Fairness: A Restatement. There he defends his principles via two “fundamental comparisons.” He again argues that parties following maximin would prefer his principles to average utilitarianism because his first principle guarantees equal rights and liberties and utilitarianism does not. He then compares his principles to constrained utilitarianism – to the principles that result from replacing the difference principle with the principle of average utility constrained by the requirement of a social minimum. Because the worst outcome of adopting constrained utilitarianism is satisfactory, the conditions for using maximin would not be met and the rule would be out of place.5 Rawls argues instead that his principles would be preferred on several grounds, including reciprocity (JF, 122–24). Because the defense of the principles in “Basic Liberties” also turns on their reciprocity, we might expect Rawls to use it again in Restatement. But he does not. The reciprocity argument for the principles in Restatement makes no mention of the good of a social union of social unions. Restatement does refer to a social union of social unions in the argument for stability found in its closing pages. That argument, like the argument of Theory, section 86, is supposed to support the conclusion that members of a well-ordered society would find it good to be just. And that argument, like the arguments of Theory, section 86, appeals to the claim that a great social good is available to citizens when everyone has and knows that everyone else has an effective sense of justice. The Rawls of Restatement says he “hint[s] at the idea” that the good in question is that of a social union of social unions and he recurs to the musical analogy to argue for its possibility.6 But in Restatement, that good is not 5 6

The conditions of maximin are given at (TJ 1999, 134). See JF, 201. The remark about hinting at the idea of a social union of social unions is found at JF, 201, note 22.

288

Paul Weithman

the “comprehensive good” referred to in “Basic Liberties” that “can greatly enlarge and sustain each person´s determinate good.” It is a strictly political good, the good of establishing and successfully maintaining reasonably just (though of course always imperfect) democratic institutions over a long period of time, perhaps gradually reforming them over generations[.] (JF, 201)

It seems Rawls’s conclusion that “the conception of a social union of social unions” as he had developed it in Theory and “Basic Liberties” “is no longer viable as a political ideal” had far-reaching ramifications. It led Rawls to drop an argument for his principles of justice. It also led him greatly to modify an argument for their stability. The question is why “recogniz[ing] the fact of reasonable pluralism” led Rawls to conclude that the conception was not viable. In Section 16.2, I venture an explanation. Recognizing reasonable religious pluralism, I suggest, raises doubts about the security assumption. Since the goods of a social union of social unions are available only to those of whom the assumption is true, raising doubts about that assumption raises doubts about the viability of the conception. If the explanation is correct, then the security assumption – though introduced without fanfare in section 79 and recovered only by drilling deeply into Rawls’s treatment of social unions – was of fundamental importance to his view. 16.2

Reasonable Religious Pluralism and the Security Assumption

Recall that the final statement of the security assumption says: SA2: When someone is secure in her basic liberties, she is disposed to enjoy those activity-specific perfections of others that are realized in social unions whose activities are salient to her and so to regard those perfections as good for her.

To see how the recognition of reasonable religious pluralism casts doubt on the assumption, consider a citizen of faith – call her Joan – who lives in a religiously diverse society and who is reasonable in Rawls’s sense. Assume that Joan’s society is well-ordered by Rawls’s two principles, so that she is secure in her basic liberties, including religious liberty. Then if SA2 is true of her, Joan must take as salient those activities in which others realize activity-specific perfections she enjoys. But why should this be so? Suppose that Joan cares deeply about religion, is deeply interested in it, and considers it one of the most important – if not the most important – forms of human activity. In that case, it seems possible that she is sensitively attuned to its various

Religious Pluralism and Social Unions

289

manifestations and practices in her society. That is, it seems possible that religious activity is salient for her. But the activity-specific perfections realized in various forms of prayer and worship, and in the development and profession of theological doctrines, include doctrines Joan believes to be false. The implication that she regards these activities as good for her – still more, the implication that she regards their development and profession as bringing to “fruition what [s]he might have been and done” (PL, 322) – raises the question of why we should grant Rawls SA2. It may be thought that SA2 is true of Joan in virtue of her reasonability. For reasonable persons recognize the burdens of judgment (PL, 54–55). Someone who recognizes the burdens of judgment recognizes that agreement on theological questions is not to be expected. She therefore accepts that “a diversity of philosophical and religious belief” is “simply part of men’s natural situation” (TJ 1999, 110). And, it may be said, the reasonable psychological consequence of recognizing the naturalness of religious diversity is to regard salient religious pluralism as on a par with other sorts of pluralism and to develop the tendencies to which the security assumption refers. Perhaps that is what the Rawls of Theory and of “Basic Liberties” thought. To see one reason why he might have ceased to think it, it will help to recur to the musical example. Imagine someone exposed to the piano from an early age, whose natural affinity for and subsequent training on the instrument have made it so much a part of her life that playing the piano is central to her conception of the good. By adolescence or adulthood, she may feel that music calls her irresistibly and that playing the piano is her response to that call. She also recognizes others who feel the musical call but respond to it differently, by making similar commitments to other instruments. While she believes that engagement with the piano makes a richness and complexity of musical experience available that cannot be had from other instruments, she also recognizes that those committed to other instruments will have similar feelings about the French horn or the cello. She recognizes that the life of music would be much less rich without the other instruments, and she believes that others experience the call of music differently. Finally, while she thinks that those to whom music is of no personal significance are missing something of great beauty and value, she also recognizes that precisely because they deny the importance of music, they are able to commit themselves to other pursuits by which they enrich human culture in ways musicians cannot. Similarly, Joan might feel irresistibly called by God to participate in religious life and might find her own tradition the most appropriate way for her to respond to that call. At the same time, she could recognize that

290

Paul Weithman

other people are called to respond in different ways, using the rituals and resources made available by different traditions. She may think that the diversity of religious traditions does not just enrich human culture but also that she learns much about God and the way God calls human beings by seeing the different ways they respond. And she may think her own religious tradition has been greatly enriched and perfected by having to adapt to the existence of, and learn from, other traditions. Finally, she may recognize that there are people, perhaps many people, who are – as Max Weber famously said of himself – “absolutely unmusical in matters religious.”7 These are people who do not feel a religious call at all and may not be able to understand why others are drawn to it. And yet precisely because they are not drawn to religion, Joan recognizes that they can make philosophical contributions from which anyone can learn and benefit. If this is Joan’s attitude, then the security assumption seems to be true of her. Someone with the view of religious diversity I have just described need not be a religious skeptic, who thinks it impossible to know what one should believe, or a religious relativist, who thinks that creeds are true for one person but not for another. She may instead think that God calls different people to lead very different lives and that the resulting religious pluralism is in accord with divine intentions.8 In this way, she develops a theological underpinning for her view of religious diversity. But it seems clear that someone who recognized the burdens of judgment need not react to religious diversity in this way. For recognizing the burdens of judgment, taking religious differences to be irresolvable for political purposes and affirming the basic liberties on principle, are all compatible with wishing that false doctrines were not as widely accepted as they are and with thinking that the prevalence of falsehood is the price we must pay for free institutions. These attitudes, in turn, are incompatible with enjoying religious pluralism in the way the musical example suggests. So even someone who is secure in her liberties and who recognizes the burdens of judgment might not have the tendency the security assumption implies that she will. Reasonability, understood as recognition of the burdens of judgment, is compatible with the falsity of that assumption. Moreover, Joan might not have the disposition because her religious view is what implies the falsity of the religious doctrines held by others. 7 8

Max Weber, letter to Ferdinand Tönnies, Feb. 19, 1909, cited by Kloppenberg (1986, 498). See Pope Francis (2019), which says “the pluralism and the diversity of religions, colour, sex, race and language are willed by God[.]”

Religious Pluralism and Social Unions

291

Recognizing the existence of views with that implication, and recognizing that someone could hold such views while acknowledging the burdens of judgment, forces us to recognize that a well-ordered society could include citizens who adhere to reasonable views but of whom the security assumption is untrue. And if, as I argued in Section 16.1, that assumption must be true of someone if she is to enjoy all the goods of a social union of social unions, then a well-ordered society could include citizens who do not enjoy the good of a social union of salient social unions. This may be why Rawls says that the conception of a social union of social unions is not a viable ideal once we recognize the fact of reasonable pluralism, understood now as including reasonable religious pluralism.9 But I do not believe Rawls came to think that in Theory and “Basic Liberties” he had merely overlooked the possibility of citizens of faith of whom the security assumption is untrue. I believe he came to think he had overlooked it because he assumed that wherever else their conceptions of the good might differ, everyone in a well-ordered society would hold liberalism as a partially comprehensive doctrine. To pinpoint a place at which he made that assumption, let us return to the suggestion that religious pluralism should be thought of as on a par with other forms of diversity to which human nature gives rise. That view of religious pluralism is roughly my own. I find it appealing because I am drawn to the conception of a social union of social unions as a way of thinking about religious pluralism. When I reflect on why I find the conception appealing for this purpose, I find that it is because it offers a way of reconciling religious pluralism with a robust idea of equality. According to that robust idea, people are not just equal in standing or equal in the respect they deserve, though they are that. In addition, many of their lives manifest different but equally worthwhile responses to the human predicament. SA2 may be true of citizens of faith who have this view of equality. For those who care deeply about religion and who have this view of equality may find religious diversity to be the object of interest and enjoyment. And they may think, with the Rawls of Theory, that “it is by 9

This hypothesis may seem to be contradicted by two passages in Political Liberalism. For Rawls says that “the fact of reasonable pluralism is not an unfortunate condition of human life” (PL, 37); he repeats the claim at PL, 144. His implication in both passages seems to be that reasonable persons do not regard the fact of reasonable pluralism as unfortunate. Both of these passages were written before the footnote in “Reply to Habermas” that expresses what I believe to be Rawls’s considered view. It is the considered view that I want to understand. Moreover, Rawls seems to imply that the relevant condition on reasonability is that one not “see reasonable pluralism as a disaster” (PL, XXIV–XXV). One could have the attitude described in the text without regarding religious pluralism as disastrous.

292

Paul Weithman

maintaining … public arrangements [which allow for such diversity] that persons best express their nature and achieve the widest regulative excellences of which each is capable” (TJ 1999, 463). But to understand the value of equality so robustly is to treat almost everyone above some threshold of deliberative capacity as capable of framing and executing plans of life that are of equal validity and not just as judged from a political point of view. This is to treat liberal equality as more than a political value and to treat liberalism as a doctrine that is at least partially comprehensive. Whether or not the conjectures of the last couple of paragraphs are correct, the dubitability of SA2 has an important upshot for the reciprocity of a well-ordered society. SA2 was put forward as a psychological claim about how it is natural to respond to diversity under favorable circumstances. If Rawls came to doubt SA2, it is because he came to recognize that there is no one response to the pluralism of a well-ordered society that is natural. Rather, there is a plurality of responses to it that are reasonable. And so reasonable pluralism is not just a first-order phenomenon brought about by the diversity of reasonable views about God and the good. It is also a higher-order one brought about by the diversity of reasonable responses to the diversity of reasonable views. If a Rawlsian well-ordered society would exhibit this higher-order pluralism, then not all of its members will view the activity-specific perfections of others as complementing their own or as good for them. In that case, the first of the two motivations to reciprocity that I identified at the end of Section 16.1.2 will still be present since each will still know that the others are committed to securing their liberty. But the second motivation may be weakened, perhaps considerably. 16.3

The Political Security Assumption and Our Current Predicament

Looking at the condition under which Rawls thinks citizens of a wellordered society can enjoy a social union of social unions helps us understand why the social unions argument of “Basic Liberties” does not turn up in Rawls’s late Justice as Fairness: A Restatement, despite its being an argument from reciprocity. It also shows why the essays in which Rawls worked out the new account of stability required by political liberalism do not make use of the social unions argument of Theory of Justice, section 86. I shall now argue that the condition under which the Rawls of Restatement thinks members of a well-ordered society can realize the political good that he likens to a social union of social unions suggests an explanation of a phenomenon I mentioned at the outset: the fact that

Religious Pluralism and Social Unions

293

many citizens of contemporary liberal democracies like the United States seem not to see great or inherent value in their civic relationships. Recall that that political good is the good of “establishing and successfully maintaining reasonably just … democratic institutions over a long period of time, perhaps gradually reforming them over generations[.]” Rawls insists that this is a social good, which, like the good of social union, is realized in the ongoing attainment of a shared final end (JF, 201). Let us call this good “the good of the political analogue of a social union of social unions” – or “the good of the political analogue,” for short. Rawls’s argument in Restatement that citizens of a well-ordered society would enjoy the good of the political analogue, like his earlier argument that they would enjoy the good of a social union of social unions, depends upon a psychological principle (JF, 201). If Rawls came to recognize that there could be reasonable members of a well-ordered society of whom the security assumption does not hold, and if he therefore ceased to rely on it, then that assumption cannot express the relevant principle. Unfortunately, Rawls does not say what that principle is. What Rawls does say is that “each citizen of the well-ordered society recognizes the others as also affirming the principles of justice” and that each also recognizes that all citizens give high priority to the end of cooperating politically with one another on terms that the representative of each would endorse in a situation in which they are all represented as free and equal, reasonable and rational. (JF, 202)

As a result, “citizens want to cooperate politically with one another in ways that satisfy the liberal principle of legitimacy” (JF, 202). From this result, it is supposed to follow that the good of the political analogue has a “public character.” It is hard to see how the result would even be relevant to the claim that is supposed to follow from it unless the activity of “cooperating politically in ways that satisfy the liberal principle of legitimacy” were the activity in which the good of the political analogue is realized. So what Rawls says is that each citizen of the wellordered society gives high priority to the activity in which that good is realized because she recognizes that everyone else affirms the principles of justice and gives high priority to engaging in that activity. The psychological principle Rawls has in mind must therefore be a principle of reciprocity. Nothing in Rawls’s texts suggests what that principle might be. If the similarity of the good of the political analogue to the good of a social union of social unions is close, perhaps that principle is a political version of the security assumption. The versions I want to consider are:

294

Paul Weithman

PSA1: When someone is secure in her basic liberties, she wants to cooperate politically with others to establish and maintain legitimate democratic institutions.

or PSA2: When someone is secure in the knowledge that everyone else is committed to justice, she wants to cooperate politically with them to establish and maintain legitimate democratic institutions.

Though Rawls discusses the good of the political analogue as part of ideal theory, the political security assumption is a general psychological claim that is not said to be true only of citizens of an ideally just society. And speaking of what I have called “the political analogue” itself, Rawls says that “this good can be highly significant even when the conditions for realizing it are quite imperfect” (JF, 201). And so I assume Rawls thinks if citizens of nonideal societies like our own are secure in their liberties, or secure in the knowledge that others are committed to justice, they want to cooperate politically with others. PSA1 and PSA2 are equivalent to the claims that if someone does not want to cooperate politically with others, she is not secure in her liberties or in the knowledge that her fellow citizens are committed to justice. Expressed this way, the two versions of the political security assumption are potentially illuminating. For as I observed at the outset, the tenor of our political life strongly suggests that many citizens do not want to cooperate with others with whom they disagree – at least if cooperation entails valuing collective self-government for its own sake or according it what the Rawls of Restatement called “high priority.” Indeed, it suggests they prefer acrimony and hostility to cooperation with their political opponents. If either PSA1 or PSA2 is correct, then we would expect to find that citizens with this preference – citizens who are, in effect, willing to give up on a cooperative political life – experience some form of political insecurity. We have seen how Rawls connects security and self-respect, so if either PSA1 or PSA2 is correct, we would also expect those citizens to experience a diminution in the social bases of their self-respect. The suggestion that the degraded quality of public life is ultimately due to insecurity, and to the diminished self-respect consequent upon it, strikes me as an interesting one. Mustering the evidence needed to confirm this suggestion is a task for social science and one that lies far beyond the scope of this chapter. I conjecture, however, that the evidence is there to be mustered. Rising inequality, stagnating middle class wages, and police violence that discriminates by race all provide reasons for doubting a widespread commitment to justice. Some religious participants in cultural skirmishes

Religious Pluralism and Social Unions

295

about changing sexual mores and family structures do not feel secure in their possession of religious liberty (Phillips 2018). Asking whether their sense of insecurity is reasonable or well-founded also lies well beyond what I can do here, though that question would have to be pressed if we are to figure out how to provide all citizens with the appropriate bases of self-respect and security. I close instead with an observation. Rawls’s treatment of a social union of social unions has received relatively little attention in the voluminous literature that has grown up around his work. Yet that treatment, though offered as part of ideal theory, ultimately suggests an explanation of our current predicament when it is examined carefully. That it does is testimony to the continuing interest and relevance of this neglected part of Rawls’s magisterial theory of justice.

17

One Person, at Least One Vote? Rawls on Political Equality … within Limits

David Estlund*

Political liberty … is not designed to satisfy the individual’s desire for self-mastery, much less his quest for power. (TJ 1999, 205)

The fundamental criterion for judging any procedure is the justice of its likely results. (TJ 1999, 202)

17.1

The Most Obvious Political Inequality

John Rawls writes in A Theory of Justice, “Perhaps the most obvious political inequality is the violation of the precept one person one vote” (TJ 1999, 203). That precept is an article of faith in contemporary democracies. It draws a red line against, among other things, traditional models of rule by the ostensibly wiser or more competent. The precept (as I will call it) has approached the status of dogma, though this quietude is recently being disrupted in the academic literature.1 It is striking, then, that Rawls did not himself hold that precept to be a fundamental requirement of justice. He introduces the precept only to introduce his explanation of how, in principle, violating it might be justified. To that purpose, Rawls sympathetically explicates John Stuart Mill’s famous defense of plural voting in which Mill (1861, chapter 8) proposes to give two or more votes to some, only one vote to the others, thereby to * Thanks for helpful comments on earlier versions to Samuel Freeman, Charles Larmore, Peter de Marneffe, Tim Sommers, Paul Weithman, and Leif Wenar. I benefited by discussion at John Rawls’s A Theory of Justice at Fifty: An Anniversary Conference, September 24–25, 2021; Panel on A Theory of Justice online, American Political Science Association, October 2, 2021; Presentation at the Moral and Political Philosophy Working Group, Yale University, March 3, 2022; Mark L. Shapiro Graduate Philosophy Conference (keynote); and Brown University, April 9, 2022. 1 See, for example, Brennan (2016); Mulligan (2018) and many others discussing the issue under the rubric of “epistocracy,” meaning rule of the wise. For my own presentation of the epistocratic challenge to democracy, and an attempt to answer it (as well as coining the term), see Estlund (2008).

296

One Person, at Least One Vote?

297

privilege the input of citizens more capable of making wise and just decisions. Call this epistemic plural voting. In what might look like a rebuttal, Rawls writes, “the precept of one elector one vote implies, when strictly adhered to, that each vote has approximately the same weight in determining the outcome of elections” (TJ 1999, 196). And plural voting, “perhaps the most obvious political inequality,” would clearly be a “violation” of that precept (TJ 1999, 203). A recent commentator is surely right that, “in a world that has come to accept ‘one person, one vote’ as a fundamental principle of democracy, Mill’s plural voting suggestions will certainly ring both undemocratic and illiberal” (Miller 2003). But in Rawls’s view, Mill’s argument for violating that precept is “of the required form” (TJ 1999 , 204). In fact, Rawls says, “plural voting may be perfectly just” (TJ 1999 205).2 An instrumental approach to political liberty such as epistemic plural voting fits naturally with Mill’s utilitarianism, but in Rawls, the great anti-utilitarian, it is more surprising. As against any temptation to think that Rawls will commend electoral democracy as a form of self-rule, as is often said, that approach is explicitly repudiated in the texts from my epigraph, texts drawn from the part of Theory addressed to plural voting: “Political liberty … is not designed to satisfy the individual’s desire for self-mastery.” Rather, “the fundamental criterion for judging any procedure is the justice of its likely results.” In this chapter, I pull together textual and doctrinal ingredients from Theory to reinforce the point that Rawls’s conception of justice, justice as fairness, is open in principle to plural voting. Commentators have often seemed to either deny this, or to argue that even if it’s so, it is a failure of Rawls to properly work out the implications of the rest of his theory.3 I will argue that this is not an errant or superficial position of Rawls’s but arises naturally out of the structure of his overall theory and method. It is just what we should expect, even though it certainly has a troubling “ring.” It is an implication of his reasoning from the original position that the precept, “one person, one vote,” is only a defeasible presumption, subsidiary to the more fundamental matter of promoting and protecting the other basic liberties.

2

3

More fully, and after a few sentences introducing the ship analogy: Now the ship of state is in some ways analogous to a ship at sea; and to the extent that this is so, the political liberties are indeed subordinate to the other freedoms that, so to say, define the intrinsic good of the passengers. Admitting these assumptions, plural voting may be perfectly just. (TJ 1999TJ, 205) Wall takes the view that Theory is open to plural voting, contrary to the precept, but he does not attempt to lay out the interpretive case, focusing instead on the basis for the fair value guarantee (Wall 2006). Two other authors (Beitz 1990) and (Krishnamurthy 2012, 2013), appear to doubt this interpretation on doctrinal grounds such as the value for Rawls of self-respect, discussed below.

298

David Estlund

Now, that view would not mean that plural voting ought to be implemented now or maybe ever, and two reasons that might weigh against doing so are the likelihood of intense brute resistance even if unwarranted and the risk of abuse of the idea if the Pandora’s box were opened in practice. But neither of those renders plural voting fundamentally unjust. As I will emphasize, they are both concessions in nonideal theory to the danger of certain misbehavior. But what, then, is political equality in Rawls’s theory of electoral democracy? Rawls speaks sometimes of “political equality” and, more often, of “equal political liberties,” but that may tell us less than it seems, for the following reasons.4 As he says in discussing the difference principle, “egalitarianism admits of degrees”(TJ 1999, 471). Even so, while the that principle is not a case of “strict egalitarianism” (TJ 1999, 472), he goes so far as to say that it “is a strongly egalitarian conception in the sense that unless there is a distribution that makes both persons better off … an equal distribution is to be preferred” (TJ 1999, 65, emphasis added).5 A principle with that defeasible presumption in favor of equality is, in his view, “strongly” though not “strictly” egalitarian. He says precisely the same thing about the principle of fair equality of opportunity – notably named a principle of “equality” – namely that, in principle, strict equality of opportunity may be violated but only for the sake of more opportunity for those with the least.6 Let’s call this recurring

4

5

The term “political equality” occurs four times in Theory. More interestingly, it does not occur at all in Political Liberalism, and just once in Justice as Fairness, saying, without further explication of the term, that the theory takes “the moral powers as the basis of political equality” (JF, sec. 7.3) In another passage focused on the difference principle he argues that the two principles of justice “are recognizably egalitarian, even though certain significant disparities are permitted” (TJ 1999, 471). At TJ, 194, he speaks, with “equal” in parentheses, of the “principle of (equal) participation.” Why parentheses? Perhaps to avoid giving the impression that it is a requirement of rather than a presumption in favor of equality. According to the “principle of participation,” the constitutional process should preserve the equal representation of the original position to the degree that this is practicable For the time being I assume that a constitutional democracy can be arranged so as to satisfy the principle of participation. (TJ 1999, 195)

6

However, “For the time being,” gives way in the following section (sec. 37), entitled, “Limitations on the Principle of Participation,” where, among other things, we get his main discussion of plural voting. For more on the analogy between the democracy and the original position see notes 18 and 20. He writes, arguments overriding … fair equality of opportunity in favor of a hierarchical class structure … have the right form when they claim (whether correctly or not) that the opportunities of the least favored sectors of the community would be still more limited if these inequalities were removed. (TJ 1999, 265, emphasis added)

One Person, at Least One Vote?

299

point the Lift All Boats exception to strict equality.7 That fits his template for being what he calls strongly but not strictly egalitarian. So, when Rawls says that justice as fairness is committed to “political equality,” all that tells us is that, like the other two cases of equality principles, it contains at least a principled presumption against political inequality, possibly one that is violable for the sake of basic liberties themselves. As it turns out, it is no more than that, so I will (continue to) argue. What, then, is the form of argument for plural voting that Rawls tells us Mill gets right? The Lift All Boats exception that he incorporates into both fair equality of opportunity and the difference principle is echoed more or less exactly. Rawls writes, unequivocally eschewing any requirement, as a matter of justice, of formal political equality: [In the case of] the justification of unequal political liberty … the priority rule requires us to show that the inequality of right would be accepted by the less favored in return for the greater protection of their other liberties that results from this restriction. (TJ 1999, 203, emphasis added)8

Properly understood, then, the guarantee of equal basic liberties, including equal political liberties and their fair value, does not rule out what Rawls himself calls “the most obvious political inequality” violation of the precept “one person one vote” any more than it rules out inequality of primary goods or of opportunity. Rawls does say “none of [the basic] liberties is absolute; but however they are adjusted to form one system, this system is to be the same for all” (TJ 1999, 54). Since he is explicit, as we have just seen, that there can be a “justification for unequal political liberty,” what might he mean here by “same for all?” This might well be analogous to the way in which the principle of fair equality of opportunity ensures a system of opportunity that is the same for all, in that offices and positions are “open to all on the

7

8

Whether or not there are arguments of that form that are “sound” is “not part of the theory of justice” (TJ 1999, 265) I can leave aside questions about the question of the principle of fair equality of opportunity as applied to political office. Whatever the answer, it’s clear that in Rawls’s view there is to be an elected assembly, and I’m concentrating on the electoral stage. The fair equality of opportunity principle insists that when the appropriate positions exist people have fair opportunity to achieve them. Whether there should be tiers of voters in the manner of epistemic plural voting is not borne upon by that principle. From the colloquial phrase, “A rising tide lifts all boats,” often used to suggest that measures to grow the overall economy, even if they are especially good for the wealthy, can be to the benefit even of the worse off. More generally, for any of the liberties, if liberty is less extensive, the representative citizen must find this a gain for his freedom on balance; and if liberty is unequal, the freedom of those with the lesser liberty must be better secured. (TJ 1999, 214–15, emphasis added)

300

David Estlund

basis of qualities and efforts reasonably related to the relevant duties and tasks” (TJ 1999, 245–46), those differences being justified by their benefits to all, especially the worst off. In that same sense, a system of liberties that allowed for plural voting would be the same for all: no one may have more formal political liberty than others without some qualification by which that would redound to the greater liberties of all, especially those with the least. Differences in authority and responsibility needn’t be unjust or fundamentally inegalitarian in either setting. It’s true that very few rights or liberties can sensibly be held to be absolute even though the heavens may fall, but the provision for plural voting is much more than that. It is not as if formally equal voting power must be guaranteed unless deviating from that precept is the only way to avoid catastrophe. Rather, unequal voting is justifiable so long as that would better protect and promote the other liberties especially for those with less. Just as in the Lift All Boats exception built into the principles of fair equality of opportunity and the difference principle, while there is a presumption in favor of such equality in all three cases, and it can’t be outweighed by considerations of just any kind, it can be overridden far short of the case of disaster-avoidance. As we saw, he calls that kind of presumption in favor of equality “strongly” egalitarian, but it is not nearly as strongly egalitarian as having only an emergency escape hatch view would be. Not only is it not absolutely egalitarian, and even if we allow Rawls’s dubbing it “strongly” egalitarian, it is not even very strongly egalitarian in the manner of Falling Heavens exceptions. So far, this provides a strong prima facie case for the justifiability of plural voting in Rawls. Maybe there’s more to it. In the remaining sections, I try to enrich this interpretation by considering some ways this result might be thought avoidable within Rawls’s theory, either according to or despite his own view of the matter. Are the political liberties special? What about the fair value guarantee? Is this feature only part of nonideal theory? What about threats to self-respect? What about political liberalism? Then, finally, would it be an embarrassment?

17.2

What Is Special about Political Liberties?

Are the political liberties “special,” so that this general form of justification for unequal basic liberty is blocked in their case?9 Well, Rawls says 9

About the fair value guarantee that had been introduced in A Theory of Justice, Rawls says in Political Liberalism that “we must … treat the equal political liberties in a special way” (PL, 327). I’m arguing that he does not mean that they are the most important liberties, and, in a way, on the contrary.

One Person, at Least One Vote?

301

that justice as fairness accepts a strand of the liberal tradition holding that the political liberties, “have less intrinsic10 value” than other basic liberties.11 Formal political equality is somewhat flexible, just as each of the basic liberties is, on what I’m calling the Lift All Boats exception to strict equality. But, in addition, rather than being less flexible than the others, formal political equality is, in an important way, more flexible given that it is primarily instrumental to the protection of the other liberties. Indeed, the value of electoral democracy is explained primarily in instrumental terms in A Theory of Justice and this flexibility of the formal political liberties is just what that should lead us to expect. Rawls writes in Theory: the procedure of majority rule … has a subordinate place as a procedural device. The justification for it rests squarely on the political ends that the constitution is designed to achieve, and therefore on the two principles of justice. I have assumed that some form of majority rule is justified as the best available way of insuring just and effective legislation. (TJ 1999, 313; see also 213, 261)

Now, the first principle adds to the equal formal political liberties a guarantee of what he calls their “fair value,” meaning the rough equality of effective means for exercising the formal political liberties, such as the

10

11

As Peter de Marneffe has pointed out to me, “intrinsic” is a potentially misleading term for Rawls to use here. The basic liberties and all the primary goods are, in their role in the theory, instrumental, mostly to the protection and promotion of important individual interests. Rawls mentions such a traditional liberal view sympathetically but without explicitly embracing it, several times. For example, see TJ, 201–5, echoed in Lecture VIII, “The Basic Liberties and their Priority” (PL, 299). And see along similar lines “Reply to Habermas” (Rawls 1995, 158, note 39). But he also goes further already in TJ to say, “various liberties are not all on a par” (TJ 1999, 271, emphasis added). Unlike the case for liberty of conscience and integrity of the person, “the case for certain political liberties and the rights of fair equality of opportunity is less compelling” (TJ 1999, 271). Again, later, in Justice as Fairness, he writes, Justice as fairness agrees with the strand of the liberal tradition (represented by Constant and Berlin) that regards the equal political liberties (the liberties of the ancients) as having in general less intrinsic value than, say, freedom of thought and liberty of conscience (the liberties of the moderns) …. The political liberties can still be counted as basic even if they are only essential institutional means to protect and preserve other basic liberties. (JF, sec. 43.3, emphases added) Also, granted, the phrase there is “even if they are only essential institutional means,” which does not say that they are. But, as we see, he said a few lines prior that justice as fairness “agrees” that the equal political liberties do have less intrinsic value than (at least) some other basic liberties. That repeats what he said, as noted above, in Theory at 271. I think the most natural reading is that the reason for the political liberties being less compelling and as having less intrinsic value is their being partly “essential institutional means” to the protection of the other basic liberties. So, we should read, “even if,” as “even though.” Thanks to Charles Larmore for highlighting this question.

302

David Estlund

political uses of unequal wealth.12 It might seem, and I believe it is widely thought, that the fair value guarantee is a way of insisting on effective political equality after all – “effective” meaning an equal distribution of each person’s formal political liberty qualified by the means they have to exercise it.13 If correct, this might seem to count against the idea that plural voting is plausibly permitted by the principles of justice in Rawls. After all, it would be bizarre for him to hold that plural voting “may be perfectly just” for epistemic reasons but then to make sure that this didn’t result in any significant effective political inequality in the end. Is the fair value guarantee enough to block plural voting, at least within ideal theory (which I treat in Section 16.3)? Even apart from the contrary textual evidence already presented, I think this argument fails. The importance of fair value in Rawls is not to insist on effective political equality as such but to block political inequality based on unequal wealth (or, more generally, superior social power of any kind).14 Not only is wealth or social power a terrible measure of superior voting competence, as Mill himself argued.15 Even if there were some correlation, there is obvious danger in exacerbating already unequal social power, multiplying it by letting the wealthy also thereby have extra formal or informal political power.16 None of this counts against epistemic plural voting in general. The fair value guarantee rests on a sharp distinction between the formal political liberties and their value or worth in light of the person’s means for exercising them, taking the arrangement of formal liberties – equal, unequal, etc. – as given. The fair value guarantee and its rationale doesn’t directly bear on the analytically distinct question of the extent or 12 13

14

15

16

In later work, Rawls sharpens “fair” in “fair value” to, “roughly equal,” but that is not said or implied in A Theory of Justice. See PL, 337. As I understand it, a fair value guarantee does not insulate the political liberties from instrumental arguments for their being formally unequal across voters, as in epistemic plural voting. I disagree with Wall (2006), who gives an otherwise mostly congenial reading on this point. I say more about both this and an important point of Wall’s below. Of course, since the basic liberties are put at risk if the wealthy are able to influence the political process more effectively, then obviously those liberties would be at risk if the wealthy in particular had more votes than others! So the theory easily and obviously opposes plural votes on grounds of property or wealth, as both Rawls and Mill emphatically agree. But both of them see that this is no general argument against epistemic plural voting but only against the overweening influence of the wealthy on the political system. Mill held it to be “entirely inadmissible, unless as a temporary makeshift, that the superiority of influence should be conferred in consideration of property” (Mill 1861, chapter 8). Contemplating that danger does not take us out of Rawls’s frame of “ideal theory,” since, as he points out in PL, tendencies of the dangerous kind can be present even without any individual failing in their own duties of justice (PL, 267).

One Person, at Least One Vote?

303

distribution of the formal political liberties, so it is no support for strict formal political equality as against epistemic plural voting nor for neutralizing any formal political inequality by adjusting people’s means to achieve effective political equality. Rawls glosses “the worth of liberty” as “the value to individuals of the rights that the first principle defines” (TJ 1999, 179).17 Since, as I’ve argued, the political liberties are special by being importantly instrumental to the other liberties, they ought to be distributed in whatever way best serves that purpose. As an instrument, the system of political liberties is to be tuned. Ensuring that formal political liberties, equal or not, are combined with fair value – roughly equal means for exercising those liberties – retains and reinforces the desired tuning. As far as principles of justice are concerned, that might be an equal or an unequal distribution of effective political liberty. As against my interpretation so far, there is some, but very little, text that might suggest a procedural fairness principle for the political system and that would seem to be at odds with my mainly epistemic interpretation. Rawls later writes: The idea [of] the guarantee of fair value for the political liberties is to incorporate into the basic structure of society an effective political procedure which mirrors in that structure the fair representation of persons achieved by the original position. (PL, 330–31)18

That statement of Rawls’s is hard to reconcile with his theory of electoral democracy.19 The analogy is hopeless between the original position and a political procedure as Rawls understand it.20 Maybe most fundamentally, the original position is a (hypothetical) case of pure procedural justice in which participants maximize their own relevant interests, whereas a political system in a just society is, according to Rawls, a case of “imperfect procedural justice”21 in which participants address not only their own interests but an impartial standard, and a standard that is independent of the procedure, for the correctness of decisions. 17

18

19 20 21

Rawls goes on, a few sentences later: Thus liberty and the worth of liberty are distinguished as follows: liberty is represented by the complete system of the liberties of equal citizenship, while the worth of liberty to persons and groups depends upon their capacity to advance their ends within the framework the system defines. (TJ 1999, 179) See also the similar passage at TJ, 194–95, discussed above, and 256 at “ideally democratic decision.” In the latter case, Rawls seems to interpret the OP as an idealized case of democracy. For reasons I give here, I believe that is untenable. I agree with Beitz, (1990, 96, note 19) on this. I explore this issue at length and as it arises in several authors in (Estlund 2003, 2008). “In framing a just constitution I assume that the two principles of justice already chosen define an independent standard of the desired outcome” (TJ 1999, 174).

304

David Estlund

Now, certainly it is intuitive to think that the political system ought to be fair, even if, as far as I can tell, there isn’t a single instance in A Theory of Justice where Rawls speaks of the political system in particular as a case of procedural fairness. Part of this intuition is met by the fact that if power over the political process is too uneven the epistemic consequence could be very bad. That is, a certain fairness to points of view is needed to prevent the scales being tipped toward outcomes that unjustly favor those with more political power. But that is an epistemic kind of fairness crucially different from each voter having some sort of right to fair – roughly equal – opportunity for influence on some non-epistemic ground, as the parties have in the OP. In Rawls as I understand him, but according to which the passage in question is errant, pure procedural fairness or justice is a principle for the overall basic social structure and not thereby for any substructures.

17.3

Is This Only Nonideal Theory?

It is important to determine whether Rawls’s moderate equanimity about formal political inequality is meant not to reside within his central project, which is an exercise in ideal theory. There is, after all, little that could not be permitted according to Rawls if conditions are sufficiently nonideal. Roughly, any injustice might be permitted, if necessary to prevent even worse injustice. This includes slavery (TJ 1999, 218). So, it wouldn’t upset the kind of political egalitarian reading of his main theory that I am disputing to note that, a fortiori, formal political inequality might be permissible in certain contexts already containing injustice. But in section 39 of Theory, Rawls addresses this question directly, and we see that this is not the structure of his view. He says, specifically about political inequality, that its justifiability does not depend on background injustice of any kind, even if the justifiability of certain other liberty inequalities does. Here is a constructive summary of the crucial passage as I understand it: the “two kinds of circumstances that justify or excuse a restriction of liberty” are, “first,” where there are constraints of “natural limitations and accidents of human life, or from historical and social contingencies. The question of the justice of these constraints does not arise” (TJ 1999, 215). This first case is introduced as compatible with strict compliance – with the absence of injustice. It is only in the “second” case that “injustice already exists. The question here is what is the just way to answer injustice” (TJ 1999, 215). That, as we know, is a defining question for nonideal theory. He writes: How justice requires us to meet injustice [DE: the second case] is a very different problem from how best to cope with the inevitable limitations and contingencies of human life [DE: the first case]. (TJ 1999, 215)

One Person, at Least One Vote?

305

This all arises, fortunately for us, to make a point about formal political inequality, such as when some have more votes than others, [so] political liberty is unequal; and … if the votes of some are weighted much more heavily, or if a segment of society is without the franchise altogether. In many historical situations a lesser political liberty may have been justified. (TJ 1999, 217, emphasis added)

“Historical conditions” is a clear reference to the first case, where there are constraints such as “historical and social contingencies” that are not owed in any way to injustice. “These constraints do not justify the loss of liberty of conscience and the rights defining the integrity of the person.” However, as we saw above (in note 11), “various liberties are not all on a par” (TJ 1999, 217). As compared to liberty of conscience and integrity of the person, “the case for certain political liberties and the rights of fair equality of opportunity is less compelling” (TJ 1999, 217). I conclude that the provision for formal political inequality in Rawls’s theory is not an element of nonideal theory but a consequence of his regarding, in ideal theory, the value of political liberties as being mainly instrumental to the protection of other basic liberties.

17.4

Threats to Self-Respect: Reasonable Not Brute

Rawls does say, immediately after explicating Mill on plural voting, that “of course, the grounds for self-government are not solely instrumental” (TJ 1999, 205). And after the two paragraphs meant to explain this, he summarizes by saying, “equal political liberty is not solely a means” (TJ 1999, 206). An initial small point is this: everything he goes on to say there does indeed treat them as instrumental and as a means in the sense of being good for their consequences – means to people seeing themselves “as associates with whom one can cooperate to advance some interpretation of the public good” (TJ 1999, 206). By “not solely instrumental,” he means not solely instrumental to more wise and just legislation – not solely epistemic in that sense. So, the political liberties also have that civic value, let’s call it, for the tenor of public life. Oddly, though, only a small part of what he says there is about political equality. He speaks of the civic value of “self-government,” “political liberty,” voting, and “taking part in political life,” and almost nothing about the value of each having as much political liberty as others.22 22

A characteristic example is that, as he says, “the public will to consult and to take everyone’s beliefs and interests into account lays the foundations for civic friendship and shapes the ethos of political culture” (TJ 1999, 205).

306

David Estlund

Political equality does make an appearance, however. He says there that, “the effect of self-government where equal political rights have their fair value is to enhance the self-esteem and the sense of political competence of the average citizen” (TJ 1999, 205). (He was using “self-esteem” and “self-respect” interchangeably, as he explained later.)23 Might he mean that this self-esteem consideration overrides the case that he admits can be made for epistemic plural voting? That reading is strained. For one thing, in laying out the civic value of political equality, he explicitly follows Mill himself, whom he has already noted to be the premier historical advocate of plural voting. More plausibly, while these further civic considerations are real and must be weighed in the balance in any case for plural voting, they may or may not block plural voting in particular conditions. After all, as he had said just two pages back: “How far [certain kinds of representation] depart from the precept one person one vote is a measure of their abstract injustice, and indicates the strength of the countervailing reasons that must be forthcoming” (TJ 1999, 203). The same point would plainly apply to the formal weights of votes and not only to the weighting effects of schemes of representation, which is his narrower topic in that sentence. Some may hold that, whatever Rawls might have thought, the threat to self-esteem or self-respect from plural voting is bound to be decisive against it from the standpoint of justice as fairness. Certainly, the “social bases of self-respect”24 are said to be among the most important of the primary goods. But special care is required in this case. The damage to a person’s self-respect from certain social arrangements is not always weighty for purposes of justice, because it is not always a reasonable or warranted reaction. The distinction is familiar in the case of concerns about external, outright opposition to a proposed arrangement, rather than this internal sense of insult. There we know that brute resistance doesn’t bear on the justice of the thing being resisted, because people might very well resist what is right. This is a deeply Rawlsian idea present throughout the corpus. And while a sense of diminution or humiliation is not outright resistance similar reasoning applies. Not just any de facto blow to self-respect can weigh against an institutional arrangement as a matter of justice. It may be that many in the United States today regard as a threat to their own self-respect the growing non-white proportion of the

23 24

Also in “Reply to Habermas” (PL, 404 note 39). Later, in “Reply to Habermas,” he says that he meant to use “self-esteem” and “selfrespect” interchangeably, though he admits that he should have chosen a single term. See PL, 404 note 39.

One Person, at Least One Vote?

307

population (Wilkins and Kaiser 2014). That doesn’t show any injustice in the underlying policies. Rawls counsels in these same few pages that Men’s sense of outrage however irrational will set boundaries upon what is politically attainable; and popular views will affect the strategies of enforcement within these limits. But questions of strategy are not to be confused with those of justice …. The force of opposing attitudes has no bearing on the question of right but only on the feasibility of arrangements of liberty. (TJ 1999, 203)

The question about justice is not whether many people would feel diminished or looked down upon but whether the proposal or arrangement in question is reasonably or appropriately25 understood to deny their status as equals.26 But might a diminished subjective sense of full and equal moral regard be warranted by all versions of epistemic plural voting? Epistemic plural voting is in no way premised on the lesser importance of anyone’s interests or their lesser worth from the standpoint of justice or on anyone having less than the full complement of what Rawls calls the two moral powers. Epistemic plural voting is justly implemented only when (and not always when) necessary to protect the basic liberties even of those with less. As I have said, one might think that plural voting should not be implemented because the idea is too likely to be abused once it is let loose. But that doesn’t indicate that it is an unsound or unjust principle. To note that this would be an abuse, an incorrect application of the idea, is to grant all that is being claimed for it. People abuse sound principles all the time. That is a practical consideration about publicly relying on any principle but not one that casts doubt on the principle’s soundness. In sum, then, the self-respect strategy is evidently not adequate to render epistemic plural voting fundamentally unjust in the Rawlsian system. 17.5

Plural Voting and Political Liberalism

Political liberalism, as Rawls called his later theory of political justification and legitimacy, seems to open up a distinctive line of argument against plural voting and, in fact, against any form of rule by the knowers or the wise (what I call epistocracy). In the book Political Liberalism, 25 26

Or excusably. See TJ 1999, 468. Wall writes: “Obviously, if the equality of the political liberties is compromised, then their fair value will not be guaranteed” (TJ 1999, 258). For reasons I explain, I disagree with this. But he continues, “so, in principle, considerations of self-respect can undercut, as well as bolster, the case for assigning special treatment to the political liberties.” This is an important point that doesn’t depend on the first claim. Just as self-respect may be at stake if some have more votes than others, it is also at stake if one’s basic liberties are unnecessarily put at risk by forgoing a voting system that would better protect them.

308

David Estlund

written twenty years after A Theory of Justice, Rawls comes to hold that (putting it roughly and briefly) political justification for basic matters of justice doesn’t have moral force over any citizen who both (a) can’t accept all the terms of the justification without contradicting important features of their worldview or “comprehensive conception” and (b) those tenets of their worldview are “reasonable” and so represent a genuine moral claim. This is another use of the idea I used in Section 17.4: that is, not all objections from self-respect are reasonable, so not all represent genuine moral claims – and yet some do. We know that if we try to imagine proposing some scheme of plural voting today in the United States it’s obvious that there would be so much controversy and opposition that it wouldn’t be remotely feasible. But so far, that’s a point about de facto obstacles to such a scheme. We might say that except for constraints of feasibility, plural voting could be justified. Opposition, after all, is not always reasonable. But, on the other hand, sometimes it is reasonable. So, political liberalism throws up one more hurdle: there would be an argument available (though not one that Rawls ever considered) that any proposed criterion for superior voting competence seems bound to be subject to reasonable disagreement. For example, education is historically correlated with things like race and class – that is, demographic features that might confound, or could at least reasonably be suspected of confounding, the epistemic advantage that is granted to be conveyed by the education itself. On such a fundamental constitutional matter, political liberalism would block any attempt to justify it that is so open to reasonable dispute. Is it true that every possible criterion for epistocracy is open to “reasonable” objection in the appropriate sense?27 It must be said that political liberalism as developed so far has not produced any clear and well-defended criterion. Maybe no such criterion is needed in cases that are, for dialectical purposes, clear enough on all sides. For example, some objections to a political proposal would be crazy, or illogical, or despicable, and thus plausibly weightless apart from whatever obstacle they might pose to feasibility. Examples include such proposals as disenfranchising members of a powerless religion or ethnicity. I think it’s fair to say that not every objection that would arise to even the most defensible of criteria for superior voting competence (perhaps this would be an educational credential of some kind) would be crazy, illogical, or despicable. Unfortunately, though, that doesn’t tell us where the line is, precisely or even approximately. It’s worth adding that even if finding the 27

I contend with these issues at length in (Estlund 2008, 2011). Wiens and Ingham (2022) fortifies the “demographic objection” considerably.

One Person, at Least One Vote?

309

line turns out to be intractable, we could still have reason to believe that the most defensible approach to political justification must take the distinctive form of political liberalism, even if its outlines are difficult to make precise, and even if this interferes with its availability in practical political contexts. As with theories in other domains, such as physics (say, quantum theory) or biology (say, evolutionary theory), this could, in principle, be settled by assessment of arguments for and against it and its competitors, even taking for granted a certain (epistemic or metaphysical) indeterminacy in its content. But at least for now we must acknowledge that there is considerable indeterminacy in the idea of the reasonable. In any case, a final point: this political liberalism gambit against plural voting or other epistocratic proposals depends on the account of justification having enough moral weight that we are required to forgo a greater extent and security of justice and liberty lest we violate the principle of justification. That is, according to political liberalism, even if plural voting in some form would, in fact, render the basic liberties more secure, that isn’t enough to justify the arrangement if the criteria for voter competence are open to reasonable disagreement in the right way. On one hand, this might comfort those who believe political equality is a deep robust requirement. On the other hand, even if the principle of justification has some weight, more would be needed to establish that it has that much weight, at the expense of justice and basic liberty. I don’t mean to imply that this couldn’t be established.28 17.6

Would Openness to Plural Voting Be an Embarrassment?

According to the account I find in Rawls, would a constitutional provision for plural voting ever be justified in real social and political conditions if it were politically feasible? That depends on the plethora of considerations that might, in the actual moment, inform the best prediction about how the other basic liberties could be expected to fare without, and then with, such a provision. And he does say that “unequal political liberty might conceivably have been a permissible adjustment to historical limitations” (TJ 1999, 217) so it is no idle detail of his theory. How troubling should that be? It’s helpful, as a start, to point out that as a matter of logic, any objector would be selecting this one basic liberty – a right to an equally weighted vote – and raise it, alone among them, and 28

Rawls says some pertinent but inconclusive things in this direction at PL, 138 and 152–53.

310

David Estlund

with lexical priority, above even the other basic liberties – as being inviolable even if the result is grave, unequal, and avoidable risk to other equal basic liberties such as conscience, association, liberty of the person, of speech, and expression. This opposing view can begin to look like the extreme one of the two. A second point is that the idea of epistemic plural voting has no special solicitude for elites. We can even say that a natural possibility would be to consider, still on epistemic grounds, rather than on some model of redress for example, plural votes for members of groups whose liberties or interests are systematically at risk in an enduring way. This analysis would not be taken up in a theory of an idealized well-ordered society such as Rawls’s, but the instrumental account of the political liberties that his ideal theory generates seems to warrant consideration of what we might call this kind of remedial epistemic plural voting in historically unjust societies such as our own and most others. This possibility, incidentally, illustrates how a group’s having more votes than others does not automatically put them on top in an overall social hierarchy, which would otherwise be a powerful general objection on grounds of the social bases of self-respect. A third helpful point is that some possible uses of the idea may not raise democratic hackles after all, and it might only be justly applicable in settings like that. For one example, so long as bias and bigotry can be avoided, even if poor voter competence were plausibly to correlate with being, say, poor or oppressed, even so stalwart an advocate of oppressed groups as W. E. B. Du Bois allows that voting power might justifiably be formally unequal – not merely in the form of plural voting but even denial of the franchise to some adults –on epistemic grounds, for some period. In an essay supporting the enfranchisement of Blacks and women in the United States, du Bois’s account of electoral democracy is mainly instrumental and epistemic, as are the accounts of Mill and, so I have argued, Rawls. Du Bois (1920, 147) wrote: Every citizen must be enfranchised. There may be temporary exclusions, until the ignorant and their children are taught, or to avoid too sudden an influx of inexperienced voters. But such exclusions can be but temporary if justice is to prevail.

Mill or Rawls may or may not have been prepared to go that far, but Du Bois’s reasoning for formally unequal voting power is the same epistemic rationale as theirs. In a similar spirit, but this time not conditional on any past injustice, consider a decision whether to expand suffrage to enfranchise all people above the age of ten. Some such measure might be optional as a matter of

One Person, at Least One Vote?

311

justice (adjust the age to whatever makes that most plausible to you) on the Rawlsian model if the extent or equality of other basic liberties were not plausibly at stake. For example, the proposal might be for the sake of other matters such as questions of public savings and debt, educational policy, encouragement of reflection and experience toward the broad duties of citizenship, and so on. If such an expansion is optional, then Du Bois’s form of argument would seem to have natural application but this time not owing to any history of unjust disenfranchisement. It might be justified or required to afford the new minor voters with lesser formal voting power than adults on epistemic grounds. If so, it is an example where Rawls’s account of epistemic plural voting might have application in realistic conditions, not necessarily only in the past, and without depending on the presence of any injustice or misbehavior. And this is also no “falling heavens” sort of exception to strict political equality, illustrating the earlier point that Rawlsian formal political equality is not very strongly egalitarian. There is a temptation, at least in the political culture, though emphatically not in Rawls, to think that the moral equality of persons directly implies a right to strict formal political equality – we’re all equal, therefore no one may have more voting power than anyone else. All that moral equality shows directly, though, is that formal inequality premised on moral inequality is forbidden. But epistemic plural voting is not premised on moral inequality, as we have seen. In Rawls, the equality of persons is represented not in that way at all but by the original position. The principles for institutions that honor that equality of persons, including but not limited to institutions making up the political system, are to be determined from that standpoint and the connected constitutional stage of argument. If strict political egalitarianism were the result, then this would obviously be a theoretical basis for a prohibition on plural voting. I have argued that according to Rawls that it is not the result. This does not mean, by the way, that equality of social relations – the absence of overall hierarchy – is not the result; plausibly, it is. But what it is to count as living together as equals without hierarchy comprises all the basic liberties considered together, not only – and not even primarily – the political liberties.29 In effect, justice as fairness allows in principle that living together as equals might, in some conditions, be best fostered by plural voting. Of course, there are any number of alternative theories about the normative basis of electoral democracy. But in assessing any theory of 29

Thanks to Paul Weithman for encouraging me to fold in the issue about hierarchy as well as distributive equality of liberties.

312

David Estlund

voting rights, it is important to see that a fundamentalism (to speak a bit tendentiously) about strict political equality would, in an important way, be similar to a different fundamentalism, namely one about an allegedly inviolable right to engage in any “capitalist acts between consenting adults” (Nozick 1974, 163). The two are similar in this way: each of these fundamentalisms would hold firm even if the result over time were to effectively put some groups and their liberties at the mercy of others. Rawls, as I understand him, argues that there is no adequate basis for either fundamentalism – the one about consensual market exchange or the one about one-person-one-vote. Rawls takes equality more seriously than that. The structure of the basic liberties, including the political liberties, is best assessed from a standpoint that also gives weight to the extent and degree of equality of all of the liberties and their protection over time, as the original position is meant to do.

18

Reflections on Democracy’s Fragility Joshua Cohen*

18.1

The Problem of Fragility

A Theory of Justice is a work of realistic utopianism, which defends “reasonable faith in the possibility of a just constitutional democratic regime” (PL, 172). Defending faith in a possibility is not arguing for a confident expectation. And Rawls’s writings also suggest concerns about the fragility of democracy, with growing intensity in his later writings.1 I want to explore the reasons for these concerns, even when we make the favorable assumptions about people and social cooperation that provide the basis for his case for the possibility of a just constitutional democracy. In focusing on fragilities that arise on favorable assumptions, this exploration provides a lens on democratic fragility different from views that emphasize the dangers of mutually reinforcing social cleavages (Achen and Bartels 2016; Abramowitz, 2018; Mason 2018; Haggard and Kaufman 2021), or tensions between political accountability and state policy autonomy (Fukuyama 2014), or popular alienation from political elites, perceived as rootlessly cosmopolitan servants of global markets. I do not mean to dismiss these concerns but to supplement them. The Rawlsian concern about democracy’s fragility proceeds as follows: Assume reasonable political pluralism, that the politically active citizens in a

* I am grateful to David Estlund, Samuel Freeman, Archon Fung, Erin Kelly, Niko Kolodny, Christopher Kutz, Rob Reich, Tim Scanlon, Karen Sipprell, Coleman Solis, Noelle Stout, Kevin Vallier, Jay Wallace, and Paul Weithman for discussion and for comments on drafts of this essay. I presented earlier versions at a conference on the fiftieth anniversary of A Theory of Justice at Notre Dame University and to Berkeley’s Kadish Workshop in Law, Philosophy, and Political Theory. 1 This mood strikes me as especially strong in a 1998 draft of a revised version of Political Liberalism (references to this draft will be abbreviated in the text as PLR). In 1998, Rawls sent two letters to Anne Miller, the editor at Columbia University Press, about the revised draft. He explains that political liberalism is focused on a family of reasonable political conceptions that satisfy “the criterion of reciprocity,” not on justice as fairness. “Justice as fairness” is replaced nearly everywhere by “a family of reasonable political conceptions.” He continues to assert that justice as fairness is the most reasonable conception, but he does not explain what role that assertion might play in political liberalism.

313

314

Joshua Cohen

society endorse competing, reasonable conceptions of justice. Each such conception requires basic liberties and opportunities, some form of democracy, and a decent level of resources for all. Still, we may see deep disagreement among these conceptions of justice, the forms of democracy they recommend, and the interests and identities of citizens who endorse those competing conceptions. Together, this polarization and fragmentation about justice and democracy, and correlated conflicts of interest and identity, may produce a downward spiral of distrust, resentment, and hostility.2 To be sure, these political dynamics may not lead directly to withdrawal of support from democratic institutions. But they may result in an attenuated commitment to democracy. Thus its fragility.

18.2

An Orientation to Democracy: Four Episodes

From early on, Rawls’s political philosophy had an orientation to democracy. He thought of his philosophical writing as a contribution to the public reasoning about justice and the common good that he saw as essential to a well-functioning democracy.

18.2.1 Dissertation Consider Rawls’s dissertation. Focused on judgments about the “moral worth of character,” the dissertation aims to refute anti-rationalist sensibilities about ethics by showing that ethical judgments can be justified through reasoned discussion (SGEK, 11). Ethical anti-rationalism comes in “positivist” (emotivist) and “authoritarian” variants (SGEK, 1–4). These views “encourage in social life just those elements which, in democratic countries, we have tried to get rid of: the authoritarian, the arbitrary, and the irrational” (SGEK, 7). To clarify the threat, Rawls sketches a conception of democracy that emphasizes rational discussion. Because rational discussion is fundamental, “democratic theory and practice must consider the process of reasoning as one of the very crucial points in its whole program.” The defense of moral reasoning is thus intended as a defense of the kind of public discussion that democracy depends on: “It is because the two views mentioned above [positivism and authoritarianism] strike at what is essential in democratic theory that 2

The literature on polarization is vast. See Fiorina (2017); Rosenfeld (2017); Abramowitz (2018); Mason (2018); McCarty (2019). Some of the literature treats polarization as disagreement in beliefs and convictions; some is social-psychological and focuses on the alignment of social cleavages. I am focused here on differences of belief and conviction.

Reflections on Democracy’s Fragility

315

the question about the rational foundation of ethical principles is worthy of our attention” (SGEK, 8). Rawls thus thought of his work from the outset as oriented to democracy and focused on a conception of democracy that depends on public reasoning about what is right and reasonable. Because such public reasoning is so important, positivism and authoritarianism are dangerous when they become widely shared views about normative discourse. This democratic orientation – with its focus on philosophy’s contribution to public reasoning – plays a central role in A Theory of Justice and Political Liberalism. And I will consider the key elements of that role in a moment. But I want first to bookend this early statement about the practical importance of public reasoning with a vivid late discussion in Rawls’s Lectures on the History of Political Philosophy.

18.2.2 The German Case3 Imagine someone saying: “Yes, of course people go in for high-minded, public rhetoric about rightness and the common good. But why take it seriously?” “Why isn’t politics,” Rawls asks, simply the struggle for power and influence – everyone trying to get their own way? Harold Lasswell said: “Politics is the study of who gets what and how.” Why isn’t that all there is to it? Are we naive, as the cynic says, to think that it could be anything else? If so, then why isn’t all the talk of justice and the common good simply the manipulation of symbols that have the psychological effect of getting people to go along with our view, not for good reasons, plainly, but somehow mesmerized by what we say?

Rawls responds in three steps. First, as in the dissertation, he affirms the reality of normative public reasoning: that such reasoning is what it seems to be, namely the exploration of requirements of justice and the common good. If principles and ideals were simply discursive lipstick on the unprepossessing pig of power, then people would have long since dismissed them: “People are not so stupid as not to discern when those norms are being appealed to by certain groups and their leaders in a purely manipulative and group-interested fashion.” Second, Rawls emphasizes the educative effects of normative public reasoning. We cannot expect broad allegiance to ideas of justice and the common good to arise simply from families, schools, and churches. Such allegiance depends also on public, political discussion. Reflecting on Wilhelmine and Weimar Germany, Rawls observes that German 3

These remarks draw on LHPP, 7–9.

316

Joshua Cohen

political parties functioned like “pressure groups” aiming to extract resources for their base. They did not make the more encompassing arguments about justice and the common good required to put together a stable working majority (PL, LXI–LXII). In a mass democracy, the attention of most citizens is not focused on politics. Democratic success depends on citizens’ acquiring a sense of justice, which happens (if at all) through public discussion of a kind that was absent in Weimar (TJ 1999, chap. 8; PL, Lecture 4). Which leads to the third point, about the profound consequences of normative public reasoning (or its absence). German officials and parties were not engaged in a public discussion focused on justice and the common good. The normal workings of the political system thus did not have the desired educative effect. A fundamental problem was the absence from public discussion of the more or less sincere and constraining arguments about principles of justice and the common good. As a result, there was not broad citizen allegiance to principles of justice, which doomed democracy. 18.2.3 A Theory of Justice When we turn back to A Theory of Justice, we find these same three themes about normative public reasoning in a democracy. First, we have the reality of normative political reasoning. A Theory of Justice aims to contribute to democracy by identifying the most suitable principles of justice for public reasoning: “The principles of justice are among the main criteria to be used [by citizens] in judging a representative’s record and the reasons he gives in defense of it” (TJ 1999, 199–200). Second, this open appeal to principles of justice in public reasoning plays an educative role in forming a public will. It is essential to the acquisition of a sense of justice, thus to the stability of the well-ordered society of justice as fairness: “The morality of association quite naturally leads up to a knowledge of the standards of justice.” By using these principles in public discussion, people acquire “a mastery of these principles” and understand “the values which they secure and the way in which they are to everyone’s advantage” (TJ 1999, 414). And third, while members of a well-ordered society have diverse, conflicting conceptions of the good, “everyone has a similar sense of justice and … [p]olitical argument appeals to this moral consensus” (TJ 1999, 232). Moreover, this moral consensus has profound consequences: it is needed in making “a viable human community” possible. “In the absence of a certain measure of agreement on what is just and unjust, it is

Reflections on Democracy’s Fragility

317

clearly more difficult for individuals to coordinate their plans efficiently in order to ensure that mutually beneficial arrangements are maintained. Distrust and resentment corrode the ties of civility, and temptation and hostility tempt men to act in ways they would otherwise avoid” (TJ 1999, 6). Like the dissertation, then, A Theory of Justice is oriented to democracy, in aiming to contribute to the public exercise of reasoning about what is right that Rawls sees as so essential to democratic will-formation and democratic success. 18.2.4 Political Liberalism In Political Liberalism, Rawls is “concerned with the survival, historically, of constitutional democracy” (CP, 616). Moreover, the three themes I have mentioned are all present but along with a new and fundamental question: How is the public reasoning about justice that is essential to democracy possible? Such reasoning seems at odds with the “pluralism of reasonable comprehensive religious, philosophical, and moral doctrines found in modern democratic societies.” (PL, 36) Democracy depends on public discussion with some measure of agreement about justice. How, then, is democracy possible, given this pluralism? To clarify the force of the question, I want to draw attention to Rawls’s distinction between three types of conflict in a democratic society: conflicts deriving from the burdens of judgment, from conflicting comprehensive doctrines, and from our social positions, interests, and identities (PL, lx; CP, 612–13; PLR, 35). Putting aside for now the burdens of judgment, we have conflicts about comprehensive doctrines, which address “the highest matters” (CP, 614) and lie at the basis of “the most intractable struggles.” And we have conflicts that come from differences in “citizens’ fundamental interests – political, economic, and social” – associated with “status, class position, or occupation, or differences in ethnicity, gender, or race” (CP, 612). Conflicts of interest and identity are not about the highest things and are assumed to be more tractable. They need to be resolved, and can be, if the society satisfies principles of justice that members recognize as “reasonable (even if not the most reasonable)” (CP, 612). To keep conflicts of interest and identity from disrupting democracy, people must recognize justice as fundamental (perhaps as the first virtue) and have some measure of agreement about the requirements of justice. The tractable conflicts can be resolved because and so long as there is a shared, superordinate allegiance to principles of justice, “the fundamental charter of a well-ordered human association” (TJ 1999, 5).

318

Joshua Cohen

But intractable conflicts about the highest things remain. Preserving an enduring and secure democratic regime – one that addresses the tractable conflicts – requires a sufficient measure of agreement on justice, consistent with those intractable disagreements. The answer is an overlapping consensus. We can remain in unresolvable disagreement about the ultimate religious and moral convictions, so long as the conflicting doctrines have the resources to support or are at least consistent with endorsing, a reasonable conception of justice – one of the family of reasonable conceptions of justice that can make the conflicts of interest and identity tractable. But is such an overlapping consensus utopian? Answering this question requires a sense of what it would take for an overlapping consensus to be possible. Thus Rawls says: In such a consensus, the reasonable doctrines endorse the political conception [justice as fairness], each from its own point of view. Social unity is based on a consensus on the political conception; and stability is possible when the doctrines making up the consensus are affirmed by the society’s politically active citizens and the requirements of justice are not too much in conflict with citizens’ essential interests as formed and encouraged by their social arrangements. (PL, 134, emphasis added)4

We have two conditions here. First, we have consensus on a political conception of justice or family of political conceptions – which is a source of social integration or social unity. Such unity bounds the conflicts that arise from our interests and identities. Though interests and identities are not naturally harmonious, there is a possibility – with sufficient social unity – of achieving a reasonably just resolution of the tractable conflicts. So Social Unity requires that there be enough agreement about justice to contain the tractable conflicts. Second, a political conception, which is the focus of an overlapping consensus, cannot make excessive demands on citizens, given their basic interests. Let’s call this the Incentive Compatibility condition. My discussion of the fragility of democracy will focus on these two conditions. Can we reasonably expect to meet these two conditions?

18.3

Pluralisms and Democracies

As a bridge to that discussion, I want now to add two ideas that may create troubles for Social Unity and Incentive Compatibility, thus for an overlapping consensus. 4

This passage changes in PLR, 105 in ways we will come back to.

Reflections on Democracy’s Fragility

319

18.3.1 Two Pluralisms Political Liberalism distinguishes two kinds of reasonable pluralism. The pluralism of reasonable comprehensive doctrines is the fact that, under conditions favorable for reflection and deliberation, reasonable people are drawn to a plurality of reasonable but incompatible comprehensive doctrines. Putting aside the complexities associated with the reasonableness of comprehensive doctrines, I will assume that reasonable comprehensive doctrines are comprehensive doctrines endorsed by reasonable citizens. Because they are endorsed by reasonable citizens, such comprehensive doctrines are at least consistent with a reasonable political conception of justice – of which more in a moment. The fact of reasonable political pluralism is the fact that, under conditions favorable for reflection and deliberation, reasonable citizens are drawn to competing reasonable political conceptions of justice (PL, XLVIII–XLIX). As I mentioned earlier, A Theory of Justice defines a well-ordered society as a society with fully shared public knowledge that everyone accepts the same fundamental principles for “assigning basic rights and duties and determining what they take to be the proper distribution of the benefits and burdens of social cooperation.” Moreover, the society satisfies and is publicly known to satisfy, as a general matter, those shared fundamental principles (TJ 1999, 4; JF, 8; 35; PLR, 26). This conception of a well-ordered society is not simply a terminological stipulation. It reflects a broader set of substantive convictions about social cooperation. Thus, “some measure” of agreement on what is just and unjust makes a “secure association together possible.” Absent a “certain measure of agreement,” it is “clearly more difficult” to solve such practical problems as social coordination, efficient cooperation, and social stability. A well-ordered society enables us to avoid or resolve the tractable conflicts growing out of differences of interest and identity. Without sufficient agreement about justice “distrust and resentment corrode the ties of civility, and suspicion and hostility tempt men to act in ways they would otherwise avoid” (TJ 1999, 5–6). I do not mean to suggest that standards of justice are foreign to our nature, which puts us constantly on the precipice of social conflict. To the contrary, there is a powerful strand of natural law thinking in Rawls’s view: norms of justice express our moral powers and grow from a reciprocity deeply rooted in our nature (TJ 1999, 433; PL, 81–86). But the tractable conflicts are always present and depend for their resolution on a shared sense of justice. Reasonable political pluralism puts pressure on this shared sense. The “two pluralisms” (PLR, 3) have a common explanation. The pluralism of conflicting reasonable comprehensive doctrines reflects the

320

Joshua Cohen

“burdens of judgment” faced by reasonable citizens (PL, 54–57). The aim in invoking the burdens of judgment is to identify sources of disagreement, while operating in a tolerant spirit. The idea is to explain disagreements consistent with acknowledging that people are reasonable: they have developed their moral powers; are able to draw inferences, consider evidence, weigh competing considerations and interpret values categories; have a settled interest in offering and respecting fair terms of cooperation; and recognize the difficulty of resolving certain fundamental questions. Specifically, reasonable people may disagree at least in part because we evaluate, weigh, interpret, and judge in light of our “total experience, our whole life course up to now” (PL, 57). “Thus, in a modern society with its numerous offices and positions, its various divisions of labor, its many social groups and their ethnic variety, citizens’ total experiences are disparate enough for their judgments to diverge, at least some degree, on many if not most cases of any significant complexity” (PL, 57). Moreover, people with shared social positions and common identities may have sufficiently common experiences for their judgments and evaluations to cluster, consistent with their being fully reasonable, not simply solidaristic members of their tribe. The same burdens of judgment that account for the pluralism of reasonable comprehensive doctrines may also help to account for reasonable political pluralism. Rawls says that the burdens of judgment apply “considerably less forcefully” to principles of justice than to comprehensive doctrines (PLR, 47). Whatever the differences, both comprehensive doctrines and conceptions of justice face the burdens of judgment. In both cases, we have disagreements that need not be explained by pointing to moral or epistemic vices. Moreover, reasonable disagreements may reflect differences in the social experience of individuals and groups. Reasonable political conceptions must all satisfy the “criterion of reciprocity,” roughly, that the conception presents terms of cooperation that adherents think others can reasonably accept as free and equal citizens (PL, xliv, 49–50, 54; CP, 578). To satisfy this criterion, the conceptions need to meet three conditions: specify certain liberties and opportunities, give “a special priority” to those liberties and opportunities, and assure “adequate all-purpose means to make intelligent and effective use of their liberties and opportunities,” including their political liberties (PL, XLVIII). With reasonable political pluralism, a well-ordered society is now understood always to involve a family of such political conceptions of justice, all regarded as at least reasonable, “even if barely so,” by proponents of other views (CP, 578). The best we can hope for by

Reflections on Democracy’s Fragility

321

way of social unity is that the basic structure meets the conditions set out by one member of this family (or some blend of them), that “an enduring majority” of citizens embrace some reasonable conception or other, and that political active citizens regard the regnant conception as reasonable. How do the members of this family differ? Rawls’s two principles of justice are an egalitarian version of liberalism. The egalitarianism is expressed in the difference principle’s requirements of justifying inequalities by showing that they work to the maximum benefit of the least well-off and in the conception of fair equality of opportunity and the fair value of the political liberties. Other views, less egalitarian views on each of those dimensions, appear to meet the three conditions on a reasonable political conception. The liberalism is expressed in the priority of a family of basic liberties: the family of basic liberties includes both personal and political liberties, and the priority is lexical. Here, there is room for reasonable disagreement on the nature of the priority rules and the substance of the liberties. Thus, lexical priority is an especially strong priority rule: an alternative, reasonable view would give the liberties great weight but not lexical priority (Sen 2009, 63). And on the substance of the liberties, some reasonable political conceptions may give especially significant weight to religious liberties because of the special reasons that adherents have for complying with religious obligations, as they understand them, and be prepared to restrict the scope of political liberty to ensure appropriate protections (PL, 310–12). Others may see those restrictions as destructive of political liberties, which have a special importance. 18.3.2 Contestability of Democracy All reasonable conceptions of justice endorse some form of democracy. To be more specific, I will assume that a “democracy” must meet at least four conditions: regular, competitive elections for law and policy making positions; protections of rights of speech and assembly; a government chosen through elections that actually makes law and policy; and universal suffrage. Those conditions leave a great deal open. With reasonable political pluralism, we should expect proponents of different political conceptions to favor different forms of democracy. So consider a well-ordered society in which different groups are drawn to different reasonable conceptions of justice. The less egalitarian conceptions might reject the difference principle in favor of an assurance of a decent minimum or reject the requirements of the fair value of political liberty or fair equality of opportunity in favor of weaker assurances of opportunities that are adequate for meeting some specified target. Or they may weaken the priority rules or offer a different account of an

322

Joshua Cohen

adequate set of basic liberties, perhaps with special emphasis on religious liberty. All these views seem consistent with the three conditions on a reasonable political conception that Rawls associates with the criterion of reciprocity. Moreover, these different political conceptions of justice lead different groups to support competing conceptions of democracy, as an authoritative form of collective decision-making. As a general matter, the selection of a just, democratic constitution needs to look both to the justice of political process and to the justice of outcomes that issue from that process. Consider the case of justice as fairness. Here, justice of process is defined by the rights and liberties covered by the first principle, especially those covered by the principle of participation: “allcitizens are to have an equal right to take part in, and determine the outcome of constitutional processes that establish the laws with which they are to comply” (TJ 1999, 194). Justice of outcomes is assessed both in terms of preserving the basic liberties and by reference to the need to ensure adequate income and wealth. Rawls’s case for political democracy is based in the first instance on the first principle – both the principle of participation and the requirement of a process that protects other basic liberties. The second principle reinforces the case for political democracy, at least if we suppose that greater equality of political power means a smaller likelihood of class legislation. But the second principle has more than a reinforcing role. It helps to define what scheme of democracy is most suitable: majoritarian or consensual, more constraints on forming stable and assertive majorities or greater ability of majorities to experiment, more representative or more direct, more proportionality to reflect social diversity and foster representativeness, or more majoritarian to better enable accountability. All of these variants are compatible with the four basic features of democracy, but they may well have very different institutional consequences. Consider now the fact of reasonable political pluralism, with a range of reasonable conceptions of justice. Those differences lead to conflicting views about what the appropriate form of democracy is. A more egalitarian view of justice might support a more majoritarian constitutional democracy, to ensure that entrenched economic power is not in a position to prevent legislation required for economic fairness. But other reasonable political conceptions of justice provide good reasons for supporting different conceptions of democracy. 18.4

Political Pluralism, Overlapping Consensus, Fragility

Let’s now consider the idea of overlapping consensus and its implications for democratic fragility in light of reasonable political pluralism and the

Reflections on Democracy’s Fragility

323

contestability of democracy. To better define the issue, I want to draw attention to three dimensions of overlapping consensus that distinguish it from what Rawls calls “constitutional consensus”: breadth, depth, and specificity. 18.4.1 Specificity A constitutional consensus is, essentially, an agreement on political procedures for authoritatively resolving disagreements. An overlapping consensus has greater breadth than a constitutional consensus because it extends beyond matters of political procedure. So an overlapping consensus will require the protection of some nonpolitical liberties – say, liberty of conscience – as well as some kind of assurances about education and material well-being. The assurances will vary across different members of the family of reasonable conceptions. But all reasonable conceptions need to develop greater breadth as they try to win adherents by speaking to “the fundamental political questions that will be debated” (PL, 166). An overlapping consensus is also deeper than a constitutional consensus. It is not simply an agreement on standards of justice but on fundamental ideas as well, such as the idea of society as a fair system of cooperation or of persons as free and equal citizens in virtue of their possessing to a sufficient degree the capacities required for being social cooperators. Once again, efforts to broaden political support through public argument pushes politically active groups to appeal to more fundamental ideas, such as the idea of fairness or a conception of citizens as free and equal. Specificity is the key condition for our purposes. The intuitive idea is that a consensus is more or less specific (or wide) depending on how different the reasonable political conceptions of justice are that win adherents in a democratic system. In the revised version of Political Liberalism, Rawls states the condition this way: “Thus, the focus of an overlapping consensus is a family of liberal conceptions that vary within the range allowed by the criterion of reciprocity. The more restricted the range, the more specific the consensus. In a political society with a consensus of this kind, several conceptions of justice will normally be political rivals, no doubt favored by different interests and strata” (PLR, 131). So the normal condition is a plurality of reasonable conceptions, with differences along both the equality and liberty dimensions. Moreover, differences of interest correlate with the differences in conception of justice. And these differences of interest are especially important in thinking about Incentive Compatibility. Assume, for example, that wealthier citizens are ordinarily less inclined than the least wealthy to support an egalitarian conception of justice, that people with

324

Joshua Cohen

lower social status may be more inclined to support a stronger formulation of fair equality of opportunity, and that people with less widely held religious convictions will be more likely to be drawn to a conception of basic liberties that gives great weight to the protection of religious freedom. Thus “different social and economic interests may be assumed to support different liberal conceptions. The differences between conceptions expresses, in part, a conflict between these interests …. The width of the range of liberal conceptions will be determined by the degree of opposition among these interests” (PL, 167; PLR, 134). This observation may seem surprising. If differences of interest draw people to different conceptions of justice, perhaps conceptions of justice are rationalizations for interest after all. That conclusion is too quick. Each of the conceptions of justice under consideration satisfies the criterion of reciprocity. The issue here is what draws people to different reasonable conceptions. Are people simply drawn to the view that best serves their interests? Perhaps not. For example, someone who has experienced the burdens of restrictions on opportunity may have a better understanding of those burdens, and be drawn to a conception of justice with a stronger formulation of fair equality of opportunity. They say that they are drawn to it because their life experience and the experience of people comparably situated provides normatively relevant insight. Similarly, someone with experience running businesses may think that they have normatively relevant experience of the burdens that come from laws and policies required to achieve more equal conditions. They think they understand something that others fail to understand. Returning then to the width or specificity of an overlapping consensus: we may be inclined to focus on variant conceptions of fair distribution, thus to think of the width as bounded by Rawls’s two principles and “mixed conceptions” of justice that offer alternatives to the difference principle. But as I said earlier, this way of bounding the width is too narrow. Reasonable political pluralism includes political conceptions that vary along both the equality and liberty dimensions. Now we know (or can assume) a few things that will be important to the exploration of the Social Unity and Incentive Compatibility discussion that follows: (1) Three Conditions. Reasonable conceptions of justice must include the three provisions identified earlier as required by the criterion of reciprocity. (2) Rivalry. Normally, we expect a rivalry among different conceptions that differ on both the liberal and egalitarian axes.

Reflections on Democracy’s Fragility

325

(3) Acquisition. People acquire conceptions of justice through political participation. (4) Correlation. Groups with different interests and identities are drawn to different conceptions. (5) Width. The greater the opposition of interests and differentiation of identities in the society, the greater the width of the overlapping consensus and the corresponding disagreements about justice. I want now to consider the reasons for concern about democratic fragility that come from the Social Unity condition and the Incentive Compatibility condition. 18.4.2 Social Unity: Religious Exemptions People (at least those with appropriate religious convictions) have reason to care about their salvation, about public security, and about the quality of the public environment that they live in and about the preservation of the earth. And their concerns about each of these can come into conflict with a commitment to democracy. They care (with good reason) about democracy, but they care about – and have reason to care about – some good whose realization may conflict with it. Let’s focus on the religious case and consider a simple example.5 To be clear, the concern I wish to explore is not about a tension between a comprehensive religious doctrine and principles of justice but about the content of the principles of justice themselves, in particular how they conceive of the protection of religious liberty. Suppose I live in a community in which most people keep a Sunday sabbath. I am a Saturday sabbatarian. A regulation is passed that denies me unemployment benefits if I refuse, from a sense of obligation, jobs that require me to work on Saturday. The administrative denial is made pursuant to a law enacted through a democratic process, and there is no evidence that the legislation is driven by animus against people like me. Of course there are other, more complex cases of religious exemptions, concerning contraception, gender equality, and marriage equality. In these cases, the claim for an exemption is motivated by a concern not to be complicit in what is judged to be a moral violation and the regulations are understood to serve the value of civic equality.6 But the simpler case suffices to clarify the logic of the issue. 5 6

My example is a simplified version of the issues in Sherbert (1963). See Employment Division (1990); Burwell (2014); Masterpiece (2018); Little Sisters (2020); Fulton (2021). For helpful discussion of the constitutional, philosophical, and political issues, see McConnell (1990); Tebbe (2021a, 2021b).

326

Joshua Cohen

I might initially complain that the decision was wrong, not that the process was undemocratic. The decision is wrong because of the nature of the burden it imposes, but the process is independently recognizable as a democratic process. But there is an alternative way to present the case, attentive both to reasonable political pluralism and the contestability of democracy. We know that there are a range of reasonable conceptions of justice, and these conceptions may disagree about their understanding of religious liberty as a basic liberty and in particular how to treat claims for religious exemptions from generally applicable law. Let’s say that some reasonable conceptions treat those exemptions as mandatory. Because of the importance of religious reasons, these conceptions require especially compelling public reasons for overriding them. Others think that burdens on religious exercise are an inevitable consequence of making general rules of conduct for a pluralistic society, with different and conflicting comprehensive doctrines. People drawn to the first view may be inclined to find it compelling because it provides them with special protections. But that does not of itself make their endorsement self-serving. Instead, in the spirit of the burdens of judgment, we can say that the nature of their experience – shared with others in their group – gives them a special awareness of the severity of burdens on religious exercise and that this awareness draws them to a view that promises more stringent protections. Moreover, because the design of a democratic constitution is guided by a conception of justice, we now have a potentially deep disagreement about the kind of democracy that we ought to have. Should the protection of religious freedom be left largely to a majoritarian process that permits burdens on religious freedom so long as they come from generally applicable laws? Or should there be a design of democratic process that gives greater weight to the protection of religious freedom, with mandatory exemptions from regulations that burden religious exercise? People disagree about the justness of regulations, which leads them also to disagree about the appropriate processes for adjudicating these disagreements. There is no shared basis because the differences in conceptions of justice provide a basis for different conceptions of what an appropriate democratic constitution should look like. Moreover, there is not only a disagreement but a basis for mutual suspicion and mistrust. Consider the person with religious convictions who faces a conception of justice and associated form of democracy that does not provide especially stringent protections for religious freedom. They see that a reasonable case can be made for this view but wonder if the secular co-citizens are drawn to it because they have an anti-religious animus. And they see that suitable protections may be provided by a

Reflections on Democracy’s Fragility

327

majoritarian political process but worry that more secular citizens do not have a sufficiently strong sense of what religious convictions are. When they are told that the regulations are adopted democratically, that does not provide them with a strong reason because they think that the kind of democracy that authorizes the rules is the wrong kind, recommended by the wrong view of justice, and may itself be an expression of the antireligious animus. They embrace reciprocity but not being suckers. Correspondingly, more secular citizens may see the reasonableness of the conception of justice and design of democracy that provides stronger protections. But they may also suspect that the disagreement about justice and the design of democracy is animated by an interest in being free of reasonable requirements that we all are bound by. They are struck by the correlation between the conception of justice and democracy advanced by the religious citizens who favor stronger protections and the interests and identities of those citizens who may wish to be freed from being burdened by general regulations. They have a sense of the burdens of judgment: perhaps the allegiance is not self-serving but is born of insight about the magnitude of a burden. But there is cause for concern. They are all for reciprocity but not for being suckers. Each side has some basis for concern about the democratic commitments of the other side. Each is concerned that the other side is bending democracy to their advantage. They do not have an agreed basis in an understanding of justice or in a conception of democracy as the basis for resolving their disagreements. Instead, they have a less structured concern about the reasonableness of the people they disagree with. Their strains are symmetrical. Each side may understandably lose confidence in the democratic convictions of the other side. Invocations of values of trust and civility may seem like invitations to a mug’s game. And they may then think it is permissible to break the democratic rules both because of their substantive commitments and because they are concerned that the other side is breaking the rules. 18.4.3 Incentive Compatibility, Inequality, and Democracy Though Rawls does not say very much in A Theory of Justice about the conduct of democratic politics, two things are clear: first, that he rejects the economic theory of democracy, which sees voters as voting for candidates on the basis of utility maximizing judgments about which candidate is closer to their preferred point in policy space (TJ 1999, 431). Instead, voters make judgments based, in the first instance, on their conception of justice: in the well-ordered society of justice as fairness, everyone is assumed to be basing those judgments on the same

328

Joshua Cohen

conception of justice. Second, Rawls thinks that democracy itself is required by justice and that the stability of democracy depends on citizens acting to uphold a democratic regime because they regard it as a requirement of justice. The story in Political Liberalism is similarly distant from an economic theory of democracy. Citizens are guided by a sense of justice, whose content is given by some member of a family of reasonable conceptions of justice. Justice as fairness, with its specifically egalitarian elements, is said to be the most reasonable conception. But the fact of reasonable political pluralism means that different citizens will endorse different conceptions, and there is no assurance that the dominant conception will be justice as fairness. How different are the competing reasonable conceptions of justice? In Political Liberalism, Rawls says “several conceptions of political justice will be political rivals and no doubt favored by different interests and different strata.” Moreover, we are assuming that this pluralism of reasonable political conceptions is not simply what we can realistically expect: instead, the focus of the overlapping consensus is a family of conceptions “that vary within the range allowed by the criterion of reciprocity.” So here, we have a plurality of reasonable political conceptions. In part, the question here is philosophical: What are the views that respect the criterion of reciprocity and can be elaborated from the fundamental ideas of society as a fair system of cooperation, free and equal persons as full participants in such a fair scheme, and a well-ordered society as a society that is effectively regulated by a public conception of justice or a family of such conceptions? As I emphasized in the discussion of reasonable political pluralism, we should expect to see, even under favorable conditions, a range of reasonable conceptions varying on both the equality and liberty axes. But there is also a question about the relationship between convictions about justice and political, economic, and social interests and identities. Thus “different social and economic interests may be assumed to support different liberal conceptions. The differences between conceptions expresses, in part, a conflict between these interests” (PL, 167; PLR, 34). This claim about economic and social interests drawing people to different, competing conceptions of justice fits well with the account of the burdens of judgment and reasonable political pluralism. Recall the idea that differences in total experience lead to disagreements about how best to evaluate evidence, weigh conflicts of value, and interpret openended categories. Such disagreements need not impugn anyone’s reasonableness. But there is a concern. Suppose a reasonable political

Reflections on Democracy’s Fragility

329

conception encourages “deeply conflicting political and economic interests” and those conflicting interests, in turn, draw people to political conceptions that further expand the width of the overlapping consensus. Perhaps, then, “a full overlapping consensus cannot, it seems, be achieved” (PL, 168). Here we have the Incentive Compatibility concern. Though Rawls does not, as I said, endorse an economic conception of democracy, his concerns here do overlap with those of an economic conception. To illustrate, consider the Acemoglu-Robinson theory of democratization and democratic stability (Acemoglu and Robinson, 2006). Focusing on democratic transitions (I will be extending their point to the stability of a consolidated democracy), they argue that democratic institutions serve as a commitment device for economic and political elites. Rather than forestalling political resistance and revolution by offering discrete payoffs, elites can avoid political revolution by institutionally – thus more or less durably – enshrining some measure of political equality for less well-off citizens. The enhanced political power is of instrumental importance – whatever other importance it may have – because it enables poorer citizens to win greater benefits than under autocratic political institutions. Thus, less well-off citizens want a greater share of resources; having more political power increases their expected share of resources; democracy enhances that power; and because democracy is an institution, the increase in power, thus in expected benefits, is relatively durable. Of course, democratic transitions do not always occur, and democratic consolidation is not permanent. One of the conditions that increases the chances of democratic transition (and consolidation) is a significant but not excessive degree of inequality. With too little inequality, there is not enough to be gained for less well-off citizens to need to fight for greater political power. With too much inequality, elites have too much to lose and will opt for repression (see also Moore, 1966; Fukuyama 2014).7 But there is a range of inequality over which different classes all find democracy in their interest: less well-off citizens, because it enhances their political power enough to win sufficiently significant advantages; elites, because they want to forestall costly revolution and will not be ruined by institutional concessions of power. To bring the argument closer to our animating concerns about democratic fragility, let’s add that people may have intrinsic commitments to democracy as well: what Acemoglu and Robinson call “ideological 7

Moore’s point is not simply about inequality but about labor-repressive agriculture. But inequality is an important part of the story. See also Acemoglu and Robinson (2006, chapter 9) on the importance of different asset classes.

330

Joshua Cohen

preferences” for democracy as a requirement of justice (Acemoglu and Robinson 2006, 211–14). Still, the basic story about the “right level” of economic inequality fostering democratic transition holds up, so long as ideological preferences are “not large enough to totally swamp the relevant economic concerns of individuals.” Thus, the greater the inequality, the more economically costly democracy is for elites. So to achieve a democratic transition, their ideological preference for democracy would need to be correspondingly stronger. Suppose then that we have a political society with a range of conflicting political conceptions of justice, including justice as fairness with its egalitarian elements and alternative views of justice that do not have those elements: all views include some account of an adequate minimum, but the alternatives to justice as fairness have a lower minimum, great dispersion, less stringent requirements on socioeconomic and political opportunity. And let’s say that the leading views politically speaking are the least egalitarian alternatives and that those views are favored by the economically advantaged groups. Because these views are, by stipulation, reasonable (if only barely so), the arrangements are viewed by adherents to justice as fairness as legitimate though seriously unjust. Now suppose I belong to the Justice as Fairness party. I have views about justice and about the design of a just, democratic constitution. And my views differ, perhaps sharply, from the views of what I see as the Party (or Parties) of Unjust Inequality. Maybe I think that a shift to proportional representation – or to some more majoritarian form of democracy – would have the effect of weakening what I see as the Party of Unjust Inequality. Focusing on proportionality: I think it fully satisfies the principle of participation, fosters representativeness, and would better advance the most reasonable principles of socioeconomic justice. I do not think there is a way to settle on the best form of constitutional democracy apart from using these standards of justice. So I fight to change the rules of the democratic process to move the results in a direction that I think is less likely to create serious injustices. I understand that the Party of Unjust Inequality will resist the changes. I understand that they will say that my efforts are animated by class interest, that they are acting in service of democracy while my efforts are distorting democracy. And I see that their views of justice and democracy are reasonable, if barely so. But I also think that maybe they are being self-serving, and I am not prepared to be – nor am I required to be – a sucker. And I am concerned that the kinds of inequalities that their views permit are precisely the kinds of inequalities that yield AcemogluRobinson temptations to push back against democracy because as the inequalities grow, democracy, even as they understand it, is too costly.

Reflections on Democracy’s Fragility

331

The “deeply conflicting political and economic interests” permitted by the dominant conception of justice strike me as a threat to democracy itself. The situation is symmetrical. The Party of Unjust Inequality thinks of me as belonging to the party of Unjust Envy. Drawing on their experience, they think I underappreciate the burdens of the kinds of restrictions and regulations I favor. They see that the results of a shift to a proportional system will work to their disadvantage, so they resist it. They reject the claim that proportionality has democratic advantages – because the conception of democracy they regard as most appropriate is (like mine) guided by their conception of justice. I worry that their resistance is a precursor to pushing back against any kind of democracy, because they see the costs to them as too great. They think we may be acting selfservingly and insist that my group is insufficiently respectful of democratic process and are concerned that we are aiming to manipulate it to our advantage, as I say of them. Our conflict about standards of socioeconomic fairness has now produced a conflict about the right way to ensure a democratic process. Neither side in the conflict thinks of themselves as indifferent to democracy. Indeed, they think of their own preferred form of democracy as the right variant of democratic institutions. So we have a conflict – perhaps of considerable width – about principles, policy, democratic process, and interests. And because the disagreements are correlated, each side has some grounds for suspicion about the democratic commitments of their opponents and some reasons for concern about political cooperation. They understand the case for the reasonableness of the commitments, but they may each regard the opposing commitments as just barely reasonable: and because of the connections between reasonableness and reciprocity, each wonders about the other’s commitment to reciprocity. And these concerns can only be fueled by the correlation of differences of experience and interest with differences of principles of justice and conceptions of democracy. Suppose I am in the Party of Unjust Inequality. I observe my opponents working hard for a proportional system. Because their convictions about proportionality have grown stronger in the wake of their troubles winning in a nonproportional system, I think they care sufficiently much about something other than democracy that they are prepared to promote a form of democracy not really because it is more just but simply because it better advances their interests. I am concerned about the conditionality of their democratic commitments. I do not think that anyone is unconditionally committed to democracy, and I worry that I am accepting the constraints of democracy as I understand those

332

Joshua Cohen

constraints and paying a price for that commitment. My opponents, whatever their public rhetoric, are treating democratic process as something to be bent to other values. And if that is what they are doing, then the strain on my democratic commitments becomes too great. Preserving democracy is costing me a great deal relative to what I think justice entitles me to, given the conception of justice that I am drawn to. And I am not confident that the other side is committed to democracy. Once more, the disagreements fuel distrust and greater caution. Moreover, the distrust and caution foster correspondingly difficulties in basic democratic functioning – like passing legislation that addresses large public problems. Here again, when the conflict about the proper form of democracy is so closely associated with a conflict about which substantive policy is the best, each side may understandably lose confidence in the democratic convictions of the other side. And they may then think it is permissible to break the democratic rules both because of their substantive commitments, because they are concerned that the other side is breaking the rules, and because they are concerned that the political logjams resulting from the disagreement and mistrust are debilitating to effective government. 18.5

Conclusion

I have described how reasonable people might develop a set of disagreements, a form of ideological polarization about basic issues of justice and an associated mistrust and democratic dysfunction that strains democracy. In conclusion, I want to make three points: about the practical relevance of the argument, about the plausibility of the theory, and about how we might constructively address the challenges I have described. My concern about practical relevance: I do think that lots of people – and I am thinking partly here of people I broadly agree with on substance – are too quick to think that other people are not only wrong, but unreasonable, and unreasonable because they are indifferent to a concern about justification to others. That said, there is lots of unreasonableness around. My concern about practical relevance is that it may be a distraction from the troubles for democracy that arise from unreasonableness to focus on the sources of fragility on favorable assumptions that have occupied me here. Perhaps these are theoretical exercises of no practical interest. Maybe. But I am not so sure. The political world is complicated and, in practice, different sources of concern are always blended together. In

Reflections on Democracy’s Fragility

333

practice, it is difficult to distinguish reasonable from unreasonable persons and views, not least because people with apparently reasonable views may enter into political coalitions with people who have, to all appearances, unreasonable views. What I have been assuming, for the sake of argument, is that we can analytically separate out different strands that are in practice intertwined – thus, assume no unreasonableness – and that doing so might help in understanding and perhaps responding to our circumstances. My concern about theoretical plausibility: maybe democratic success, as social-psychologically inflected views of politics and democracy suggest, is less about agreement on justice and more about cross-cutting social cleavages that dampen social conflict. Correspondingly, maybe democratic fragility is more about mutually reinforcing cleavages than about disagreement on standards of justice. Democracy has troubles now, but – if much democratic theory and much study of American politics is right – Rawls does not illuminate them, because he has always tied democratic success to an implausibly elevated picture of public reasoning. I do not propose to address this disagreement here. The socialpsychologically inflected views are important and worth exploring. The same is true for normatively elevated views of public reasoning of the kind that helped to define Rawls’s orientation to democracy. A fuller understanding will, I think, come from exploring them in tandem. Finally, a constructive gesture: What kind of Rawlsian resources are available for addressing democracy’s fragility? The response must be founded on the idea of reciprocity, which belongs both to the world of justification and to the world of motivation. In the world of justification, Rawls emphasizes that reasonable political conceptions must satisfy the criterion of reciprocity. They must present terms of cooperation that adherents think others can reasonably accept as free and equal citizens, with moral powers and a concern with their own good. In the world of motivation, reciprocity is essential in developing the political motivations – including trust and civility – that preserve a legitimate, democratic society in the face of disagreements in conceptions of the good, comprehensive doctrines, and conceptions of justice. Reciprocity links the world of norm and the world of fact. In the normative order, “the theory of right and justice is founded on the notion of reciprocity which reconciles the points of view of the self and of others as equal moral persons” (TJ 1999, 424). In the world of fact, reciprocity is central to moral psychology, where “the basic idea is one of reciprocity, a tendency to answer in kind …. A capacity for a sense of justice built up by responses in kind would appear to be a condition of human sociability” (TJ 1999, 433).

334

Joshua Cohen

In explaining the role of reciprocity in motivation, Rawls emphasizes the importance of observing the evident intention of others to take your good into account. Thus a crucial step in the acquisition of a sense of justice is that people see others acting on principles of justice that arguably include their good (TJ 1999, 154–56). Because they see this in how others act and explain their conduct, they respond in kind, offering terms that others can similarly see as embracing their good. Moreover, the experience of cooperation of these terms fosters in turn a sense of trust and confidence in others. Under conditions of reasonable political pluralism, the story about reciprocity, trust, and confidence is more complicated, more strained. If people who have a different understanding of justice from mine treat me as their conception of justice requires, I am less likely to see the treatment as manifesting an intention to ensure my good as an equal moral person. The difficulty grows when I observe the correlation between their interests and their principles. And our situations are entirely symmetrical: they are also less likely to see me as manifesting an intention to embrace their good. Reciprocity in the formation of motivations – along with the associated trust and confidence – requires that I understand how their conception of justice, misguided by my lights, meets the criterion of reciprocity. I must come to see that it is not unreasonable for them to think that they are acting on and advancing a conception of justice that I could reasonably accept as suited to a society of equals. If I do see how their conception of justice meets the criterion of reciprocity – that in view of the burdens of judgment, it is a view that a reasonable person might advance as a basis for social cooperation among equal persons – then I may be prepared to “respond in kind,” not by accepting their views of justice but by treating them as a good faith, trustworthy partner in our public political life. In short, it is not sufficient that they have a reasonable view. I must understand it as reasonable and as genuinely constraining. How might I and others come to see this? The answer imposes great expectations on public reason and the associated duty of civility. The duty of civility is here understood not as a requirement of politeness but as a requirement of explaining political views and conduct by reference to public values and listening to the explanations offered by others (PL, 217). Citizens in a democracy must be prepared not only to explain their views on fundamental political questions by reference to public values. There also must be occasions for them to explain why they think their views meet the criterion of reciprocity: how they offer terms of cooperation that others can reasonably accept as free and equal citizens. Rawls emphasizes that ideas of fair cooperation and of moral persons as free and

Reflections on Democracy’s Fragility

335

equal are fundamental ideas in the public culture of democratic societies. I am suggesting now that an appreciation of reasonable disagreement on the fundamentals of justice – of the fact of reasonable political pluralism – must itself have a genuine practical grip on the public, political culture for reciprocity to work its effects. While social cooperation requires that people take justice very seriously, it also requires that they acquire through public discussion a tolerant and patient understanding of disagreements about justice. That is a great deal to expect. It sounds like turning politics into philosophy. Gramsci said that we are all philosophers. Let’s hope that he was right.

19

A Society of Self-Respect Leif Wenar

19.1

Americans and the Original Position

The year 2021 marked the fiftieth anniversary of the publication of Rawls’s A Theory of Justice. What parts of justice as fairness might spread beyond the academy into American public culture by the 100th anniversary, in 2071? This question is vital because justice as fairness cannot succeed in its own terms unless it is endorsed by ordinary citizens. In what Rawls calls “the third level of the full publicity condition,” all citizens of a wellordered society affirm the full justification of justice as fairness or at least its justification is accessible to all. Rawls’s publicity condition seems attractive for any country that aims to be democratic. Citizens who can affirm the reasons for their institutions can be autonomous political actors, and societies based on a shared understanding of basic laws need not be shifting battlegrounds of ideology. Rather, publicity allows citizens to interact on the basis of mutual respect and encourages genuine ties of community among them. Yet while Rawlsian publicity is appealing, it is also demanding. Rawls’s publicity condition seems to imply that for America to become wellordered, the justifications for its institutions will have to be accessible enough, say, for a President to invoke them in a speech from the Oval Office, for an Alabama high-school teacher to teach them in civics class, or for an Uber driver to explain them to foreign visitors should the conversation turn to politics. What in justice as fairness could become part of America’s public political culture like that? Let us find an example of what we’re looking for – an example of a political touchstone that meets this publicity condition today. We can think of how Americans might justify religious freedom to each other. They might say, “Americans have religious freedom because we respect that everybody needs to follow their own conscience.” That is a deep idea, an idea that a judicial decision could spell out at length. It is also an 336

A Society of Self-Respect

337

accessible idea that Americans today can offer to each other. What ideas in Rawls might become a political touchstone like that? We might think that the veil of ignorance has a chance. The veil of ignorance is a useful field test for the fairness of public policies: “Would you support tax cuts if you didn’t know your income?” and “Would you support equal pay if you didn’t know your gender?” and so on. It is also what most of our students remember from their classes on Rawls, and we might take it as a promising sign for its chances in the public culture that it has occasionally been mentioned in American mass media.1 What about the full original position (OP)? That is, what about the argument meant to prove that the basic structure should be ordered by Rawls’s two principles and not by utilitarianism or any other conception of justice? By the full OP argument, I mean what is represented in Figure 19.1.2

Figure 19.1 The OP (in JF)

I suspect that the full OP argument has no chance of becoming accessible to most Americans in the next fifty years, and so it cannot become the public justification for justice as fairness during this time. What especially gives me pause is that from 1971 to 2001, Rawls seems to have become increasingly concerned about the accessibility of the OP and so, I believe, he progressively weakened his statements on what publicity demands. 1 2

For example, The West Wing, season 4, episode 17; David Wolpe, “Here’s a Better Strategy for Picking a President,” Time, June 8, 2016. The figure summarizes the OP as described in JF.

338

Leif Wenar

In A Theory of Justice, Rawls says that the OP should be built from “widely accepted but weak premises … each of the presumptions should by itself be natural and plausible” (TJ 1999, 16, 12, 214). He emphasizes “the minimal nature of the conditions” defining the OP, partly because of the importance of publicity (TJ 1999, 510). Thus, he says, “what is important is that the various features of the OP should be expressed in the simplest and most compelling way …. The crucial thing is not to use principles that are contested …. The idea of the initial agreement can only succeed if its conditions are in fact widely recognized, or can become so” (TJ 1999, 512–13). In 1971, Rawls was satisfied that the OP meets these conditions: it is “reasonably simple,” he says, and is fit for a society whose “members have a lucid grasp of the public conception of justice upon which their relations are founded” (TJ 1999, 234; 501). In the mid-1990s, Rawls renewed his emphasis on the importance of publicity, saying, for example, that “the knowledge and ways of reasoning that ground our affirming the principles of justice … are to rest on the plain truth now widely accepted, or available, to citizens generally” (PL, 225, emphasis added). However, the qualification “or available” appears to signal Rawls’s growing unease with the requirement of widespread acceptance of his arguments. This is reflected in his description of the third level of the full publicity condition in Political Liberalism: At this level I suppose this full justification also to be publicly known, or better, at least to be publicly available. This weaker condition (that full justification be available) allows for the possibility that some will not want to carry philosophical reflection about political life so far, and certainly no one is required to. (PL, 67, emphases added)

The idea of a society that is fully public, Rawls says, “may seem much too strong.” Yet “It is adopted … because it is appropriate for a political conception of justice for reasonable and rational citizens who are free and equal” (PL, 67). My sense that Rawls was having doubts about the OP increases in The Law of Peoples, published in 1999. Here, the set-up of the (international) OP is barely sketched in. More, Rawls does not even allow the parties in this OP to perform their main function, which is to show the reasoning that favors certain principles over others. Rather, Rawls just announces that the principles he favors “are superior to any others” and sets the parties the task of reflecting on their advantages without considering alternatives (LP, 41; 57, 69). This OP is summarized in Figure 19.2. By Justice as Fairness in 2001, Rawls’s hope for a (domestic) OP based on uncontested premises appears to have gone completely. He no longer characterizes even the crucial “basic liberties” argument for his two

A Society of Self-Respect

339

Figure 19.2 The OP (in Law of Peoples)

principles as resting on uncontroversial ideas. “Many points,” he says about this argument, “are highly controversial”: among them the assumptions made about probability and the basis of the aversion to uncertainty; the assertion that in the circumstances of justice, even under reasonably favorable conditions, there are situations in which the principle of utility requires at least the restriction if not the suppression of basic rights and liberties: and finally, the idea that some things are not negotiable. (JF, 110)

From “widely accepted but weak premises” in 1971 to “highly controversial points” in 2001 is a major change in how Rawls describes the OP. How Rawls squared what he came to see as the controversial nature of the OP with his desire for publicity I do not know. By the time he renders the publicity condition in Justice as Fairness, he is even more pessimistic than he was in Political Liberalism that most citizens of a well-ordered society will know the justification for justice as fairness: “Of course,” he says, “that [citizens] will carry reflection so far is unlikely; still, the full justification is available in the public culture for them to consider if they wish” (JF, 121, emphasis added). In a poignant footnote, Rawls adds, “here I entertain the fantasy that works like this restatement are known in the public culture” (JF, 121). The “fantasy” in this footnote seems to express more than Rawls’s famous personal modesty. I believe that Rawls came to see serious tensions between the OP and the hope that the justification of justice as fairness will be accessible to most citizens in a meaningful way. I believe that he died with that tension unresolved.

340

Leif Wenar

What might Rawls’s worries have been about using the OP as the public justification for his principles? He might have had at least two concerns. The first is that the OP is cognitively inaccessible to most citizens in a country like the United States, now and in the foreseeable future. The arguments are simply too difficult for most people to understand. Rawls might have noticed that even Americans with advanced degrees in philosophy sometimes seem not to grasp how the OP arguments are meant to work. If some of the most intellectually gifted Americans can get this wrong even after specialist training, he might have thought, how much can we expect of other citizens? Is the OP comprehensible by only a few thousand, out of hundreds of millions? Looking back at Figure 19.1, we might wonder whether we could expect this argument to be cognitively accessible to most Americans. As Rawls says (TJ 1999, 123), “we … recognize an intricate theoretical construction when we meet one,” and this seems to be such. More, even Figure 19.1 does not capture many of the subtleties of the OP argument. Recall, for example, that in the argument for the superiority of Rawls’s two principles over utilitarianism, much hangs on maximin being the uniquely best decision rule in circumstances of uncertainty (which many experts doubt). Even more importantly, the whole argument turns on the initial situation being properly defined as a situation of uncertainty, and not as a Harsanyian situation of risk, so that maximin (instead of equiprobability) is relevant as a decision rule. What would we need to believe about most Americans, Rawls might have wondered, for these kinds of abstruse considerations to be cognitively accessible to them? At points in his work, Rawls reveals a highly intellectualized image of citizens of a well-ordered society, which does not describe Americans today.3 Two-thirds of today’s Americans, for example, never get a fouryear college degree. (Only 2 percent get a doctorate.) And even if Rawls were picturing an idealized United States with universal high-quality education, he might have come to reflect on the near-tautology that, after all, many Americans will be of average intelligence, and many will be below. Even if Rawls’s books became “known in the public culture” in some sense, how many citizens of this idealized America could read them with good comprehension if they wanted to? Perhaps Rawls came to see that the very subtlety and sweep that made his work so celebrated in the 3

For example, in a well-ordered society each citizen will achieve wide reflective equilibrium, having “considered the leading conceptions of political justice found in our philosophical tradition … and has weighed the force of the different philosophical and other reasons for them” (JF, 31).

A Society of Self-Respect

341

academy also put it beyond the ken of most Americans. He might have come to feel trapped in a certain paradox of democratic intellectuals: once they have done work sophisticated enough to interest each other, they find that their work is too sophisticated to be understood by most of their fellow citizens. Although I think there might be something to this first concern, I also believe it likely that Rawls came to have a second worry: not that the OP was cognitively inaccessible but that it was culturally inaccessible. Complex arguments can be simplified, after all, main points can be distilled, big ideas can be publicized by journalists, political commentators, TV shows, and so on. As I noted above, I believe that there is hope that the veil of ignorance might become something like a field test for the fairness of specific public policies in this way. But the OP? The concern that Rawls may have come to have is not that his fellow citizens could not understand the OP arguments but that they would not care about them. I have tried to distill an OP argument to its simplest – perhaps you can do better. Here is my best attempt at an accessible OP argument for the basic liberties (BL): BL: If we Americans were in a totally fair situation, where none of us knew anything distinctive about ourselves or about our form of government, we’d all agree to equal rights instead of to raising the average condition of people as high as possible.

Could something like BL become a widely accepted public justification in America, say by 2071? This seems to me to be beyond comprehension. It is hard for me to imagine an Alabama civics teacher explaining BL to his students, or a presidential candidate extolling it on the stump, or an Uber driver explaining it to foreign visitors. The form of reasoning in the antecedent of BL is too disembodied, too disconnected from daily concerns. If you would like to reflect on this question, perhaps ask whether we find this form of reasoning anywhere in American mass politics, or in the mass politics of any democracy that we know. In order to be culturally accessible, a form of reasoning needs enough to latch onto that is already in the public culture, and I do not see enough in American culture that BL could latch onto. I cannot see how most Americans could develop a strong conviction that BL defines how citizens should think about politics at the deepest levels – or not in the next fifty years, at least. Rawls was unusually firm in insisting that in a democracy, the audience for political philosophy must be most citizens, or more precisely, all voters (LHPP, 1, 6). He labels “completely mistaken” the view in which political philosophers find the truth about justice and then seek “a

342

Leif Wenar

political agent to realize that truth in institutions, irrespective of whether that truth is freely accepted, or even understood” (LHPP, 2, 3). Rawls held strongly that most citizens should be able to accept the reasons for their society’s institutions, meaning that these reasons should be cognitively and culturally accessible to them. And yet, it appears, the famous argument that he offers for his own principles of justice cannot satisfy this democratic demand. I think we may have to conclude with regret that the OP is a hothouse flower, which has grown luxuriously within the academy but that is unlikely to take root in the soil of American democratic culture. If this is right, it would not be the end of justice as fairness. After all, Rawls’s publicity condition only requires that the justification for justice as fairness be public, not that it be the OP. And I believe that there are ideas in Rawls’s work that might spread widely and grow deeply in America’s public culture, which our attention to the OP has left underdeveloped. Rawls’s work contains a model of social relations that presidential candidates and civics teachers and Uber drivers might offer as a political justification of a future well-ordered American society. The rest of this chapter will set out this model of social relations, which I will call a society of self-respect. Recentering justice as fairness on self-respect will be work of sympathetic reconstruction, filling the justificatory hole left by the OP with elements that Rawls used for other purposes. The main reason to explore this reconstructed justification of justice as fairness is to ask whether it might be attractive to Americans as they are or might become, and so whether it might offer a public basis for a shared political life.

19.2

The Significance of Distributions

As we begin to look for a public justification for Rawls’s principles of justice, let me first note that Rawls did not see distributions of social goods as important only in how they map onto citizens’ characters and choices but also in how they impact them. Unlike some theorists today, Rawls was always looking through distributions to their effects: A theory of justice must take into account how the aims and aspirations of people are formed …. The institutional form of society affects its members and determines in large part the kind of persons they want to be as well as the kind of persons they are …. So an economic regime, say, is not only an institutional scheme for satisfying existing desires and aspirations but a way of fashioning desires and aspirations in the future. (PL, 269; TJ 1999, 229; JF, 56)

The significance of any distribution, Rawls holds, includes how it affects citizens’ self-conceptions and their conscious relations to each other. This

A Society of Self-Respect

343

line of thought places Rawls in what Appiah (2010, 155–212) calls the “soul-making” tradition of political thought, a tradition that includes Plato, Aristotle, Montesquieu, Burke, Rousseau, Wollstonecraft, and Marx. Rawls’s concern with the effects of distributions is one reason that publicity is so important to him. “Publicity ensures, so far as practical measures allow, that citizens are in a position to know and to accept the pervasive influences of the basic structure that shape their conceptions of themselves, their character and ends” (PL, 68). The justification of a distribution must be in this way reflexive: citizens must be able to affirm the institutions that determine the kind of persons they want to be as well as the kind of persons they are. Rawls’s view here differs markedly from what we might call the intrinsic view of justice, which is that a society can be judged to be just without considering what effects the distribution of social goods has on citizens or whether citizens even know the distribution exists. To take a well-known example, G. A. Cohen (2008, 323–27, 344–71) argues that publicity is no part of what is required for a society to be just. Justice simply requires that a particular (luck-egalitarian) pattern obtain. A society can be just even when no one knows that it is, or when no one could know that it is, or when no one cares that it is, or even when most citizens are hostile to justice so conceived. Indeed, justice could be realized even when most citizens are leading anxious, petulant lives of haughty incivility, and when the desired pattern works to reproduce and reinforce those tendencies. On this intrinsic view, a just distribution need have no particular effects, or positive effects, on people’s lives. For Rawls, justice is quite otherwise. The justice of a distribution depends, at least in part, on its nondistributive effects. Above all, Rawls was concerned about how distributions affect citizens’ self-respect. “We may define self-respect,” he says, as having two aspects. First of all … it includes a person’s sense of his own value, his secure conviction that his conception of the good, his plan of life, is worth carrying out. And second, self-respect implies a confidence in one’s ability, so far as it is within one’s power, to fulfill one’s intentions. When we feel that our plans are of little value, we cannot pursue them with pleasure or take delight in their execution. Nor plagued by failure and self-doubt can we continue in our endeavors …. Without [self-respect] nothing may seem worth doing, or if some things have value for us, we lack the will to strive for them. All desire and activity become empty and vain, and we sink into apathy and cynicism. (TJ 1999, 386; PL, 318–19)

Rawls says repeatedly in Theory that self-respect is “perhaps the most important primary good” – and one time he drops the “perhaps” (TJ

344

Leif Wenar

1999, 54, 155–56, 348, 380, 386, 477; 468). If self-respect is the most important primary good in justice as fairness, then it is more important in this theory than are the liberty and integrity of the person, freedom of thought and association, the rule of law, the political liberties, and of course more important than income and wealth (PL, 291). The justification of justice as fairness that I develop will take self-respect as seriously as that. On the model of social relations set out below, the central justification of Rawls’s principles of justice is that they structure a society where everyone can have self-respect. I will describe a Rawlsian society of self-respect after motivating it by describing a contemporary political pathology that Rawls diagnoses. We scholars have perhaps overstudied the prescriptions that Rawls wrote, while neglecting the ills that he meant to treat. Rawls was a social theorist in the tradition of Rousseau, Marx, and Nietzsche. Applying his diagnosis of a political disorder to contemporary politics will show why a society of self-respect may have special relevance for America today.

19.3

Rawls’s Critique of Meritocracy

In Sections 19.3 and 19.4, I will first describe and then apply Rawls’s critique of a leading justification of America’s distributive system, which is that it is a meritocracy. The American meritocratic model says that if one has talent and works hard, one will win the competition for money and political power. As we know, America is very far from being a meritocracy, but let us leave that aside for now. Even if America became a perfect meritocracy, Rawls would have grave concerns. For he thought that meritocracy can lead to a “politics of resentment” and then to what he called “hostile outbreaks of envy.” Rawls also thought that these politics of resentment can lead to a destructive racial nationalism that might endanger constitutional government itself. Rawls’s analysis of these instabilities urges us to move beyond meritocracy, for the sake both of justice and prudence, toward a society of self-respect, which will be set out in Sections 19.5–19.9. In a meritocracy, Rawls says, There exists a marked disparity between the upper and lower classes in both the means of life and the rights and privileges of organizational authority. The culture of the poorer strata is impoverished while that of the governing technocratic elite is securely based on the service of the national ends of power and wealth. Equality of opportunity means an equal chance to leave the less fortunate behind in the personal quest for influence and social position. (TJ 1999, 91)

Worse, the American meritocracy is a status competition for money and power – and for money and power as positional goods, where one’s

A Society of Self-Respect

345

level of public esteem turns on whether one has more than others. A status competition for such positional goods is a “great misfortune,” Rawls says, in this very Rousseauian passage: Everyone cannot have the highest status, and to improve one person’s position is to lower that of someone else. Social cooperation to increase the conditions of self-respect is impossible. The means of status, so to speak, are fixed, and each man’s gain is another’s loss …. Persons are set at odds with one another in the pursuit of their self-esteem. (TJ 1999, 478)

In this zero-sum status-competitive society, the upper classes enjoy a rich cultural life that they believe their merit has won for them, and they provide their children with vastly superior education and opportunities that ensure as much as possible that their children will win the next round of the status competition. Meanwhile, the lower classes are trapped in an impoverished culture, with limited education and opportunities, and with little hope that they or their children can better their situation. In the national competition for the most socially desired goods, the worse-off must regard themselves – and know that others regard them – as “losers” and as losers because they lack “merit.” More, if good, stable jobs become scarce, as in America today, a meritocracy becomes even worse: Lacking a sense of long-term security and the opportunity for meaningful work and occupation is not only destructive of citizens’ self-respect but of their sense that they are members of society and not simply caught in it. This leads to selfhatred, bitterness, and resentment. (PL, LVII)

Finally, in a society where those with greater private means are able to control the course of public affairs, “political power rapidly accumulates and becomes unequal; and making use of the coercive apparatus of the state and its law, those who gain the advantage can often assure themselves of a favored position” (TJ 1999, 199). In such a society, lessfavored citizens see no way out through politics: “Having been effectively prevented by their lack of means from exercising their fair degree of influence, [they] withdraw into apathy and resentment” (TJ 1999, 198).

19.4

A Politics of Resentment and Hostile Outbreaks of Envy

Thus Rawls fears that meritocracy leads to a politics of resentment: resentment in lower classes who cannot see the system as built for them, and who struggle to find self-respect in their constitutional identities as citizens. In this section, I will go beyond Rawls’s texts and apply this analysis of the politics of resentment to America’s white working class

346

Leif Wenar

since 2016 (Williams 2020). This analysis only aims to capture one aspect of the complex reality of recent American politics. Let me especially emphasize that this analysis is not meant to apply to communities of people of color, where today’s politics have entirely different dynamics. Rawls’s analysis of the politics of resentment is here used to describe only one causal vector in recent American politics, though an important one. Rawls thinks that a politics of resentment may lead to political instability. This may seem surprising. After all, meritocratic status hierarchies like America’s can be stable for long periods, not least because the lower classes’ perception of their own political impotence helps the hierarchy to continue undisturbed. However, conditions in any society can get so dire for the lower classes that they no longer believe that their fellow citizens value them at all. Citizens’ “self-respect and their confidence in the value of their own system of ends cannot withstand the indifference much less the contempt of others” (TJ 1999, 297). When social inequalities become exceptionally glaring, the assault on the self-respect of the lower classes may rouse them to shake off their political impotence; then, their resentment erupts into “hostile outbreaks of envy”: Three conditions … encourage hostile outbreaks of envy. [First,] persons lack a sure confidence in their own value and in their ability to do anything worthwhile. Second … many occasions arise when this psychological condition is experienced as painful and humiliating. The discrepancy between oneself and others is made visible by the social structure and style of life of one’s society. The less fortunate are therefore often forcefully reminded of their situation, sometimes leading them to an even lower estimation of themselves and their mode of living. And third, they see their social position as allowing no constructive alternative to opposing the favored circumstances of the more advantaged. To alleviate their feelings of anguish and inferiority, they believe they have no choice but to impose a loss on those better placed even at some cost to themselves, unless of course they are to relapse into resignation and apathy. (TJ 1999, 469)

After decades of rising inequality in the United States, Rawls would likely have seen hostile outbreaks of envy in recent American politics. Scheffler (2019), for example, gives a Rawlsian analysis of the 2016 presidential election, where, as it is said, the white working class voted with their middle fingers. The effort of the less advantaged to bring down the more advantaged, even at some cost to themselves, can be disastrous for the polity. In this struggle, basic civility may break down – as Rawls writes about such periods, “much political debate betrays the marks of warfare. It consists of rallying the troops and intimidating the other side, which must now increase its efforts or back down. In all this one may find the

A Society of Self-Respect

347

thought that to have character is to have firm convictions and be ready to proclaim them defiantly to others. To be is to confront” (JF, 118). The mention of the white working class brings a further quite dangerous potential to the politics of resentment. Those who cannot find selfrespect in their constitutional identities may turn ever more to race and nation, to feel self-worth in superiority over other racial groups and foreigners. A resentful rise of racial nationalism is what Rawls saw in Weimar Germany (Weithman 2016). Ordinary Germans during the Weimar period, having suffered the humiliating end of World War I and the economic hardships of the 1920s, found themselves ruled by an indifferent, status-obsessed elite. So they turned to their national and racial identities as the main sources of their self-worth. We can hear Rawls speaking of Germany when he writes of a people “inflamed by what Rousseau diagnosed as arrogant or wounded pride or by lack of due self-respect” (LP, 47). German political passions were then captured by a demagogic nativist who dismantled the country’s remaining liberal institutions – and who then launched a racialnationalist war that destabilized the entire international order. The analysis of the politics of resentment centers on a social class that has little hope in succeeding in a decisive status competition for money and political power. The members of this class feel dominated by political elites, so when given an opportunity, they may support political actions that make themselves worse off, so long as they can also bring down those elites who are the objects of their intense resentment. Since they can gain so little self-respect from their constitutional identities as citizens, they may turn ever more to race and nation as sources of their self-worth. If this resentful class is powerful enough, it can seriously damage whatever liberal institutions exist, and if it becomes very powerful it can even destroy the constitution altogether. The politics of resentment in a meritocracy like the United States are driven by zero-sum status competitions that many millions must lose in ways obvious to all. These losers, trapped in an impoverished culture, deprived of political influence, lacking opportunities for advancement and often opportunities for meaningful work, struggle to see how their society affirms their value, and therefore may lose their allegiance to the system as a whole. Let me emphasize this point by noting that some today seem to be hoping that the white working class will eventually be overwhelmed by America’s demographics and so will no longer be able to cause so much trouble. Yet if Rawls is right that the politics of resentment are built into every meritocratic system, America may suffer these pathologies as long as it remains a meritocracy, the resentment festering in whatever social groups lose out in the status competition for positional goods and so for self-respect.

348

Leif Wenar

19.5

A Society of Self-Respect

Let us return to Rawls’s publicity condition. What ideas in justice as fairness could become a touchstone for political justification in America’s public culture, say by 2071? We are looking for an organizing rationale for the two principles of justice that politicians and civics teachers and many other Americans could use in explaining their society to each other. Self-respect is such a touchstone. In a future America well-ordered by Rawls’s two principles, Americans could say to each other: “Our country works for everyone. Our laws affirm the value of each citizen regardless of their gender, race, class, religion, or inborn abilities. In this country, everyone can have self-respect.” Justice as fairness can be reframed as an accessible and attractive alternative to meritocracy, which can keep the politics of resentment from getting started. We can call this model of social relations “a society of self-respect.” Sections 19.6–19.9 set out this model and the conclusion offers it as a model of social unity. Just as Rawls’s diagnosis of the politics of resentment is essentially Rousseauian, so his solution to the problem of resentment is also Rousseauian (LHPP, 218–35, 244–48). For both Rawls and Rousseau, the solution to status competition over positional goods is to secure the self-worth of all through equality – especially equality “at the highest level” in the fundamental role of citizenship (LHPP, 247–48). Equality is the only relation that can tame enflamed desires for ever-more positional goods, and as Rawls says, “in a well-ordered society … self-respect is secured by the public affirmation of the status of equal citizenship for all” (TJ 1999, 478). Recall the bases of self-respect in the meritocratic society that Rawls’s well-ordered society is meant to replace. In a meritocracy, the upper classes gain self-respect from having left the less fortunate behind in their personal quests for influence and social position and enjoy the political power and wealth that they display as markers of their success. The lower classes know that others see them as losers in the national merit-based competition for power and money, and are consigned to an impoverished culture that challenges their attempts to find respect in their daily lives. Rawls’s strategy for avoiding the pathologies of meritocracy has three stages. First, the well-ordered society places strong public emphasis on political rights and liberties, so that equal citizenship can be a foundation of self-respect for all. Second, the basic structure distributes socioeconomic goods in ways meant to enrich the life of each citizen and especially the culture of the worse-off. Third, inequalities in wealth and income are publicly justified by how they support the self-respect of those who have the least. Thus a well-ordered society will support the self-

A Society of Self-Respect

349

respect of citizens as they live their daily lives: at their work, in their worship, in their communities and associations, and during their participation in public affairs. We can enumerate the principles of justice as fairness to show how they are constructed to strengthen self-respect, drawing on some less-commonly cited parts of Rawls’s texts. Rawls’s first principle bolsters every citizen’s self-respect by securing the equal basic rights and liberties. The best solution to a status competition for positional goods, Rawls says, is “to support the primary good of self-respect as far as possible by the assignment of the basic liberties that can indeed be made equal, defining the same status for all” (TJ 1999, 478). In a just society, citizens are not “disposed to acknowledge a less than equal liberty,” which would “have the effect of publicly establishing their inferiority as defined by the basic structure of society” (TJ 1999, 477). The fair value of the political liberties adds a dimension of substantive equality to the formal equality of the other basic liberties. Rawls believed that the opportunity for active and consequential participation in public affairs is critical for the self-respect of citizens in a democratic society, which is one reason why he was unusually outspoken in his calls for campaign finance reform in the United States (TJ 1999, 198; PL, 449). In a well-ordered society, equal political liberty when assured its fair value is bound to have a profound effect on the moral quality of civic life …. The public will to consult and to take everyone’s beliefs and interests into account lays the foundation for civic friendship and shapes the ethos of political culture … [and enhances] the selfesteem and the sense of political competence of the average citizen. His awareness of his own worth … is confirmed in the constitution of the whole society. (TJ 1999, 205)

Rawls’s emphasis on the equal liberties and the fair value of the political liberties is essential to his strategy of deemphasizing socioeconomic goods as public bases for self-respect. Thus the priority of liberty is critical: When it is the position of equal citizenship that answers to the need for status, the precedence of the equal liberties becomes all the more necessary. Having chosen a conception of justice that seeks to eliminate the significance of relative economic and social advantages as supports for men’s self-confidence, it is essential that the priority of liberty be firmly maintained. (TJ 1999, 478)

To “eliminate the significance of relative economic and social advantages as supports for men’s self-confidence” is an immense challenge, relative to current meritocratic public norms. The second stage of Rawls’s strategy goes some way to meeting this challenge, by arranging

350

Leif Wenar

socioeconomic institutions so that they enrich the lives of all and especially the culture of the worse-off. Fair equality of opportunity requires investment in education and training that allows all “to develop their native endowments and to acquire socially productive skills” (JF, 67). Fair equality also requires constraints on economic inequalities and limitations of inheritance and bequest, so as to “put in the hands of citizens generally, and not only of a few, sufficient productive means for them to be fully cooperating members of society on a footing of equality … [including] knowledge and an understanding of institutions, educated abilities, and trained skills” (JF, 140). Regardless of their class background, then, all will have an equal chance for work that suits their developed talents – the work that is most likely to be inherently satisfying (TJ 1999, 372–80). When fair opportunity is satisfied, Rawls says that the rewards of work are not only “certain external rewards of office” such as wealth and privilege. Rather, meaningful work increases citizens’ self-respect through “the realization of the self which comes from a skillful and devoted exercise of social duties … one of the main forms of human good” (TJ 1999, 73). The difference principle helps to replace the impoverished culture of the less-advantaged in a meritocracy with one that enriches the personal and social lives of these citizens, for example, through education: The difference principle transforms the aims of society in fundamental respects …. The confident sense of their own worth should be sought for the least favored and this limits the forms of hierarchy and the degrees of inequality that justice permits. Thus, for example, resources for education are not to be allotted solely or necessarily mainly according to their [economic] return … but also according to their worth in enriching the personal and social life of citizens, including here the less favored. (TJ 1999, 91–92)

In a society of self-respect, the equal value of each citizen is proclaimed and prioritized by the protection of their equal basic rights and liberties, and by the securing of their equal voice in matters of public concern. Highquality education and training enable all to develop their natural talents and skills, yielding the daily satisfactions of meaningful work and enriching the personal and cultural lives of citizens in all segments of society. Each citizen is also (and this is a new element) free to follow her particular interests to find associations where her gifts and abilities will be publicly affirmed by the other members (TJ 1999, 388, 470–71). So long as every citizen finds at least one group, club, or community where her participation is valued, everyone will enjoy associational esteem (TJ 1999, 470). Rawls’s vision is that in the well-ordered society, “members take little interest in their relative position as such …. They are not much affected

A Society of Self-Respect

351

by envy and jealousy, and for the most part they do what seems best to them as judged by their own plan of life, and those of their associates, without being dismayed by the greater amenities and enjoyments of others socially more distant” (TJ 1999, 477). Still, citizens will not be oblivious to the national distribution of wealth and income or to their place in it. The third stage of Rawls’s strategy for avoiding the pathologies of meritocracy is to transform the economic distribution from a battleground of condescension and resentment to a meeting ground of mutual recognition. “By arranging inequalities for reciprocal advantage and by abstaining from the exploitation of the contingencies of nature and social circumstance,” he says, “persons express their respect for one another in the very constitution of their society. In this way they ensure their self-respect” (TJ 1999, 156). This is the public role of the difference principle. At a deeper level, the difference principle defines strong bonds of social unity. While a meritocracy uses talent and effort to divide winners from losers, in a society structured by the difference principle citizens “agree to share one another’s fate. In designing institutions they undertake to avail themselves of the accidents of nature and social circumstance only when doing so is for the common benefit” (TJ 1971, 102). Living in a society organized by the difference principle would transform how the less fortunate would be viewed by the more fortunate: The least advantaged are not, if all goes well, the unfortunate and unlucky – objects of our charity and compassion, much less our pity – but those to whom reciprocity is owed as a matter of political justice among those who are free and equal citizens along with everyone else. Although they control fewer resources, they are doing their full share on terms recognized by all as mutually advantageous and consistent with everyone’s self-respect. (JF, 139)

Living in this society would also transform how the less fortunate would view themselves. With the difference principle, less-advantaged citizens believe that the economy is also built for them, with the equal value of their lives in mind. While they know that their particular skills are in less demand in the economy right now, they also know that they benefit as much as possible from the efforts of those whose skills are in more demand. As Rawls says, in a well-ordered society, the greater advantages of some are in return for compensating benefits for the less favored; and no one supposes that those who have a larger share are more deserving from a moral point of view …. Regardless of the excellences that persons or associations display, their claims to social resources are always adjudicated by principles of mutual justice. For all these reasons the less fortunate have no cause to consider themselves inferior and the public principles generally accepted underwrite their self-assurance. (TJ 1999, 470).

352

Leif Wenar

With every part of his two principles, Rawls built a model of a just society that affirms the self-respect of each citizen. “The basis for selfrespect in a just society is not then one’s income share but the publicly affirmed distribution of fundamental rights and liberties …. No one is inclined to look beyond the constitutional affirmation of equality for further political ways of securing his status” (TJ 1999, 477). This reciprocal affirmation of the worth of all citizens will lead to greater fellowfeeling among citizens: A more unconditional caring for our good and a clearer refusal by others to take advantage of accidents and happenstance, must strengthen our self-esteem; and this greater good must in turn lead to a closer affiliation with persons and institutions by way of an answer in kind. (TJ 1999, 437; JF, 117)

Finally, the reciprocal affirmation of the worth of all citizens helps to win the allegiance of citizens to the system as a whole: “persons tend to love, cherish, and support whatever affirms their own good. Since everyone’s good is affirmed, all acquire inclinations to uphold the scheme” (TJ 1999, 154–55) .

19.6

The Uber Driver’s Speech to the Foreign Visitors

“… which is why we’re so proud of this country.” “Why’s that?” “You see, in this country, everyone can have self-respect. We’re all free, we’re all equal, and the system’s fair to everybody.” “Well, sure but … what does that even mean?” “It’s like what we learn in school. We’re all free to live our lives. Everybody’s got about an equal shot no matter where they start. And even if I end up with less money, I know the economy’s built for me.” “OK, but you know … what does that really come to, day-to-day?” “All right, we’re all free. Whatever god you believe in, whatever you want to say, whoever you want to spend time with and whatever you want to do with your life, we all respect that. That’s the main thing. “We’re all equal. Whether you’re born rich or poor, whatever your race or gender or religion, you’ve got an equal say in politics. And in the economy we make sure you’re trained up to find good work that suits you. “And the system’s fair. Some people have got more looks or brains or whatever. That’s great for them, but we’re all in this together. So the lucky ones can use their gifts to lift themselves up but only if that lifts everyone up. The economy makes poorer people as well-off as we know how to, because they’re just as good as everyone else.”

A Society of Self-Respect

19.7

353

Rawls’s Philosophical Anthropology

If America became a well-ordered society, the politics of resentment that characterize a meritocracy would not get started. All Americans could then see how their institutions are good for them and could say to each other: “Our county works for everyone –here everyone can have selfrespect.” A meritocrat might looks at Rawls’s society of self-respect and object that its economy will be less competitive and so may generate less growth. Rawls would agree with the premise and reply that the goal of endless growth is mere bourgeois ideology (LP, 107; JF, 159). To achieve a just and good society, Rawls says, great wealth is not necessary. In fact, beyond some point it is more likely to be a positive hindrance, a meaningless distraction at best if not a temptation to indulgence and emptiness. (TJ 1999, 257–58)

The conviction that a preoccupation with money damages both societies and individuals runs throughout Rawls’s work. It is part of Rawls’s reply to a second meritocratic objection, which is that the well-off might think that their economic prospects were being sacrificed for the sake of the worse-off. Here Rawls would first emphasize that the better-off in a well-off society will gain self-respect from their status as equal citizens, from their associational lives and the satisfactions of their work, and from their favorable position in the economic distribution (JF, 124). Yet he would add that the desire to gain self-respect by having more than others is itself pathological. Positional competitions for money and power are bad not only for the losers, they are also bad for the winners (CP, 277; BI, 193–206). What healthy people in modern societies really want, Rawls holds, is not more money or power. Rather, what men want is meaningful work in free association with others, these associations regulating their relations to one another within a framework of just basic institutions. (TJ 1999, 257)

What people in modern societies really want, Rawls thinks, is to enjoy the exercise of their developed capacities and to associate freely within a social order that treats them fairly as equal citizens. To do any of these things – to engage with life successfully – people need self-respect, which is why Rawls says that self-respect is so important to his theory. And selfrespect will be secured, as much as it can be, by the very institutions that enable people to pursue the lives they want. The distributions of primary goods described by the two principles are the social bases of self-respect:

354

Leif Wenar

these distributions create the conditions for all citizens to have selfrespect, as much as institutions can do so (JF, 60). In a liberal democracy, Rawls says, the best way to secure self-respect for all is through politics that publicly proclaim and secure the freedom and equality of all, within institutions that all can see are fair.

19.8

Public Philosophy

Everything to here has flowed from publicity. Most citizens should be able to accept the reasons for their society’s institutions, meaning that these reasons should be cognitively and culturally accessible to them. This seems such a natural aim for any democratic society, one that embodies weighty democratic values. Publicity tells against “government house” utilitarianism. It also presses against the use of an intrinsic view of justice where an intellectual elite agrees on the distribution that is just, then passes this information to the political elite so that this pattern can be imposed on most citizens – whether they know it or not, whether they like it or not. Meritocracy satisfies the publicity condition – that is not its problem. Its problem is that it fails to support the self-respect of so many citizens, which threatens instability. Rawls instead aims for stability “for the right reasons”: a stability of transparency (PL, 458–59; xl–xli, 389–92, 458–62). Rawls’s hope is that citizen can know the social order as it really is and can accept it without the need for indoctrination or false consciousness. Achieving this transparency means that the justification of the social order can be reflexive: all citizens can affirm the institutions that determine the kind of persons they want to be as well as the kind of persons they are (PL, 68–69). In a well-ordered society, every citizen understands how her society works – and how it works for her. The institutions of a well-ordered society will exert their soul-making powers, shaping the kinds of people that citizens want to be. A citizen raised in a well-ordered society will want to be fulfilled in her work, to be esteemed by her close associates, and to participate in a rich culture. She will want her most important lifechoices to be protected by the state, and her views to be respected in public affairs. She will want recognition of her economic contributions, wherever she ends up in the economic distribution. In a well-ordered society, all citizens will have these desires – and each will correctly believe that her society is designed to satisfy them. Each citizen will see her world clearly, and see how it is built for her, given who she understands herself to be. She will affirm her society freely, because her society affirms her.

A Society of Self-Respect

355

Rawls’s vision is a society in which everyone has self-respect because the social order embodies what each citizen thinks she is, what she thinks her fellow citizens are, and what she thinks her society should be. And how does Rawls know that he has found the right conceptions to ground this reflexive justification –that citizens should be free and equal and that society should be fair? Here Rawls will say that he has done his best to find the deepest bases of agreement within us – to find the ideals that best capture our understandings of ourselves, our world, and the relations we want with each other. The floor remains open for anyone who can offer a better model of social relations than his. The test of Rawls’s attempt at social interpretation is whether we, here and now, find his society of self-respect attractive. The test is whether we and most of our fellow citizens would, on reflection, be proud to live in the society described by the Uber driver’s speech. A well-ordered society is stable because everyone can – knowing all the facts – have self-respect. As Rawls puts it, “the most stable conception of justice, therefore, is presumably one that is perspicuous to our reason, congruent with our good, and rooted not in abnegation but in affirmation of the self” (TJ 1999, 436; PL, 317–18). 19.9

Self-Respect and Social Unity

Let me close with a thought you may have had already – that for America to become a society of self-respect, many Americans would need to have quite different attitudes toward each other than they do now. Many Americans would need a more robust sense of unity with one another: a greater respect for the equal value of all, and a greater willingness to share each other’s fate, regardless of everything that distinguishes them from each other. Many Americans may seem not to feel such a deep sense of social unity today but instead to accept meritocratic norms, or more rough-and-tumble ideals of self-reliance, or even hateful ideologies of difference. So it might seem unlikely, now, that many Americans could come to want to share each other’s fate at the deep level that Rawls describes. Americans have been in a fight recently, even a truce is hard to see. Yet recall the dangers of the politics of resentment that Rawls describes, where citizens who lose in competitions for public esteem launch destabilizing attacks on the social order, possibly tipping into a racial nationalism that undermines the constitution, perhaps even engendering the great evils of human history (LP, 6–7). If Rawls were here with us, looking out over America’s recent politics, I believe he would say that sharing each other’s fate is now the country’s best hope. If Rawls were here, I believe he would say that Americans have a choice between dividing further and forging a deeper unity with each other.

References

Abramowitz, Alan. 2018. The Great Alignment. New Haven: Yale University Press. Acemoglu, Daron, and Robinson, James A. 2006. Economic Origins of Dictatorship and Democracy. Cambridge: Cambridge University Press. Achen, Chris, and Bartels, Larry. 2016. Democracy for Realists. Princeton: Princeton University Press. Alvey, James E. 2011. “James M. Buchanan on the Ethics of Public Debt and Default.” The Journal of Markets & Morality 14: 85–101. Anderson, Elizabeth. 1999. “What Is the Point of Equality?” Ethics 109: 287–337. 2008. “How Should Egalitarians Cope with Market Risks?” Theoretical Inquiries in Law 9: 61–92. 2010. The Imperative of Integration. Princeton: Princeton University Press. 2012. “The Fundamental Disagreement between Luck Egalitarians and Relational Egalitarians.” Canadian Journal of Philosophy S36: 1–23. Appiah, Kwame Anthony. 2010. The Ethics of Identity. Princeton: Princeton University Press. Aquinas, Thomas. 1947 [c. 1260–73]. Summa Theologiae, trans. Fathers of the English Dominican Province. New York: Benziger Brothers. Aristotle. 2019 [350 bce]. Nicomachean Ethics, trans. T. Irwin, 3rd ed. Indianapolis: Hackett. Augustine. 2004 [426]. City of God, trans. Henry Bettenson. London: Penguin Classics. Baier, Annette. 1985. Postures of the Mind: Essays on Mind and Morals. Minneapolis. University of Minnesota Press. Bastiat, Frédéric. 1995. “Protectionism and Communism.” In George Huszar (ed.), Seymour Cain (trans.), Selected Essays on Political Economy. Irvingtonon-Hudson. Foundation for Economic Education, 194–228. Beitz, Charles. 1990. Political Equality. Princeton: Princeton University Press. Bell, Stephanie. 2000. “Do Taxes and Bonds Finance Government Spending?” Journal of Economic Issues 34: 603–20. Berman, Sheri. 2006. The Primacy of Politics: Social Democracy and the Making of Europe’s Twentieth Century. Cambridge: Cambridge University Press. Bok, P. MacKenzie. 2017. “To the Mountaintop Again: The Early Rawls and Post-Protestant Ethics in Postwar America.” Modern Intellectual History 14: 153–85. Boxill, Bernard R. 1992. Blacks and Social Justice, rev. ed. Lanham, MD. Rowman and Littlefield. 356

References

357

2003. “A Lockean Argument for Black Reparations.” The Journal of Ethics 7, no. 1: 63–91. Brennan, Jason. 2016. Against Democracy. Princeton: Princeton University Press, 2017. Broadwalter, Luke. 2012. “Wells Fargo Agrees to Pay $175M Settlement in Pricing Discrimination Suit,” Baltimore Sun, July 12. www.baltimoresun .com/news/bs-xpm-2012-07-12-bs-md-ci-wells-fargo-20120712-story.html. Brown, Alexander. 2015. “Principles of Stakes Fairness in Sport.” Politics, Philosophy & Economics 14: 1–35. Brudney, Daniel. 1997. “Community and Completion.” In Andrews Reath, Barbara Herman, and Christine Korsgaard (eds.), Reclaiming the History of Ethics: Essays for John Rawls. Cambridge: Cambridge University Press, 388–418. 1998. Marx’s Attempt to Leave Philosophy. Cambridge, MA: Harvard University Press. 2001. “Justifying a Conception of the Good Life: The Problem of the 1844 Marx.” Political Theory 29: 364–94. 2010. “Producing for Others.” In Christopher Zurn and Hans-Christoph Schmidt am Busch (eds.), The Philosophy of Recognition: Historical and Contemporary Perspectives. Lanham, MD: Rowman and Littlefield, 151–88. 2014. “The Young Marx and the Middle-Aged Rawls.” In Jon Mandel and David Reidy (eds.), A Companion to Rawls. Chichester; West Sussex: WileyBlackwell, 450–71. Bruni, Luigino, and Sugden, Robert. 2008. “Fraternity: Why the Market Need Not Be a Morally Free Zone.” Economics and Philosophy 24: 35–64. Buchanan, Allen. 1990. “Justice as Reciprocity versus Subject-Centered Justice.” Philosophy & Public Affairs 19: 227–52. Buchanan, James, and Tullock, Gordon. 1962. The Calculus of Consent. Ann Arbor: University of Michigan Press. Burke, Edmund. 1999 [1800]. “Thoughts and Details on Scarcity.” In Francis Canavan (ed.), Select Works of Edmund Burke, vol. 4. Indianapolis: Liberty Fund. Burwell v. Hobby Lobby Stores, Inc. 2014. 573 U.S. 682. Carcieri, Martin D. 2010. “Rawls and Reparations.” Michigan Journal of Race and Law 15: 267–316. Catholic Church. 2012. Catechism of the Catholic Church. Vatican City: Libreria Editrice Vaticana. https://penandthepad.com/cite-catholic-catechism7852709.html. Clark, John Bates. 1899. The Distribution of Wealth: A Theory of Wages, Interest and Profits. New York: Macmillan. Clark, Kenneth B. 1989. Dark Ghetto: Dilemmas of Social Power, 2nd ed. Middletown, CT: Wesleyan University Press. Cohen, G. A. 2001. “Facts and Principles.” Philosophy & Public Affairs 31: 211–45. 2008. Rescuing Justice and Equality. Cambridge, MA: Harvard University Press. Curley, Edwin. 1994. Introduction to Hobbes’s Leviathan with Selected Variants from the Latin Edition of 1668. Indianapolis: Hackett.

358

References

D’Agostino, Fred. 1996. Free Public Reason: Making It up as We Go. Oxford: Oxford University Press. 2010. Naturalizing Epistemology: Thomas Kuhn and the Essential Tension. London: Palgrave MacMillan. Daniels, Norman (ed.). 1975. Reading Rawls: Critical Studies on Rawls’ “A Theory of Justice.” Stanford: Stanford University Press. Darby, Derrick. 2010. “Reparations and Racial Inequality.” Philosophy Compass 5: 55–66. Darity, William A. Jr., and Mullen, A. Kirsten. 2020. From Here to Equality: Reparations for Black Americans in the Twenty-First Century. Chapel Hill: University of North Carolina Press. Darwall, Stephen. 1977. “Two Kinds of Respect.” Ethics 88: 36–49. 2002a. Contractarianism/Contractualism. Oxford: Blackwell. 2002b. Welfare and Rational Care. Princeton: Princeton University Press. 2006. The Second-Person Standpoint: Morality, Respect, and Accountability. Cambridge, MA: Harvard University Press. 2013a. Morality, Authority, and Law: Essays in Second-Personal Ethics I. Oxford: Oxford University Press. 2013b. Honor, History, and Relationship: Essays in Second-Personal Ethics II. Oxford: Oxford University Press. 2016. “Making the ‘Hard’ Problem of Moral Normativity Easier.” In Erroll Lord and Barry Maguire (eds.), Weighing Reasons. Oxford: Oxford University Press. 2018. “Contempt as an Other-Characterizing, ‘Hierarchizing’ Attitude.” In Michelle Mason (ed.), The Moral Psychology of Contempt. Lanham, MD: Rowman and Littlefield. 2019. “What Are Moral Reasons?,” The 2017 Amherst Lecture in Philosophy. www.amherstlecture.org/darwall2017/index.html. Forthcoming. “Commanding Reasons and ‘Commands of Reason’: Kant on Moral Obligation.” In Stefano Bacin and Oliver Sensen (eds.), Kant’s Moral Philosophy in Context. Cambridge: Cambridge University Press. Diamond, Cora. 1991. “The Importance of Being Human.” Royal Institute of Philosophy Supplement 29: 35–62. Dietsch, Peter. 2021. “Money Creation, Debt, and Justice.” Politics, Philosophy, and Economics 20: 151–79. Dreben, Burton. 2003. “On Rawls and Political Liberalism.” In Samuel Freeman (ed.), The Cambridge Companion to Rawls. Cambridge: Cambridge University Press, 316–46. Du Bois, W. E. B. 1920. Darkwater: Voices from Within the Veil. New York: Washington Square Press. 2009 [1903].The Souls of Black Folk (ed.), Brent Hayes Edwards. Oxford: Oxford University Press. Dworkin, Ronald. 1977. Taking Rights Seriously. Cambridge, MA: Harvard University Press. 1981. “What Is Equality? Part 2: Equality of Resources.” Philosophy & Public Affairs 10: 283–345. 2000. Sovereign Virtue: The Theory and Practice of Equality. Cambridge, MA: Harvard University Press.

References

359

Eberle, Christopher. 2002. Religious Conviction in Liberal Politics. Cambridge: Cambridge University Press. Edmundson, William. 2017. John Rawls: Reticent Socialist. Cambridge: Cambridge University Press. Employment Division, Department of Human Resources of Oregon v. Smith. 1990. 494 U.S. 872. Engels, Friedrich 1975 [1845]. The Condition of the Working Class in England. In Marx-Engels Collected Works, vol. 4, 295–583. Enoch, David. 2013. “The Disorder of Public Reason.” Ethics 124: 141–76. 2015. “Against Public Reason.” In David Sobel, Peter Vallentyne, and Steven Wall (eds.), Oxford Studies in Political Philosophy, vol. 1. Oxford: Oxford University Press, 112–44. Espindola, Juan, and Vaca, Moises. 2014. “The Problem of Historical Rectification for Rawlsian Theory.” Res Publica 20: 227–43. Estlund, David. 2003. “The Democracy/Contractualism Analogy.” Philosophy and Public Affairs 31: 387–412. 2008. Democratic Authority: A Philosophical Framework. Princeton: Princeton University Press. 2011. “Replies to Commentators.” Ethics 121: 354–89. 2015. “G. A. Cohen’s Critique of the Original Position.” In Timothy Hinton (ed.), The Original Position. Cambridge: Cambridge University Press, 139–58. 2020. Utopophobia. Princeton: Princeton University Press. Unpublished. “What’s Unjust About Structural Injustice?” Fiorina, Morris. 2017. Unstable Majorities: Polarization, Party Sorting, and Political Stalemate. Stanford: Hoover Institution Press. Fleischacker, Samuel. 2004. A Short History of Distributive Justice. Cambridge, MA: Harvard University Press. Forrester, Katrina. 2019. In the Shadow of Justice: Postwar Liberalism and the Remaking of Political Philosophy. Princeton: Princeton University Press. Forst, Rainer. 2002. Contexts of Justice: Political Philosophy Beyond Liberalism and Communitarianism, trans. J. Farrell. Berkeley; Los Angeles: University of California Press. 2011. The Right to Justification: Elements of a Constructivist Theory of Justice, trans. J. Flynn. New York: Columbia University Press. 2013. Toleration in Conflict: Past and Present, trans. C. Cronin. Cambridge: Cambridge University Press. 2017. “Political Liberalism: A Kantian View.” Ethics 128: 123–44. Foucault, Michel. 1978. The History of Sexuality, Vol. 1: An Introduction, trans. Robert Hurley. New York: Pantheon. 1983. “The Subject and Power.” In Herbert L. Dreyfus and Paul Rabinow (eds.), Beyond Structuralism and Hermeneutics. Chicago: University of Chicago Press. Pope Francis. 2019. “A Document on Human Fraternity for World Peace and Living Together.” http://w2.vatican.va/content/francesco/en/travels/2019/ outside/documents/papa-francesco_20190204_documento-fratellanzaumana.html.

360

References

Frankena, William. 1963. Ethics. Englewood Cliffs, NJ: Prentice-Hall. Freeman, Samuel (ed.). 2002. The Cambridge Companion to Rawls. Cambridge: Cambridge University Press. 2003. “John Rawls: An Overview.” In Samuel Freeman (ed.), The Cambridge Companion to Rawls. Cambridge: Cambridge University Press, 1–59. 2007. Rawls. London: Routledge. 2013. “Property-Owning Democracy and the Difference Principle.” Analyse & Kritik 35: 9–36. Fricker, Miranda. 2007. Epistemic Injustice: Power and the Ethics of Knowing. Oxford: Oxford University Press. Fried, Barbara H. 1998. The Progressive Assault on Laissez Faire: Robert Hale and the First Law and Economics Movement. Cambridge, MA: Harvard University Press. Friedman, Milton. 1947. “Lerner on the Economics of Control.” Journal of Political Economy 55: 405–16. 1948. “A Monetary and Fiscal Framework for Economic Stability.” American Economic Review 38: 245–64. 1968. “The Role of Monetary Policy.” The American Economic Review 58: 1–17 Fukuyama, Francis. 2014. Political Order and Political Decay. New York: Farrar, Straus and Giroux. Fulton v. City of Philadelphia. 2021. 593 U.S. ___ (2021) Gaus, Gerald. 1986. “Subjective Value and Justificatory Political Theory.” Nomos 28: 241–69. 1990. Value and Justification: The Foundations of Liberal Theory. Cambridge: Cambridge University Press. 1996. Justificatory Liberalism: An Essay on Epistemology and Political Theory. Oxford: Oxford University Press. 2011a. The Order of Public Reason. Cambridge: Cambridge University Press. 2011b. “A Tale of Two Sets: Public Reason in Equilibrium.” Public Affairs Quarterly 25: 305–25. 2016. The Tyranny of the Ideal: Justice in a Diverse Society. Princeton: Princeton University Press. 2018. “The Complexity of a Diverse Moral Order.” Georgetown Journal of Law and Public Policy 16: 645–80. 2021. The Open Society and Its Complexities. Oxford: Oxford University Press. Gaus, Gerald, and Van Schoelandt, Chad. 2017. “Consensus on What? Convergence for What? Four Models of Political Liberalism.” Ethics 128: 145–72. Gauthier, David 1969. The Logic of Leviathan. Oxford: Oxford University Press. 1986. Morals by Agreement. Oxford: Oxford University Press. George, Henry. 1912. Progress and Poverty: An Inquiry into the Cause of Industrial Depressions and of Increase of Want with Increase of Wealth: The Remedy. Garden City, NJ: Doubleday, Page & Co. Gibbard, Allan. 1999. “Morality as Consistency in Living: Korsgaard’s Kantian Lectures.” Ethics 110: 140–64.

References

361

Goldman, Alvin, and Kim, Jaegwon (eds.). 1979. Values and Morals: Essays in Honor of William Frankena, Charles Stevenson, and Richard Brandt. Dordrecht: Reidel. Godley, Wynne, and Lavoie, Marc. 2006. Monetary Economics: An Integrated Approach to Credit, Money, Income, Production and Wealth. New York: Palgrave Macmillan. Gøsta Esping-Anderson. 1990. The Three Worlds of Welfare Capitalism. Princeton: Princeton University Press. Greenwald, Daniel, Lettau, Martin, and Ludvigson, Sydney. 2021. “How the Wealth Was Won: Factors Shares as Market Fundamentals,” NBER Working Paper 25769. www.nber.org/system/files/working_papers/w25769/ w25769.pdf. Habermas, Jürgen. 1998. Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy, trans. William Rehg. Cambridge, MA: MIT Press. Haggard Stephen, and Kaufman, Robert. 2021. Backsliding. Cambridge: Cambridge University Press. Hampton, Jean. 1986. Hobbes and the Social Contract Tradition. Cambridge: Cambridge University Press. Harsanyi, John. 1955. “Cardinal Welfare, Individualistic Ethics, and Interpersonal Comparisons of Utility.” Journal of Political Economy 63: 309–21. Haslanger, Sally. 2012. Resisting Reality: Social Construction and Social Critique. Oxford: Oxford University Press. Hegel, G. W. F. 1991 [1820]. Elements of the Philosophy of Right. Cambridge: Cambridge University Press. Hill, Renee A. 2002. “Compensatory Justice: Over Time and Between Groups.” The Journal of Political Philosophy 10: 392–415. Hinton, Timothy (ed.). 2015. The Original Position. Cambridge: Cambridge University Press. Hobbes, Thomas. 1839 [1656]. Six Lessons to the Savilian Professors of the Mathematics. In The English works of Thomas Hobbes of Malmesbury, vol. VII. (ed.), J. Bohn. 1985 [1651]. Leviathan (ed.), C. B. Macpherson. New York: Penguin. 1990 [1681]. Behemoth or The Long Parliament. Chicago: University of Chicago Press. Hobson, John. 1891. “The Law of Three Rents.” Quarterly Journal of Economics 5: 263–88. 1909. The Industrial System: An Inquiry into Earned and Unearned Income. London: Longmans, Green, and Co. Hocket, Robert. 2013. “Bretton Woods 1.0: A Constructive Retrieval for Sustainable Finance.” Cornell Law Faculty Publications 921. https:// scholarship.law.cornell.edu/facpub/921. Hockett, Robert. 2021. “The Once and Future Fed – And Treasury: From Credit Modulo-Allocative Separation to Re-Consolidation.” https://ssrn .com/abstract=3808792. Hockett, Robert, and Omarova, Saule T. 2017. “The Finance Franchise.” Cornell Law Review 102: 1143–1218.

362

References

Hockett, Robert, and James, Aaron. 2020. Money from Nothing: Or Why We Should Learn to Stop Worrying about Debt and Love the Federal Reserve. New York: Melville House Press. Hodgskin, Thomas. 1825. Labour Defended against the Claims of Capital: Or, the Unproductiveness of Capital Proved with Reference to the Present Combinations Amongst Journeymen. London: Knight and Lacey. Hohfeld, Wesley. 1919. Fundamental Legal Conceptions as Applied in Legal Reasoning. New Haven: Yale University Press. Hume, David. 1978 [1739]. A Treatise of Human Nature, 2nd ed., L. A. SelbyBigge (ed.), with text revised and notes by P. H. Nidditch. Oxford: Oxford University Press. 1987 [1741]. “Of the First Principles of Government.” In Essays Moral and Political. Indianapolis, IN: Liberty Fund, 32–36. Hussain, Waheed. 2020. “Pitting People against Each Other.” Philosophy & Public Affairs 48: 79–113. Irwin, Terence. 1990. “The Good of Political Activity.” In Günther Patzig (ed.), Aristoteles’ Politik. Göttingen: Vandenhoeck and Ruprecht, 73–98. James, Aaron. 2005. “Constructing Justice for Existing Practice: Rawls and the Status Quo.” Philosophy & Public Affairs 33: 281–316. (2022) “What Trade’s Pioneers Missed: Money.” Moral Philosophy and Politics 9 no. 1: 89–106. Kandiyali, Jan. 2020. “The Importance of Others: Marx on Unalienated Production.” Ethics 130: 555–87. Kant, Immanuel. 1965 [1797]. The Metaphysical Elements of Justice, trans. John Ladd. New York: Bobbs-Merrill. 1991 [1797]. The Metaphysics of Morals, trans. Mary Gregor. New York: Cambridge University Press. 1996 [1788]. Critique of Practical Reason, in Practical Philosophy, trans. and ed. Mary Gregor. Cambridge: Cambridge University Press. Kelly, Erin. 2017. “The Historical Justice Problem for Political Liberalism.” Ethics 128: 75–94. Keynes, J. M. 1930. A Treatise on Money. London: Macmillan & Co. 1933. “A Monetary Theory of Production.” www.hetwebsite.net/het/texts/key nes/keynes1933mtp.htm. King, Martin Luther Jr. 1961. “The Time for Freedom Has Come.” The New York Times. September 10. 1963. “Letter from a Birmingham Jail.” Philadelphia: American Friends Service Committee. 1964. “Letter from a Birmingham Jail.” Why We Can’t Wait. Boston, MA: Beacon Press. 1967. “Where Are We?” In Where Do We Go From Here: Chaos or Community? Boston, MA: Beacon Press, 1–22. Kloppenberg, James T. 1986. Uncertain Victory: Social Democracy and Progressivism in European and American Thought, 1870–1920. Oxford: Oxford University Press. Krishnamurthy, Meena. 2013. “Completing Rawls’s arguments for equal political liberty and its fair value: the argument from self-respect.” Canadian Journal of Philosophy 43: 179–205.

References

363

Kutz, Christopher. 2004. “Justice in Reparations: The Cost of Memory and the Value of Talk.” Philosophy & Public Affairs 32: 277–312. Larmore, Charles. 1987. Patterns of Moral Complexity. Cambridge: Cambridge University Press. 2008. The Autonomy of Morality. Cambridge: Cambridge University Press. Laurence, Ben. 2021. Agents of Change: Political Philosophy in Practice. Cambridge, MA: Harvard University Press. Lawford-Smith, Holly. 2010. “Debate: Ideal Theory – A Reply to Valentini.” Journal of Political Philosophy 18: 357–68. Lee, Trymaine. 2019. “A Vast Wealth Gap, Driven by Segregation, Redlining, Evictions and Exclusion Separates Black and White America.” New York Times Magazine. August 14. www.nytimes.com/interactive/2019/08/14/maga zine/racial-wealth-gap.html. Lerner, Abba. 1941. “The Economic Steering Wheel.” The University Review 7: 257–63. 1943. “Functional Finance and the Federal Debt.” Social Research 10: 38–51. 1944. The Economics of Control. New York: Kelley. List, Christian, and Pettit, Phillip. 2013. Group Agency: The Possibility, Design, and Status of Corporate Agents. Oxford: Oxford University Press. List, Christian, and Valentini, Laura. 2016. “The Methodology of Political Theory.” In Herman Cappelen, Tamar Gendler, and John Hawthorne (eds.), Oxford Handbook of Philosophical Methodology. Oxford: Oxford University Press, 525–53. Little, Daniel. 2014. “Rawls and Economics.” In Jon Mandle and David Reidy (eds.), A Companion to Rawls. Sussex: Wiley Blackwell. Little Sisters of the Poor Saints Peter and Paul Home v. Pennsylvania. 2020. 591 U.S. ___. Lloyd, S.A. 1992. Ideals as Interests in Hobbes’s Leviathan. Cambridge: Cambridge University Press. 2009. Morality in the Philosophy of Thomas Hobbes: Cases in the Law of Nature. Cambridge: Cambridge University Press. 2020. “All the Mind’s Pleasure: Glory, Self-Admiration, and Moral Motivation in On the Citizen and Leviathan.” In Robin Douglass and John Olsthoorn, (eds.), Hobbes’s On the Citizen: A Critical Guide. Cambridge: Cambridge University Press, 51–70. Locke, John. 1959 [1689]. An Essay Concerning Human Understanding. A. C. Fraser (ed.). New York: Dover. 1983 [1689]. A Letter Concerning Toleration. James H. Tully (ed.). Indianapolis: Hackett. Lukács, Georg. 1972. History and Class Consciousness: Studies in Marxist Dialectics, trans. Rodney Livingstone. Cambridge, MA: MIT Press. Lukes, Steven. 2018. “‘Getting and Spending, We Lay Waste Our Powers’: On the Expanding Reach of the Market.” In Arthur M. Melzer and Steven J. Kautz (eds.), Are Markets Moral?. Philadelphia: University of Pennsylvania Press, 68–84. Lyons, David. 2013. Confronting Injustice: Moral History and Political Theory. Oxford: Oxford University Press.

364

References

MacIntyre, Alasdair. 1994 “A Partial Response to My Critics.” In John Horton and Susan Mendus (eds.), After MacIntyre: Critical Perspectives on the Work of Alasdair MacIntyre. South Bend: University of Notre Dame Press. MacKinnon, Catharine. 1989. Toward a Feminist Theory of the State. Cambridge, MA: Harvard University Press. Marglin, Stephen A. 2021. Raising Keynes: A Twenty-First Century General Theory. Cambridge, MA: Harvard University Press. Mason, Liliana. 2018. Uncivil Agreement: How Politics Became Our Identity. Chicago: University of Chicago Press. Masterpiece Cakeshop v. Colorado Civil Rights Commission. 2018. 584 U.S. __. Martin, Rex. 2014. “Overlapping Consensus.” In Jon Mandel and David Reidy (eds.) A Companion to Rawls. Sussex: Wiley Blackwell, 281–96. Martinich, A. P. 2021. Hobbes’s Political Philosophy: Interpretation and Interpretations. Oxford: Oxford University Press. Marx, Karl. 1967 [1867]. Capital: A Critique of Political Economy. New York: International. 1973 [1857]. Grundrisse, trans. Martin Nicolaus. New York: Vintage Books. 1975a [1844]. “Comments on James Mill.” In Marx-Engels Collected Works, vol. 3, 211–28. 1975b [1844]. “Economic and Philosophic Manuscripts of 1844.” In MarxEngels Collected Works, vol. 3, 229–348. 1989 [1875]. “The Critique of the Gotha Programme.” In Marx-Engels Collected Works, vol. 24, 75–76. Marx, Karl, and Engels, Friedrich. 1975–. Marx-Engels Collected Works (MECW), New York: International. 1976 [1845]. “The German Ideology.” In Marx-Engels Collected Works, vol. 5, 42–43. McAward, Jennifer Mason. 2012. “Defining the Badges and Incidents of Slavery.” Journal of Constitutional Law 14: 561–630. McCarthy, Thomas. 2004. “Political Philosophy and Racial Injustice: From Normative to Critical Theory.” In Seyla Benhabib and Nancy Fraser (eds.), Pragmatism, Critique, Judgment. Cambridge, MA: MIT Press, 147–68. McCarty, Nolan. 2019. Polarization: What Everyone Needs to Know. Oxford: Oxford University Press. McConnell, Michael. 1990. “The Origins and Historical Understanding of Free Exercise of Religion.” Harvard Law Review 103: 1409–1519. McDaniel, W. Caleb. 2019. Sweet Taste of Liberty: A True Story of Slavery and Restitution in America. Oxford: Oxford University Press. McDermott, Daniel. 2008. “Analytic Political Philosophy.” In David Leopold and Marc Stears (eds.), Political Theory: Methods and Approaches. Oxford: Oxford University Press, 11–28. McGary, Howard. 1999. Race and Social Justice. Malden, MA: Blackwell. 2010. “Reconciliation and Reparations.” Metaphilosophy 41: 546–62. Meade, James. 1965. Efficiency, Equality and the Ownership of Property. Cambridge, MA: Harvard University Press.

References

365

Mehrling, Perry. 2014. “MIT and Money.” History of Political Economy 46: 177–97. 2019. “The Monetary Education of John Hicks.” In Annalisa Roselli, Nerio Naldi, and Eleonora Sanfilippo (eds.), Money, Finance and Crises in Economic History. London: Routledge, 107–16. Meyer, Lukas H. 2006. “Reparations and Symbolic Restitution.” Journal of Social Philosophy 37: 406–22. Milanovic, Branko. 2016. Global Inequality: A New Approach for the Age of Globalization. Cambridge, MA: Harvard University Press. Mill, John Stuart. 1979 [1861]. Utilitarianism. George Sheer (ed.). Indianapolis: Hackett. 1861. Considerations on Representative Government. London: Parker, Son & Bourn. Miller, J. Joseph. 2003. “J. S. Mill on Plural Voting, Competence and Participation.” History of Political Thought 24: 647–67. Mills, Charles W. 1997. The Racial Contract. Ithaca, NY: Cornell University Press. 2003. “White Supremacy as a Sociopolitical System.” In Charles W. Mills (ed.), From Class to Race: Essays in White Marxism and Black Radicalism. Lanham, MD: Rowman & Littlefield, 177–94. 2005. “Ideal Theory as Ideology.” Hypatia, 20: 165–83. 2007. “Contract of Breach: Repairing the Racial Contract.” In Carole Pateman and Charles W. Mills (eds.), Contract and Domination. Cambridge, MA: Polity Press, 106–33. 2017a. “Racial Liberalism.” In Black Rights/White Wrongs: The Critique of Racial Liberalism. Oxford: Oxford University Press, 28–48. 2017b. Black Rights/White Wrongs: The Critique of Racial Liberalism. Oxford: Oxford University Press. “Epilogue (As Prologue): Towards a Black Radical Liberalism.” In Black Rights/White Wrongs: The Critique of Racial Liberalism. Oxford: Oxford University Press, 201–16. Mills, Charles W., and Flikschuh, Katrin. 2018. “Racial Justice.” Aristotelian Society Supplementary Volume XCII: 69–89. 2018. “Through a Glass, Whitely: Ideal Theory as Epistemic Injustice.” Proceedings and Addresses of the American Philosophical Association 92: 43–77. “Theorizing Racial Justice.” https://tannerlectures.utah.edu/ MILLSTANNERLECTURE.pdf Mizruchi, Mark S. 2013. The Fracturing of the American Corporate Elite. Cambridge, MA: Harvard University Press. Moore, Barrington, 1966. Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World. Boston: Beacon Press. Mulligan, Thomas. 2018. “Plural Voting for the Twenty-First Century.” Philosophical Quarterly 68: 286–306. Murdoch, Iris. 1985. The Sovereignty of Good. New York: Routledge and Kegan Paul. Murphy, Colleen. The Conceptual Foundations of Transitional Justice. Cambridge: Cambridge University Press, 2017.

366

References

Murphy, Liam, and Nagel, Thomas. 2002. The Myth of Ownership: Taxes and Justice. Oxford: Oxford University Press. Nagel, Thomas. 1973. “Rawls on Justice.” The Philosophical Review 82: 220–34. 2005. “The Problem of Global Justice.” Philosophy & Public Affairs 33: 114–47. Nelson, Eric. 2019. The Theology of Liberalism: Political Philosophy and the Justice of God. Cambridge, MA: Belknap Press. Nozick, Robert. 1974. Anarchy, State, and Utopia. New York: Basic Books. Ockham, William of. 1495. Quaestiones et decisiones in quattuor libros Sententiarum Petri Lombardi. Lyon. O’Neill, Martin, and Williamson, Thad. 2012a. “The Promise of Predistribution.” Published online by Policy Network. www.policy-network .net/pno_detail.aspx?ID=4262&title=The-Promise-of-Pre-distribution. 2012. Property-Owning Democracy: Rawls and Beyond. Malden, MA: WileyBlackwell. O’Neill, Onora. 1987. “Abstraction, Idealization and Ideology in Ethics.” Royal Institute of Philosophy Lectures 22: 55–69. 1996. Towards Justice and Virtue: A Constructive Account of Practical Reasoning. Cambridge: Cambridge University Press. Paine, Thomas. 1792. Rights of Man. Part the Second. Combining Principle and Practice. London: H. D. Symonds. 1796. “Agrarian Justice.” www.ssa.gov/history/paine4.html. Pallikkathayil, Japa. 2017. “Persons and Bodies.” In Sari Kisilevsky and Martin J. Stone (eds.), Freedom and Force: Essays on Kant’s Legal Philosophy. Oxford: Hart, 35–54. 2019. “Disagreement and the Duties of Citizenship.” American Philosophical Quarterly 56: 71–82. Parfit Derek. 1973. “Later Selves and Moral Principles.” In Alan Montefiore (ed.), Philosophy and Personal Relations. London: Routledge & Kegan Paul, 137–69. Pettit, Phillip. 1997. Republicanism. Oxford: Oxford University Press. Phillips, Jack. 2018. “I’m the Masterpiece Cakeshop Baker. Will the Supreme Court Uphold My Freedom?” The Washington Post, April 26. Piketty, Thomas. 2014. CAPITAL in the Twenty-First Century, trans. Arthur Goldhammer. Cambridge, MA: Harvard University Press. Pinkard, Terry. 2017. Does History Make Sense? Hegel on the Historical Shapes of Justice. Cambridge, MA: Harvard University Press. Plato. 1992 [c. 375 bce]. Republic, trans. G. M. A. Grube, rev. C. D. C. Reeve. Indianapolis: Hackett. Prichard, H. A. 2002. “Does Moral Philosophy Rest on a Mistake?” in Jim McAdam (ed.), Moral Writings. Oxford: Oxford University Press, 7–20. Quong, Jonathan. 2011. Liberalism without Perfection. Oxford: Oxford University Press. Reidy, David A. 2014. “From Philosophical Theology to Democratic Theory.” In Jon Mandle and David Reidy (eds.), A Companion to Rawls. Malden, MA: Wiley-Blackwell, 9–30.

References

367

Richardson, Henry S. 2021. “Taking Multidimensionality Seriously: Capabilities, Rawls’s Primary Goods and Guiding Action.” In Enrica Chiappero-Martinetti, Siddiqur Osmani, and Mozaffar Qizilbash (eds.), The Cambridge Handbook of the Capability Approach. Cambridge: Cambridge University Press, 146–64. Ripstein, Arthur. 2009. Force and Freedom: Kant’s Legal and Political Philosophy. Cambridge, MA: Harvard University Press. Robeyns, Ingrid. 2012. “Care, Gender, and Property-Owning Democracy.” In Martin O’Neill and Thad Williamson (eds.), Property-Owning Democracy: Rawls and Beyond. Malden, MA: Wiley-Blackwell, 163–79. Rödl, Sebastian. 2007. Self-Consciousness. Cambridge, MA: Harvard University Press. Rosenfeld, Sam. 2017. The Polarizers. Chicago: University of Chicago Press. Ruml, Beardsley. 1946. “Taxes for Revenue are Obsolete.” American Affairs VIII: 35–39. Ronzoni, Miriam. 2010. “Constructivism and Practical Reason: On Intersubjectivity, Abstraction, and Judgment.” Journal of Moral Philosophy 7: 74–104. Rousseau, Jean-Jacques. 1997 [1762]. “Of the Social Contract.” In The Social Contract and Other Later Political Writings, trans. Victor Gourevitch. Cambridge: Cambridge University Press, 38–155. 2019 [1755]. “Discourse on the Origin and Foundations of Inequality Among Men.” In The Discourses and Other Early Political Writings, trans. Victor Gourevitch. Cambridge: Cambridge University Press, 111–88. Samuelson, Paul. 1972. “Money, Interest Rates, and Economic Activity: Their Interrelationship in a Market Economy.” In Robert C. Merton (ed.), The Collected Scientific Papers of Paul A. Samuelson, vol. 3. Cambridge, MA: MIT Press, 550–70. 2014. Interview with Mark Blaug in his film, “John Maynard Keynes: Life/ Ideas/Legacy,” selection at www.youtube.com/watch?v=4_pasHodJ-8. Sangiovanni, Andrea. 2016. “How Practices Matter.” Journal of Political Philosophy 24: 3–23. Sandel, Michael. 1982. Liberalism and the Limits of Justice. Cambridge: Cambridge University Press. 1984. “The Procedural Republic and the Unencumbered Self.” Political Theory 12: 81–96. Sandel, Michal. 2020. The Tyranny of Merit: What’s Become of the Common Good? New York: Farrar, Straus and Giroux. Scanlon, T. M. 1998. What We Owe to Each Other. Cambridge, MA: Harvard University Press. 2003a. “Rawls on Justification.” In Samuel Freeman (ed.), The Cambridge Companion to Rawls. Cambridge: Cambridge University Press. 2003b. “The Difficulty of Tolerance.” In The Difficulty of Tolerance: Essays in Moral and Political Philosophy. Cambridge: Cambridge University Press, 187–201. 2008. Moral Dimensions: Permissibility, Meaning, Blame. Cambridge, MA: Harvard University Press.

368

References

2018. Why Does Inequality Matter? Oxford: Oxford University Press. Scheffler, Samuel. 1979. “Moral Independence and the Original Position.” Philosophical Studies 35: 397–403. 1982. The Rejection of Consequentialism. Oxford: Oxford University Press. 2018. “Membership and Obligation.” The Journal of Political Philosophy 26: 3–23. 2019. “The Rawlsian Diagnosis of Donald Trump.” Boston Review, February 1. Schultz-Bergin, Marcus. 2021. “Does Convergence Liberalism Risk Anarchy?” Journal of Ethics and Social Philosophy 20 no. 1. www.jesp.org/index.php/jesp/ article/view/1072. Schmidtz, David. 2002. “How to Deserve.” Political Theory 30: 774–99. Schwartz, Adina. 1973. “Moral Neutrality and Primary Goods.” Ethics 83: 294–307. Schweickart, David. 2012. “Property-Owning Democracy or Economic Democracy?” In Martin O’Neill and Thad Williamson (eds.), PropertyOwning Democracy: Rawls and Beyond. Malden, MA: Wiley-Blackwell, 201–21. Sen, Amartya, and Williams, Bernard. 1982. Utilitarianism and Beyond. Cambridge: Cambridge University Press. Sen, Amartya. 2009. The Idea of Justice. Cambridge, MA: Harvard University Press. Senior, Nassau. 1836. An Outline of the Science of Political Economy. London: W. Clowes and Sons. Shelby, Tommie. 2004. “Race and Social Justice: Rawlsian Considerations.” Fordham Law Review 72: 1697–1714. 2013. “Racial Realities and Corrective Justice: A Reply to Charles Mills.” Critical Philosophy of Race 1: 145–62. 2016. Dark Ghettos: Injustice, Dissent, and Reform. Cambridge, MA: Harvard University Press. Sher, George. 1979. “Effort, Ability, and Personal Desert.” Philosophy & Public Affairs 8: 361–76. Sherbert v. Verner. 1963. 374 U.S. 398 Shiffrin, Seana Valentine. 2021. Democratic Law. Oxford: Oxford University Press. Shklar, Judith. 1990. The Faces of Injustice. New Haven: Yale University Press. Sidgwick, Henry. 1981 [1874]. The Methods of Ethics. Indianapolis: Hackett. Simmons, A. John. 2010. “Ideal and Nonideal Theory.” Philosophy & Public Affairs 38: 5–36. Skinner, Quentin. 1984. “The Idea of Negative Liberty: Philosophical and Historical Perspectives.” In Richard Rorty, Jerome B. Schneewind, and Quentin Skinner (eds.), Philosophy in History: Essays in the Historiography of Philosophy. Cambridge: Cambridge University Press, 139–222. Smith, Adam. 1869 [1759]. “The Theory of Moral Sentiments.” in The Essays of Adam Smith. London: Alex Murray, 9–304. Smith, Sophie, Bejan, Teresa M., and Zimmermann, Annette (eds.). 2021. The Historical Rawls (Special Forum), Modern Intellectual History 18, no. 4.

References

369

Spencer, Herbert. 1891. Principles of Ethics. London: Williams and Norgate. 1981. “The Coming Slavery.” In Eric Mack (ed.), The Man versus the State, with Six Essays on Government, Society, and Freedom. Indianapolis: Liberty Classics. Spinner-Halev, Jeff. 2007. “From Historical to Enduring Injustice.” Political Theory 35: 574–97. Stein, Ben. 2006. “In Class Warfare, Guess Which Class Is Winning.” New York Times, November 26. www.nytimes.com/2006/11/26/business/yourmoney/ 26every.html. Stephen, James Fitzjames. 1992 [1873]. “Liberty, Equality, Fraternity.” In Liberty, Equality, Fraternity and Three Brief Essays. Chicago: The University of Chicago Press, 23–282. Strawson, P. F. 1962. “Freedom and Resentment.” Proceedings of the British Academy 48: 1–25. Sumner, William Graham. 1911. What Social Classes Owe to Each Other. New York: Harper and Brothers. Táíwò, Olúfemi O. 2019. “States Are Not Basic Structures: Against State_ Centric Political Theory.” Philosophical Papers 48: 59–82. Taylor, Charles. 1985. “Atomism.” In Charles Taylor (ed.), Philosophical Papers: Volume 2: Philosophy and the Human Sciences. Cambridge: Cambridge University Press, 187–210. 1989. Sources of the Self. Cambridge, MA: Harvard University Press. Tebbe, Nelson. 2021a. “The Principle and Politics of Equal Value.” Columbia Law Review 121: 2397–482. 2021b. “The Principle and Politics of Liberty of Conscience.” Harvard Law Review 135: 267–321 Teitelman, Michael. 1972. “The Limits of Individualism.” Journal of Philosophy 69: 545–56. Terry, Brandon M. 2021. “Conscription and the Color Line: Rawls, Race, and Vietnam.” Modern Intellectual History 18: 960–83. Thomas, Alan. 2017. Republic of Equals: Predistribution and Property-Owning Democracy. Oxford: Oxford University Press. Thompson, Janna. 2001. “Historical Injustice and Reparation: Justifying Claims of Descendants.” Ethics 112: 114–35. Thompson, William. 1827. Labor Rewarded. The Claims of Labor and Capital Conciliated: Or, How to Secure to Labor the Whole Product of Its Exertions, by One of the Idle Classes. London: Hunt and Clarke. Tilly, Charles. 1999. Durable Inequality. Berkeley: University of California Press. Tomasi, John. 2012. Free-Market Fairness. Princeton: Princeton University Press. Ture, Kwame (aka Stokely Carmichael), and Hamilton, Charles V. 1992. Black Power: The Politics of Liberation. New York: Vintage Books. Turner, Piers Norris, and Gaus, Gerald (eds.). 2018. Public Reason in Political Philosophy: Classic Sources and Contemporary Controversies. London: Routledge. Valentini. Laura. 2009. “On the Apparent Paradox of Ideal Theory.” Journal of Political Philosophy 17: 332–55.

370

References

2016. “On the Messy Utopophobia vs Factophobia Controversy.” In Kevin Vallier and Michael Weber (eds.), Political Utopias. Oxford University Press, 11–34. Vallier, Kevin. 2018. “Public Justification.” Stanford Encyclopedia of Philosophy. http://plato.stanford.edu/entries/justification-public/. 2019. “Political Liberalism and the Radical Consequences of Justice Pluralism.” Journal of Social Philosophy 50: 212–31. 2022. The Social Philosophy of Gerald Gaus: Moral Relations Amid Control, Contestation, and Complexity. Journal of the American Philosophical Association, 1–23. doi:10.1017/apa.2022.16. Vallier, Kevin, and Ryan Muldoon. 2021. “In Public Reason, Diversity Trumps Coherence.” Journal of Political Philosophy 29: 211–30. Van ‘T Klooster, Jens. 2019. “Central Banking in Rawls’s Property-Owning Democracy.” Political Theory 47: 674–98. Vincent, Michael. 2021. “A Sympathetic Critique of Gaus’s The Order of Public Reason.” A dissertation submitted to Brisbane: University of Queensland. Waldron, Jeremy. 1992. “Superseding Historic Injustice.” Ethics 103: 4–28. Walker, David. 2015 [1830]. David Walker’s Appeal in Four Articles. Mansfield Centre, CT: Martino. Walker, Margaret Urban. 2006. “Restorative Justice and Reparation.” Journal of Social Philosophy 37: 377–95. Wallace, R. J. 2019. The Moral Nexus. Princeton: Princeton University Press. Watson, Alan, ed. 1998. The Digest of Justinian. Philadelphia: University of Pennsylvania Press. Wall, Steven. 2006. “Rawls and the Status of Political Liberty.” Pacific Philosophical Quarterly 87: 245–70. Walzer, Michael. 1983. Spheres of Justice: A Defense of Pluralism and Equality. New York: Basic Books. Webb, Sidney. 1888. “The Rate of Interest and the Laws of Distribution.” Quarterly Journal of Economics 2: 188–208. Weber, Max. 1978 [1921]. Guenther Roth and Claus Wittich (eds.), Economy and Society, vol. 2. Berkeley; Los Angeles: University of California Press. 1909. “Letter to Ferdinand Tönnies, February 19, 1909.” In James T. Kloppenberg. 1986. Uncertain Victory: Social Democracy and Progressivism in European and American Thought, 1870–1920. Oxford: Oxford University Press. Weithman, Paul. 2010. Why Political Liberalism? On John Rawls’s Political Turn. Oxford: Oxford University Press. 2015. “Inclusivism, Stability, and Assurance.” In Tom Bailey and Valentina Gentile (eds.), Rawls and Religion. New York: Columbia University Press, 75–96. 2016. “John Rawls and the Task of Political Philosophy.” In Paul Weithman (ed.), Rawls, Political Liberalism and Reasonable Faith. Cambridge: Cambridge University Press, 29–40. 2017. “Autonomy and Disagreement about Justice in Political Liberalism.” Ethics 128: 95–122. Wenar, Leif. 2006. “Reparations for the Future.” Journal of Social Philosophy 37: 396–405.

References

371

White, Stuart. 2014. “Property-Owning Democracy and Republican Citizenship.” In Martin O’Neill and Thad Williamson (eds.), PropertyOwning Democracy: Rawls and Beyond. Oxford: Wiley-Blackwell, 129–44. Wiens, David, and Ingham, Sean. 2021. “Demographic Objections to Epistocracy: A Generalization.” Philosophy & Public Affairs 49: 323–49. Wilkins, Clara L., and Kaiser, Cheryl R. 2014. “Racial Progress as Threat to the Status Hierarchy: Implications for Perceptions of Anti-White Bias.” Psychological Science 25: 439–46. Williams, Joan C. 2020. White Working Class: Overcoming Class Cluelessness in America. Boston: Harvard Business Review Press. Wolff, Jonathan. 1998. “Fairness, Respect, and the Egalitarian Ethos.” Philosophy & Public Affairs 27: 97–122. Wray, L. Randall. 1988. Understanding Modern Money. Cheltenham: Edward Elgar. 2002. “A Monetary and Fiscal Framework for Economic Stability: A Friedmanian Approach to Restoring Growth.” dx.doi.org/10.2139/ssrn .1010174. Young, Iris Marion. 2011a. Justice and the Politics of Difference. Princeton: Princeton University Press. 2011b. Responsibility for Justice. Oxford: Oxford University Press.

Index

American politics, contemporary, 27, 333, 346 Anderson, Elizabeth, 4, 5, 8, 13, 14, 104–5, 182, 191–92 Aquinas, Thomas, 5, 96 Aristotelian Principle, 280–83 Aristotle, 5, 36, 38, 49, 50, 95–96, 98–99, 343 Arrow, Kenneth, 75 Augustine, 277 autonomy, 44, 111, 145, 148, 149, 224, 313 (see also: freedom) doctrinal autonomy, 7, 141, 147, 150 full autonomy, 145, 148–49 Kantian autonomy, 145, 153 political autonomy, 149, 263 background justice, 187 basic justice, 214, 217, 240, 273 basic liberties, 102, 194, 204, 283, 297, 299, 302, 305, 307, 309–12, 324 argument for, 338, 341 do not include market liberties, 57, 270 formal basic liberties, 266, 267 includes liberty of conscience, 269, 301, 305, 323 none is absolute, 299 part of reasonable political conceptions, 267–70, 322 political liberties are among, 270, 299, 301, 314, 321 principle of equal basic liberties, 90, 92, 191, 200, 223, 263, 266, 272, 290 and reciprocity, 270 security of, 11, 184, 283–85, 288, 294 and self-respect, 284, 294–95, 349 basic structure, 6, 8–9, 37, 47, 55–59, 80–82, 95, 98–99, 101, 103–4, 147, 152, 157, 162, 170, 179, 183–90, 193–94, 196–99, 200–1, 203–4, 211–13, 215–17, 222, 258, 264–67, 281, 303, 321, 337, 343, 348–49

372

Baumol, William, 69, 70, 75, 76 Bentham, Jeremy, 18 Berlin, Isaiah, 301 Brennan, Jason, 11, 258, 269, 296 Buchanan, James, 11, 70, 254, 258, 269 burdens of judgment, 12, 148, 242, 245, 258, 259, 261, 273, 289, 290–91, 317, 320, 326–28, 334 capacity for a sense of justice, 117, 268, 333 capitalism, 38–39, 43, 68, 77, 84–85, 213, 245–46, 272 Categorical Imperative, 225 Catholicism, 242, 246 children, 27, 176, 216, 263, 310, 345 Christianity, 291 circumstances of justice, 7–9, 43–44, 151–52, 155, 159–64, 169, 171–75, 177–79, 182, 339 civic friendship, 49, 201, 305, 349 civil disobedience, 202, 211 civility, 12, 317, 319, 327, 333, 346 duty of civility, 13, 103, 166, 334 coercion, 21, 30, 54, 162 Cohen, G.A. (Gerald), 7, 9–10, 49, 57, 151, 154–59, 159–63, 185–86, 188, 190, 229–31, 343 colonialism, 200, 207 common good, 38, 40, 42, 96, 263, 270, 314–16 communitarian critique, 129 communitarians, 219, 228 community, 29, 39, 40–41, 48, 51–53, 54, 57, 58, 88, 145, 221, 235, 250, 278, 283, 298, 316, 325, 336, 350 justificatory community, 57 comprehensive doctrines (includes: comprehensive view), 7, 20, 23, 34, 58, 109, 127, 137, 146–50, 153–54, 165–66, 195, 241, 248, 262, 317, 326

Index justice as fairness as a comprehensive doctrine, 136 Kantian comprehensive doctrine, 258 non-comprehensive doctrines, 149 partially comprehensive doctrines, 142, 291 reasonable comprehensive doctrines (includes: reasonable comprehensive view), 12, 146–47, 153, 242–43, 244, 263, 266, 319–20, 333 congruence, 13, 122, 158 considered judgments, 142–44, 224, 226 Constant, Benjamin, 301 constitution (US), 263 constitutional consensus, 323 constitutional essentials, 240 constitutional stage (of the four-stage sequence), 51, 55, 187, 311 constitutionalism, 55 constructivism, 6, 9, 109, 113, 133–34, 136, 148, 157 Kantian constructivism, 109–11, 116–17, 134, 135, 144 moral constructivism, 109, 134–35, 137 political constructivism, 109, 134, 137 social constructivism, 143 contractualism, 120 Darwall, Stephen, 5–7, 97, 105 Debreu, Gérard, 75 Declaration of Independence, 48, 145, democracy (10–13, 19, 89, 154, 171, 176, 202, 246, 251, 270, 272, 279, 296–98, 301, 303, 310–11, 313–14, 314–18, 321–23, 325–32, 332–34, 341 (see also: property-owning democracy, social democracy) deliberative democracy, 264, 268 parliamentary democracy, 88 deontic concepts, 113–14 deontology, 114, 116, 141, 146, 200, 255 desert, 4, 13, 78–83, 83–90, 273 Dewey Lectures, Rawls’s, 5–7, 109, 110–11, 112–13, 117, 120, 129–31, 133–36, 141, 144–45, 147 difference principle, 13, 45, 50–51, 56–57, 61, 64, 66, 74, 76–77, 79–81, 83, 89–90, 92–94, 100, 116, 184–85, 190–91, 194–96, 213–14, 223, 244, 266, 273, 287, 298–300, 321, 324, 350–51 dignity, 27, 90, 120 Du Bois, W.E.B., 182, 310–11

373 duty, natural (includes: natural duties), 9, 22, 97, 101–4, 203, 210–11 Dworkin, Ronald, 47, 89, 220 education, see moral education efficiency, 76, 87, 92, 222, 229, 235, 264, 271 egalitarianism, 89, 161, 298, 321 democratic egalitarianism, 89 luck egalitarianism, 79, 89, 104 political egalitarianism, 311 egoism, 17, 24, 27 Engels, Friedrich, 39, 46, 51 envy, 13, 58, 91, 190, 331, 344–46, 351 Epstein, Richard, 11, 258, 269 equality of opportunity, 217, 263, 344 fair equality of opportunity, 11, 13, 64, 76–77, 80, 82, 90, 92–93, 100, 105, 178–80, 184, 190–91, 193–94, 200, 203, 207, 211, 213–14, 216–17, 223, 263, 266–69, 271, 273–74, 298–300, 301, 305, 321, 324, 350 formal equality of opportunity, 91 equilibrium, 156 (see also: reflective equilibrium) general equilibrium, 75–76 market equilibrium, 85–86 evident intention, 334 fact of reasonable pluralism, 11, 195, 242, 259–60, 278, 288, 291 fact-sensitivity critique, 221, 227, 229, 231, 234–36 fairness (does not include: justice as fairness), 96, 232, 281, 304, 323, 341 economic fairness, 271, 322, 331 Free-market fairness, 245 procedural fairness, 303–4 family, 53, 103, 179, 187–88, 205, 207, 220, 263–64, 281, 295 Foucault, Michel, 190, 197 Francis, Pope, 290 fraternity, 45, 93 freedom, 26–27, 44, 65, 83, 90, 110, 113–15, 117, 118, 120, 172, 176, 217, 224, 240, 245, 251, 255, 263, 268–72, 275, 297, 299, 301, 344, 354 (see also: autonomy) religious freedom, 324, 326, 336 Friedman, Milton, 11, 61–62, 67–68, 72–74, 75, 258, 269 friendship, 249 (see also: civic friendship) game theory, 34–35

374

Index

Gaus, Gerald, 1, 10, 11, 239–40, 243–56, 258, 263, 269, 276 Gauthier, David, 34, 254, 269 Gibbard, Allan, 111 goodness, 11, 146, 247–48 globalization, 78 guilt, 9, 95, 98, 101, 112–13, 118, 201

pure procedural justice, 76, 81–82, 90, 193, 258, 303 reparative justice, 178–80, 181, 198, 200–1, 204, 205–7, 208–9, 210–11, 212–14, 216, 218 restorative justice, 208–9, 217 transitional justice, 172

Habermas, Jürgen, 11, 149, 254, 264, 278, 291, 301, 306 Hampton, Jean, 34 Harsanyi, John, 254 Hart, H.L.A., 186 Haslanger, Sally, 190, 195, 197 Hayek, Friedrich, 11, 257, 269 Hegel, G.W.F., 49, 172 heteronomy, 145 hierarchy, 4, 8, 78, 91, 94, 176, 217, 310–11, 346, 350 Hobbes, Thomas, 2, 5, 17–35, 97, 152, 162, 164–65, 189, 192, 254, 276 Howison Lecture, Rawls’s, 129–31, 133–35 Hume, David, 5, 18, 49, 97, 159, 169, 173

Kant, Immanuel, 5, 36, 97, 111, 112–13, 118–19, 120, 135, 148, 164–65, 179, 254 Kant Lecture, Rawls’s, 129–31 Kantian constructivism, (see: constructivism, Kantian) Kantian interpretation, 5, 23, 40, 103, 110, 111, 112, 114–15, 117–18, 120 Kantian metaphysics, 144–45 Kavka, Gregory, 34 Keynes, John Maynard, 75, 77 King Jr., Martin Luther, 174–75

ideal theory, 8–9, 70–71, 178–79, 181–85, 193–94, 197–98, 199–201, 211, 228–29, 233, 294–95, 302, 304–5, 310 indignation, 21, 95, 98, 101, 249 inequality (includes: inequalities), 11–13, 62, 77, 78–81, 89, 90–94, 100, 104, 163, 169–70, 174–76, 178–79, 191, 213–15, 217, 223, 257–58, 267, 269, 271–73, 274–76, 294, 296–97, 298, 299, 302–3, 304–5, 311, 321, 329–31, 346, 348, 350–51 intuitionism (includes: rational intuitionism), 109, 131, 134 Jefferson, Thomas, 176 Jim Crow, 173, 178, 203, 212, 216–17 justice concept of justice, 7, 95–96, 98–100, 105, 146, 159, 163 corrective justice, 179, 203–4, 206, 216, 218 (see also: rectification distributive justice, 4, 44, 79–82, 84, 86, 88, 94, 96, 98–99, 170, 177, 180, 206, 212–13, 271, 273, 275 imperfect procedural justice, 258, 303 justice pluralism, 10–11, 13, 239–40, 243–46 liberal political conceptions of, 11, 55, 245, 257–58, 260–62, 265, 267, 269, 272–73, 275–76

least advantaged, 80, 89, 93, 100, 215, 223, 271, 274–75, 351 (see also: worst off) legitimacy, 79, 224, 245–48, 258, 260, 266, 275–76, 307 liberal principle of legitimacy (includes: liberal principle of political legitimacy), 258, 262, 265, 293 Lerner, Abba, 3, 60, 65–68, 69, 72, 73 liberal democracy, 154, 279, 354 liberal equality, 268, 271 liberal political conceptions of justice, see: justice, liberal political conceptions of liberal socialism, 184, 213, 244 liberalism, 18, 172, 194, 262, 266–67, 292, 321 public reason liberalism, 239, 256 libertarian (includes: libertarianism), 11, 78, 232, 247, 257, 267, 269–70, 272–73, 274–75 liberty of conscience, 18, 172, 269, 301, 305, 323 life plan (includes: plan of life, plans of life), 40, 125, 158, 160, 171, 223, 286, 292, 343, 351 Locke. John, 5, 97, 172, 179, 254 Lyons, David, 217 MacKinnon, Catherine, 190, 197 Marx, Karl, 2, 36–40, 40–49, 49, 57, 84, 343–44 merit, 13, 96, 345, 348 meritocracy, 78–79, 90, 190, 344–45, 347, 348, 350–51, 353, 354

Index metaphysics, 137, 142, 144–45 Mill, John Stuart, 39, 46, 83, 96, 296, 299, 302, 305–6, 310 Mills, Charles, 1, 8–10, 178–79, 181–83, 185, 188, 190, 197, 200, 219, 229, 233–34 modus vivendi, 23, 153–54, 165–66, 172, 242, 276 Montesquieu, Charles de, 343 moral education, 59 moral personality, 40, 110, 145, 249 moral repair, 8, 171, 174, 177, 208 morality of association, 316 moral powers, 40, 117, 143, 199, 264, 268, 270, 272, 282, 298, 307, 319–20, 333 moral psychology, 54, 151, 153, 195, 201, 261, 333 Murphy, Colleen, 170 Murphy, Liam, 53, 61 Musgrave, Richard, 71, 74 mutual advantage (includes: mutual benefit), 89, 152, 161–63, 182, 265 Nagel, Thomas, 53, 61, 101, 128 Native Americans, 200, 206–7 natural rights, 214 natural duties, see duty, natural natural lottery, 110 New Deal, 4, 78 Nietzsche, Friedrich, 344 non-ideal theory, 8, 70–71, 170, 179, 181–82, 185, 198, 199–202, 211, 298, 300, 304–5 Nozick, Robert, 3–4, 61–63, 63–64, 104, 214, 220, 267, 269, 312 Objectivity, 109, 134, 137 Ockham, William of, 140–41 O’Neill, Onora, 227–28, 232 Original Position (OP), 6–7, 8–9, 12, 13, 48, 58, 102, 109, 110–12, 114–15, 116, 120, 125, 126, 128, 129, 145, 151, 155, 158, 169–70, 177–79, 187, 199, 200, 211, 220, 222, 231, 236, 261, 265–66, 270, 286–87, 297–98, 303, 311–12, 337 and expressing our nature, 5, 112 metaphysical interpretation of (Sandel), 129 vs. state of nature, 13 overlapping consensus, 20, 23, 127, 146, 154, 166, 171–72, 242, 246, 258–59, 266, 318, 322–25, 328–29 Pareto Improvements, 155–56 Pareto Principle, 250

375 Parfit, Derek, 124, 126, 129 perfectionism, 124, 131 persons, conception of, 6, 20, 26, 28, 33, 109, 126–29, 131–37, 139, 143, 144 model conception of the person, 109 noumenal conception of the person, 114 political conception of the person, 2, 6, 19, 23, 24–26, 33, 129, 134, 137–38 personal identity, 6, 121, 124–27, 132, 137, 142–43 philosophy of mind, 6, 123–24, 126–28, 142 Piketty, Thomas, 275 Plato, 5, 53, 96, 343 political values, 147–48, 149, 171–72, 262–64, 266, 269, 270–71, 273, 274–75 practical reason, 118, 148, 149, 157, 261 object of practical reason, 119 pure practical reason, 119, 120, 145, 147 unity of practical reason, 145 Prichard, H.A., 112 primary goods, 94, 116, 128, 264, 299, 301, 306, 353 prisoner’s dilemma, 13, 34, 35, 241 priority of liberty, 191, 349 priority of right, 117–18, 120, 142, 149 property-owning democracy, 51, 59, 77, 81, 90, 92–94, 184, 213 public reason, 3, 7, 10, 11, 20, 58, 141, 146–48, 149–50, 165–66, 200, 210, 239, 260–66, 269–71, 273, 274–75, 334 public reason project, 10, 239–40, 242, 243, 246–47, 252–53, 254, 256 public reasoning, 10, 180, 314–17, 333 publicity condition, 135, 336, 338–39, 342, 348, 354 complete publicity condition, 135–36 full publicity condition, 336 Quong, Jonathan, 245, 254 rational (used substantively as “the rational”), 25, 29, 117, 145, 148 rational autonomy, see: autonomy, rational rational choice theory, 34 rational intuitionism, see: intuitionism reasonable (used substantively as “the reasonable”), 20, 117, 145–46, 147, 149, 239, 260, 309

376

Index

reasonable (used substantively as “the reasonable”) (cont.) reasonable comprehensive doctrine, see: comprehensive doctrine, rational reasonable disagreement, 13, 23, 148, 239, 308, 309, 320, 321, 335 reasonable persons, 29, 260–62, 266, 279, 273, 275, 289, 291, 333–34 reasonable pluralism of comprehensive doctrines, 11, 19, 137, 195, 242, 259–60, 278, 288, 291–92, 319 reasonable pluralism about justice, 10, 319 (see also: Justice pluralism) reciprocity, 11, 20, 30, 93, 177, 241, 263, 269, 271, 274, 275, 279, 282, 283, 286–87, 292, 327, 331, 333–35, 351 capacity of reciprocity, 20, 25, 28, 33 criterion of reciprocity (includes: Condition of reciprocity), 11, 257–58, 260–63, 265–66, 266–68, 269–71, 273, 275–76, 313, 320, 322–24, 328, 333, 334 economic reciprocity, 263, 269, 271–75 principle of reciprocity, 93, 275, 279, 293 requirement of reciprocity, 24, 28–29, 30–33, 35, 264 tendency to reciprocity, 286, 292, 319, 333–34 reflective equilibrium, 57, 141, 143, 145, 200, 220, 224, 226, 233, 265 general reflective equilibrium, 259 wide reflective equilibrium, 141, 259, 340 reparations, 9, 200–10, 212–19 Rousseau, Jean-Jacques, 17, 46, 49, 51, 254, 343–44, 347, 348 Sandel, Michael, 90, 128–29, 228 Scanlon, T.M. - 18, 49, 56, 61, 97, 103, 105, 143, 200, 277–78 Schmidtz, David, 11, 80, 258, 269 segregation, 173, 178, 199, 212, 216 self-respect, 52, 57–58, 216, 263, 268–69, 270–73, 294–95, 297, 300, 305–7, 308, 343–44, 344–45, 345–47, 348–52, 353–54, 354–55 social bases of self-respect, 51–53, 56, 90–91, 173, 178–80, 263, 285, 294, 306, 310, 349, 353 ‘society of self-respect’ defined, 342 sense of justice, 35, 42, 44, 50, 55, 117, 143, 152–54, 158, 172, 210–11, 232–33, 261–62, 268, 275, 278, 282–83, 285–87, 316, 319, 328, 333–34

shame, 112–13, 118 Shklar, Judith, 50 Sidgwick, Henry, 98, 111–12 slavery, 102, 110, 115, 173, 176, 178–79, 194, 199, 200, 202, 205, 212, 215, 216–17, 224, 304 Smith, Adam, 5, 18, 75, 97, 140 social democracy, 4, 79 social justice, 95, 98–100, 105, 132, 183, 185, 198, 203, 220, 222, 235 social union (includes: social union of social unions), 11, 40, 41, 50, 51, 52, 278–79, 279, 281–84, 286–88, 291, 292–95 socialism, 88, 245 (see also: liberal socialism) Spencer, Herbert, 85, 89 stability, 7, 10, 21, 35, 41, 73, 122–23, 135, 151–54, 154–59, 165, 183, 201, 212, 222, 229, 235, 241, 242, 258, 278–79, 286–88, 292, 316, 318–19, 328–29, 354 (see also: modus vivendi) imposed stability, 152, 165 inherent stability (includes: inherently stable), 7, 151–52, 165–66, 241 price stability, 73–75 stability for the right reasons (includes: stable for the right reasons), 20, 153–54, 183, 242, 243, 259, 267, 354 stability of possession, 97 state of nature, 17, 24, 26, 33, 34, 35, 169, 173, 222 taxation, 3, 61–63, 64, 66, 69, 76, 78, 87–88, 90, 263 Taylor, Charles, 38, 42, 228 theoretical reason, 149 thin theory of the good, 35, 116, 158 toleration, 18–19, 146, 147–48, 172, 224 (see also: liberty of conscience) Tomasi, John, 11, 245, 258, 269 Tullock, Gordon, 254 utilitarianism, 23, 121, 124–26, 131, 136, 141, 287, 297, 337, 340, 354 Veil of Ignorance, 17, 54, 177, 179, 187, 199, 222, 227, 232, 265, 274, 337, 341 voting, 42, 102, 213, 269, 300, 302, 305, 307, 308, 310–12, 327 plural voting, 12, 296–300, 302–3, 305–7, 307–9, 309–11 Waldron, Jeremy, 209, 218 Weber, Max, 195, 290

Index welfare state (includes: welfare state capitalism), 77, 78, 90–91, 93, 213, 272, 274 well-ordered society, 3, 8–9, 11, 12–13, 35, 37, 40–42, 45–49, 50, 52–53, 54–59, 103–4, 124, 136, 143, 145, 152,

377 156–58, 181–84, 186, 193, 194–95, 197–98, 199, 203–4, 211, 240–41, 242–43, 244, 257, 258–60, 274, 278, 279, 281–83, 286–87, 291–92, 292–93, 310, 316, 319–21, 327, 339–40, 348–51, 353, 354–55